{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.252539912917271,
  "eval_steps": 3000,
  "global_step": 102000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.742568969726562,
      "epoch": 0.00045355587808418,
      "grad_norm": 5.4375,
      "learning_rate": 2e-06,
      "loss": 10.7846,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10827.0,
      "step": 5
    },
    {
      "entropy": 10.742588996887207,
      "epoch": 0.00090711175616836,
      "grad_norm": 5.25,
      "learning_rate": 4.5e-06,
      "loss": 10.7602,
      "mean_token_accuracy": 6.963788182474673e-05,
      "num_tokens": 21025.0,
      "step": 10
    },
    {
      "entropy": 10.742573070526124,
      "epoch": 0.00136066763425254,
      "grad_norm": 5.25,
      "learning_rate": 7e-06,
      "loss": 10.7437,
      "mean_token_accuracy": 9.000900317914783e-05,
      "num_tokens": 31678.0,
      "step": 15
    },
    {
      "entropy": 10.742661285400391,
      "epoch": 0.00181422351233672,
      "grad_norm": 5.125,
      "learning_rate": 9.5e-06,
      "loss": 10.6591,
      "mean_token_accuracy": 0.0005137635511346161,
      "num_tokens": 41025.0,
      "step": 20
    },
    {
      "entropy": 10.742550945281982,
      "epoch": 0.0022677793904208998,
      "grad_norm": 4.875,
      "learning_rate": 1.2e-05,
      "loss": 10.5792,
      "mean_token_accuracy": 0.006170097610447556,
      "num_tokens": 51270.0,
      "step": 25
    },
    {
      "entropy": 10.742267036437989,
      "epoch": 0.00272133526850508,
      "grad_norm": 3.953125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4637,
      "mean_token_accuracy": 0.027165642101317646,
      "num_tokens": 62142.0,
      "step": 30
    },
    {
      "entropy": 10.741279792785644,
      "epoch": 0.00317489114658926,
      "grad_norm": 3.390625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.3199,
      "mean_token_accuracy": 0.04252696111798286,
      "num_tokens": 72709.0,
      "step": 35
    },
    {
      "entropy": 10.738855934143066,
      "epoch": 0.00362844702467344,
      "grad_norm": 2.625,
      "learning_rate": 1.95e-05,
      "loss": 10.1477,
      "mean_token_accuracy": 0.041494901105761525,
      "num_tokens": 82313.0,
      "step": 40
    },
    {
      "entropy": 10.734966659545899,
      "epoch": 0.00408200290275762,
      "grad_norm": 2.484375,
      "learning_rate": 2.2e-05,
      "loss": 10.0172,
      "mean_token_accuracy": 0.048007883876562116,
      "num_tokens": 92450.0,
      "step": 45
    },
    {
      "entropy": 10.731274795532226,
      "epoch": 0.0045355587808417995,
      "grad_norm": 2.125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.9997,
      "mean_token_accuracy": 0.0419012438505888,
      "num_tokens": 103978.0,
      "step": 50
    },
    {
      "entropy": 10.729759407043456,
      "epoch": 0.00498911465892598,
      "grad_norm": 1.96875,
      "learning_rate": 2.7e-05,
      "loss": 9.8946,
      "mean_token_accuracy": 0.04565211050212383,
      "num_tokens": 114134.0,
      "step": 55
    },
    {
      "entropy": 10.727651214599609,
      "epoch": 0.00544267053701016,
      "grad_norm": 2.015625,
      "learning_rate": 2.95e-05,
      "loss": 9.8734,
      "mean_token_accuracy": 0.04470008127391338,
      "num_tokens": 125037.0,
      "step": 60
    },
    {
      "entropy": 10.725145721435547,
      "epoch": 0.00589622641509434,
      "grad_norm": 1.9375,
      "learning_rate": 3.2e-05,
      "loss": 9.7559,
      "mean_token_accuracy": 0.04514850843697786,
      "num_tokens": 134721.0,
      "step": 65
    },
    {
      "entropy": 10.723241519927978,
      "epoch": 0.00634978229317852,
      "grad_norm": 2.03125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.6935,
      "mean_token_accuracy": 0.04249380584806204,
      "num_tokens": 144448.0,
      "step": 70
    },
    {
      "entropy": 10.721195030212403,
      "epoch": 0.006803338171262699,
      "grad_norm": 2.015625,
      "learning_rate": 3.7e-05,
      "loss": 9.635,
      "mean_token_accuracy": 0.04518970362842083,
      "num_tokens": 154343.0,
      "step": 75
    },
    {
      "entropy": 10.7159348487854,
      "epoch": 0.00725689404934688,
      "grad_norm": 2.0,
      "learning_rate": 3.95e-05,
      "loss": 9.6121,
      "mean_token_accuracy": 0.041934683173894885,
      "num_tokens": 165650.0,
      "step": 80
    },
    {
      "entropy": 10.709981918334961,
      "epoch": 0.007710449927431059,
      "grad_norm": 1.828125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.6128,
      "mean_token_accuracy": 0.03882604259997606,
      "num_tokens": 177098.0,
      "step": 85
    },
    {
      "entropy": 10.703641796112061,
      "epoch": 0.00816400580551524,
      "grad_norm": 2.015625,
      "learning_rate": 4.45e-05,
      "loss": 9.4133,
      "mean_token_accuracy": 0.04838715270161629,
      "num_tokens": 187903.0,
      "step": 90
    },
    {
      "entropy": 10.691870975494385,
      "epoch": 0.00861756168359942,
      "grad_norm": 1.9296875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.3442,
      "mean_token_accuracy": 0.04516914151608944,
      "num_tokens": 198077.0,
      "step": 95
    },
    {
      "entropy": 10.677101039886475,
      "epoch": 0.009071117561683599,
      "grad_norm": 1.859375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3156,
      "mean_token_accuracy": 0.04309112168848515,
      "num_tokens": 208641.0,
      "step": 100
    },
    {
      "entropy": 10.661001300811767,
      "epoch": 0.009524673439767779,
      "grad_norm": 1.9296875,
      "learning_rate": 5.2e-05,
      "loss": 9.1598,
      "mean_token_accuracy": 0.05524700954556465,
      "num_tokens": 218398.0,
      "step": 105
    },
    {
      "entropy": 10.64128761291504,
      "epoch": 0.00997822931785196,
      "grad_norm": 1.7578125,
      "learning_rate": 5.45e-05,
      "loss": 9.035,
      "mean_token_accuracy": 0.06018046177923679,
      "num_tokens": 229059.0,
      "step": 110
    },
    {
      "entropy": 10.614240169525146,
      "epoch": 0.01043178519593614,
      "grad_norm": 1.796875,
      "learning_rate": 5.7e-05,
      "loss": 8.9594,
      "mean_token_accuracy": 0.05966283418238163,
      "num_tokens": 239775.0,
      "step": 115
    },
    {
      "entropy": 10.567374324798584,
      "epoch": 0.01088534107402032,
      "grad_norm": 1.9140625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.8042,
      "mean_token_accuracy": 0.06268535517156124,
      "num_tokens": 249954.0,
      "step": 120
    },
    {
      "entropy": 10.49941110610962,
      "epoch": 0.011338896952104499,
      "grad_norm": 1.71875,
      "learning_rate": 6.2e-05,
      "loss": 8.7019,
      "mean_token_accuracy": 0.06550404466688634,
      "num_tokens": 260056.0,
      "step": 125
    },
    {
      "entropy": 10.433360958099366,
      "epoch": 0.01179245283018868,
      "grad_norm": 1.6015625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.5999,
      "mean_token_accuracy": 0.06795549392700195,
      "num_tokens": 269904.0,
      "step": 130
    },
    {
      "entropy": 10.364013862609863,
      "epoch": 0.01224600870827286,
      "grad_norm": 1.703125,
      "learning_rate": 6.7e-05,
      "loss": 8.4687,
      "mean_token_accuracy": 0.06760800294578076,
      "num_tokens": 280900.0,
      "step": 135
    },
    {
      "entropy": 10.2804931640625,
      "epoch": 0.01269956458635704,
      "grad_norm": 1.59375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.3637,
      "mean_token_accuracy": 0.067552350461483,
      "num_tokens": 290844.0,
      "step": 140
    },
    {
      "entropy": 10.172641277313232,
      "epoch": 0.013153120464441219,
      "grad_norm": 1.3046875,
      "learning_rate": 7.2e-05,
      "loss": 8.2257,
      "mean_token_accuracy": 0.07797299772500992,
      "num_tokens": 301345.0,
      "step": 145
    },
    {
      "entropy": 10.0338849067688,
      "epoch": 0.013606676342525399,
      "grad_norm": 1.484375,
      "learning_rate": 7.45e-05,
      "loss": 8.0893,
      "mean_token_accuracy": 0.07597564421594143,
      "num_tokens": 312681.0,
      "step": 150
    },
    {
      "entropy": 9.874010562896729,
      "epoch": 0.01406023222060958,
      "grad_norm": 1.203125,
      "learning_rate": 7.7e-05,
      "loss": 7.9627,
      "mean_token_accuracy": 0.08046026974916458,
      "num_tokens": 323492.0,
      "step": 155
    },
    {
      "entropy": 9.684037780761718,
      "epoch": 0.01451378809869376,
      "grad_norm": 1.234375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.9031,
      "mean_token_accuracy": 0.07810574918985366,
      "num_tokens": 334521.0,
      "step": 160
    },
    {
      "entropy": 9.470422744750977,
      "epoch": 0.014967343976777939,
      "grad_norm": 1.015625,
      "learning_rate": 8.2e-05,
      "loss": 7.8175,
      "mean_token_accuracy": 0.08020211830735206,
      "num_tokens": 345468.0,
      "step": 165
    },
    {
      "entropy": 9.22849702835083,
      "epoch": 0.015420899854862119,
      "grad_norm": 1.1015625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.6913,
      "mean_token_accuracy": 0.07718781009316444,
      "num_tokens": 355911.0,
      "step": 170
    },
    {
      "entropy": 9.006134128570556,
      "epoch": 0.0158744557329463,
      "grad_norm": 1.03125,
      "learning_rate": 8.7e-05,
      "loss": 7.6633,
      "mean_token_accuracy": 0.07722515687346458,
      "num_tokens": 367537.0,
      "step": 175
    },
    {
      "entropy": 8.818270492553712,
      "epoch": 0.01632801161103048,
      "grad_norm": 0.94921875,
      "learning_rate": 8.95e-05,
      "loss": 7.7245,
      "mean_token_accuracy": 0.07779238075017929,
      "num_tokens": 377918.0,
      "step": 180
    },
    {
      "entropy": 8.593470573425293,
      "epoch": 0.01678156748911466,
      "grad_norm": 0.85546875,
      "learning_rate": 9.2e-05,
      "loss": 7.5619,
      "mean_token_accuracy": 0.07740958072245122,
      "num_tokens": 388461.0,
      "step": 185
    },
    {
      "entropy": 8.449785518646241,
      "epoch": 0.01723512336719884,
      "grad_norm": 0.984375,
      "learning_rate": 9.45e-05,
      "loss": 7.5162,
      "mean_token_accuracy": 0.07913253530859947,
      "num_tokens": 399771.0,
      "step": 190
    },
    {
      "entropy": 8.267493534088135,
      "epoch": 0.01768867924528302,
      "grad_norm": 0.8125,
      "learning_rate": 9.7e-05,
      "loss": 7.3798,
      "mean_token_accuracy": 0.08124997988343238,
      "num_tokens": 410254.0,
      "step": 195
    },
    {
      "entropy": 8.230424785614014,
      "epoch": 0.018142235123367198,
      "grad_norm": 0.84765625,
      "learning_rate": 9.95e-05,
      "loss": 7.3917,
      "mean_token_accuracy": 0.07860286422073841,
      "num_tokens": 419792.0,
      "step": 200
    },
    {
      "entropy": 8.091243362426757,
      "epoch": 0.018595791001451378,
      "grad_norm": 0.96875,
      "learning_rate": 0.000102,
      "loss": 7.3759,
      "mean_token_accuracy": 0.08409927114844322,
      "num_tokens": 429264.0,
      "step": 205
    },
    {
      "entropy": 8.078143405914307,
      "epoch": 0.019049346879535557,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.3197,
      "mean_token_accuracy": 0.08545555844902993,
      "num_tokens": 439624.0,
      "step": 210
    },
    {
      "entropy": 8.048319149017335,
      "epoch": 0.01950290275761974,
      "grad_norm": 0.984375,
      "learning_rate": 0.000107,
      "loss": 7.3071,
      "mean_token_accuracy": 0.08367115408182144,
      "num_tokens": 449556.0,
      "step": 215
    },
    {
      "entropy": 7.937560939788819,
      "epoch": 0.01995645863570392,
      "grad_norm": 1.125,
      "learning_rate": 0.0001095,
      "loss": 7.3073,
      "mean_token_accuracy": 0.08580821454524994,
      "num_tokens": 460243.0,
      "step": 220
    },
    {
      "entropy": 7.98973650932312,
      "epoch": 0.0204100145137881,
      "grad_norm": 1.015625,
      "learning_rate": 0.000112,
      "loss": 7.3523,
      "mean_token_accuracy": 0.08153139427304268,
      "num_tokens": 469767.0,
      "step": 225
    },
    {
      "entropy": 7.94960503578186,
      "epoch": 0.02086357039187228,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001145,
      "loss": 7.2779,
      "mean_token_accuracy": 0.08782213702797889,
      "num_tokens": 480329.0,
      "step": 230
    },
    {
      "entropy": 7.937411499023438,
      "epoch": 0.02131712626995646,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.2332,
      "mean_token_accuracy": 0.08278771415352822,
      "num_tokens": 490096.0,
      "step": 235
    },
    {
      "entropy": 7.9301097869873045,
      "epoch": 0.02177068214804064,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.2751,
      "mean_token_accuracy": 0.0801933728158474,
      "num_tokens": 501231.0,
      "step": 240
    },
    {
      "entropy": 7.883290576934814,
      "epoch": 0.022224238026124818,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000122,
      "loss": 7.1802,
      "mean_token_accuracy": 0.09071151316165924,
      "num_tokens": 510722.0,
      "step": 245
    },
    {
      "entropy": 7.763730192184449,
      "epoch": 0.022677793904208998,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0001245,
      "loss": 7.225,
      "mean_token_accuracy": 0.08499881774187087,
      "num_tokens": 521452.0,
      "step": 250
    },
    {
      "entropy": 7.808817529678345,
      "epoch": 0.023131349782293177,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000127,
      "loss": 7.2137,
      "mean_token_accuracy": 0.08842154070734978,
      "num_tokens": 532397.0,
      "step": 255
    },
    {
      "entropy": 7.740368890762329,
      "epoch": 0.02358490566037736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001295,
      "loss": 7.1927,
      "mean_token_accuracy": 0.08906038105487823,
      "num_tokens": 542271.0,
      "step": 260
    },
    {
      "entropy": 7.7880730628967285,
      "epoch": 0.02403846153846154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000132,
      "loss": 7.1677,
      "mean_token_accuracy": 0.09267213717103004,
      "num_tokens": 552411.0,
      "step": 265
    },
    {
      "entropy": 7.73106369972229,
      "epoch": 0.02449201741654572,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.1675,
      "mean_token_accuracy": 0.09035667106509208,
      "num_tokens": 561518.0,
      "step": 270
    },
    {
      "entropy": 7.755526113510132,
      "epoch": 0.0249455732946299,
      "grad_norm": 1.046875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.1846,
      "mean_token_accuracy": 0.09218391999602318,
      "num_tokens": 571361.0,
      "step": 275
    },
    {
      "entropy": 7.656434154510498,
      "epoch": 0.02539912917271408,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001395,
      "loss": 7.0893,
      "mean_token_accuracy": 0.09728347510099411,
      "num_tokens": 580926.0,
      "step": 280
    },
    {
      "entropy": 7.710359334945679,
      "epoch": 0.025852685050798258,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1891,
      "mean_token_accuracy": 0.08993546590209008,
      "num_tokens": 592282.0,
      "step": 285
    },
    {
      "entropy": 7.69242033958435,
      "epoch": 0.026306240928882438,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001445,
      "loss": 7.1904,
      "mean_token_accuracy": 0.09070137366652489,
      "num_tokens": 602767.0,
      "step": 290
    },
    {
      "entropy": 7.715097570419312,
      "epoch": 0.026759796806966617,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000147,
      "loss": 7.1064,
      "mean_token_accuracy": 0.0925739623606205,
      "num_tokens": 612763.0,
      "step": 295
    },
    {
      "entropy": 7.563499784469604,
      "epoch": 0.027213352685050797,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0001495,
      "loss": 7.0617,
      "mean_token_accuracy": 0.09330770447850227,
      "num_tokens": 622796.0,
      "step": 300
    },
    {
      "entropy": 7.630935192108154,
      "epoch": 0.027666908563134977,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000152,
      "loss": 7.0426,
      "mean_token_accuracy": 0.09754194170236588,
      "num_tokens": 633136.0,
      "step": 305
    },
    {
      "entropy": 7.510959100723267,
      "epoch": 0.02812046444121916,
      "grad_norm": 1.109375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.0335,
      "mean_token_accuracy": 0.09690061807632447,
      "num_tokens": 643347.0,
      "step": 310
    },
    {
      "entropy": 7.634088945388794,
      "epoch": 0.02857402031930334,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000157,
      "loss": 7.0192,
      "mean_token_accuracy": 0.10197105258703232,
      "num_tokens": 653096.0,
      "step": 315
    },
    {
      "entropy": 7.5339111328125,
      "epoch": 0.02902757619738752,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001595,
      "loss": 7.0519,
      "mean_token_accuracy": 0.09302655458450318,
      "num_tokens": 663557.0,
      "step": 320
    },
    {
      "entropy": 7.588706541061401,
      "epoch": 0.0294811320754717,
      "grad_norm": 1.515625,
      "learning_rate": 0.000162,
      "loss": 7.1186,
      "mean_token_accuracy": 0.0957950733602047,
      "num_tokens": 674395.0,
      "step": 325
    },
    {
      "entropy": 7.544366073608399,
      "epoch": 0.029934687953555878,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.0165,
      "mean_token_accuracy": 0.0989912673830986,
      "num_tokens": 684776.0,
      "step": 330
    },
    {
      "entropy": 7.589376211166382,
      "epoch": 0.030388243831640058,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.9936,
      "mean_token_accuracy": 0.09702750965952874,
      "num_tokens": 694886.0,
      "step": 335
    },
    {
      "entropy": 7.451715087890625,
      "epoch": 0.030841799709724237,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.0626,
      "mean_token_accuracy": 0.09623553976416588,
      "num_tokens": 705426.0,
      "step": 340
    },
    {
      "entropy": 7.509190368652344,
      "epoch": 0.03129535558780842,
      "grad_norm": 0.984375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.0567,
      "mean_token_accuracy": 0.09196493923664092,
      "num_tokens": 716271.0,
      "step": 345
    },
    {
      "entropy": 7.5583083152771,
      "epoch": 0.0317489114658926,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.0486,
      "mean_token_accuracy": 0.09727451354265212,
      "num_tokens": 726875.0,
      "step": 350
    },
    {
      "entropy": 7.49442138671875,
      "epoch": 0.032202467343976776,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000177,
      "loss": 6.9698,
      "mean_token_accuracy": 0.09514860436320305,
      "num_tokens": 737825.0,
      "step": 355
    },
    {
      "entropy": 7.501305818557739,
      "epoch": 0.03265602322206096,
      "grad_norm": 0.953125,
      "learning_rate": 0.0001795,
      "loss": 6.9607,
      "mean_token_accuracy": 0.0989458367228508,
      "num_tokens": 747424.0,
      "step": 360
    },
    {
      "entropy": 7.407667875289917,
      "epoch": 0.033109579100145135,
      "grad_norm": 1.046875,
      "learning_rate": 0.000182,
      "loss": 6.9748,
      "mean_token_accuracy": 0.09646827578544617,
      "num_tokens": 757649.0,
      "step": 365
    },
    {
      "entropy": 7.480729293823242,
      "epoch": 0.03356313497822932,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001845,
      "loss": 6.946,
      "mean_token_accuracy": 0.09825383350253106,
      "num_tokens": 768556.0,
      "step": 370
    },
    {
      "entropy": 7.410275650024414,
      "epoch": 0.034016690856313495,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000187,
      "loss": 6.9456,
      "mean_token_accuracy": 0.0978932335972786,
      "num_tokens": 778331.0,
      "step": 375
    },
    {
      "entropy": 7.457346200942993,
      "epoch": 0.03447024673439768,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001895,
      "loss": 6.9326,
      "mean_token_accuracy": 0.10083384290337563,
      "num_tokens": 788065.0,
      "step": 380
    },
    {
      "entropy": 7.39845495223999,
      "epoch": 0.03492380261248186,
      "grad_norm": 1.078125,
      "learning_rate": 0.000192,
      "loss": 6.9717,
      "mean_token_accuracy": 0.10045002475380897,
      "num_tokens": 798912.0,
      "step": 385
    },
    {
      "entropy": 7.406559228897095,
      "epoch": 0.03537735849056604,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001945,
      "loss": 6.9143,
      "mean_token_accuracy": 0.10309505760669709,
      "num_tokens": 809580.0,
      "step": 390
    },
    {
      "entropy": 7.444182062149048,
      "epoch": 0.03583091436865022,
      "grad_norm": 1.015625,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0008,
      "mean_token_accuracy": 0.09738668501377105,
      "num_tokens": 820207.0,
      "step": 395
    },
    {
      "entropy": 7.290267753601074,
      "epoch": 0.036284470246734396,
      "grad_norm": 0.984375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8228,
      "mean_token_accuracy": 0.0949432410299778,
      "num_tokens": 831038.0,
      "step": 400
    },
    {
      "entropy": 7.429319906234741,
      "epoch": 0.03673802612481858,
      "grad_norm": 0.90625,
      "learning_rate": 0.000202,
      "loss": 6.9129,
      "mean_token_accuracy": 0.10030955821275711,
      "num_tokens": 842144.0,
      "step": 405
    },
    {
      "entropy": 7.347342252731323,
      "epoch": 0.037191582002902755,
      "grad_norm": 1.015625,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.905,
      "mean_token_accuracy": 0.10373404398560523,
      "num_tokens": 851315.0,
      "step": 410
    },
    {
      "entropy": 7.323047733306884,
      "epoch": 0.03764513788098694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000207,
      "loss": 6.912,
      "mean_token_accuracy": 0.10021007284522057,
      "num_tokens": 861233.0,
      "step": 415
    },
    {
      "entropy": 7.350712013244629,
      "epoch": 0.038098693759071114,
      "grad_norm": 1.0,
      "learning_rate": 0.0002095,
      "loss": 6.8978,
      "mean_token_accuracy": 0.09711670130491257,
      "num_tokens": 871131.0,
      "step": 420
    },
    {
      "entropy": 7.294861841201782,
      "epoch": 0.0385522496371553,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000212,
      "loss": 6.8749,
      "mean_token_accuracy": 0.09577203691005706,
      "num_tokens": 882195.0,
      "step": 425
    },
    {
      "entropy": 7.353318929672241,
      "epoch": 0.03900580551523948,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002145,
      "loss": 6.8527,
      "mean_token_accuracy": 0.10149310603737831,
      "num_tokens": 891806.0,
      "step": 430
    },
    {
      "entropy": 7.362502574920654,
      "epoch": 0.03945936139332366,
      "grad_norm": 0.953125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.9129,
      "mean_token_accuracy": 0.09761691987514495,
      "num_tokens": 901350.0,
      "step": 435
    },
    {
      "entropy": 7.301689004898071,
      "epoch": 0.03991291727140784,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002195,
      "loss": 6.8293,
      "mean_token_accuracy": 0.09842056706547737,
      "num_tokens": 912061.0,
      "step": 440
    },
    {
      "entropy": 7.250009202957154,
      "epoch": 0.040366473149492016,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000222,
      "loss": 6.7954,
      "mean_token_accuracy": 0.09848268181085587,
      "num_tokens": 922220.0,
      "step": 445
    },
    {
      "entropy": 7.2643632888793945,
      "epoch": 0.0408200290275762,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002245,
      "loss": 6.8122,
      "mean_token_accuracy": 0.1012580119073391,
      "num_tokens": 932319.0,
      "step": 450
    },
    {
      "entropy": 7.2017487525939945,
      "epoch": 0.041273584905660375,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.8351,
      "mean_token_accuracy": 0.10266126319766045,
      "num_tokens": 943254.0,
      "step": 455
    },
    {
      "entropy": 7.330859661102295,
      "epoch": 0.04172714078374456,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8445,
      "mean_token_accuracy": 0.09818541109561921,
      "num_tokens": 953921.0,
      "step": 460
    },
    {
      "entropy": 7.193669176101684,
      "epoch": 0.042180696661828734,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.7701,
      "mean_token_accuracy": 0.10723233297467231,
      "num_tokens": 963832.0,
      "step": 465
    },
    {
      "entropy": 7.234699535369873,
      "epoch": 0.04263425253991292,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8614,
      "mean_token_accuracy": 0.09933211356401443,
      "num_tokens": 974066.0,
      "step": 470
    },
    {
      "entropy": 7.233931684494019,
      "epoch": 0.0430878084179971,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000237,
      "loss": 6.8441,
      "mean_token_accuracy": 0.10467069000005721,
      "num_tokens": 984402.0,
      "step": 475
    },
    {
      "entropy": 7.256547212600708,
      "epoch": 0.04354136429608128,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002395,
      "loss": 6.7629,
      "mean_token_accuracy": 0.1010866865515709,
      "num_tokens": 993555.0,
      "step": 480
    },
    {
      "entropy": 7.208513498306274,
      "epoch": 0.04399492017416546,
      "grad_norm": 1.015625,
      "learning_rate": 0.000242,
      "loss": 6.7524,
      "mean_token_accuracy": 0.10970519632101058,
      "num_tokens": 1002920.0,
      "step": 485
    },
    {
      "entropy": 7.203861713409424,
      "epoch": 0.044448476052249636,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002445,
      "loss": 6.8285,
      "mean_token_accuracy": 0.09806114360690117,
      "num_tokens": 1013611.0,
      "step": 490
    },
    {
      "entropy": 7.1150740623474125,
      "epoch": 0.04490203193033382,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000247,
      "loss": 6.7083,
      "mean_token_accuracy": 0.10193079337477684,
      "num_tokens": 1024121.0,
      "step": 495
    },
    {
      "entropy": 7.323940992355347,
      "epoch": 0.045355587808417995,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002495,
      "loss": 6.9503,
      "mean_token_accuracy": 0.10487301424145698,
      "num_tokens": 1033845.0,
      "step": 500
    },
    {
      "entropy": 7.164136552810669,
      "epoch": 0.04580914368650218,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000252,
      "loss": 6.7388,
      "mean_token_accuracy": 0.10410365164279937,
      "num_tokens": 1044642.0,
      "step": 505
    },
    {
      "entropy": 7.172935676574707,
      "epoch": 0.046262699564586354,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002545,
      "loss": 6.8158,
      "mean_token_accuracy": 0.10116696879267692,
      "num_tokens": 1054728.0,
      "step": 510
    },
    {
      "entropy": 7.138863849639892,
      "epoch": 0.04671625544267054,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000257,
      "loss": 6.7643,
      "mean_token_accuracy": 0.10025905445218086,
      "num_tokens": 1065743.0,
      "step": 515
    },
    {
      "entropy": 7.10414080619812,
      "epoch": 0.04716981132075472,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0002595,
      "loss": 6.6969,
      "mean_token_accuracy": 0.10323216244578362,
      "num_tokens": 1076798.0,
      "step": 520
    },
    {
      "entropy": 7.174085855484009,
      "epoch": 0.047623367198838897,
      "grad_norm": 1.03125,
      "learning_rate": 0.000262,
      "loss": 6.8163,
      "mean_token_accuracy": 0.0997597634792328,
      "num_tokens": 1087445.0,
      "step": 525
    },
    {
      "entropy": 7.094452857971191,
      "epoch": 0.04807692307692308,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7197,
      "mean_token_accuracy": 0.10675485357642174,
      "num_tokens": 1098379.0,
      "step": 530
    },
    {
      "entropy": 7.202598142623901,
      "epoch": 0.048530478955007256,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8289,
      "mean_token_accuracy": 0.0993705414235592,
      "num_tokens": 1108016.0,
      "step": 535
    },
    {
      "entropy": 7.089155960083008,
      "epoch": 0.04898403483309144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.701,
      "mean_token_accuracy": 0.10358337983489037,
      "num_tokens": 1117887.0,
      "step": 540
    },
    {
      "entropy": 7.104513120651245,
      "epoch": 0.049437590711175615,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6948,
      "mean_token_accuracy": 0.09933766350150108,
      "num_tokens": 1128497.0,
      "step": 545
    },
    {
      "entropy": 7.055957365036011,
      "epoch": 0.0498911465892598,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002745,
      "loss": 6.7188,
      "mean_token_accuracy": 0.10837908834218979,
      "num_tokens": 1138168.0,
      "step": 550
    },
    {
      "entropy": 7.0923104763031,
      "epoch": 0.050344702467343974,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000277,
      "loss": 6.691,
      "mean_token_accuracy": 0.10448997020721436,
      "num_tokens": 1148357.0,
      "step": 555
    },
    {
      "entropy": 7.086644268035888,
      "epoch": 0.05079825834542816,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002795,
      "loss": 6.7185,
      "mean_token_accuracy": 0.10321378186345101,
      "num_tokens": 1159964.0,
      "step": 560
    },
    {
      "entropy": 7.139619493484497,
      "epoch": 0.05125181422351233,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.7781,
      "mean_token_accuracy": 0.10231135338544846,
      "num_tokens": 1170683.0,
      "step": 565
    },
    {
      "entropy": 7.115566444396973,
      "epoch": 0.051705370101596516,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002845,
      "loss": 6.7343,
      "mean_token_accuracy": 0.10421703606843949,
      "num_tokens": 1181680.0,
      "step": 570
    },
    {
      "entropy": 7.0173659324646,
      "epoch": 0.0521589259796807,
      "grad_norm": 1.078125,
      "learning_rate": 0.000287,
      "loss": 6.6227,
      "mean_token_accuracy": 0.10645965486764908,
      "num_tokens": 1191619.0,
      "step": 575
    },
    {
      "entropy": 6.977677965164185,
      "epoch": 0.052612481857764876,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002895,
      "loss": 6.6222,
      "mean_token_accuracy": 0.10420064777135848,
      "num_tokens": 1202264.0,
      "step": 580
    },
    {
      "entropy": 7.002698135375977,
      "epoch": 0.05306603773584906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000292,
      "loss": 6.6254,
      "mean_token_accuracy": 0.10796518102288247,
      "num_tokens": 1212119.0,
      "step": 585
    },
    {
      "entropy": 7.092545223236084,
      "epoch": 0.053519593613933235,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002945,
      "loss": 6.7588,
      "mean_token_accuracy": 0.10538023859262466,
      "num_tokens": 1222317.0,
      "step": 590
    },
    {
      "entropy": 7.060644435882568,
      "epoch": 0.05397314949201742,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000297,
      "loss": 6.703,
      "mean_token_accuracy": 0.1080887496471405,
      "num_tokens": 1232914.0,
      "step": 595
    },
    {
      "entropy": 6.919224500656128,
      "epoch": 0.054426705370101594,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0002995,
      "loss": 6.5519,
      "mean_token_accuracy": 0.11217400431632996,
      "num_tokens": 1244214.0,
      "step": 600
    },
    {
      "entropy": 7.022872066497802,
      "epoch": 0.05488026124818578,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000302,
      "loss": 6.6491,
      "mean_token_accuracy": 0.10687957927584649,
      "num_tokens": 1254626.0,
      "step": 605
    },
    {
      "entropy": 6.948216915130615,
      "epoch": 0.05533381712626995,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003045,
      "loss": 6.7394,
      "mean_token_accuracy": 0.10631354376673699,
      "num_tokens": 1265461.0,
      "step": 610
    },
    {
      "entropy": 7.030997228622437,
      "epoch": 0.055787373004354136,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000307,
      "loss": 6.6204,
      "mean_token_accuracy": 0.10827788710594177,
      "num_tokens": 1275810.0,
      "step": 615
    },
    {
      "entropy": 6.895572900772095,
      "epoch": 0.05624092888243832,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003095,
      "loss": 6.6276,
      "mean_token_accuracy": 0.11254601404070855,
      "num_tokens": 1285170.0,
      "step": 620
    },
    {
      "entropy": 7.06603364944458,
      "epoch": 0.056694484760522496,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000312,
      "loss": 6.5641,
      "mean_token_accuracy": 0.11045938208699227,
      "num_tokens": 1295460.0,
      "step": 625
    },
    {
      "entropy": 6.950644588470459,
      "epoch": 0.05714804063860668,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003145,
      "loss": 6.687,
      "mean_token_accuracy": 0.10515375509858131,
      "num_tokens": 1306656.0,
      "step": 630
    },
    {
      "entropy": 7.004617834091187,
      "epoch": 0.057601596516690855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000317,
      "loss": 6.6603,
      "mean_token_accuracy": 0.10941670313477517,
      "num_tokens": 1318019.0,
      "step": 635
    },
    {
      "entropy": 6.8719371318817135,
      "epoch": 0.05805515239477504,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003195,
      "loss": 6.5802,
      "mean_token_accuracy": 0.11324906200170518,
      "num_tokens": 1328120.0,
      "step": 640
    },
    {
      "entropy": 7.002581357955933,
      "epoch": 0.058508708272859214,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000322,
      "loss": 6.6842,
      "mean_token_accuracy": 0.10679537355899811,
      "num_tokens": 1338354.0,
      "step": 645
    },
    {
      "entropy": 7.0219018936157225,
      "epoch": 0.0589622641509434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.6503,
      "mean_token_accuracy": 0.10833998396992683,
      "num_tokens": 1348666.0,
      "step": 650
    },
    {
      "entropy": 6.952841186523438,
      "epoch": 0.05941582002902757,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6921,
      "mean_token_accuracy": 0.11318251714110375,
      "num_tokens": 1359713.0,
      "step": 655
    },
    {
      "entropy": 6.950813102722168,
      "epoch": 0.059869375907111756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5549,
      "mean_token_accuracy": 0.11367055401206017,
      "num_tokens": 1369507.0,
      "step": 660
    },
    {
      "entropy": 6.984795665740966,
      "epoch": 0.06032293178519594,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.7111,
      "mean_token_accuracy": 0.10242682620882988,
      "num_tokens": 1380726.0,
      "step": 665
    },
    {
      "entropy": 7.01419677734375,
      "epoch": 0.060776487663280115,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6325,
      "mean_token_accuracy": 0.11175172179937362,
      "num_tokens": 1391423.0,
      "step": 670
    },
    {
      "entropy": 6.930062580108642,
      "epoch": 0.0612300435413643,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000337,
      "loss": 6.6054,
      "mean_token_accuracy": 0.11288354322314262,
      "num_tokens": 1402492.0,
      "step": 675
    },
    {
      "entropy": 6.939368772506714,
      "epoch": 0.061683599419448475,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003395,
      "loss": 6.6504,
      "mean_token_accuracy": 0.1099183864891529,
      "num_tokens": 1412497.0,
      "step": 680
    },
    {
      "entropy": 6.999884557723999,
      "epoch": 0.06213715529753266,
      "grad_norm": 1.15625,
      "learning_rate": 0.000342,
      "loss": 6.6204,
      "mean_token_accuracy": 0.11455832123756408,
      "num_tokens": 1423214.0,
      "step": 685
    },
    {
      "entropy": 6.856761169433594,
      "epoch": 0.06259071117561683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6202,
      "mean_token_accuracy": 0.11103508621454239,
      "num_tokens": 1434044.0,
      "step": 690
    },
    {
      "entropy": 6.9463762760162355,
      "epoch": 0.06304426705370102,
      "grad_norm": 1.015625,
      "learning_rate": 0.000347,
      "loss": 6.5826,
      "mean_token_accuracy": 0.11131571531295777,
      "num_tokens": 1444387.0,
      "step": 695
    },
    {
      "entropy": 6.848030471801758,
      "epoch": 0.0634978229317852,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003495,
      "loss": 6.5282,
      "mean_token_accuracy": 0.10915688425302505,
      "num_tokens": 1455157.0,
      "step": 700
    },
    {
      "entropy": 6.915933418273926,
      "epoch": 0.06395137880986937,
      "grad_norm": 1.0,
      "learning_rate": 0.000352,
      "loss": 6.6147,
      "mean_token_accuracy": 0.11529198288917542,
      "num_tokens": 1466145.0,
      "step": 705
    },
    {
      "entropy": 6.937824773788452,
      "epoch": 0.06440493468795355,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003545,
      "loss": 6.6053,
      "mean_token_accuracy": 0.1114728607237339,
      "num_tokens": 1476713.0,
      "step": 710
    },
    {
      "entropy": 6.905196857452393,
      "epoch": 0.06485849056603774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000357,
      "loss": 6.6113,
      "mean_token_accuracy": 0.11422201618552208,
      "num_tokens": 1486908.0,
      "step": 715
    },
    {
      "entropy": 6.866687297821045,
      "epoch": 0.06531204644412192,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003595,
      "loss": 6.5389,
      "mean_token_accuracy": 0.11563119068741798,
      "num_tokens": 1498185.0,
      "step": 720
    },
    {
      "entropy": 6.873838233947754,
      "epoch": 0.0657656023222061,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000362,
      "loss": 6.5814,
      "mean_token_accuracy": 0.10893168300390244,
      "num_tokens": 1508376.0,
      "step": 725
    },
    {
      "entropy": 6.905275201797485,
      "epoch": 0.06621915820029027,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003645,
      "loss": 6.6695,
      "mean_token_accuracy": 0.11235518902540206,
      "num_tokens": 1519953.0,
      "step": 730
    },
    {
      "entropy": 6.795944786071777,
      "epoch": 0.06667271407837445,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000367,
      "loss": 6.5408,
      "mean_token_accuracy": 0.11527841538190842,
      "num_tokens": 1529793.0,
      "step": 735
    },
    {
      "entropy": 6.8445906162261965,
      "epoch": 0.06712626995645864,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003695,
      "loss": 6.5363,
      "mean_token_accuracy": 0.11552815213799476,
      "num_tokens": 1540859.0,
      "step": 740
    },
    {
      "entropy": 6.8253929138183596,
      "epoch": 0.06757982583454282,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000372,
      "loss": 6.4732,
      "mean_token_accuracy": 0.11656396687030793,
      "num_tokens": 1549783.0,
      "step": 745
    },
    {
      "entropy": 6.756074047088623,
      "epoch": 0.06803338171262699,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003745,
      "loss": 6.5266,
      "mean_token_accuracy": 0.11289834827184678,
      "num_tokens": 1560522.0,
      "step": 750
    },
    {
      "entropy": 6.810197973251343,
      "epoch": 0.06848693759071117,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000377,
      "loss": 6.5008,
      "mean_token_accuracy": 0.11671004444360733,
      "num_tokens": 1570668.0,
      "step": 755
    },
    {
      "entropy": 6.83150315284729,
      "epoch": 0.06894049346879536,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003795,
      "loss": 6.6122,
      "mean_token_accuracy": 0.1193370096385479,
      "num_tokens": 1581523.0,
      "step": 760
    },
    {
      "entropy": 6.850863075256347,
      "epoch": 0.06939404934687954,
      "grad_norm": 1.125,
      "learning_rate": 0.000382,
      "loss": 6.5253,
      "mean_token_accuracy": 0.11707687377929688,
      "num_tokens": 1591356.0,
      "step": 765
    },
    {
      "entropy": 6.784787607192993,
      "epoch": 0.06984760522496372,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003845,
      "loss": 6.5113,
      "mean_token_accuracy": 0.11673457846045494,
      "num_tokens": 1601779.0,
      "step": 770
    },
    {
      "entropy": 6.848218965530395,
      "epoch": 0.07030116110304789,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5091,
      "mean_token_accuracy": 0.11470222994685172,
      "num_tokens": 1611995.0,
      "step": 775
    },
    {
      "entropy": 6.800413322448731,
      "epoch": 0.07075471698113207,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5369,
      "mean_token_accuracy": 0.11278748363256455,
      "num_tokens": 1621834.0,
      "step": 780
    },
    {
      "entropy": 6.780804920196533,
      "epoch": 0.07120827285921626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5788,
      "mean_token_accuracy": 0.1113444909453392,
      "num_tokens": 1632881.0,
      "step": 785
    },
    {
      "entropy": 6.707678127288818,
      "epoch": 0.07166182873730044,
      "grad_norm": 1.0,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5541,
      "mean_token_accuracy": 0.11073025166988373,
      "num_tokens": 1642870.0,
      "step": 790
    },
    {
      "entropy": 6.871548318862915,
      "epoch": 0.07211538461538461,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4402,
      "mean_token_accuracy": 0.12203826904296874,
      "num_tokens": 1653390.0,
      "step": 795
    },
    {
      "entropy": 6.7886035442352295,
      "epoch": 0.07256894049346879,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003995,
      "loss": 6.5612,
      "mean_token_accuracy": 0.11627909317612647,
      "num_tokens": 1663382.0,
      "step": 800
    },
    {
      "entropy": 6.804583787918091,
      "epoch": 0.07302249637155298,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000402,
      "loss": 6.5652,
      "mean_token_accuracy": 0.1136283002793789,
      "num_tokens": 1673601.0,
      "step": 805
    },
    {
      "entropy": 6.703377294540405,
      "epoch": 0.07347605224963716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004045,
      "loss": 6.4891,
      "mean_token_accuracy": 0.12096939235925674,
      "num_tokens": 1683565.0,
      "step": 810
    },
    {
      "entropy": 6.758061265945434,
      "epoch": 0.07392960812772134,
      "grad_norm": 1.78125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.45,
      "mean_token_accuracy": 0.11798930242657661,
      "num_tokens": 1693636.0,
      "step": 815
    },
    {
      "entropy": 6.882167196273803,
      "epoch": 0.07438316400580551,
      "grad_norm": 1.125,
      "learning_rate": 0.0004095,
      "loss": 6.5623,
      "mean_token_accuracy": 0.11476152688264847,
      "num_tokens": 1703415.0,
      "step": 820
    },
    {
      "entropy": 6.755157995223999,
      "epoch": 0.0748367198838897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000412,
      "loss": 6.5396,
      "mean_token_accuracy": 0.11611356288194656,
      "num_tokens": 1713579.0,
      "step": 825
    },
    {
      "entropy": 6.7464526176452635,
      "epoch": 0.07529027576197388,
      "grad_norm": 1.0,
      "learning_rate": 0.0004145,
      "loss": 6.5394,
      "mean_token_accuracy": 0.11386027708649635,
      "num_tokens": 1724583.0,
      "step": 830
    },
    {
      "entropy": 6.758458185195923,
      "epoch": 0.07574383164005806,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000417,
      "loss": 6.4527,
      "mean_token_accuracy": 0.12356605678796768,
      "num_tokens": 1735293.0,
      "step": 835
    },
    {
      "entropy": 6.7750757217407225,
      "epoch": 0.07619738751814223,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004195,
      "loss": 6.4786,
      "mean_token_accuracy": 0.12075236141681671,
      "num_tokens": 1745623.0,
      "step": 840
    },
    {
      "entropy": 6.725871658325195,
      "epoch": 0.07665094339622641,
      "grad_norm": 0.96875,
      "learning_rate": 0.000422,
      "loss": 6.5145,
      "mean_token_accuracy": 0.11627071723341942,
      "num_tokens": 1755713.0,
      "step": 845
    },
    {
      "entropy": 6.776280164718628,
      "epoch": 0.0771044992743106,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004245,
      "loss": 6.506,
      "mean_token_accuracy": 0.11393775343894959,
      "num_tokens": 1765264.0,
      "step": 850
    },
    {
      "entropy": 6.824870538711548,
      "epoch": 0.07755805515239478,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000427,
      "loss": 6.511,
      "mean_token_accuracy": 0.11867988780140877,
      "num_tokens": 1775769.0,
      "step": 855
    },
    {
      "entropy": 6.691818761825561,
      "epoch": 0.07801161103047896,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004295,
      "loss": 6.4441,
      "mean_token_accuracy": 0.11789206489920616,
      "num_tokens": 1786245.0,
      "step": 860
    },
    {
      "entropy": 6.627203035354614,
      "epoch": 0.07846516690856313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000432,
      "loss": 6.4773,
      "mean_token_accuracy": 0.1227649137377739,
      "num_tokens": 1796040.0,
      "step": 865
    },
    {
      "entropy": 6.691270112991333,
      "epoch": 0.07891872278664731,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004345,
      "loss": 6.4935,
      "mean_token_accuracy": 0.11697295233607292,
      "num_tokens": 1806007.0,
      "step": 870
    },
    {
      "entropy": 6.88080358505249,
      "epoch": 0.0793722786647315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437,
      "loss": 6.5665,
      "mean_token_accuracy": 0.10995617434382439,
      "num_tokens": 1816935.0,
      "step": 875
    },
    {
      "entropy": 6.631897354125977,
      "epoch": 0.07982583454281568,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004395,
      "loss": 6.4244,
      "mean_token_accuracy": 0.11615645289421081,
      "num_tokens": 1827419.0,
      "step": 880
    },
    {
      "entropy": 6.766531658172608,
      "epoch": 0.08027939042089985,
      "grad_norm": 1.15625,
      "learning_rate": 0.000442,
      "loss": 6.533,
      "mean_token_accuracy": 0.11622850596904755,
      "num_tokens": 1837933.0,
      "step": 885
    },
    {
      "entropy": 6.714184427261353,
      "epoch": 0.08073294629898403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004445,
      "loss": 6.5854,
      "mean_token_accuracy": 0.11893809214234352,
      "num_tokens": 1848551.0,
      "step": 890
    },
    {
      "entropy": 6.7861744403839115,
      "epoch": 0.08118650217706821,
      "grad_norm": 1.125,
      "learning_rate": 0.000447,
      "loss": 6.4577,
      "mean_token_accuracy": 0.11503675356507301,
      "num_tokens": 1858380.0,
      "step": 895
    },
    {
      "entropy": 6.660554695129394,
      "epoch": 0.0816400580551524,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3853,
      "mean_token_accuracy": 0.11978982239961625,
      "num_tokens": 1868291.0,
      "step": 900
    },
    {
      "entropy": 6.629644966125488,
      "epoch": 0.08209361393323658,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4839,
      "mean_token_accuracy": 0.11984607204794884,
      "num_tokens": 1879219.0,
      "step": 905
    },
    {
      "entropy": 6.754321718215943,
      "epoch": 0.08254716981132075,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4652,
      "mean_token_accuracy": 0.1155819833278656,
      "num_tokens": 1889809.0,
      "step": 910
    },
    {
      "entropy": 6.659244441986084,
      "epoch": 0.08300072568940493,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4349,
      "mean_token_accuracy": 0.11736063361167907,
      "num_tokens": 1899986.0,
      "step": 915
    },
    {
      "entropy": 6.727784872055054,
      "epoch": 0.08345428156748912,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4192,
      "mean_token_accuracy": 0.11866439804434777,
      "num_tokens": 1909314.0,
      "step": 920
    },
    {
      "entropy": 6.709451198577881,
      "epoch": 0.0839078374455733,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000462,
      "loss": 6.4667,
      "mean_token_accuracy": 0.1235547974705696,
      "num_tokens": 1918443.0,
      "step": 925
    },
    {
      "entropy": 6.680801582336426,
      "epoch": 0.08436139332365747,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004645,
      "loss": 6.4235,
      "mean_token_accuracy": 0.12087266743183137,
      "num_tokens": 1928683.0,
      "step": 930
    },
    {
      "entropy": 6.541482257843017,
      "epoch": 0.08481494920174165,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000467,
      "loss": 6.3312,
      "mean_token_accuracy": 0.11764199584722519,
      "num_tokens": 1938293.0,
      "step": 935
    },
    {
      "entropy": 6.629144906997681,
      "epoch": 0.08526850507982583,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004695,
      "loss": 6.4449,
      "mean_token_accuracy": 0.11702702939510345,
      "num_tokens": 1948233.0,
      "step": 940
    },
    {
      "entropy": 6.660422420501709,
      "epoch": 0.08572206095791002,
      "grad_norm": 1.046875,
      "learning_rate": 0.000472,
      "loss": 6.3444,
      "mean_token_accuracy": 0.12124838382005691,
      "num_tokens": 1958611.0,
      "step": 945
    },
    {
      "entropy": 6.661191082000732,
      "epoch": 0.0861756168359942,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004745,
      "loss": 6.4983,
      "mean_token_accuracy": 0.1220276914536953,
      "num_tokens": 1968957.0,
      "step": 950
    },
    {
      "entropy": 6.65334038734436,
      "epoch": 0.08662917271407837,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000477,
      "loss": 6.3551,
      "mean_token_accuracy": 0.12157930657267571,
      "num_tokens": 1978503.0,
      "step": 955
    },
    {
      "entropy": 6.552236032485962,
      "epoch": 0.08708272859216255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004795,
      "loss": 6.3482,
      "mean_token_accuracy": 0.12029609382152558,
      "num_tokens": 1988458.0,
      "step": 960
    },
    {
      "entropy": 6.655647468566895,
      "epoch": 0.08753628447024674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482,
      "loss": 6.491,
      "mean_token_accuracy": 0.11715313494205475,
      "num_tokens": 1999984.0,
      "step": 965
    },
    {
      "entropy": 6.510547399520874,
      "epoch": 0.08798984034833092,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004845,
      "loss": 6.3364,
      "mean_token_accuracy": 0.12651536986231804,
      "num_tokens": 2010645.0,
      "step": 970
    },
    {
      "entropy": 6.751053190231323,
      "epoch": 0.08844339622641509,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000487,
      "loss": 6.4506,
      "mean_token_accuracy": 0.11614105999469757,
      "num_tokens": 2020196.0,
      "step": 975
    },
    {
      "entropy": 6.587527418136597,
      "epoch": 0.08889695210449927,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004895,
      "loss": 6.3423,
      "mean_token_accuracy": 0.1243398666381836,
      "num_tokens": 2030101.0,
      "step": 980
    },
    {
      "entropy": 6.5954345703125,
      "epoch": 0.08935050798258345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492,
      "loss": 6.4263,
      "mean_token_accuracy": 0.1206196092069149,
      "num_tokens": 2040454.0,
      "step": 985
    },
    {
      "entropy": 6.529821062088013,
      "epoch": 0.08980406386066764,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004945,
      "loss": 6.3602,
      "mean_token_accuracy": 0.11969397887587548,
      "num_tokens": 2050378.0,
      "step": 990
    },
    {
      "entropy": 6.656999588012695,
      "epoch": 0.09025761973875182,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497,
      "loss": 6.4186,
      "mean_token_accuracy": 0.11806954741477967,
      "num_tokens": 2060282.0,
      "step": 995
    },
    {
      "entropy": 6.559673261642456,
      "epoch": 0.09071117561683599,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995,
      "loss": 6.4115,
      "mean_token_accuracy": 0.12180240452289581,
      "num_tokens": 2070563.0,
      "step": 1000
    },
    {
      "entropy": 6.661390542984009,
      "epoch": 0.09116473149492017,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999999985110217,
      "loss": 6.4743,
      "mean_token_accuracy": 0.12212796434760094,
      "num_tokens": 2080153.0,
      "step": 1005
    },
    {
      "entropy": 6.525332069396972,
      "epoch": 0.09161828737300436,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999999924620471,
      "loss": 6.3263,
      "mean_token_accuracy": 0.12526472061872482,
      "num_tokens": 2090590.0,
      "step": 1010
    },
    {
      "entropy": 6.613137102127075,
      "epoch": 0.09207184325108854,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999999817600155,
      "loss": 6.4047,
      "mean_token_accuracy": 0.11499982699751854,
      "num_tokens": 2101945.0,
      "step": 1015
    },
    {
      "entropy": 6.613913679122925,
      "epoch": 0.09252539912917271,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999999664049268,
      "loss": 6.4031,
      "mean_token_accuracy": 0.12405626773834229,
      "num_tokens": 2112514.0,
      "step": 1020
    },
    {
      "entropy": 6.498817586898804,
      "epoch": 0.09297895500725689,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999463967816,
      "loss": 6.3365,
      "mean_token_accuracy": 0.1212070882320404,
      "num_tokens": 2122296.0,
      "step": 1025
    },
    {
      "entropy": 6.740924310684204,
      "epoch": 0.09343251088534107,
      "grad_norm": 0.921875,
      "learning_rate": 0.00049999992173558,
      "loss": 6.4899,
      "mean_token_accuracy": 0.11208800077438355,
      "num_tokens": 2134512.0,
      "step": 1030
    },
    {
      "entropy": 6.578291606903076,
      "epoch": 0.09388606676342526,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999998924213228,
      "loss": 6.4201,
      "mean_token_accuracy": 0.12014863640069962,
      "num_tokens": 2143934.0,
      "step": 1035
    },
    {
      "entropy": 6.489139556884766,
      "epoch": 0.09433962264150944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999998584540105,
      "loss": 6.3024,
      "mean_token_accuracy": 0.12084058821201324,
      "num_tokens": 2155730.0,
      "step": 1040
    },
    {
      "entropy": 6.6251708984375,
      "epoch": 0.09479317851959361,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999998198336437,
      "loss": 6.3417,
      "mean_token_accuracy": 0.1186050683259964,
      "num_tokens": 2166422.0,
      "step": 1045
    },
    {
      "entropy": 6.549373722076416,
      "epoch": 0.09524673439767779,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999997765602233,
      "loss": 6.4281,
      "mean_token_accuracy": 0.11772651746869087,
      "num_tokens": 2177842.0,
      "step": 1050
    },
    {
      "entropy": 6.614485788345337,
      "epoch": 0.09570029027576198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999997286337502,
      "loss": 6.4994,
      "mean_token_accuracy": 0.12015130147337913,
      "num_tokens": 2188614.0,
      "step": 1055
    },
    {
      "entropy": 6.645506238937378,
      "epoch": 0.09615384615384616,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999996760542254,
      "loss": 6.3042,
      "mean_token_accuracy": 0.12434962838888168,
      "num_tokens": 2197886.0,
      "step": 1060
    },
    {
      "entropy": 6.450875282287598,
      "epoch": 0.09660740203193033,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049999961882165,
      "loss": 6.3515,
      "mean_token_accuracy": 0.12345030084252358,
      "num_tokens": 2209930.0,
      "step": 1065
    },
    {
      "entropy": 6.616359043121338,
      "epoch": 0.09706095791001451,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999995569360251,
      "loss": 6.3704,
      "mean_token_accuracy": 0.11785119250416756,
      "num_tokens": 2219767.0,
      "step": 1070
    },
    {
      "entropy": 6.475496864318847,
      "epoch": 0.0975145137880987,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499999490397352,
      "loss": 6.2989,
      "mean_token_accuracy": 0.1218901552259922,
      "num_tokens": 2230082.0,
      "step": 1075
    },
    {
      "entropy": 6.539343738555909,
      "epoch": 0.09796806966618288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999994192056321,
      "loss": 6.3308,
      "mean_token_accuracy": 0.1269037939608097,
      "num_tokens": 2240066.0,
      "step": 1080
    },
    {
      "entropy": 6.536786890029907,
      "epoch": 0.09842162554426705,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999993433608669,
      "loss": 6.3427,
      "mean_token_accuracy": 0.12274787575006485,
      "num_tokens": 2250732.0,
      "step": 1085
    },
    {
      "entropy": 6.519191551208496,
      "epoch": 0.09887518142235123,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999992628630579,
      "loss": 6.2367,
      "mean_token_accuracy": 0.12317501232028008,
      "num_tokens": 2261452.0,
      "step": 1090
    },
    {
      "entropy": 6.529998350143432,
      "epoch": 0.09932873730043541,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999991777122069,
      "loss": 6.3514,
      "mean_token_accuracy": 0.12049490809440613,
      "num_tokens": 2271929.0,
      "step": 1095
    },
    {
      "entropy": 6.48491473197937,
      "epoch": 0.0997822931785196,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999990879083156,
      "loss": 6.3314,
      "mean_token_accuracy": 0.12258056998252868,
      "num_tokens": 2282987.0,
      "step": 1100
    },
    {
      "entropy": 6.453391408920288,
      "epoch": 0.10023584905660378,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999989934513857,
      "loss": 6.3044,
      "mean_token_accuracy": 0.12670305147767066,
      "num_tokens": 2292796.0,
      "step": 1105
    },
    {
      "entropy": 6.591043186187744,
      "epoch": 0.10068940493468795,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999988943414193,
      "loss": 6.3516,
      "mean_token_accuracy": 0.12217535674571992,
      "num_tokens": 2303057.0,
      "step": 1110
    },
    {
      "entropy": 6.426148223876953,
      "epoch": 0.10114296081277213,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999987905784184,
      "loss": 6.334,
      "mean_token_accuracy": 0.12347318157553673,
      "num_tokens": 2314168.0,
      "step": 1115
    },
    {
      "entropy": 6.5130480289459225,
      "epoch": 0.10159651669085631,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999986821623853,
      "loss": 6.2937,
      "mean_token_accuracy": 0.12410474196076393,
      "num_tokens": 2325241.0,
      "step": 1120
    },
    {
      "entropy": 6.5921142578125,
      "epoch": 0.1020500725689405,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999985690933219,
      "loss": 6.3255,
      "mean_token_accuracy": 0.12369295954704285,
      "num_tokens": 2334866.0,
      "step": 1125
    },
    {
      "entropy": 6.381993150711059,
      "epoch": 0.10250362844702467,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999984513712311,
      "loss": 6.3279,
      "mean_token_accuracy": 0.12018040791153908,
      "num_tokens": 2345422.0,
      "step": 1130
    },
    {
      "entropy": 6.5917377948760985,
      "epoch": 0.10295718432510885,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999983289961146,
      "loss": 6.3473,
      "mean_token_accuracy": 0.12131484672427177,
      "num_tokens": 2356285.0,
      "step": 1135
    },
    {
      "entropy": 6.436669492721558,
      "epoch": 0.10341074020319303,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999982019679755,
      "loss": 6.2615,
      "mean_token_accuracy": 0.12863264828920365,
      "num_tokens": 2366861.0,
      "step": 1140
    },
    {
      "entropy": 6.494740104675293,
      "epoch": 0.10386429608127722,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999980702868164,
      "loss": 6.2841,
      "mean_token_accuracy": 0.12933464497327804,
      "num_tokens": 2376851.0,
      "step": 1145
    },
    {
      "entropy": 6.544817733764648,
      "epoch": 0.1043178519593614,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999979339526396,
      "loss": 6.3588,
      "mean_token_accuracy": 0.13027909770607948,
      "num_tokens": 2386857.0,
      "step": 1150
    },
    {
      "entropy": 6.5093776226043705,
      "epoch": 0.10477140783744557,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999977929654484,
      "loss": 6.2838,
      "mean_token_accuracy": 0.12735775336623192,
      "num_tokens": 2397035.0,
      "step": 1155
    },
    {
      "entropy": 6.470782279968262,
      "epoch": 0.10522496371552975,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999976473252453,
      "loss": 6.3559,
      "mean_token_accuracy": 0.12582893818616867,
      "num_tokens": 2408177.0,
      "step": 1160
    },
    {
      "entropy": 6.572890186309815,
      "epoch": 0.10567851959361393,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999974970320338,
      "loss": 6.3173,
      "mean_token_accuracy": 0.12205304279923439,
      "num_tokens": 2418445.0,
      "step": 1165
    },
    {
      "entropy": 6.472396755218506,
      "epoch": 0.10613207547169812,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999973420858165,
      "loss": 6.3095,
      "mean_token_accuracy": 0.1242362380027771,
      "num_tokens": 2428990.0,
      "step": 1170
    },
    {
      "entropy": 6.47665491104126,
      "epoch": 0.10658563134978229,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999971824865968,
      "loss": 6.3435,
      "mean_token_accuracy": 0.11929563954472541,
      "num_tokens": 2439643.0,
      "step": 1175
    },
    {
      "entropy": 6.585198831558228,
      "epoch": 0.10703918722786647,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999970182343782,
      "loss": 6.3812,
      "mean_token_accuracy": 0.12422708719968796,
      "num_tokens": 2450691.0,
      "step": 1180
    },
    {
      "entropy": 6.567696046829224,
      "epoch": 0.10749274310595065,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999968493291638,
      "loss": 6.3221,
      "mean_token_accuracy": 0.12464427575469017,
      "num_tokens": 2460706.0,
      "step": 1185
    },
    {
      "entropy": 6.306331729888916,
      "epoch": 0.10794629898403484,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999966757709573,
      "loss": 6.2099,
      "mean_token_accuracy": 0.1348014645278454,
      "num_tokens": 2470947.0,
      "step": 1190
    },
    {
      "entropy": 6.511581468582153,
      "epoch": 0.10839985486211902,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999964975597622,
      "loss": 6.1938,
      "mean_token_accuracy": 0.1275109112262726,
      "num_tokens": 2481132.0,
      "step": 1195
    },
    {
      "entropy": 6.452719688415527,
      "epoch": 0.10885341074020319,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999963146955821,
      "loss": 6.2243,
      "mean_token_accuracy": 0.12854906544089317,
      "num_tokens": 2491701.0,
      "step": 1200
    },
    {
      "entropy": 6.320997333526611,
      "epoch": 0.10930696661828737,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499996127178421,
      "loss": 6.2922,
      "mean_token_accuracy": 0.12324495241045952,
      "num_tokens": 2502360.0,
      "step": 1205
    },
    {
      "entropy": 6.524869346618653,
      "epoch": 0.10976052249637155,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999959350082825,
      "loss": 6.1951,
      "mean_token_accuracy": 0.13170049339532852,
      "num_tokens": 2512140.0,
      "step": 1210
    },
    {
      "entropy": 6.4425348281860355,
      "epoch": 0.11021407837445574,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999957381851709,
      "loss": 6.3032,
      "mean_token_accuracy": 0.12556114420294762,
      "num_tokens": 2522721.0,
      "step": 1215
    },
    {
      "entropy": 6.40459713935852,
      "epoch": 0.1106676342525399,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999955367090901,
      "loss": 6.1739,
      "mean_token_accuracy": 0.12824239879846572,
      "num_tokens": 2531786.0,
      "step": 1220
    },
    {
      "entropy": 6.501996803283691,
      "epoch": 0.11112119013062409,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999953305800441,
      "loss": 6.2788,
      "mean_token_accuracy": 0.12112325578927993,
      "num_tokens": 2542895.0,
      "step": 1225
    },
    {
      "entropy": 6.3701423645019535,
      "epoch": 0.11157474600870827,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999951197980374,
      "loss": 6.2802,
      "mean_token_accuracy": 0.12341128513216973,
      "num_tokens": 2554120.0,
      "step": 1230
    },
    {
      "entropy": 6.539844799041748,
      "epoch": 0.11202830188679246,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999949043630744,
      "loss": 6.313,
      "mean_token_accuracy": 0.12485537752509117,
      "num_tokens": 2563935.0,
      "step": 1235
    },
    {
      "entropy": 6.355460834503174,
      "epoch": 0.11248185776487664,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999946842751593,
      "loss": 6.2219,
      "mean_token_accuracy": 0.12797416001558304,
      "num_tokens": 2574465.0,
      "step": 1240
    },
    {
      "entropy": 6.420122289657593,
      "epoch": 0.11293541364296081,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999944595342968,
      "loss": 6.1786,
      "mean_token_accuracy": 0.13132491633296012,
      "num_tokens": 2584694.0,
      "step": 1245
    },
    {
      "entropy": 6.382951498031616,
      "epoch": 0.11338896952104499,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999942301404916,
      "loss": 6.1569,
      "mean_token_accuracy": 0.1340431585907936,
      "num_tokens": 2594540.0,
      "step": 1250
    },
    {
      "entropy": 6.32035346031189,
      "epoch": 0.11384252539912917,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999939960937484,
      "loss": 6.0947,
      "mean_token_accuracy": 0.13378870859742165,
      "num_tokens": 2604896.0,
      "step": 1255
    },
    {
      "entropy": 6.4604442596435545,
      "epoch": 0.11429608127721336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999937573940721,
      "loss": 6.2717,
      "mean_token_accuracy": 0.13821497261524202,
      "num_tokens": 2614143.0,
      "step": 1260
    },
    {
      "entropy": 6.316441440582276,
      "epoch": 0.11474963715529753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999935140414674,
      "loss": 6.1643,
      "mean_token_accuracy": 0.12574966847896576,
      "num_tokens": 2623999.0,
      "step": 1265
    },
    {
      "entropy": 6.501933288574219,
      "epoch": 0.11520319303338171,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999932660359395,
      "loss": 6.3783,
      "mean_token_accuracy": 0.12551918104290963,
      "num_tokens": 2636208.0,
      "step": 1270
    },
    {
      "entropy": 6.424142551422119,
      "epoch": 0.11565674891146589,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999930133774936,
      "loss": 6.2436,
      "mean_token_accuracy": 0.13297036662697792,
      "num_tokens": 2645854.0,
      "step": 1275
    },
    {
      "entropy": 6.461622619628907,
      "epoch": 0.11611030478955008,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999927560661348,
      "loss": 6.2337,
      "mean_token_accuracy": 0.13188307881355285,
      "num_tokens": 2655877.0,
      "step": 1280
    },
    {
      "entropy": 6.463164520263672,
      "epoch": 0.11656386066763426,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999924941018685,
      "loss": 6.2551,
      "mean_token_accuracy": 0.13034509047865867,
      "num_tokens": 2666157.0,
      "step": 1285
    },
    {
      "entropy": 6.322438955307007,
      "epoch": 0.11701741654571843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999922274847,
      "loss": 6.1759,
      "mean_token_accuracy": 0.12801435366272926,
      "num_tokens": 2676647.0,
      "step": 1290
    },
    {
      "entropy": 6.35759539604187,
      "epoch": 0.11747097242380261,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499991956214635,
      "loss": 6.2134,
      "mean_token_accuracy": 0.13719891905784606,
      "num_tokens": 2686417.0,
      "step": 1295
    },
    {
      "entropy": 6.4111240863800045,
      "epoch": 0.1179245283018868,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499991680291679,
      "loss": 6.2127,
      "mean_token_accuracy": 0.12441546320915223,
      "num_tokens": 2697540.0,
      "step": 1300
    },
    {
      "entropy": 6.4009637355804445,
      "epoch": 0.11837808417997098,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999913997158378,
      "loss": 6.2549,
      "mean_token_accuracy": 0.11752010956406593,
      "num_tokens": 2709397.0,
      "step": 1305
    },
    {
      "entropy": 6.34979658126831,
      "epoch": 0.11883164005805515,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999911144871169,
      "loss": 6.2137,
      "mean_token_accuracy": 0.1235676109790802,
      "num_tokens": 2720430.0,
      "step": 1310
    },
    {
      "entropy": 6.358980894088745,
      "epoch": 0.11928519593613933,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999908246055226,
      "loss": 6.1748,
      "mean_token_accuracy": 0.12498210072517395,
      "num_tokens": 2730648.0,
      "step": 1315
    },
    {
      "entropy": 6.358032989501953,
      "epoch": 0.11973875181422351,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999905300710607,
      "loss": 6.1528,
      "mean_token_accuracy": 0.12824709638953208,
      "num_tokens": 2739979.0,
      "step": 1320
    },
    {
      "entropy": 6.410165405273437,
      "epoch": 0.1201923076923077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999902308837373,
      "loss": 6.2777,
      "mean_token_accuracy": 0.1253784976899624,
      "num_tokens": 2750329.0,
      "step": 1325
    },
    {
      "entropy": 6.337530565261841,
      "epoch": 0.12064586357039188,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999899270435584,
      "loss": 6.1525,
      "mean_token_accuracy": 0.13555875271558762,
      "num_tokens": 2760535.0,
      "step": 1330
    },
    {
      "entropy": 6.363765001296997,
      "epoch": 0.12109941944847605,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999896185505307,
      "loss": 6.1902,
      "mean_token_accuracy": 0.12614675164222716,
      "num_tokens": 2770470.0,
      "step": 1335
    },
    {
      "entropy": 6.35250883102417,
      "epoch": 0.12155297532656023,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999893054046603,
      "loss": 6.2192,
      "mean_token_accuracy": 0.12356962338089943,
      "num_tokens": 2781506.0,
      "step": 1340
    },
    {
      "entropy": 6.300785732269287,
      "epoch": 0.12200653120464441,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999889876059537,
      "loss": 6.0809,
      "mean_token_accuracy": 0.13543135821819305,
      "num_tokens": 2791245.0,
      "step": 1345
    },
    {
      "entropy": 6.336817073822021,
      "epoch": 0.1224600870827286,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999886651544176,
      "loss": 6.0946,
      "mean_token_accuracy": 0.13693180531263352,
      "num_tokens": 2801680.0,
      "step": 1350
    },
    {
      "entropy": 6.373439359664917,
      "epoch": 0.12291364296081277,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999883380500584,
      "loss": 6.1846,
      "mean_token_accuracy": 0.1356772594153881,
      "num_tokens": 2812805.0,
      "step": 1355
    },
    {
      "entropy": 6.366350793838501,
      "epoch": 0.12336719883889695,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999880062928831,
      "loss": 6.2141,
      "mean_token_accuracy": 0.13203392773866654,
      "num_tokens": 2822834.0,
      "step": 1360
    },
    {
      "entropy": 6.361548852920532,
      "epoch": 0.12382075471698113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999876698828985,
      "loss": 6.1554,
      "mean_token_accuracy": 0.1361112043261528,
      "num_tokens": 2832312.0,
      "step": 1365
    },
    {
      "entropy": 6.335978317260742,
      "epoch": 0.12427431059506532,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999873288201116,
      "loss": 6.1665,
      "mean_token_accuracy": 0.13540629148483277,
      "num_tokens": 2842998.0,
      "step": 1370
    },
    {
      "entropy": 6.394313335418701,
      "epoch": 0.1247278664731495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004999869831045294,
      "loss": 6.1961,
      "mean_token_accuracy": 0.13257036805152894,
      "num_tokens": 2854218.0,
      "step": 1375
    },
    {
      "entropy": 6.33300895690918,
      "epoch": 0.12518142235123367,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999866327361589,
      "loss": 6.0892,
      "mean_token_accuracy": 0.13703012093901634,
      "num_tokens": 2864484.0,
      "step": 1380
    },
    {
      "entropy": 6.370788526535034,
      "epoch": 0.12563497822931785,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999862777150077,
      "loss": 6.2028,
      "mean_token_accuracy": 0.13148538768291473,
      "num_tokens": 2873800.0,
      "step": 1385
    },
    {
      "entropy": 6.390060663223267,
      "epoch": 0.12608853410740203,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999859180410829,
      "loss": 6.1965,
      "mean_token_accuracy": 0.13013821244239807,
      "num_tokens": 2884247.0,
      "step": 1390
    },
    {
      "entropy": 6.341064691543579,
      "epoch": 0.12654208998548622,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999855537143919,
      "loss": 6.1096,
      "mean_token_accuracy": 0.13113176226615905,
      "num_tokens": 2894739.0,
      "step": 1395
    },
    {
      "entropy": 6.234110736846924,
      "epoch": 0.1269956458635704,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999851847349425,
      "loss": 6.1616,
      "mean_token_accuracy": 0.13198944479227065,
      "num_tokens": 2904376.0,
      "step": 1400
    },
    {
      "entropy": 6.4315849304199215,
      "epoch": 0.12744920174165458,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999848111027419,
      "loss": 6.2265,
      "mean_token_accuracy": 0.12843010425567628,
      "num_tokens": 2914304.0,
      "step": 1405
    },
    {
      "entropy": 6.319644594192505,
      "epoch": 0.12790275761973874,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999844328177983,
      "loss": 6.0959,
      "mean_token_accuracy": 0.13285723477602004,
      "num_tokens": 2924400.0,
      "step": 1410
    },
    {
      "entropy": 6.456520843505859,
      "epoch": 0.12835631349782292,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999840498801191,
      "loss": 6.3401,
      "mean_token_accuracy": 0.1338380053639412,
      "num_tokens": 2935040.0,
      "step": 1415
    },
    {
      "entropy": 6.279810905456543,
      "epoch": 0.1288098693759071,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999836622897126,
      "loss": 6.1154,
      "mean_token_accuracy": 0.1344787932932377,
      "num_tokens": 2945536.0,
      "step": 1420
    },
    {
      "entropy": 6.234977674484253,
      "epoch": 0.1292634252539913,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999832700465865,
      "loss": 6.0487,
      "mean_token_accuracy": 0.13450392782688142,
      "num_tokens": 2954969.0,
      "step": 1425
    },
    {
      "entropy": 6.315559005737304,
      "epoch": 0.12971698113207547,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999828731507491,
      "loss": 6.1625,
      "mean_token_accuracy": 0.13059850856661798,
      "num_tokens": 2966218.0,
      "step": 1430
    },
    {
      "entropy": 6.379030179977417,
      "epoch": 0.13017053701015965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999824716022085,
      "loss": 6.1894,
      "mean_token_accuracy": 0.127619219571352,
      "num_tokens": 2976640.0,
      "step": 1435
    },
    {
      "entropy": 6.313711643218994,
      "epoch": 0.13062409288824384,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999820654009731,
      "loss": 6.1533,
      "mean_token_accuracy": 0.1332421600818634,
      "num_tokens": 2985883.0,
      "step": 1440
    },
    {
      "entropy": 6.321325731277466,
      "epoch": 0.13107764876632802,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999816545470513,
      "loss": 6.1541,
      "mean_token_accuracy": 0.12692199870944024,
      "num_tokens": 2996695.0,
      "step": 1445
    },
    {
      "entropy": 6.334502124786377,
      "epoch": 0.1315312046444122,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999812390404514,
      "loss": 6.16,
      "mean_token_accuracy": 0.1343579739332199,
      "num_tokens": 3007500.0,
      "step": 1450
    },
    {
      "entropy": 6.184367370605469,
      "epoch": 0.13198476052249636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999808188811823,
      "loss": 6.0132,
      "mean_token_accuracy": 0.13415692150592803,
      "num_tokens": 3017299.0,
      "step": 1455
    },
    {
      "entropy": 6.403480911254883,
      "epoch": 0.13243831640058054,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999803940692524,
      "loss": 6.3011,
      "mean_token_accuracy": 0.12249834388494492,
      "num_tokens": 3028389.0,
      "step": 1460
    },
    {
      "entropy": 6.317180013656616,
      "epoch": 0.13289187227866472,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999799646046707,
      "loss": 6.1135,
      "mean_token_accuracy": 0.1368947669863701,
      "num_tokens": 3038883.0,
      "step": 1465
    },
    {
      "entropy": 6.403723573684692,
      "epoch": 0.1333454281567489,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999795304874461,
      "loss": 6.2419,
      "mean_token_accuracy": 0.13426837772130967,
      "num_tokens": 3049904.0,
      "step": 1470
    },
    {
      "entropy": 6.2098509788513185,
      "epoch": 0.1337989840348331,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999790917175873,
      "loss": 6.0755,
      "mean_token_accuracy": 0.13349713161587715,
      "num_tokens": 3059788.0,
      "step": 1475
    },
    {
      "entropy": 6.351387977600098,
      "epoch": 0.13425253991291727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999786482951036,
      "loss": 6.0716,
      "mean_token_accuracy": 0.1320554032921791,
      "num_tokens": 3070413.0,
      "step": 1480
    },
    {
      "entropy": 6.270417833328247,
      "epoch": 0.13470609579100146,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999782002200041,
      "loss": 6.1959,
      "mean_token_accuracy": 0.12884705662727355,
      "num_tokens": 3081093.0,
      "step": 1485
    },
    {
      "entropy": 6.342434740066528,
      "epoch": 0.13515965166908564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999777474922982,
      "loss": 6.129,
      "mean_token_accuracy": 0.13011036217212676,
      "num_tokens": 3090594.0,
      "step": 1490
    },
    {
      "entropy": 6.322659873962403,
      "epoch": 0.13561320754716982,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999772901119951,
      "loss": 6.2158,
      "mean_token_accuracy": 0.13068215921521187,
      "num_tokens": 3101061.0,
      "step": 1495
    },
    {
      "entropy": 6.294349813461304,
      "epoch": 0.13606676342525398,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999768280791043,
      "loss": 6.1383,
      "mean_token_accuracy": 0.13425096422433852,
      "num_tokens": 3111715.0,
      "step": 1500
    },
    {
      "entropy": 6.270824289321899,
      "epoch": 0.13652031930333816,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999763613936352,
      "loss": 6.0633,
      "mean_token_accuracy": 0.13947287648916246,
      "num_tokens": 3121378.0,
      "step": 1505
    },
    {
      "entropy": 6.246769809722901,
      "epoch": 0.13697387518142234,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999758900555979,
      "loss": 5.9928,
      "mean_token_accuracy": 0.1401647448539734,
      "num_tokens": 3131229.0,
      "step": 1510
    },
    {
      "entropy": 6.208933019638062,
      "epoch": 0.13742743105950653,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999754140650017,
      "loss": 6.0878,
      "mean_token_accuracy": 0.13210750818252565,
      "num_tokens": 3140859.0,
      "step": 1515
    },
    {
      "entropy": 6.351792240142823,
      "epoch": 0.1378809869375907,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999749334218567,
      "loss": 6.0861,
      "mean_token_accuracy": 0.13336024880409242,
      "num_tokens": 3151692.0,
      "step": 1520
    },
    {
      "entropy": 6.221499061584472,
      "epoch": 0.1383345428156749,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999744481261726,
      "loss": 6.1524,
      "mean_token_accuracy": 0.13277293592691422,
      "num_tokens": 3161879.0,
      "step": 1525
    },
    {
      "entropy": 6.278037405014038,
      "epoch": 0.13878809869375908,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999739581779597,
      "loss": 6.0267,
      "mean_token_accuracy": 0.13772518262267114,
      "num_tokens": 3171282.0,
      "step": 1530
    },
    {
      "entropy": 6.244936561584472,
      "epoch": 0.13924165457184326,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499973463577228,
      "loss": 6.0779,
      "mean_token_accuracy": 0.13918796479701995,
      "num_tokens": 3181398.0,
      "step": 1535
    },
    {
      "entropy": 6.294873476028442,
      "epoch": 0.13969521044992744,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999729643239877,
      "loss": 6.1991,
      "mean_token_accuracy": 0.13689608871936798,
      "num_tokens": 3191431.0,
      "step": 1540
    },
    {
      "entropy": 6.297893095016479,
      "epoch": 0.1401487663280116,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999724604182492,
      "loss": 6.1937,
      "mean_token_accuracy": 0.1275126539170742,
      "num_tokens": 3202147.0,
      "step": 1545
    },
    {
      "entropy": 6.362968301773071,
      "epoch": 0.14060232220609578,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999719518600229,
      "loss": 6.073,
      "mean_token_accuracy": 0.13779982998967172,
      "num_tokens": 3212428.0,
      "step": 1550
    },
    {
      "entropy": 6.222197484970093,
      "epoch": 0.14105587808417996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999714386493192,
      "loss": 6.1203,
      "mean_token_accuracy": 0.13455342650413513,
      "num_tokens": 3222099.0,
      "step": 1555
    },
    {
      "entropy": 6.321711540222168,
      "epoch": 0.14150943396226415,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999709207861489,
      "loss": 6.0726,
      "mean_token_accuracy": 0.13806343004107474,
      "num_tokens": 3232079.0,
      "step": 1560
    },
    {
      "entropy": 6.153911828994751,
      "epoch": 0.14196298984034833,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999703982705227,
      "loss": 5.9721,
      "mean_token_accuracy": 0.14627765491604805,
      "num_tokens": 3241762.0,
      "step": 1565
    },
    {
      "entropy": 6.244314336776734,
      "epoch": 0.1424165457184325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999698711024513,
      "loss": 6.0669,
      "mean_token_accuracy": 0.1394437588751316,
      "num_tokens": 3252571.0,
      "step": 1570
    },
    {
      "entropy": 6.215821599960327,
      "epoch": 0.1428701015965167,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999693392819456,
      "loss": 6.1098,
      "mean_token_accuracy": 0.13461477681994438,
      "num_tokens": 3263542.0,
      "step": 1575
    },
    {
      "entropy": 6.340892887115478,
      "epoch": 0.14332365747460088,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999688028090166,
      "loss": 6.1811,
      "mean_token_accuracy": 0.13378583863377572,
      "num_tokens": 3273295.0,
      "step": 1580
    },
    {
      "entropy": 6.3619109153747555,
      "epoch": 0.14377721335268506,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999682616836755,
      "loss": 6.1952,
      "mean_token_accuracy": 0.136061330139637,
      "num_tokens": 3283605.0,
      "step": 1585
    },
    {
      "entropy": 6.243741178512574,
      "epoch": 0.14423076923076922,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999677159059334,
      "loss": 6.0415,
      "mean_token_accuracy": 0.14292106181383132,
      "num_tokens": 3293837.0,
      "step": 1590
    },
    {
      "entropy": 6.130685424804687,
      "epoch": 0.1446843251088534,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999671654758016,
      "loss": 6.0417,
      "mean_token_accuracy": 0.14042141437530517,
      "num_tokens": 3304131.0,
      "step": 1595
    },
    {
      "entropy": 6.302355670928955,
      "epoch": 0.14513788098693758,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999666103932915,
      "loss": 6.0746,
      "mean_token_accuracy": 0.13077147156000138,
      "num_tokens": 3313543.0,
      "step": 1600
    },
    {
      "entropy": 6.222482824325562,
      "epoch": 0.14559143686502177,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999660506584145,
      "loss": 6.0608,
      "mean_token_accuracy": 0.1355305939912796,
      "num_tokens": 3324487.0,
      "step": 1605
    },
    {
      "entropy": 6.281767272949219,
      "epoch": 0.14604499274310595,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999654862711823,
      "loss": 6.0873,
      "mean_token_accuracy": 0.1388099044561386,
      "num_tokens": 3335094.0,
      "step": 1610
    },
    {
      "entropy": 6.182447385787964,
      "epoch": 0.14649854862119013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999649172316065,
      "loss": 6.0205,
      "mean_token_accuracy": 0.13585399314761162,
      "num_tokens": 3344252.0,
      "step": 1615
    },
    {
      "entropy": 6.219208812713623,
      "epoch": 0.14695210449927432,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499964343539699,
      "loss": 6.0905,
      "mean_token_accuracy": 0.13364078626036643,
      "num_tokens": 3354265.0,
      "step": 1620
    },
    {
      "entropy": 6.3048234462738035,
      "epoch": 0.1474056603773585,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999637651954714,
      "loss": 6.182,
      "mean_token_accuracy": 0.13599163964390754,
      "num_tokens": 3364616.0,
      "step": 1625
    },
    {
      "entropy": 6.33318829536438,
      "epoch": 0.14785921625544268,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999631821989358,
      "loss": 6.1253,
      "mean_token_accuracy": 0.14277874156832696,
      "num_tokens": 3373650.0,
      "step": 1630
    },
    {
      "entropy": 6.114496469497681,
      "epoch": 0.14831277213352684,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999625945501043,
      "loss": 6.0331,
      "mean_token_accuracy": 0.13961085602641105,
      "num_tokens": 3384590.0,
      "step": 1635
    },
    {
      "entropy": 6.34743185043335,
      "epoch": 0.14876632801161102,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499962002248989,
      "loss": 6.0701,
      "mean_token_accuracy": 0.13647577166557312,
      "num_tokens": 3395616.0,
      "step": 1640
    },
    {
      "entropy": 6.165446662902832,
      "epoch": 0.1492198838896952,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999614052956024,
      "loss": 6.0688,
      "mean_token_accuracy": 0.1349581092596054,
      "num_tokens": 3405810.0,
      "step": 1645
    },
    {
      "entropy": 6.321514320373535,
      "epoch": 0.1496734397677794,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999608036899563,
      "loss": 6.1055,
      "mean_token_accuracy": 0.1329289957880974,
      "num_tokens": 3415932.0,
      "step": 1650
    },
    {
      "entropy": 6.197605752944947,
      "epoch": 0.15012699564586357,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999601974320636,
      "loss": 6.0631,
      "mean_token_accuracy": 0.13647494986653327,
      "num_tokens": 3425626.0,
      "step": 1655
    },
    {
      "entropy": 6.155983877182007,
      "epoch": 0.15058055152394775,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999595865219367,
      "loss": 6.0609,
      "mean_token_accuracy": 0.1348869502544403,
      "num_tokens": 3434837.0,
      "step": 1660
    },
    {
      "entropy": 6.258736085891724,
      "epoch": 0.15103410740203194,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999589709595881,
      "loss": 6.0701,
      "mean_token_accuracy": 0.14022651612758635,
      "num_tokens": 3445238.0,
      "step": 1665
    },
    {
      "entropy": 6.191891717910766,
      "epoch": 0.15148766328011612,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999583507450308,
      "loss": 6.0521,
      "mean_token_accuracy": 0.13351622000336646,
      "num_tokens": 3456738.0,
      "step": 1670
    },
    {
      "entropy": 6.110352945327759,
      "epoch": 0.1519412191582003,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999577258782774,
      "loss": 5.9378,
      "mean_token_accuracy": 0.14009969010949136,
      "num_tokens": 3466296.0,
      "step": 1675
    },
    {
      "entropy": 6.319712066650391,
      "epoch": 0.15239477503628446,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499957096359341,
      "loss": 6.1345,
      "mean_token_accuracy": 0.13245827108621597,
      "num_tokens": 3476989.0,
      "step": 1680
    },
    {
      "entropy": 6.156819915771484,
      "epoch": 0.15284833091436864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999564621882343,
      "loss": 5.97,
      "mean_token_accuracy": 0.1411023445427418,
      "num_tokens": 3487050.0,
      "step": 1685
    },
    {
      "entropy": 6.2730926990509035,
      "epoch": 0.15330188679245282,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999558233649708,
      "loss": 6.188,
      "mean_token_accuracy": 0.12993267178535461,
      "num_tokens": 3497777.0,
      "step": 1690
    },
    {
      "entropy": 6.238672304153442,
      "epoch": 0.153755442670537,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999551798895636,
      "loss": 6.0905,
      "mean_token_accuracy": 0.13315979093313218,
      "num_tokens": 3509066.0,
      "step": 1695
    },
    {
      "entropy": 6.142438983917236,
      "epoch": 0.1542089985486212,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999545317620258,
      "loss": 6.026,
      "mean_token_accuracy": 0.13728305473923683,
      "num_tokens": 3519567.0,
      "step": 1700
    },
    {
      "entropy": 6.209616088867188,
      "epoch": 0.15466255442670537,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999538789823711,
      "loss": 6.0481,
      "mean_token_accuracy": 0.13899025991559028,
      "num_tokens": 3529402.0,
      "step": 1705
    },
    {
      "entropy": 6.3146570205688475,
      "epoch": 0.15511611030478956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999532215506127,
      "loss": 6.0703,
      "mean_token_accuracy": 0.1440271407365799,
      "num_tokens": 3539805.0,
      "step": 1710
    },
    {
      "entropy": 6.143733882904053,
      "epoch": 0.15556966618287374,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999525594667645,
      "loss": 5.9303,
      "mean_token_accuracy": 0.14397806972265242,
      "num_tokens": 3550477.0,
      "step": 1715
    },
    {
      "entropy": 6.110523653030396,
      "epoch": 0.15602322206095792,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049995189273084,
      "loss": 6.0405,
      "mean_token_accuracy": 0.139139074832201,
      "num_tokens": 3560729.0,
      "step": 1720
    },
    {
      "entropy": 6.220071601867676,
      "epoch": 0.15647677793904208,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499951221342853,
      "loss": 6.0973,
      "mean_token_accuracy": 0.13195881322026254,
      "num_tokens": 3571538.0,
      "step": 1725
    },
    {
      "entropy": 6.214411115646362,
      "epoch": 0.15693033381712626,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999505453028174,
      "loss": 6.1078,
      "mean_token_accuracy": 0.13914671614766122,
      "num_tokens": 3583128.0,
      "step": 1730
    },
    {
      "entropy": 6.246854066848755,
      "epoch": 0.15738388969521044,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999498646107472,
      "loss": 6.0354,
      "mean_token_accuracy": 0.14036523252725602,
      "num_tokens": 3594358.0,
      "step": 1735
    },
    {
      "entropy": 6.198376893997192,
      "epoch": 0.15783744557329463,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999491792666566,
      "loss": 5.959,
      "mean_token_accuracy": 0.14713839516043664,
      "num_tokens": 3604345.0,
      "step": 1740
    },
    {
      "entropy": 6.148045206069947,
      "epoch": 0.1582910014513788,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999484892705597,
      "loss": 6.0217,
      "mean_token_accuracy": 0.13767840191721917,
      "num_tokens": 3616129.0,
      "step": 1745
    },
    {
      "entropy": 6.18104510307312,
      "epoch": 0.158744557329463,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999477946224705,
      "loss": 6.0602,
      "mean_token_accuracy": 0.13979990035295486,
      "num_tokens": 3627458.0,
      "step": 1750
    },
    {
      "entropy": 6.226989650726319,
      "epoch": 0.15919811320754718,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999470953224037,
      "loss": 5.9442,
      "mean_token_accuracy": 0.14392665922641754,
      "num_tokens": 3636047.0,
      "step": 1755
    },
    {
      "entropy": 6.051787710189819,
      "epoch": 0.15965166908563136,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999463913703735,
      "loss": 5.9397,
      "mean_token_accuracy": 0.1413258582353592,
      "num_tokens": 3646930.0,
      "step": 1760
    },
    {
      "entropy": 6.218035459518433,
      "epoch": 0.16010522496371554,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999456827663948,
      "loss": 6.0338,
      "mean_token_accuracy": 0.14086730852723123,
      "num_tokens": 3657279.0,
      "step": 1765
    },
    {
      "entropy": 6.123547697067261,
      "epoch": 0.1605587808417997,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499944969510482,
      "loss": 5.9464,
      "mean_token_accuracy": 0.1423609510064125,
      "num_tokens": 3667070.0,
      "step": 1770
    },
    {
      "entropy": 6.166107845306397,
      "epoch": 0.16101233671988388,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999442516026498,
      "loss": 5.9613,
      "mean_token_accuracy": 0.1381991073489189,
      "num_tokens": 3677560.0,
      "step": 1775
    },
    {
      "entropy": 6.090453720092773,
      "epoch": 0.16146589259796806,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999435290429133,
      "loss": 5.9268,
      "mean_token_accuracy": 0.1474891871213913,
      "num_tokens": 3686701.0,
      "step": 1780
    },
    {
      "entropy": 6.152267122268677,
      "epoch": 0.16191944847605225,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999428018312872,
      "loss": 6.0228,
      "mean_token_accuracy": 0.14551690965890884,
      "num_tokens": 3696338.0,
      "step": 1785
    },
    {
      "entropy": 6.150699281692505,
      "epoch": 0.16237300435413643,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999420699677867,
      "loss": 5.9993,
      "mean_token_accuracy": 0.14381766989827155,
      "num_tokens": 3705752.0,
      "step": 1790
    },
    {
      "entropy": 6.229558277130127,
      "epoch": 0.1628265602322206,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999413334524269,
      "loss": 6.084,
      "mean_token_accuracy": 0.13718133196234703,
      "num_tokens": 3716225.0,
      "step": 1795
    },
    {
      "entropy": 6.1720422267913815,
      "epoch": 0.1632801161103048,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999405922852229,
      "loss": 5.9581,
      "mean_token_accuracy": 0.14193314164876938,
      "num_tokens": 3726593.0,
      "step": 1800
    },
    {
      "entropy": 6.101333570480347,
      "epoch": 0.16373367198838898,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999398464661901,
      "loss": 6.0091,
      "mean_token_accuracy": 0.13523678332567216,
      "num_tokens": 3737563.0,
      "step": 1805
    },
    {
      "entropy": 6.307489013671875,
      "epoch": 0.16418722786647316,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999390959953441,
      "loss": 6.0708,
      "mean_token_accuracy": 0.1370485432446003,
      "num_tokens": 3747752.0,
      "step": 1810
    },
    {
      "entropy": 6.057960081100464,
      "epoch": 0.16464078374455732,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999383408727002,
      "loss": 6.0019,
      "mean_token_accuracy": 0.139044688642025,
      "num_tokens": 3758795.0,
      "step": 1815
    },
    {
      "entropy": 6.2216832637786865,
      "epoch": 0.1650943396226415,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999375810982741,
      "loss": 5.9714,
      "mean_token_accuracy": 0.14233563393354415,
      "num_tokens": 3768690.0,
      "step": 1820
    },
    {
      "entropy": 6.090987873077393,
      "epoch": 0.16554789550072568,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999368166720815,
      "loss": 6.0494,
      "mean_token_accuracy": 0.13934776037931443,
      "num_tokens": 3779018.0,
      "step": 1825
    },
    {
      "entropy": 6.219239854812622,
      "epoch": 0.16600145137880987,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999360475941382,
      "loss": 5.9469,
      "mean_token_accuracy": 0.14526350796222687,
      "num_tokens": 3788985.0,
      "step": 1830
    },
    {
      "entropy": 6.173433303833008,
      "epoch": 0.16645500725689405,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049993527386446,
      "loss": 6.0821,
      "mean_token_accuracy": 0.13157572448253632,
      "num_tokens": 3800787.0,
      "step": 1835
    },
    {
      "entropy": 6.16374773979187,
      "epoch": 0.16690856313497823,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999344954830633,
      "loss": 5.9647,
      "mean_token_accuracy": 0.13687085136771202,
      "num_tokens": 3811381.0,
      "step": 1840
    },
    {
      "entropy": 6.157071971893311,
      "epoch": 0.16736211901306242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999337124499638,
      "loss": 6.0685,
      "mean_token_accuracy": 0.14529063701629638,
      "num_tokens": 3822055.0,
      "step": 1845
    },
    {
      "entropy": 6.190087890625,
      "epoch": 0.1678156748911466,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999329247651777,
      "loss": 6.0321,
      "mean_token_accuracy": 0.1372029162943363,
      "num_tokens": 3831900.0,
      "step": 1850
    },
    {
      "entropy": 6.127886390686035,
      "epoch": 0.16826923076923078,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999321324287213,
      "loss": 5.9863,
      "mean_token_accuracy": 0.14112438708543779,
      "num_tokens": 3842702.0,
      "step": 1855
    },
    {
      "entropy": 6.237508106231689,
      "epoch": 0.16872278664731494,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999313354406114,
      "loss": 5.9599,
      "mean_token_accuracy": 0.14055541008710862,
      "num_tokens": 3852690.0,
      "step": 1860
    },
    {
      "entropy": 6.039614915847778,
      "epoch": 0.16917634252539912,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499930533800864,
      "loss": 5.9466,
      "mean_token_accuracy": 0.1505817547440529,
      "num_tokens": 3863040.0,
      "step": 1865
    },
    {
      "entropy": 6.180584526062011,
      "epoch": 0.1696298984034833,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999297275094958,
      "loss": 6.0401,
      "mean_token_accuracy": 0.13932662308216096,
      "num_tokens": 3873613.0,
      "step": 1870
    },
    {
      "entropy": 6.042843770980835,
      "epoch": 0.1700834542815675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999289165665235,
      "loss": 5.8485,
      "mean_token_accuracy": 0.1480574920773506,
      "num_tokens": 3883936.0,
      "step": 1875
    },
    {
      "entropy": 6.060999536514283,
      "epoch": 0.17053701015965167,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999281009719639,
      "loss": 5.8154,
      "mean_token_accuracy": 0.14890880212187768,
      "num_tokens": 3893788.0,
      "step": 1880
    },
    {
      "entropy": 6.056200647354126,
      "epoch": 0.17099056603773585,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999272807258339,
      "loss": 5.8885,
      "mean_token_accuracy": 0.14825719892978667,
      "num_tokens": 3903434.0,
      "step": 1885
    },
    {
      "entropy": 6.1081467628479,
      "epoch": 0.17144412191582004,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999264558281504,
      "loss": 5.895,
      "mean_token_accuracy": 0.1487080954015255,
      "num_tokens": 3913476.0,
      "step": 1890
    },
    {
      "entropy": 5.991112613677979,
      "epoch": 0.17189767779390422,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999256262789305,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1485322192311287,
      "num_tokens": 3923301.0,
      "step": 1895
    },
    {
      "entropy": 6.099672555923462,
      "epoch": 0.1723512336719884,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999247920781913,
      "loss": 6.0042,
      "mean_token_accuracy": 0.13689797297120093,
      "num_tokens": 3933335.0,
      "step": 1900
    },
    {
      "entropy": 6.197177267074585,
      "epoch": 0.17280478955007256,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999239532259502,
      "loss": 6.0293,
      "mean_token_accuracy": 0.13568304777145385,
      "num_tokens": 3944171.0,
      "step": 1905
    },
    {
      "entropy": 6.177489757537842,
      "epoch": 0.17325834542815674,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999231097222244,
      "loss": 6.0829,
      "mean_token_accuracy": 0.14412558674812317,
      "num_tokens": 3954160.0,
      "step": 1910
    },
    {
      "entropy": 6.171565389633178,
      "epoch": 0.17371190130624092,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999222615670312,
      "loss": 5.8844,
      "mean_token_accuracy": 0.14805472120642663,
      "num_tokens": 3964110.0,
      "step": 1915
    },
    {
      "entropy": 6.134349346160889,
      "epoch": 0.1741654571843251,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999214087603885,
      "loss": 5.9675,
      "mean_token_accuracy": 0.1452759712934494,
      "num_tokens": 3974293.0,
      "step": 1920
    },
    {
      "entropy": 6.078616380691528,
      "epoch": 0.1746190130624093,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999205513023135,
      "loss": 5.9293,
      "mean_token_accuracy": 0.15142350345849992,
      "num_tokens": 3984346.0,
      "step": 1925
    },
    {
      "entropy": 6.053181409835815,
      "epoch": 0.17507256894049347,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999196891928245,
      "loss": 5.866,
      "mean_token_accuracy": 0.15365748405456542,
      "num_tokens": 3992897.0,
      "step": 1930
    },
    {
      "entropy": 6.05509238243103,
      "epoch": 0.17552612481857766,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999188224319388,
      "loss": 5.897,
      "mean_token_accuracy": 0.15057138204574586,
      "num_tokens": 4002575.0,
      "step": 1935
    },
    {
      "entropy": 6.070607757568359,
      "epoch": 0.17597968069666184,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999179510196746,
      "loss": 5.9925,
      "mean_token_accuracy": 0.14099667817354203,
      "num_tokens": 4013752.0,
      "step": 1940
    },
    {
      "entropy": 6.107510566711426,
      "epoch": 0.17643323657474602,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999170749560498,
      "loss": 5.8739,
      "mean_token_accuracy": 0.14898938685655594,
      "num_tokens": 4023718.0,
      "step": 1945
    },
    {
      "entropy": 6.064271402359009,
      "epoch": 0.17688679245283018,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999161942410827,
      "loss": 5.9743,
      "mean_token_accuracy": 0.1369916170835495,
      "num_tokens": 4034210.0,
      "step": 1950
    },
    {
      "entropy": 6.2005760192871096,
      "epoch": 0.17734034833091436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999153088747913,
      "loss": 5.9622,
      "mean_token_accuracy": 0.1424011543393135,
      "num_tokens": 4044868.0,
      "step": 1955
    },
    {
      "entropy": 5.97287278175354,
      "epoch": 0.17779390420899854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499914418857194,
      "loss": 5.9603,
      "mean_token_accuracy": 0.14433022439479828,
      "num_tokens": 4055049.0,
      "step": 1960
    },
    {
      "entropy": 6.208456897735596,
      "epoch": 0.17824746008708273,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999135241883092,
      "loss": 6.0087,
      "mean_token_accuracy": 0.14180623665452002,
      "num_tokens": 4065889.0,
      "step": 1965
    },
    {
      "entropy": 6.090062475204467,
      "epoch": 0.1787010159651669,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999126248681555,
      "loss": 6.0282,
      "mean_token_accuracy": 0.1386457771062851,
      "num_tokens": 4077288.0,
      "step": 1970
    },
    {
      "entropy": 6.084208917617798,
      "epoch": 0.1791545718432511,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999117208967513,
      "loss": 5.9802,
      "mean_token_accuracy": 0.1426486611366272,
      "num_tokens": 4087402.0,
      "step": 1975
    },
    {
      "entropy": 6.152999305725098,
      "epoch": 0.17960812772133528,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999108122741154,
      "loss": 6.027,
      "mean_token_accuracy": 0.13648915737867356,
      "num_tokens": 4098150.0,
      "step": 1980
    },
    {
      "entropy": 6.015424156188965,
      "epoch": 0.18006168359941946,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999098990002666,
      "loss": 5.8866,
      "mean_token_accuracy": 0.1459066838026047,
      "num_tokens": 4108134.0,
      "step": 1985
    },
    {
      "entropy": 6.190234994888305,
      "epoch": 0.18051523947750364,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999089810752237,
      "loss": 6.0277,
      "mean_token_accuracy": 0.13577395901083947,
      "num_tokens": 4118251.0,
      "step": 1990
    },
    {
      "entropy": 6.0451069355010985,
      "epoch": 0.1809687953555878,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499908058499006,
      "loss": 5.8609,
      "mean_token_accuracy": 0.1469724267721176,
      "num_tokens": 4129228.0,
      "step": 1995
    },
    {
      "entropy": 6.0451616764068605,
      "epoch": 0.18142235123367198,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999071312716321,
      "loss": 5.9319,
      "mean_token_accuracy": 0.15002663657069207,
      "num_tokens": 4139391.0,
      "step": 2000
    },
    {
      "entropy": 6.0800574779510494,
      "epoch": 0.18187590711175616,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999061993931215,
      "loss": 5.8526,
      "mean_token_accuracy": 0.1524651437997818,
      "num_tokens": 4148891.0,
      "step": 2005
    },
    {
      "entropy": 6.2835156440734865,
      "epoch": 0.18232946298984035,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999052628634934,
      "loss": 6.2064,
      "mean_token_accuracy": 0.138189584761858,
      "num_tokens": 4159551.0,
      "step": 2010
    },
    {
      "entropy": 6.138812065124512,
      "epoch": 0.18278301886792453,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999043216827671,
      "loss": 6.0055,
      "mean_token_accuracy": 0.14381420388817787,
      "num_tokens": 4170055.0,
      "step": 2015
    },
    {
      "entropy": 6.033186531066894,
      "epoch": 0.1832365747460087,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999033758509621,
      "loss": 5.8591,
      "mean_token_accuracy": 0.14449694454669954,
      "num_tokens": 4180767.0,
      "step": 2020
    },
    {
      "entropy": 6.07762713432312,
      "epoch": 0.1836901306240929,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499902425368098,
      "loss": 5.9878,
      "mean_token_accuracy": 0.14416737630963325,
      "num_tokens": 4190654.0,
      "step": 2025
    },
    {
      "entropy": 6.103671360015869,
      "epoch": 0.18414368650217708,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999014702341944,
      "loss": 5.9002,
      "mean_token_accuracy": 0.1454018011689186,
      "num_tokens": 4200753.0,
      "step": 2030
    },
    {
      "entropy": 6.093059635162353,
      "epoch": 0.18459724238026126,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999005104492712,
      "loss": 5.9798,
      "mean_token_accuracy": 0.14127395525574685,
      "num_tokens": 4210994.0,
      "step": 2035
    },
    {
      "entropy": 6.107717561721802,
      "epoch": 0.18505079825834542,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998995460133481,
      "loss": 5.9256,
      "mean_token_accuracy": 0.1447222650051117,
      "num_tokens": 4220883.0,
      "step": 2040
    },
    {
      "entropy": 6.123315095901489,
      "epoch": 0.1855043541364296,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998985769264451,
      "loss": 5.9518,
      "mean_token_accuracy": 0.13728086575865744,
      "num_tokens": 4231269.0,
      "step": 2045
    },
    {
      "entropy": 6.056053113937378,
      "epoch": 0.18595791001451378,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998976031885821,
      "loss": 5.9715,
      "mean_token_accuracy": 0.14049455970525743,
      "num_tokens": 4241559.0,
      "step": 2050
    },
    {
      "entropy": 6.045488786697388,
      "epoch": 0.18641146589259797,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998966247997794,
      "loss": 5.8466,
      "mean_token_accuracy": 0.14596481025218963,
      "num_tokens": 4250870.0,
      "step": 2055
    },
    {
      "entropy": 6.027891731262207,
      "epoch": 0.18686502177068215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998956417600572,
      "loss": 5.8125,
      "mean_token_accuracy": 0.1532222166657448,
      "num_tokens": 4260848.0,
      "step": 2060
    },
    {
      "entropy": 6.0480344772338865,
      "epoch": 0.18731857764876633,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998946540694358,
      "loss": 5.9379,
      "mean_token_accuracy": 0.14905711114406586,
      "num_tokens": 4270973.0,
      "step": 2065
    },
    {
      "entropy": 6.159943771362305,
      "epoch": 0.18777213352685052,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998936617279357,
      "loss": 6.0104,
      "mean_token_accuracy": 0.14108132272958757,
      "num_tokens": 4281211.0,
      "step": 2070
    },
    {
      "entropy": 6.0061546802520756,
      "epoch": 0.1882256894049347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998926647355773,
      "loss": 5.8728,
      "mean_token_accuracy": 0.14745167195796965,
      "num_tokens": 4291425.0,
      "step": 2075
    },
    {
      "entropy": 6.123989725112915,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998916630923813,
      "loss": 5.9712,
      "mean_token_accuracy": 0.1410026654601097,
      "num_tokens": 4301079.0,
      "step": 2080
    },
    {
      "entropy": 6.104276657104492,
      "epoch": 0.18913280116110304,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998906567983685,
      "loss": 5.9629,
      "mean_token_accuracy": 0.14320143908262253,
      "num_tokens": 4312374.0,
      "step": 2085
    },
    {
      "entropy": 6.043295049667359,
      "epoch": 0.18958635703918722,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998896458535595,
      "loss": 5.8714,
      "mean_token_accuracy": 0.15539209097623824,
      "num_tokens": 4322810.0,
      "step": 2090
    },
    {
      "entropy": 6.002667427062988,
      "epoch": 0.1900399129172714,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998886302579752,
      "loss": 5.8651,
      "mean_token_accuracy": 0.14834146052598954,
      "num_tokens": 4332082.0,
      "step": 2095
    },
    {
      "entropy": 6.0972264289855955,
      "epoch": 0.19049346879535559,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499887610011637,
      "loss": 5.9636,
      "mean_token_accuracy": 0.14356174021959306,
      "num_tokens": 4342632.0,
      "step": 2100
    },
    {
      "entropy": 6.016445589065552,
      "epoch": 0.19094702467343977,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998865851145656,
      "loss": 5.8886,
      "mean_token_accuracy": 0.1432491272687912,
      "num_tokens": 4353544.0,
      "step": 2105
    },
    {
      "entropy": 6.031062507629395,
      "epoch": 0.19140058055152395,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998855555667822,
      "loss": 6.02,
      "mean_token_accuracy": 0.14226981848478318,
      "num_tokens": 4363636.0,
      "step": 2110
    },
    {
      "entropy": 6.133910322189331,
      "epoch": 0.19185413642960814,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998845213683084,
      "loss": 5.8763,
      "mean_token_accuracy": 0.1426288031041622,
      "num_tokens": 4373672.0,
      "step": 2115
    },
    {
      "entropy": 5.938132858276367,
      "epoch": 0.19230769230769232,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998834825191652,
      "loss": 5.8381,
      "mean_token_accuracy": 0.14870256185531616,
      "num_tokens": 4383199.0,
      "step": 2120
    },
    {
      "entropy": 6.071506214141846,
      "epoch": 0.1927612481857765,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998824390193743,
      "loss": 5.9109,
      "mean_token_accuracy": 0.14334884583950042,
      "num_tokens": 4394446.0,
      "step": 2125
    },
    {
      "entropy": 6.126976585388183,
      "epoch": 0.19321480406386066,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998813908689573,
      "loss": 5.9646,
      "mean_token_accuracy": 0.1414870209991932,
      "num_tokens": 4405882.0,
      "step": 2130
    },
    {
      "entropy": 5.956241512298584,
      "epoch": 0.19366835994194484,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998803380679359,
      "loss": 5.8182,
      "mean_token_accuracy": 0.15368922352790831,
      "num_tokens": 4415280.0,
      "step": 2135
    },
    {
      "entropy": 6.030566740036011,
      "epoch": 0.19412191582002902,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998792806163317,
      "loss": 6.0265,
      "mean_token_accuracy": 0.13928496092557907,
      "num_tokens": 4425755.0,
      "step": 2140
    },
    {
      "entropy": 6.020366859436035,
      "epoch": 0.1945754716981132,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998782185141667,
      "loss": 5.8808,
      "mean_token_accuracy": 0.14801339209079742,
      "num_tokens": 4436259.0,
      "step": 2145
    },
    {
      "entropy": 6.056359338760376,
      "epoch": 0.1950290275761974,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998771517614629,
      "loss": 5.917,
      "mean_token_accuracy": 0.14185069352388383,
      "num_tokens": 4446423.0,
      "step": 2150
    },
    {
      "entropy": 6.050050067901611,
      "epoch": 0.19548258345428157,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998760803582422,
      "loss": 5.871,
      "mean_token_accuracy": 0.15231530964374543,
      "num_tokens": 4456678.0,
      "step": 2155
    },
    {
      "entropy": 6.006132411956787,
      "epoch": 0.19593613933236576,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998750043045269,
      "loss": 5.8574,
      "mean_token_accuracy": 0.15113169997930526,
      "num_tokens": 4467031.0,
      "step": 2160
    },
    {
      "entropy": 6.0697108745574955,
      "epoch": 0.19638969521044994,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998739236003392,
      "loss": 5.938,
      "mean_token_accuracy": 0.13495565131306647,
      "num_tokens": 4477771.0,
      "step": 2165
    },
    {
      "entropy": 6.000240993499756,
      "epoch": 0.1968432510885341,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998728382457014,
      "loss": 5.9664,
      "mean_token_accuracy": 0.14246328622102739,
      "num_tokens": 4488426.0,
      "step": 2170
    },
    {
      "entropy": 6.17234845161438,
      "epoch": 0.19729680696661828,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998717482406362,
      "loss": 5.9372,
      "mean_token_accuracy": 0.13979932889342309,
      "num_tokens": 4499735.0,
      "step": 2175
    },
    {
      "entropy": 6.042126989364624,
      "epoch": 0.19775036284470246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998706535851657,
      "loss": 5.8795,
      "mean_token_accuracy": 0.14715081751346587,
      "num_tokens": 4509500.0,
      "step": 2180
    },
    {
      "entropy": 6.014921045303344,
      "epoch": 0.19820391872278664,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998695542793129,
      "loss": 5.8802,
      "mean_token_accuracy": 0.14634579196572303,
      "num_tokens": 4520431.0,
      "step": 2185
    },
    {
      "entropy": 6.034388875961303,
      "epoch": 0.19865747460087083,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998684503231003,
      "loss": 5.9784,
      "mean_token_accuracy": 0.14680968075990677,
      "num_tokens": 4532294.0,
      "step": 2190
    },
    {
      "entropy": 6.020555639266968,
      "epoch": 0.199111030478955,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499867341716551,
      "loss": 5.8983,
      "mean_token_accuracy": 0.14305620416998863,
      "num_tokens": 4542488.0,
      "step": 2195
    },
    {
      "entropy": 6.039434766769409,
      "epoch": 0.1995645863570392,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998662284596878,
      "loss": 5.8402,
      "mean_token_accuracy": 0.13846864104270934,
      "num_tokens": 4552933.0,
      "step": 2200
    },
    {
      "entropy": 6.0450187683105465,
      "epoch": 0.20001814223512338,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998651105525335,
      "loss": 5.8223,
      "mean_token_accuracy": 0.15138240456581115,
      "num_tokens": 4562373.0,
      "step": 2205
    },
    {
      "entropy": 5.943200492858887,
      "epoch": 0.20047169811320756,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998639879951115,
      "loss": 5.9439,
      "mean_token_accuracy": 0.14522400572896005,
      "num_tokens": 4572715.0,
      "step": 2210
    },
    {
      "entropy": 6.120235109329224,
      "epoch": 0.2009252539912917,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499862860787445,
      "loss": 5.8307,
      "mean_token_accuracy": 0.14360565692186356,
      "num_tokens": 4582782.0,
      "step": 2215
    },
    {
      "entropy": 6.041302442550659,
      "epoch": 0.2013788098693759,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998617289295571,
      "loss": 5.9352,
      "mean_token_accuracy": 0.13825485333800316,
      "num_tokens": 4594108.0,
      "step": 2220
    },
    {
      "entropy": 6.074282169342041,
      "epoch": 0.20183236574746008,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998605924214714,
      "loss": 5.8911,
      "mean_token_accuracy": 0.14501106441020967,
      "num_tokens": 4604732.0,
      "step": 2225
    },
    {
      "entropy": 6.039686012268066,
      "epoch": 0.20228592162554426,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998594512632114,
      "loss": 6.0324,
      "mean_token_accuracy": 0.13512664213776587,
      "num_tokens": 4614998.0,
      "step": 2230
    },
    {
      "entropy": 6.029949378967285,
      "epoch": 0.20273947750362845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998583054548006,
      "loss": 5.8536,
      "mean_token_accuracy": 0.14373639225959778,
      "num_tokens": 4624665.0,
      "step": 2235
    },
    {
      "entropy": 6.060518264770508,
      "epoch": 0.20319303338171263,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998571549962628,
      "loss": 5.8374,
      "mean_token_accuracy": 0.15260313004255294,
      "num_tokens": 4633731.0,
      "step": 2240
    },
    {
      "entropy": 6.0299561500549315,
      "epoch": 0.2036465892597968,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998559998876216,
      "loss": 5.8706,
      "mean_token_accuracy": 0.1454593777656555,
      "num_tokens": 4643421.0,
      "step": 2245
    },
    {
      "entropy": 6.054778814315796,
      "epoch": 0.204100145137881,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998548401289012,
      "loss": 5.9553,
      "mean_token_accuracy": 0.1441769301891327,
      "num_tokens": 4653795.0,
      "step": 2250
    },
    {
      "entropy": 6.058734989166259,
      "epoch": 0.20455370101596518,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998536757201253,
      "loss": 5.9276,
      "mean_token_accuracy": 0.1450178414583206,
      "num_tokens": 4664840.0,
      "step": 2255
    },
    {
      "entropy": 6.073224878311157,
      "epoch": 0.20500725689404933,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998525066613181,
      "loss": 5.931,
      "mean_token_accuracy": 0.14578341245651244,
      "num_tokens": 4675508.0,
      "step": 2260
    },
    {
      "entropy": 5.989795446395874,
      "epoch": 0.20546081277213352,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998513329525038,
      "loss": 5.8523,
      "mean_token_accuracy": 0.14992504566907883,
      "num_tokens": 4685995.0,
      "step": 2265
    },
    {
      "entropy": 6.086735725402832,
      "epoch": 0.2059143686502177,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998501545937066,
      "loss": 5.8789,
      "mean_token_accuracy": 0.14490637332201003,
      "num_tokens": 4696659.0,
      "step": 2270
    },
    {
      "entropy": 5.985673189163208,
      "epoch": 0.20636792452830188,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499848971584951,
      "loss": 5.8403,
      "mean_token_accuracy": 0.15447261780500413,
      "num_tokens": 4706438.0,
      "step": 2275
    },
    {
      "entropy": 5.944012832641602,
      "epoch": 0.20682148040638607,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998477839262611,
      "loss": 5.8513,
      "mean_token_accuracy": 0.1460217833518982,
      "num_tokens": 4716402.0,
      "step": 2280
    },
    {
      "entropy": 6.024405479431152,
      "epoch": 0.20727503628447025,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499846591617662,
      "loss": 5.9409,
      "mean_token_accuracy": 0.14774347990751266,
      "num_tokens": 4728488.0,
      "step": 2285
    },
    {
      "entropy": 6.027887296676636,
      "epoch": 0.20772859216255443,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998453946591779,
      "loss": 5.9638,
      "mean_token_accuracy": 0.14974903911352158,
      "num_tokens": 4738693.0,
      "step": 2290
    },
    {
      "entropy": 6.020720100402832,
      "epoch": 0.20818214804063861,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998441930508339,
      "loss": 5.8436,
      "mean_token_accuracy": 0.1517803430557251,
      "num_tokens": 4748895.0,
      "step": 2295
    },
    {
      "entropy": 6.089075136184692,
      "epoch": 0.2086357039187228,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998429867926547,
      "loss": 5.9548,
      "mean_token_accuracy": 0.14135059341788292,
      "num_tokens": 4760135.0,
      "step": 2300
    },
    {
      "entropy": 6.095640230178833,
      "epoch": 0.20908925979680695,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998417758846651,
      "loss": 5.9916,
      "mean_token_accuracy": 0.15003142952919007,
      "num_tokens": 4769935.0,
      "step": 2305
    },
    {
      "entropy": 6.164234828948975,
      "epoch": 0.20954281567489114,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998405603268903,
      "loss": 5.9117,
      "mean_token_accuracy": 0.14601739197969438,
      "num_tokens": 4779520.0,
      "step": 2310
    },
    {
      "entropy": 5.891173362731934,
      "epoch": 0.20999637155297532,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998393401193554,
      "loss": 5.8264,
      "mean_token_accuracy": 0.1525731474161148,
      "num_tokens": 4791256.0,
      "step": 2315
    },
    {
      "entropy": 5.922054576873779,
      "epoch": 0.2104499274310595,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998381152620857,
      "loss": 5.7778,
      "mean_token_accuracy": 0.15643645375967025,
      "num_tokens": 4801334.0,
      "step": 2320
    },
    {
      "entropy": 5.940741872787475,
      "epoch": 0.21090348330914369,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998368857551067,
      "loss": 5.7694,
      "mean_token_accuracy": 0.15321901589632034,
      "num_tokens": 4811361.0,
      "step": 2325
    },
    {
      "entropy": 5.995629167556762,
      "epoch": 0.21135703918722787,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998356515984433,
      "loss": 5.8362,
      "mean_token_accuracy": 0.14930329322814942,
      "num_tokens": 4821190.0,
      "step": 2330
    },
    {
      "entropy": 5.996551895141602,
      "epoch": 0.21181059506531205,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998344127921213,
      "loss": 5.8616,
      "mean_token_accuracy": 0.15249975472688676,
      "num_tokens": 4830694.0,
      "step": 2335
    },
    {
      "entropy": 5.9884954452514645,
      "epoch": 0.21226415094339623,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998331693361666,
      "loss": 5.8478,
      "mean_token_accuracy": 0.14835911467671395,
      "num_tokens": 4841344.0,
      "step": 2340
    },
    {
      "entropy": 5.958621692657471,
      "epoch": 0.21271770682148042,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998319212306046,
      "loss": 5.8214,
      "mean_token_accuracy": 0.1466705918312073,
      "num_tokens": 4851161.0,
      "step": 2345
    },
    {
      "entropy": 5.9765862941741945,
      "epoch": 0.21317126269956457,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499830668475461,
      "loss": 5.8198,
      "mean_token_accuracy": 0.1497928962111473,
      "num_tokens": 4860740.0,
      "step": 2350
    },
    {
      "entropy": 5.9823229789733885,
      "epoch": 0.21362481857764876,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499829411070762,
      "loss": 5.8482,
      "mean_token_accuracy": 0.14942808002233504,
      "num_tokens": 4871733.0,
      "step": 2355
    },
    {
      "entropy": 6.0018017292022705,
      "epoch": 0.21407837445573294,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998281490165335,
      "loss": 5.9064,
      "mean_token_accuracy": 0.14177072122693063,
      "num_tokens": 4883617.0,
      "step": 2360
    },
    {
      "entropy": 6.005364894866943,
      "epoch": 0.21453193033381712,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998268823128016,
      "loss": 5.8483,
      "mean_token_accuracy": 0.14455199390649795,
      "num_tokens": 4894519.0,
      "step": 2365
    },
    {
      "entropy": 5.98149266242981,
      "epoch": 0.2149854862119013,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998256109595923,
      "loss": 5.8014,
      "mean_token_accuracy": 0.14887023940682412,
      "num_tokens": 4904965.0,
      "step": 2370
    },
    {
      "entropy": 5.936092567443848,
      "epoch": 0.2154390420899855,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998243349569323,
      "loss": 5.7256,
      "mean_token_accuracy": 0.1545273095369339,
      "num_tokens": 4914618.0,
      "step": 2375
    },
    {
      "entropy": 5.868344449996949,
      "epoch": 0.21589259796806967,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998230543048477,
      "loss": 5.7769,
      "mean_token_accuracy": 0.15782049298286438,
      "num_tokens": 4924341.0,
      "step": 2380
    },
    {
      "entropy": 5.880142879486084,
      "epoch": 0.21634615384615385,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499821769003365,
      "loss": 5.7439,
      "mean_token_accuracy": 0.15210559889674186,
      "num_tokens": 4934231.0,
      "step": 2385
    },
    {
      "entropy": 5.974342632293701,
      "epoch": 0.21679970972423804,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998204790525109,
      "loss": 5.7777,
      "mean_token_accuracy": 0.1491807922720909,
      "num_tokens": 4944303.0,
      "step": 2390
    },
    {
      "entropy": 5.959984636306762,
      "epoch": 0.2172532656023222,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998191844523119,
      "loss": 5.8578,
      "mean_token_accuracy": 0.15104133933782576,
      "num_tokens": 4954571.0,
      "step": 2395
    },
    {
      "entropy": 5.957349252700806,
      "epoch": 0.21770682148040638,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499817885202795,
      "loss": 5.8266,
      "mean_token_accuracy": 0.14804138243198395,
      "num_tokens": 4964418.0,
      "step": 2400
    },
    {
      "entropy": 5.975922727584839,
      "epoch": 0.21816037735849056,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998165813039869,
      "loss": 5.8163,
      "mean_token_accuracy": 0.1464727595448494,
      "num_tokens": 4975036.0,
      "step": 2405
    },
    {
      "entropy": 5.98774995803833,
      "epoch": 0.21861393323657474,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998152727559145,
      "loss": 5.9215,
      "mean_token_accuracy": 0.14540747702121734,
      "num_tokens": 4985272.0,
      "step": 2410
    },
    {
      "entropy": 6.043728685379028,
      "epoch": 0.21906748911465893,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499813959558605,
      "loss": 5.8498,
      "mean_token_accuracy": 0.15086397528648376,
      "num_tokens": 4995187.0,
      "step": 2415
    },
    {
      "entropy": 6.0442393779754635,
      "epoch": 0.2195210449927431,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998126417120856,
      "loss": 5.8919,
      "mean_token_accuracy": 0.15320269912481307,
      "num_tokens": 5005013.0,
      "step": 2420
    },
    {
      "entropy": 6.054208087921142,
      "epoch": 0.2199746008708273,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998113192163835,
      "loss": 5.9139,
      "mean_token_accuracy": 0.1428740605711937,
      "num_tokens": 5016789.0,
      "step": 2425
    },
    {
      "entropy": 5.934090423583984,
      "epoch": 0.22042815674891147,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499809992071526,
      "loss": 5.8588,
      "mean_token_accuracy": 0.147943913936615,
      "num_tokens": 5026913.0,
      "step": 2430
    },
    {
      "entropy": 5.962672376632691,
      "epoch": 0.22088171262699566,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998086602775406,
      "loss": 5.9471,
      "mean_token_accuracy": 0.1470245487987995,
      "num_tokens": 5037501.0,
      "step": 2435
    },
    {
      "entropy": 6.027901649475098,
      "epoch": 0.2213352685050798,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998073238344547,
      "loss": 5.7987,
      "mean_token_accuracy": 0.14490654170513154,
      "num_tokens": 5048478.0,
      "step": 2440
    },
    {
      "entropy": 5.919871520996094,
      "epoch": 0.221788824383164,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998059827422962,
      "loss": 5.859,
      "mean_token_accuracy": 0.1465262070298195,
      "num_tokens": 5059118.0,
      "step": 2445
    },
    {
      "entropy": 6.006969404220581,
      "epoch": 0.22224238026124818,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998046370010926,
      "loss": 5.8236,
      "mean_token_accuracy": 0.15497831404209136,
      "num_tokens": 5070180.0,
      "step": 2450
    },
    {
      "entropy": 5.935058832168579,
      "epoch": 0.22269593613933236,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998032866108718,
      "loss": 5.8128,
      "mean_token_accuracy": 0.14882795214653016,
      "num_tokens": 5079912.0,
      "step": 2455
    },
    {
      "entropy": 6.052799558639526,
      "epoch": 0.22314949201741655,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998019315716618,
      "loss": 5.9009,
      "mean_token_accuracy": 0.1513648509979248,
      "num_tokens": 5090549.0,
      "step": 2460
    },
    {
      "entropy": 5.9224653244018555,
      "epoch": 0.22360304789550073,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998005718834905,
      "loss": 5.7505,
      "mean_token_accuracy": 0.15711326152086258,
      "num_tokens": 5100880.0,
      "step": 2465
    },
    {
      "entropy": 5.927716445922852,
      "epoch": 0.2240566037735849,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997992075463862,
      "loss": 5.8622,
      "mean_token_accuracy": 0.14909446835517884,
      "num_tokens": 5111796.0,
      "step": 2470
    },
    {
      "entropy": 6.045079755783081,
      "epoch": 0.2245101596516691,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997978385603769,
      "loss": 5.862,
      "mean_token_accuracy": 0.1490783780813217,
      "num_tokens": 5123468.0,
      "step": 2475
    },
    {
      "entropy": 5.968772125244141,
      "epoch": 0.22496371552975328,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997964649254911,
      "loss": 5.7749,
      "mean_token_accuracy": 0.15539813786745071,
      "num_tokens": 5132925.0,
      "step": 2480
    },
    {
      "entropy": 5.98187403678894,
      "epoch": 0.22541727140783743,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997950866417572,
      "loss": 5.8555,
      "mean_token_accuracy": 0.14465042054653168,
      "num_tokens": 5143445.0,
      "step": 2485
    },
    {
      "entropy": 6.025763702392578,
      "epoch": 0.22587082728592162,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997937037092034,
      "loss": 5.7817,
      "mean_token_accuracy": 0.14337050616741182,
      "num_tokens": 5153176.0,
      "step": 2490
    },
    {
      "entropy": 6.005899715423584,
      "epoch": 0.2263243831640058,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997923161278587,
      "loss": 5.8267,
      "mean_token_accuracy": 0.1517674669623375,
      "num_tokens": 5162765.0,
      "step": 2495
    },
    {
      "entropy": 5.946051549911499,
      "epoch": 0.22677793904208998,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997909238977514,
      "loss": 5.8804,
      "mean_token_accuracy": 0.15172123089432715,
      "num_tokens": 5173390.0,
      "step": 2500
    },
    {
      "entropy": 5.946356534957886,
      "epoch": 0.22723149492017417,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997895270189108,
      "loss": 5.7811,
      "mean_token_accuracy": 0.14956301152706147,
      "num_tokens": 5184212.0,
      "step": 2505
    },
    {
      "entropy": 5.886882019042969,
      "epoch": 0.22768505079825835,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997881254913653,
      "loss": 5.7728,
      "mean_token_accuracy": 0.15003687813878058,
      "num_tokens": 5194311.0,
      "step": 2510
    },
    {
      "entropy": 5.916078996658325,
      "epoch": 0.22813860667634253,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997867193151441,
      "loss": 5.7712,
      "mean_token_accuracy": 0.15247755497694016,
      "num_tokens": 5204708.0,
      "step": 2515
    },
    {
      "entropy": 5.87827501296997,
      "epoch": 0.22859216255442671,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997853084902763,
      "loss": 5.7469,
      "mean_token_accuracy": 0.15128237307071685,
      "num_tokens": 5214566.0,
      "step": 2520
    },
    {
      "entropy": 5.943019533157349,
      "epoch": 0.2290457184325109,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997838930167911,
      "loss": 5.7878,
      "mean_token_accuracy": 0.15266707986593248,
      "num_tokens": 5224762.0,
      "step": 2525
    },
    {
      "entropy": 5.9829565525054935,
      "epoch": 0.22949927431059505,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997824728947176,
      "loss": 5.8435,
      "mean_token_accuracy": 0.15028240010142327,
      "num_tokens": 5235146.0,
      "step": 2530
    },
    {
      "entropy": 6.023767328262329,
      "epoch": 0.22995283018867924,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997810481240853,
      "loss": 5.8108,
      "mean_token_accuracy": 0.14803099185228347,
      "num_tokens": 5243867.0,
      "step": 2535
    },
    {
      "entropy": 5.893141078948974,
      "epoch": 0.23040638606676342,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997796187049238,
      "loss": 5.7234,
      "mean_token_accuracy": 0.1500532016158104,
      "num_tokens": 5254279.0,
      "step": 2540
    },
    {
      "entropy": 6.080553483963013,
      "epoch": 0.2308599419448476,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997781846372623,
      "loss": 5.8382,
      "mean_token_accuracy": 0.14982936531305313,
      "num_tokens": 5264769.0,
      "step": 2545
    },
    {
      "entropy": 5.93544192314148,
      "epoch": 0.23131349782293179,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997767459211307,
      "loss": 5.7903,
      "mean_token_accuracy": 0.14275677353143693,
      "num_tokens": 5274291.0,
      "step": 2550
    },
    {
      "entropy": 5.883300113677978,
      "epoch": 0.23176705370101597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997753025565587,
      "loss": 5.7326,
      "mean_token_accuracy": 0.15632595419883727,
      "num_tokens": 5284366.0,
      "step": 2555
    },
    {
      "entropy": 5.95240535736084,
      "epoch": 0.23222060957910015,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997738545435763,
      "loss": 5.8456,
      "mean_token_accuracy": 0.15440944731235504,
      "num_tokens": 5294494.0,
      "step": 2560
    },
    {
      "entropy": 5.928846740722657,
      "epoch": 0.23267416545718433,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997724018822133,
      "loss": 5.8394,
      "mean_token_accuracy": 0.15485012233257295,
      "num_tokens": 5305606.0,
      "step": 2565
    },
    {
      "entropy": 5.960556077957153,
      "epoch": 0.23312772133526852,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997709445724996,
      "loss": 5.8465,
      "mean_token_accuracy": 0.150348149985075,
      "num_tokens": 5315677.0,
      "step": 2570
    },
    {
      "entropy": 5.986758232116699,
      "epoch": 0.23358127721335267,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997694826144656,
      "loss": 5.9049,
      "mean_token_accuracy": 0.1463898465037346,
      "num_tokens": 5326453.0,
      "step": 2575
    },
    {
      "entropy": 5.92100830078125,
      "epoch": 0.23403483309143686,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004997680160081414,
      "loss": 5.7906,
      "mean_token_accuracy": 0.15171819776296616,
      "num_tokens": 5337591.0,
      "step": 2580
    },
    {
      "entropy": 5.9538902759552,
      "epoch": 0.23448838896952104,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997665447535573,
      "loss": 5.7791,
      "mean_token_accuracy": 0.15522343516349793,
      "num_tokens": 5349400.0,
      "step": 2585
    },
    {
      "entropy": 6.00141224861145,
      "epoch": 0.23494194484760522,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997650688507438,
      "loss": 5.7634,
      "mean_token_accuracy": 0.1506360337138176,
      "num_tokens": 5359984.0,
      "step": 2590
    },
    {
      "entropy": 5.916184520721435,
      "epoch": 0.2353955007256894,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997635882997314,
      "loss": 5.845,
      "mean_token_accuracy": 0.15854744762182235,
      "num_tokens": 5369254.0,
      "step": 2595
    },
    {
      "entropy": 5.9528069496154785,
      "epoch": 0.2358490566037736,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997621031005507,
      "loss": 5.7798,
      "mean_token_accuracy": 0.15245212465524674,
      "num_tokens": 5379648.0,
      "step": 2600
    },
    {
      "entropy": 5.849475145339966,
      "epoch": 0.23630261248185777,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997606132532325,
      "loss": 5.7372,
      "mean_token_accuracy": 0.154351457208395,
      "num_tokens": 5390321.0,
      "step": 2605
    },
    {
      "entropy": 5.919327926635742,
      "epoch": 0.23675616835994195,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997591187578075,
      "loss": 5.7926,
      "mean_token_accuracy": 0.15511986911296843,
      "num_tokens": 5401078.0,
      "step": 2610
    },
    {
      "entropy": 5.961600828170776,
      "epoch": 0.23720972423802614,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997576196143067,
      "loss": 5.7453,
      "mean_token_accuracy": 0.1566541612148285,
      "num_tokens": 5411977.0,
      "step": 2615
    },
    {
      "entropy": 5.893222713470459,
      "epoch": 0.2376632801161103,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499756115822761,
      "loss": 5.7368,
      "mean_token_accuracy": 0.15138195902109147,
      "num_tokens": 5422967.0,
      "step": 2620
    },
    {
      "entropy": 5.9912902355194095,
      "epoch": 0.23811683599419448,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997546073832015,
      "loss": 5.7835,
      "mean_token_accuracy": 0.16118090003728866,
      "num_tokens": 5432260.0,
      "step": 2625
    },
    {
      "entropy": 5.888670158386231,
      "epoch": 0.23857039187227866,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997530942956595,
      "loss": 5.7564,
      "mean_token_accuracy": 0.15361348688602447,
      "num_tokens": 5441706.0,
      "step": 2630
    },
    {
      "entropy": 5.94991717338562,
      "epoch": 0.23902394775036284,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997515765601662,
      "loss": 5.7622,
      "mean_token_accuracy": 0.1588932454586029,
      "num_tokens": 5452397.0,
      "step": 2635
    },
    {
      "entropy": 5.880772686004638,
      "epoch": 0.23947750362844702,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004997500541767531,
      "loss": 5.7218,
      "mean_token_accuracy": 0.15416701436042785,
      "num_tokens": 5463762.0,
      "step": 2640
    },
    {
      "entropy": 5.901000833511352,
      "epoch": 0.2399310595065312,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997485271454516,
      "loss": 5.7915,
      "mean_token_accuracy": 0.14654496908187867,
      "num_tokens": 5474622.0,
      "step": 2645
    },
    {
      "entropy": 5.944158935546875,
      "epoch": 0.2403846153846154,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997469954662933,
      "loss": 5.7691,
      "mean_token_accuracy": 0.15286272019147873,
      "num_tokens": 5484697.0,
      "step": 2650
    },
    {
      "entropy": 5.847146940231323,
      "epoch": 0.24083817126269957,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997454591393097,
      "loss": 5.7742,
      "mean_token_accuracy": 0.15464416444301604,
      "num_tokens": 5496433.0,
      "step": 2655
    },
    {
      "entropy": 5.972603416442871,
      "epoch": 0.24129172714078376,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997439181645329,
      "loss": 5.8895,
      "mean_token_accuracy": 0.1483437992632389,
      "num_tokens": 5507917.0,
      "step": 2660
    },
    {
      "entropy": 6.07913875579834,
      "epoch": 0.2417452830188679,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997423725419945,
      "loss": 5.8717,
      "mean_token_accuracy": 0.14761385768651963,
      "num_tokens": 5519159.0,
      "step": 2665
    },
    {
      "entropy": 5.928588962554931,
      "epoch": 0.2421988388969521,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997408222717267,
      "loss": 5.8249,
      "mean_token_accuracy": 0.15743801593780518,
      "num_tokens": 5529180.0,
      "step": 2670
    },
    {
      "entropy": 5.884641027450561,
      "epoch": 0.24265239477503628,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997392673537612,
      "loss": 5.7382,
      "mean_token_accuracy": 0.16192516684532166,
      "num_tokens": 5539363.0,
      "step": 2675
    },
    {
      "entropy": 5.909084510803223,
      "epoch": 0.24310595065312046,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997377077881306,
      "loss": 5.843,
      "mean_token_accuracy": 0.15110581144690513,
      "num_tokens": 5549841.0,
      "step": 2680
    },
    {
      "entropy": 5.940350866317749,
      "epoch": 0.24355950653120464,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997361435748668,
      "loss": 5.7156,
      "mean_token_accuracy": 0.161916483938694,
      "num_tokens": 5560299.0,
      "step": 2685
    },
    {
      "entropy": 5.962108039855957,
      "epoch": 0.24401306240928883,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997345747140024,
      "loss": 5.8001,
      "mean_token_accuracy": 0.15469845086336137,
      "num_tokens": 5571081.0,
      "step": 2690
    },
    {
      "entropy": 5.871194362640381,
      "epoch": 0.244466618287373,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997330012055695,
      "loss": 5.7317,
      "mean_token_accuracy": 0.15739612132310868,
      "num_tokens": 5581389.0,
      "step": 2695
    },
    {
      "entropy": 5.881601572036743,
      "epoch": 0.2449201741654572,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997314230496011,
      "loss": 5.779,
      "mean_token_accuracy": 0.14633584022521973,
      "num_tokens": 5592347.0,
      "step": 2700
    },
    {
      "entropy": 5.889206314086914,
      "epoch": 0.24537373004354138,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997298402461296,
      "loss": 5.7548,
      "mean_token_accuracy": 0.15528458803892137,
      "num_tokens": 5603411.0,
      "step": 2705
    },
    {
      "entropy": 5.930398607254029,
      "epoch": 0.24582728592162553,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997282527951876,
      "loss": 5.7452,
      "mean_token_accuracy": 0.15865262895822524,
      "num_tokens": 5614615.0,
      "step": 2710
    },
    {
      "entropy": 5.811875200271606,
      "epoch": 0.24628084179970972,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997266606968081,
      "loss": 5.7026,
      "mean_token_accuracy": 0.15606526732444764,
      "num_tokens": 5624582.0,
      "step": 2715
    },
    {
      "entropy": 5.876440620422363,
      "epoch": 0.2467343976777939,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997250639510241,
      "loss": 5.7258,
      "mean_token_accuracy": 0.15305925607681276,
      "num_tokens": 5634310.0,
      "step": 2720
    },
    {
      "entropy": 5.932286167144776,
      "epoch": 0.24718795355587808,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997234625578684,
      "loss": 5.8186,
      "mean_token_accuracy": 0.15080375373363494,
      "num_tokens": 5644018.0,
      "step": 2725
    },
    {
      "entropy": 5.862547779083252,
      "epoch": 0.24764150943396226,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997218565173742,
      "loss": 5.7485,
      "mean_token_accuracy": 0.15690971165895462,
      "num_tokens": 5653592.0,
      "step": 2730
    },
    {
      "entropy": 5.973826742172241,
      "epoch": 0.24809506531204645,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997202458295748,
      "loss": 5.7929,
      "mean_token_accuracy": 0.155655175447464,
      "num_tokens": 5662812.0,
      "step": 2735
    },
    {
      "entropy": 5.936570167541504,
      "epoch": 0.24854862119013063,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997186304945034,
      "loss": 5.7758,
      "mean_token_accuracy": 0.1513756588101387,
      "num_tokens": 5672939.0,
      "step": 2740
    },
    {
      "entropy": 5.83735408782959,
      "epoch": 0.24900217706821481,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997170105121935,
      "loss": 5.7543,
      "mean_token_accuracy": 0.15465778708457947,
      "num_tokens": 5683350.0,
      "step": 2745
    },
    {
      "entropy": 5.977609300613404,
      "epoch": 0.249455732946299,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997153858826786,
      "loss": 5.8559,
      "mean_token_accuracy": 0.1457457385957241,
      "num_tokens": 5694516.0,
      "step": 2750
    },
    {
      "entropy": 5.897079944610596,
      "epoch": 0.24990928882438315,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997137566059922,
      "loss": 5.6687,
      "mean_token_accuracy": 0.15890194922685624,
      "num_tokens": 5705443.0,
      "step": 2755
    },
    {
      "entropy": 5.876714992523193,
      "epoch": 0.25036284470246734,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997121226821681,
      "loss": 5.772,
      "mean_token_accuracy": 0.1513215646147728,
      "num_tokens": 5715933.0,
      "step": 2760
    },
    {
      "entropy": 5.92222113609314,
      "epoch": 0.25081640058055155,
      "grad_norm": 0.953125,
      "learning_rate": 0.00049971048411124,
      "loss": 5.835,
      "mean_token_accuracy": 0.14457671642303466,
      "num_tokens": 5726630.0,
      "step": 2765
    },
    {
      "entropy": 6.002235460281372,
      "epoch": 0.2512699564586357,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997088408932418,
      "loss": 5.7697,
      "mean_token_accuracy": 0.1539665475487709,
      "num_tokens": 5736890.0,
      "step": 2770
    },
    {
      "entropy": 5.778037357330322,
      "epoch": 0.25172351233671986,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997071930282076,
      "loss": 5.6726,
      "mean_token_accuracy": 0.1656890943646431,
      "num_tokens": 5747042.0,
      "step": 2775
    },
    {
      "entropy": 5.874831628799439,
      "epoch": 0.25217706821480407,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997055405161712,
      "loss": 5.7902,
      "mean_token_accuracy": 0.1548820838332176,
      "num_tokens": 5757367.0,
      "step": 2780
    },
    {
      "entropy": 5.933564472198486,
      "epoch": 0.2526306240928882,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997038833571671,
      "loss": 5.8466,
      "mean_token_accuracy": 0.14902109056711196,
      "num_tokens": 5769557.0,
      "step": 2785
    },
    {
      "entropy": 5.887054443359375,
      "epoch": 0.25308417997097243,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997022215512295,
      "loss": 5.7514,
      "mean_token_accuracy": 0.15811785459518432,
      "num_tokens": 5780615.0,
      "step": 2790
    },
    {
      "entropy": 5.939122200012207,
      "epoch": 0.2535377358490566,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997005550983925,
      "loss": 5.7361,
      "mean_token_accuracy": 0.1549833208322525,
      "num_tokens": 5791271.0,
      "step": 2795
    },
    {
      "entropy": 5.814103412628174,
      "epoch": 0.2539912917271408,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499698883998691,
      "loss": 5.6636,
      "mean_token_accuracy": 0.15836629867553711,
      "num_tokens": 5801117.0,
      "step": 2800
    },
    {
      "entropy": 5.859375476837158,
      "epoch": 0.25444484760522496,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996972082521591,
      "loss": 5.6955,
      "mean_token_accuracy": 0.15782571583986282,
      "num_tokens": 5811696.0,
      "step": 2805
    },
    {
      "entropy": 5.879450893402099,
      "epoch": 0.25489840348330917,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996955278588319,
      "loss": 5.7319,
      "mean_token_accuracy": 0.16295181214809418,
      "num_tokens": 5822207.0,
      "step": 2810
    },
    {
      "entropy": 5.838560342788696,
      "epoch": 0.2553519593613933,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996938428187437,
      "loss": 5.6753,
      "mean_token_accuracy": 0.16222673356533052,
      "num_tokens": 5831927.0,
      "step": 2815
    },
    {
      "entropy": 5.844130134582519,
      "epoch": 0.2558055152394775,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996921531319297,
      "loss": 5.7406,
      "mean_token_accuracy": 0.15631944462656974,
      "num_tokens": 5842093.0,
      "step": 2820
    },
    {
      "entropy": 5.862317848205566,
      "epoch": 0.2562590711175617,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996904587984246,
      "loss": 5.6799,
      "mean_token_accuracy": 0.1637733742594719,
      "num_tokens": 5850551.0,
      "step": 2825
    },
    {
      "entropy": 5.831214189529419,
      "epoch": 0.25671262699564584,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996887598182636,
      "loss": 5.6693,
      "mean_token_accuracy": 0.15734255760908128,
      "num_tokens": 5860651.0,
      "step": 2830
    },
    {
      "entropy": 5.736221647262573,
      "epoch": 0.25716618287373005,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996870561914819,
      "loss": 5.6508,
      "mean_token_accuracy": 0.15711580514907836,
      "num_tokens": 5871500.0,
      "step": 2835
    },
    {
      "entropy": 5.837002754211426,
      "epoch": 0.2576197387518142,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996853479181145,
      "loss": 5.6483,
      "mean_token_accuracy": 0.15363348871469498,
      "num_tokens": 5881523.0,
      "step": 2840
    },
    {
      "entropy": 5.969505023956299,
      "epoch": 0.2580732946298984,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996836349981969,
      "loss": 5.8168,
      "mean_token_accuracy": 0.15440904647111892,
      "num_tokens": 5891162.0,
      "step": 2845
    },
    {
      "entropy": 5.825331020355224,
      "epoch": 0.2585268505079826,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996819174317645,
      "loss": 5.6989,
      "mean_token_accuracy": 0.15304600223898887,
      "num_tokens": 5901818.0,
      "step": 2850
    },
    {
      "entropy": 5.837537527084351,
      "epoch": 0.2589804063860668,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996801952188528,
      "loss": 5.7802,
      "mean_token_accuracy": 0.15388137996196746,
      "num_tokens": 5913889.0,
      "step": 2855
    },
    {
      "entropy": 5.898041200637818,
      "epoch": 0.25943396226415094,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996784683594973,
      "loss": 5.7331,
      "mean_token_accuracy": 0.15773886442184448,
      "num_tokens": 5923429.0,
      "step": 2860
    },
    {
      "entropy": 5.9350752353668215,
      "epoch": 0.2598875181422351,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996767368537338,
      "loss": 5.7756,
      "mean_token_accuracy": 0.15665830671787262,
      "num_tokens": 5933847.0,
      "step": 2865
    },
    {
      "entropy": 5.846282577514648,
      "epoch": 0.2603410740203193,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996750007015983,
      "loss": 5.6519,
      "mean_token_accuracy": 0.15802753269672393,
      "num_tokens": 5943874.0,
      "step": 2870
    },
    {
      "entropy": 5.893306922912598,
      "epoch": 0.26079462989840346,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996732599031264,
      "loss": 5.7207,
      "mean_token_accuracy": 0.1564636617898941,
      "num_tokens": 5953653.0,
      "step": 2875
    },
    {
      "entropy": 5.912632322311401,
      "epoch": 0.2612481857764877,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996715144583543,
      "loss": 5.6787,
      "mean_token_accuracy": 0.1598701536655426,
      "num_tokens": 5964039.0,
      "step": 2880
    },
    {
      "entropy": 5.934096145629883,
      "epoch": 0.26170174165457183,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499669764367318,
      "loss": 5.7888,
      "mean_token_accuracy": 0.15126357972621918,
      "num_tokens": 5974083.0,
      "step": 2885
    },
    {
      "entropy": 5.802870368957519,
      "epoch": 0.26215529753265604,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996680096300537,
      "loss": 5.7352,
      "mean_token_accuracy": 0.1535191372036934,
      "num_tokens": 5984629.0,
      "step": 2890
    },
    {
      "entropy": 5.837651777267456,
      "epoch": 0.2626088534107402,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996662502465978,
      "loss": 5.6433,
      "mean_token_accuracy": 0.16809626668691635,
      "num_tokens": 5994295.0,
      "step": 2895
    },
    {
      "entropy": 5.847111940383911,
      "epoch": 0.2630624092888244,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996644862169864,
      "loss": 5.6911,
      "mean_token_accuracy": 0.15560346096754074,
      "num_tokens": 6004156.0,
      "step": 2900
    },
    {
      "entropy": 5.842682647705078,
      "epoch": 0.26351596516690856,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996627175412563,
      "loss": 5.7724,
      "mean_token_accuracy": 0.15826719254255295,
      "num_tokens": 6014188.0,
      "step": 2905
    },
    {
      "entropy": 5.951080179214477,
      "epoch": 0.2639695210449927,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499660944219444,
      "loss": 5.8318,
      "mean_token_accuracy": 0.15130359828472137,
      "num_tokens": 6024879.0,
      "step": 2910
    },
    {
      "entropy": 5.973435401916504,
      "epoch": 0.2644230769230769,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004996591662515861,
      "loss": 5.8978,
      "mean_token_accuracy": 0.1513467513024807,
      "num_tokens": 6036374.0,
      "step": 2915
    },
    {
      "entropy": 5.852985286712647,
      "epoch": 0.2648766328011611,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996573836377193,
      "loss": 5.7384,
      "mean_token_accuracy": 0.15605844408273697,
      "num_tokens": 6046432.0,
      "step": 2920
    },
    {
      "entropy": 5.9294966697692875,
      "epoch": 0.2653301886792453,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996555963778806,
      "loss": 5.7428,
      "mean_token_accuracy": 0.14988288134336472,
      "num_tokens": 6056864.0,
      "step": 2925
    },
    {
      "entropy": 5.9699132442474365,
      "epoch": 0.26578374455732945,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499653804472107,
      "loss": 5.7612,
      "mean_token_accuracy": 0.1611533209681511,
      "num_tokens": 6067294.0,
      "step": 2930
    },
    {
      "entropy": 5.882857704162598,
      "epoch": 0.26623730043541366,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996520079204353,
      "loss": 5.7721,
      "mean_token_accuracy": 0.15410346388816834,
      "num_tokens": 6077587.0,
      "step": 2935
    },
    {
      "entropy": 5.921375894546509,
      "epoch": 0.2666908563134978,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499650206722903,
      "loss": 5.7142,
      "mean_token_accuracy": 0.15302591323852538,
      "num_tokens": 6089306.0,
      "step": 2940
    },
    {
      "entropy": 5.801258993148804,
      "epoch": 0.267144412191582,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996484008795471,
      "loss": 5.6718,
      "mean_token_accuracy": 0.15564401000738143,
      "num_tokens": 6099651.0,
      "step": 2945
    },
    {
      "entropy": 5.831615734100342,
      "epoch": 0.2675979680696662,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499646590390405,
      "loss": 5.6786,
      "mean_token_accuracy": 0.15340599417686462,
      "num_tokens": 6110063.0,
      "step": 2950
    },
    {
      "entropy": 5.866582584381104,
      "epoch": 0.26805152394775034,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996447752555142,
      "loss": 5.7069,
      "mean_token_accuracy": 0.1569851815700531,
      "num_tokens": 6120031.0,
      "step": 2955
    },
    {
      "entropy": 5.845797681808472,
      "epoch": 0.26850507982583455,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996429554749122,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1586455672979355,
      "num_tokens": 6131274.0,
      "step": 2960
    },
    {
      "entropy": 5.888718032836914,
      "epoch": 0.2689586357039187,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996411310486367,
      "loss": 5.7833,
      "mean_token_accuracy": 0.15850588232278823,
      "num_tokens": 6140878.0,
      "step": 2965
    },
    {
      "entropy": 5.895853567123413,
      "epoch": 0.2694121915820029,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004996393019767253,
      "loss": 5.7917,
      "mean_token_accuracy": 0.15040715038776398,
      "num_tokens": 6152229.0,
      "step": 2970
    },
    {
      "entropy": 5.802438020706177,
      "epoch": 0.26986574746008707,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004996374682592158,
      "loss": 5.6873,
      "mean_token_accuracy": 0.15607988238334655,
      "num_tokens": 6162311.0,
      "step": 2975
    },
    {
      "entropy": 5.942750644683838,
      "epoch": 0.2703193033381713,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996356298961463,
      "loss": 5.7936,
      "mean_token_accuracy": 0.15298964083194733,
      "num_tokens": 6172806.0,
      "step": 2980
    },
    {
      "entropy": 5.87895131111145,
      "epoch": 0.27077285921625543,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996337868875549,
      "loss": 5.7587,
      "mean_token_accuracy": 0.15776319652795792,
      "num_tokens": 6183479.0,
      "step": 2985
    },
    {
      "entropy": 5.886080932617188,
      "epoch": 0.27122641509433965,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996319392334792,
      "loss": 5.7299,
      "mean_token_accuracy": 0.15342071503400803,
      "num_tokens": 6194201.0,
      "step": 2990
    },
    {
      "entropy": 5.82788724899292,
      "epoch": 0.2716799709724238,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499630086933958,
      "loss": 5.6724,
      "mean_token_accuracy": 0.1564711168408394,
      "num_tokens": 6204774.0,
      "step": 2995
    },
    {
      "entropy": 5.886050224304199,
      "epoch": 0.27213352685050796,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996282299890292,
      "loss": 5.6831,
      "mean_token_accuracy": 0.1526743471622467,
      "num_tokens": 6215056.0,
      "step": 3000
    },
    {
      "epoch": 0.27213352685050796,
      "eval_entropy": 5.7014565256386325,
      "eval_loss": 5.748072624206543,
      "eval_mean_token_accuracy": 0.1622600189245799,
      "eval_num_tokens": 6215056.0,
      "eval_runtime": 20.702,
      "eval_samples_per_second": 1708.048,
      "eval_steps_per_second": 213.506,
      "step": 3000
    },
    {
      "entropy": 5.934528923034668,
      "epoch": 0.27258708272859217,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004996263683987316,
      "loss": 5.786,
      "mean_token_accuracy": 0.1579520136117935,
      "num_tokens": 6226264.0,
      "step": 3005
    },
    {
      "entropy": 5.856761884689331,
      "epoch": 0.2730406386066763,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996245021631033,
      "loss": 5.7038,
      "mean_token_accuracy": 0.161579892039299,
      "num_tokens": 6235137.0,
      "step": 3010
    },
    {
      "entropy": 5.824236536026001,
      "epoch": 0.27349419448476053,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499622631282183,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15795171409845352,
      "num_tokens": 6245411.0,
      "step": 3015
    },
    {
      "entropy": 5.915331649780273,
      "epoch": 0.2739477503628447,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996207557560095,
      "loss": 5.7743,
      "mean_token_accuracy": 0.1543523669242859,
      "num_tokens": 6255633.0,
      "step": 3020
    },
    {
      "entropy": 5.906075763702392,
      "epoch": 0.2744013062409289,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996188755846216,
      "loss": 5.7388,
      "mean_token_accuracy": 0.1594541311264038,
      "num_tokens": 6265109.0,
      "step": 3025
    },
    {
      "entropy": 5.813056755065918,
      "epoch": 0.27485486211901305,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499616990768058,
      "loss": 5.7737,
      "mean_token_accuracy": 0.1503493830561638,
      "num_tokens": 6275241.0,
      "step": 3030
    },
    {
      "entropy": 5.902055835723877,
      "epoch": 0.27530841799709727,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996151013063579,
      "loss": 5.8127,
      "mean_token_accuracy": 0.15131743252277374,
      "num_tokens": 6286603.0,
      "step": 3035
    },
    {
      "entropy": 5.917137479782104,
      "epoch": 0.2757619738751814,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996132071995601,
      "loss": 5.6609,
      "mean_token_accuracy": 0.16706326305866243,
      "num_tokens": 6297219.0,
      "step": 3040
    },
    {
      "entropy": 5.804540967941284,
      "epoch": 0.2762155297532656,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996113084477042,
      "loss": 5.6991,
      "mean_token_accuracy": 0.16197659075260162,
      "num_tokens": 6307694.0,
      "step": 3045
    },
    {
      "entropy": 5.795589399337769,
      "epoch": 0.2766690856313498,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996094050508291,
      "loss": 5.6305,
      "mean_token_accuracy": 0.16191300749778748,
      "num_tokens": 6317384.0,
      "step": 3050
    },
    {
      "entropy": 5.88090558052063,
      "epoch": 0.27712264150943394,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996074970089741,
      "loss": 5.7749,
      "mean_token_accuracy": 0.15503887385129927,
      "num_tokens": 6327558.0,
      "step": 3055
    },
    {
      "entropy": 5.878848552703857,
      "epoch": 0.27757619738751815,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996055843221789,
      "loss": 5.6264,
      "mean_token_accuracy": 0.16327428370714187,
      "num_tokens": 6337828.0,
      "step": 3060
    },
    {
      "entropy": 5.789236450195313,
      "epoch": 0.2780297532656023,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996036669904831,
      "loss": 5.6907,
      "mean_token_accuracy": 0.16211704909801483,
      "num_tokens": 6348678.0,
      "step": 3065
    },
    {
      "entropy": 5.864919281005859,
      "epoch": 0.2784833091436865,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996017450139261,
      "loss": 5.783,
      "mean_token_accuracy": 0.14863884598016738,
      "num_tokens": 6359670.0,
      "step": 3070
    },
    {
      "entropy": 5.968852424621582,
      "epoch": 0.2789368650217707,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995998183925478,
      "loss": 5.7405,
      "mean_token_accuracy": 0.15753814876079558,
      "num_tokens": 6371034.0,
      "step": 3075
    },
    {
      "entropy": 5.818412113189697,
      "epoch": 0.2793904208998549,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995978871263881,
      "loss": 5.6874,
      "mean_token_accuracy": 0.15982511192560195,
      "num_tokens": 6380672.0,
      "step": 3080
    },
    {
      "entropy": 5.8274352073669435,
      "epoch": 0.27984397677793904,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995959512154868,
      "loss": 5.7991,
      "mean_token_accuracy": 0.14778632521629334,
      "num_tokens": 6390799.0,
      "step": 3085
    },
    {
      "entropy": 5.859681892395019,
      "epoch": 0.2802975326560232,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499594010659884,
      "loss": 5.6602,
      "mean_token_accuracy": 0.1630452647805214,
      "num_tokens": 6400965.0,
      "step": 3090
    },
    {
      "entropy": 5.886337471008301,
      "epoch": 0.2807510885341074,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995920654596199,
      "loss": 5.6873,
      "mean_token_accuracy": 0.1525515004992485,
      "num_tokens": 6411567.0,
      "step": 3095
    },
    {
      "entropy": 5.864609479904175,
      "epoch": 0.28120464441219156,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995901156147345,
      "loss": 5.7302,
      "mean_token_accuracy": 0.16160865426063536,
      "num_tokens": 6421193.0,
      "step": 3100
    },
    {
      "entropy": 5.852170133590699,
      "epoch": 0.2816582002902758,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995881611252684,
      "loss": 5.7431,
      "mean_token_accuracy": 0.16411930173635483,
      "num_tokens": 6430281.0,
      "step": 3105
    },
    {
      "entropy": 5.838845443725586,
      "epoch": 0.28211175616835993,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995862019912619,
      "loss": 5.7401,
      "mean_token_accuracy": 0.1545553356409073,
      "num_tokens": 6441055.0,
      "step": 3110
    },
    {
      "entropy": 5.85880446434021,
      "epoch": 0.28256531204644414,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995842382127555,
      "loss": 5.665,
      "mean_token_accuracy": 0.15411336421966554,
      "num_tokens": 6451082.0,
      "step": 3115
    },
    {
      "entropy": 5.826427602767945,
      "epoch": 0.2830188679245283,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995822697897897,
      "loss": 5.7501,
      "mean_token_accuracy": 0.15211194902658462,
      "num_tokens": 6461617.0,
      "step": 3120
    },
    {
      "entropy": 5.894758319854736,
      "epoch": 0.2834724238026125,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995802967224056,
      "loss": 5.7542,
      "mean_token_accuracy": 0.15588101148605346,
      "num_tokens": 6472780.0,
      "step": 3125
    },
    {
      "entropy": 5.758359670639038,
      "epoch": 0.28392597968069666,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995783190106434,
      "loss": 5.6652,
      "mean_token_accuracy": 0.15255783200263978,
      "num_tokens": 6484078.0,
      "step": 3130
    },
    {
      "entropy": 5.893905305862427,
      "epoch": 0.2843795355587808,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995763366545445,
      "loss": 5.7515,
      "mean_token_accuracy": 0.15216615051031113,
      "num_tokens": 6494366.0,
      "step": 3135
    },
    {
      "entropy": 5.754382371902466,
      "epoch": 0.284833091436865,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004995743496541497,
      "loss": 5.6839,
      "mean_token_accuracy": 0.16558870375156404,
      "num_tokens": 6505461.0,
      "step": 3140
    },
    {
      "entropy": 5.835577821731567,
      "epoch": 0.2852866473149492,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995723580095001,
      "loss": 5.6808,
      "mean_token_accuracy": 0.15959053486585617,
      "num_tokens": 6516141.0,
      "step": 3145
    },
    {
      "entropy": 5.781411981582641,
      "epoch": 0.2857402031930334,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499570361720637,
      "loss": 5.6851,
      "mean_token_accuracy": 0.154303365200758,
      "num_tokens": 6525536.0,
      "step": 3150
    },
    {
      "entropy": 5.864716672897339,
      "epoch": 0.28619375907111755,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004995683607876015,
      "loss": 5.7363,
      "mean_token_accuracy": 0.15480921417474747,
      "num_tokens": 6535852.0,
      "step": 3155
    },
    {
      "entropy": 5.880465698242188,
      "epoch": 0.28664731494920176,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995663552104352,
      "loss": 5.7094,
      "mean_token_accuracy": 0.16345915347337722,
      "num_tokens": 6546353.0,
      "step": 3160
    },
    {
      "entropy": 5.76881217956543,
      "epoch": 0.2871008708272859,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995643449891793,
      "loss": 5.6586,
      "mean_token_accuracy": 0.15820098519325257,
      "num_tokens": 6557782.0,
      "step": 3165
    },
    {
      "entropy": 5.869258546829224,
      "epoch": 0.2875544267053701,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995623301238757,
      "loss": 5.6772,
      "mean_token_accuracy": 0.16326069831848145,
      "num_tokens": 6568100.0,
      "step": 3170
    },
    {
      "entropy": 5.809826850891113,
      "epoch": 0.2880079825834543,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995603106145657,
      "loss": 5.7354,
      "mean_token_accuracy": 0.15384810715913771,
      "num_tokens": 6578698.0,
      "step": 3175
    },
    {
      "entropy": 5.877403926849365,
      "epoch": 0.28846153846153844,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995582864612913,
      "loss": 5.7211,
      "mean_token_accuracy": 0.15826957821846008,
      "num_tokens": 6588908.0,
      "step": 3180
    },
    {
      "entropy": 5.767029571533203,
      "epoch": 0.28891509433962265,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995562576640943,
      "loss": 5.577,
      "mean_token_accuracy": 0.16004678606987,
      "num_tokens": 6599445.0,
      "step": 3185
    },
    {
      "entropy": 5.799728536605835,
      "epoch": 0.2893686502177068,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995542242230167,
      "loss": 5.6578,
      "mean_token_accuracy": 0.16209590137004853,
      "num_tokens": 6610246.0,
      "step": 3190
    },
    {
      "entropy": 5.798572778701782,
      "epoch": 0.289822206095791,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995521861381006,
      "loss": 5.6769,
      "mean_token_accuracy": 0.15961335003376007,
      "num_tokens": 6620501.0,
      "step": 3195
    },
    {
      "entropy": 5.898326778411866,
      "epoch": 0.29027576197387517,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499550143409388,
      "loss": 5.7474,
      "mean_token_accuracy": 0.15129139870405198,
      "num_tokens": 6631206.0,
      "step": 3200
    },
    {
      "entropy": 5.829407978057861,
      "epoch": 0.2907293178519594,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004995480960369211,
      "loss": 5.6945,
      "mean_token_accuracy": 0.1549079790711403,
      "num_tokens": 6641600.0,
      "step": 3205
    },
    {
      "entropy": 5.8490814685821535,
      "epoch": 0.29118287373004353,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995460440207425,
      "loss": 5.7631,
      "mean_token_accuracy": 0.15415688455104828,
      "num_tokens": 6652319.0,
      "step": 3210
    },
    {
      "entropy": 5.883116722106934,
      "epoch": 0.29163642960812775,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995439873608945,
      "loss": 5.7008,
      "mean_token_accuracy": 0.1500185400247574,
      "num_tokens": 6663702.0,
      "step": 3215
    },
    {
      "entropy": 5.808498239517212,
      "epoch": 0.2920899854862119,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995419260574196,
      "loss": 5.684,
      "mean_token_accuracy": 0.1507360652089119,
      "num_tokens": 6673908.0,
      "step": 3220
    },
    {
      "entropy": 5.82448444366455,
      "epoch": 0.29254354136429606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995398601103603,
      "loss": 5.7033,
      "mean_token_accuracy": 0.16093184649944306,
      "num_tokens": 6683237.0,
      "step": 3225
    },
    {
      "entropy": 5.836985397338867,
      "epoch": 0.29299709724238027,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995377895197597,
      "loss": 5.6351,
      "mean_token_accuracy": 0.1617561772465706,
      "num_tokens": 6693451.0,
      "step": 3230
    },
    {
      "entropy": 5.736857318878174,
      "epoch": 0.2934506531204644,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995357142856602,
      "loss": 5.5948,
      "mean_token_accuracy": 0.16023087352514268,
      "num_tokens": 6703479.0,
      "step": 3235
    },
    {
      "entropy": 5.751489543914795,
      "epoch": 0.29390420899854863,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499533634408105,
      "loss": 5.5968,
      "mean_token_accuracy": 0.16008537858724595,
      "num_tokens": 6715321.0,
      "step": 3240
    },
    {
      "entropy": 5.791557550430298,
      "epoch": 0.2943577648766328,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499531549887137,
      "loss": 5.6655,
      "mean_token_accuracy": 0.1540275439620018,
      "num_tokens": 6725541.0,
      "step": 3245
    },
    {
      "entropy": 5.8105682849884035,
      "epoch": 0.294811320754717,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995294607227993,
      "loss": 5.7043,
      "mean_token_accuracy": 0.16713897138834,
      "num_tokens": 6735165.0,
      "step": 3250
    },
    {
      "entropy": 5.884285402297974,
      "epoch": 0.29526487663280115,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995273669151351,
      "loss": 5.701,
      "mean_token_accuracy": 0.16163005083799362,
      "num_tokens": 6744245.0,
      "step": 3255
    },
    {
      "entropy": 5.83087010383606,
      "epoch": 0.29571843251088537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499525268464188,
      "loss": 5.6645,
      "mean_token_accuracy": 0.16589844822883607,
      "num_tokens": 6754894.0,
      "step": 3260
    },
    {
      "entropy": 5.85270414352417,
      "epoch": 0.2961719883889695,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995231653700009,
      "loss": 5.7307,
      "mean_token_accuracy": 0.15026739612221718,
      "num_tokens": 6764442.0,
      "step": 3265
    },
    {
      "entropy": 5.814293384552002,
      "epoch": 0.2966255442670537,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995210576326176,
      "loss": 5.6807,
      "mean_token_accuracy": 0.1561569720506668,
      "num_tokens": 6775054.0,
      "step": 3270
    },
    {
      "entropy": 5.85151162147522,
      "epoch": 0.2970791001451379,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995189452520816,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1682630956172943,
      "num_tokens": 6784825.0,
      "step": 3275
    },
    {
      "entropy": 5.8156262874603275,
      "epoch": 0.29753265602322204,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995168282284366,
      "loss": 5.7081,
      "mean_token_accuracy": 0.16142811328172685,
      "num_tokens": 6795928.0,
      "step": 3280
    },
    {
      "entropy": 5.77383861541748,
      "epoch": 0.29798621190130625,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004995147065617263,
      "loss": 5.6295,
      "mean_token_accuracy": 0.16005017310380937,
      "num_tokens": 6805456.0,
      "step": 3285
    },
    {
      "entropy": 5.866624546051026,
      "epoch": 0.2984397677793904,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995125802519946,
      "loss": 5.7057,
      "mean_token_accuracy": 0.1569285973906517,
      "num_tokens": 6816215.0,
      "step": 3290
    },
    {
      "entropy": 5.855666875839233,
      "epoch": 0.2988933236574746,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995104492992857,
      "loss": 5.7427,
      "mean_token_accuracy": 0.15623879730701445,
      "num_tokens": 6826545.0,
      "step": 3295
    },
    {
      "entropy": 5.702891540527344,
      "epoch": 0.2993468795355588,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995083137036434,
      "loss": 5.6479,
      "mean_token_accuracy": 0.16189349293708802,
      "num_tokens": 6836184.0,
      "step": 3300
    },
    {
      "entropy": 5.938313961029053,
      "epoch": 0.299800435413643,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995061734651121,
      "loss": 5.7388,
      "mean_token_accuracy": 0.15550311654806137,
      "num_tokens": 6846646.0,
      "step": 3305
    },
    {
      "entropy": 5.832134151458741,
      "epoch": 0.30025399129172714,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995040285837358,
      "loss": 5.6397,
      "mean_token_accuracy": 0.16627662777900695,
      "num_tokens": 6857597.0,
      "step": 3310
    },
    {
      "entropy": 5.757911109924317,
      "epoch": 0.3007075471698113,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995018790595589,
      "loss": 5.7228,
      "mean_token_accuracy": 0.15299862772226333,
      "num_tokens": 6867523.0,
      "step": 3315
    },
    {
      "entropy": 5.863043165206909,
      "epoch": 0.3011611030478955,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994997248926261,
      "loss": 5.8306,
      "mean_token_accuracy": 0.15482011437416077,
      "num_tokens": 6878190.0,
      "step": 3320
    },
    {
      "entropy": 5.831428050994873,
      "epoch": 0.30161465892597966,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994975660829817,
      "loss": 5.6124,
      "mean_token_accuracy": 0.17094137519598007,
      "num_tokens": 6888483.0,
      "step": 3325
    },
    {
      "entropy": 5.726930236816406,
      "epoch": 0.3020682148040639,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994954026306705,
      "loss": 5.611,
      "mean_token_accuracy": 0.16160572692751884,
      "num_tokens": 6898337.0,
      "step": 3330
    },
    {
      "entropy": 5.853848600387574,
      "epoch": 0.30252177068214803,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994932345357371,
      "loss": 5.6829,
      "mean_token_accuracy": 0.1558000400662422,
      "num_tokens": 6909131.0,
      "step": 3335
    },
    {
      "entropy": 5.797689771652221,
      "epoch": 0.30297532656023224,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994910617982264,
      "loss": 5.6578,
      "mean_token_accuracy": 0.15806320160627366,
      "num_tokens": 6919790.0,
      "step": 3340
    },
    {
      "entropy": 5.781959342956543,
      "epoch": 0.3034288824383164,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994888844181833,
      "loss": 5.673,
      "mean_token_accuracy": 0.16309484243392944,
      "num_tokens": 6930072.0,
      "step": 3345
    },
    {
      "entropy": 5.762999105453491,
      "epoch": 0.3038824383164006,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994867023956529,
      "loss": 5.5666,
      "mean_token_accuracy": 0.1615067183971405,
      "num_tokens": 6940646.0,
      "step": 3350
    },
    {
      "entropy": 5.712483549118042,
      "epoch": 0.30433599419448476,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994845157306803,
      "loss": 5.5689,
      "mean_token_accuracy": 0.16492611765861512,
      "num_tokens": 6951645.0,
      "step": 3355
    },
    {
      "entropy": 5.875040483474732,
      "epoch": 0.3047895500725689,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994823244233107,
      "loss": 5.7363,
      "mean_token_accuracy": 0.157509508728981,
      "num_tokens": 6963020.0,
      "step": 3360
    },
    {
      "entropy": 5.842069149017334,
      "epoch": 0.3052431059506531,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994801284735895,
      "loss": 5.6612,
      "mean_token_accuracy": 0.16126682311296464,
      "num_tokens": 6972895.0,
      "step": 3365
    },
    {
      "entropy": 5.881510400772095,
      "epoch": 0.3056966618287373,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004994779278815619,
      "loss": 5.8137,
      "mean_token_accuracy": 0.14684042036533357,
      "num_tokens": 6985363.0,
      "step": 3370
    },
    {
      "entropy": 5.853707885742187,
      "epoch": 0.3061502177068215,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994757226472736,
      "loss": 5.7436,
      "mean_token_accuracy": 0.15787117779254914,
      "num_tokens": 6997040.0,
      "step": 3375
    },
    {
      "entropy": 5.848391389846801,
      "epoch": 0.30660377358490565,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994735127707702,
      "loss": 5.6733,
      "mean_token_accuracy": 0.16261067390441894,
      "num_tokens": 7007591.0,
      "step": 3380
    },
    {
      "entropy": 5.860173940658569,
      "epoch": 0.30705732946298986,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994712982520972,
      "loss": 5.6754,
      "mean_token_accuracy": 0.16760361343622207,
      "num_tokens": 7017858.0,
      "step": 3385
    },
    {
      "entropy": 5.727178287506104,
      "epoch": 0.307510885341074,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994690790913008,
      "loss": 5.6346,
      "mean_token_accuracy": 0.1600530058145523,
      "num_tokens": 7027365.0,
      "step": 3390
    },
    {
      "entropy": 5.76168851852417,
      "epoch": 0.3079644412191582,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994668552884265,
      "loss": 5.6351,
      "mean_token_accuracy": 0.1617230698466301,
      "num_tokens": 7038728.0,
      "step": 3395
    },
    {
      "entropy": 5.881175231933594,
      "epoch": 0.3084179970972424,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994646268435204,
      "loss": 5.5729,
      "mean_token_accuracy": 0.16643185764551163,
      "num_tokens": 7048736.0,
      "step": 3400
    },
    {
      "entropy": 5.75899338722229,
      "epoch": 0.30887155297532654,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994623937566286,
      "loss": 5.7002,
      "mean_token_accuracy": 0.1575309082865715,
      "num_tokens": 7059205.0,
      "step": 3405
    },
    {
      "entropy": 5.799173402786255,
      "epoch": 0.30932510885341075,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994601560277974,
      "loss": 5.6896,
      "mean_token_accuracy": 0.15942423939704894,
      "num_tokens": 7069513.0,
      "step": 3410
    },
    {
      "entropy": 5.790976524353027,
      "epoch": 0.3097786647314949,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499457913657073,
      "loss": 5.6225,
      "mean_token_accuracy": 0.15869100689888,
      "num_tokens": 7079940.0,
      "step": 3415
    },
    {
      "entropy": 5.831319952011109,
      "epoch": 0.3102322206095791,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994556666445016,
      "loss": 5.6193,
      "mean_token_accuracy": 0.16156373918056488,
      "num_tokens": 7090792.0,
      "step": 3420
    },
    {
      "entropy": 5.755025625228882,
      "epoch": 0.31068577648766327,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994534149901299,
      "loss": 5.577,
      "mean_token_accuracy": 0.16105773597955703,
      "num_tokens": 7101964.0,
      "step": 3425
    },
    {
      "entropy": 5.7506568908691404,
      "epoch": 0.3111393323657475,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994511586940044,
      "loss": 5.6442,
      "mean_token_accuracy": 0.15435996800661086,
      "num_tokens": 7113332.0,
      "step": 3430
    },
    {
      "entropy": 5.91636700630188,
      "epoch": 0.31159288824383163,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994488977561717,
      "loss": 5.7921,
      "mean_token_accuracy": 0.15468702465295792,
      "num_tokens": 7123481.0,
      "step": 3435
    },
    {
      "entropy": 5.73750057220459,
      "epoch": 0.31204644412191584,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994466321766786,
      "loss": 5.5585,
      "mean_token_accuracy": 0.1646214559674263,
      "num_tokens": 7133694.0,
      "step": 3440
    },
    {
      "entropy": 5.804030513763427,
      "epoch": 0.3125,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499444361955572,
      "loss": 5.6664,
      "mean_token_accuracy": 0.1619983956217766,
      "num_tokens": 7144491.0,
      "step": 3445
    },
    {
      "entropy": 5.945888328552246,
      "epoch": 0.31295355587808416,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994420870928988,
      "loss": 5.76,
      "mean_token_accuracy": 0.15487642139196395,
      "num_tokens": 7154716.0,
      "step": 3450
    },
    {
      "entropy": 5.8428285121917725,
      "epoch": 0.31340711175616837,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994398075887059,
      "loss": 5.7553,
      "mean_token_accuracy": 0.15003324151039124,
      "num_tokens": 7164258.0,
      "step": 3455
    },
    {
      "entropy": 5.82437686920166,
      "epoch": 0.3138606676342525,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994375234430407,
      "loss": 5.6108,
      "mean_token_accuracy": 0.15920650959014893,
      "num_tokens": 7175098.0,
      "step": 3460
    },
    {
      "entropy": 5.79690523147583,
      "epoch": 0.31431422351233673,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994352346559504,
      "loss": 5.6851,
      "mean_token_accuracy": 0.15969553738832473,
      "num_tokens": 7186480.0,
      "step": 3465
    },
    {
      "entropy": 5.710554122924805,
      "epoch": 0.3147677793904209,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994329412274821,
      "loss": 5.6593,
      "mean_token_accuracy": 0.16162224411964415,
      "num_tokens": 7197056.0,
      "step": 3470
    },
    {
      "entropy": 5.846120882034302,
      "epoch": 0.3152213352685051,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994306431576835,
      "loss": 5.7725,
      "mean_token_accuracy": 0.15470817685127258,
      "num_tokens": 7206040.0,
      "step": 3475
    },
    {
      "entropy": 5.931183242797852,
      "epoch": 0.31567489114658925,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499428340446602,
      "loss": 5.6142,
      "mean_token_accuracy": 0.16317909955978394,
      "num_tokens": 7216344.0,
      "step": 3480
    },
    {
      "entropy": 5.784151315689087,
      "epoch": 0.31612844702467346,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994260330942851,
      "loss": 5.696,
      "mean_token_accuracy": 0.15416476279497146,
      "num_tokens": 7226729.0,
      "step": 3485
    },
    {
      "entropy": 5.753327083587647,
      "epoch": 0.3165820029027576,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994237211007808,
      "loss": 5.6509,
      "mean_token_accuracy": 0.1596106082201004,
      "num_tokens": 7237616.0,
      "step": 3490
    },
    {
      "entropy": 5.826973915100098,
      "epoch": 0.3170355587808418,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994214044661367,
      "loss": 5.5765,
      "mean_token_accuracy": 0.16474129110574723,
      "num_tokens": 7247857.0,
      "step": 3495
    },
    {
      "entropy": 5.698789978027344,
      "epoch": 0.317489114658926,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994190831904007,
      "loss": 5.5503,
      "mean_token_accuracy": 0.16544652432203294,
      "num_tokens": 7257858.0,
      "step": 3500
    },
    {
      "entropy": 5.665239334106445,
      "epoch": 0.31794267053701014,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499416757273621,
      "loss": 5.5582,
      "mean_token_accuracy": 0.16463755071163177,
      "num_tokens": 7268440.0,
      "step": 3505
    },
    {
      "entropy": 5.83346905708313,
      "epoch": 0.31839622641509435,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994144267158454,
      "loss": 5.6371,
      "mean_token_accuracy": 0.16650597602128983,
      "num_tokens": 7278560.0,
      "step": 3510
    },
    {
      "entropy": 5.785221576690674,
      "epoch": 0.3188497822931785,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994120915171224,
      "loss": 5.6364,
      "mean_token_accuracy": 0.16446116715669631,
      "num_tokens": 7288422.0,
      "step": 3515
    },
    {
      "entropy": 5.840118741989135,
      "epoch": 0.3193033381712627,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994097516775002,
      "loss": 5.7106,
      "mean_token_accuracy": 0.1576638028025627,
      "num_tokens": 7298613.0,
      "step": 3520
    },
    {
      "entropy": 5.828368854522705,
      "epoch": 0.3197568940493469,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994074071970271,
      "loss": 5.7422,
      "mean_token_accuracy": 0.15291873514652252,
      "num_tokens": 7309509.0,
      "step": 3525
    },
    {
      "entropy": 5.860109329223633,
      "epoch": 0.3202104499274311,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994050580757515,
      "loss": 5.6445,
      "mean_token_accuracy": 0.15940439999103545,
      "num_tokens": 7319015.0,
      "step": 3530
    },
    {
      "entropy": 5.721933031082154,
      "epoch": 0.32066400580551524,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994027043137223,
      "loss": 5.61,
      "mean_token_accuracy": 0.1672150120139122,
      "num_tokens": 7329026.0,
      "step": 3535
    },
    {
      "entropy": 5.855372905731201,
      "epoch": 0.3211175616835994,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004994003459109879,
      "loss": 5.7137,
      "mean_token_accuracy": 0.16032420545816423,
      "num_tokens": 7339638.0,
      "step": 3540
    },
    {
      "entropy": 5.837479209899902,
      "epoch": 0.3215711175616836,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993979828675972,
      "loss": 5.624,
      "mean_token_accuracy": 0.16279072165489197,
      "num_tokens": 7349979.0,
      "step": 3545
    },
    {
      "entropy": 5.702495336532593,
      "epoch": 0.32202467343976776,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499395615183599,
      "loss": 5.5517,
      "mean_token_accuracy": 0.17009643316268921,
      "num_tokens": 7359398.0,
      "step": 3550
    },
    {
      "entropy": 5.7116045475006105,
      "epoch": 0.32247822931785197,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993932428590422,
      "loss": 5.5945,
      "mean_token_accuracy": 0.15939887911081313,
      "num_tokens": 7370418.0,
      "step": 3555
    },
    {
      "entropy": 5.781202697753907,
      "epoch": 0.3229317851959361,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993908658939762,
      "loss": 5.6239,
      "mean_token_accuracy": 0.15674008578062057,
      "num_tokens": 7380461.0,
      "step": 3560
    },
    {
      "entropy": 5.814039325714111,
      "epoch": 0.32338534107402034,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004993884842884496,
      "loss": 5.647,
      "mean_token_accuracy": 0.16048161089420318,
      "num_tokens": 7390646.0,
      "step": 3565
    },
    {
      "entropy": 5.786873054504395,
      "epoch": 0.3238388969521045,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993860980425121,
      "loss": 5.7203,
      "mean_token_accuracy": 0.1618356540799141,
      "num_tokens": 7402106.0,
      "step": 3570
    },
    {
      "entropy": 5.809912443161011,
      "epoch": 0.3242924528301887,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993837071562131,
      "loss": 5.7043,
      "mean_token_accuracy": 0.16131191849708557,
      "num_tokens": 7412936.0,
      "step": 3575
    },
    {
      "entropy": 5.8115393161773685,
      "epoch": 0.32474600870827286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993813116296017,
      "loss": 5.5708,
      "mean_token_accuracy": 0.16675995141267777,
      "num_tokens": 7422303.0,
      "step": 3580
    },
    {
      "entropy": 5.715788507461548,
      "epoch": 0.325199564586357,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993789114627275,
      "loss": 5.6511,
      "mean_token_accuracy": 0.16085287630558015,
      "num_tokens": 7432829.0,
      "step": 3585
    },
    {
      "entropy": 5.937649488449097,
      "epoch": 0.3256531204644412,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993765066556403,
      "loss": 5.6713,
      "mean_token_accuracy": 0.15823337584733962,
      "num_tokens": 7443218.0,
      "step": 3590
    },
    {
      "entropy": 5.773650455474853,
      "epoch": 0.3261066763425254,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004993740972083899,
      "loss": 5.62,
      "mean_token_accuracy": 0.16021817028522492,
      "num_tokens": 7454531.0,
      "step": 3595
    },
    {
      "entropy": 5.790258073806763,
      "epoch": 0.3265602322206096,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993716831210259,
      "loss": 5.6799,
      "mean_token_accuracy": 0.16317007392644883,
      "num_tokens": 7464993.0,
      "step": 3600
    },
    {
      "entropy": 5.729147958755493,
      "epoch": 0.32701378809869375,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993692643935983,
      "loss": 5.7096,
      "mean_token_accuracy": 0.1584396131336689,
      "num_tokens": 7475755.0,
      "step": 3605
    },
    {
      "entropy": 5.771151685714722,
      "epoch": 0.32746734397677796,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993668410261571,
      "loss": 5.6945,
      "mean_token_accuracy": 0.16145843863487244,
      "num_tokens": 7486778.0,
      "step": 3610
    },
    {
      "entropy": 5.907090902328491,
      "epoch": 0.3279208998548621,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993644130187525,
      "loss": 5.6771,
      "mean_token_accuracy": 0.16191897690296173,
      "num_tokens": 7497439.0,
      "step": 3615
    },
    {
      "entropy": 5.734846353530884,
      "epoch": 0.3283744557329463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993619803714347,
      "loss": 5.7075,
      "mean_token_accuracy": 0.15616535544395446,
      "num_tokens": 7508274.0,
      "step": 3620
    },
    {
      "entropy": 5.782626390457153,
      "epoch": 0.3288280116110305,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993595430842539,
      "loss": 5.6874,
      "mean_token_accuracy": 0.16056278795003892,
      "num_tokens": 7519315.0,
      "step": 3625
    },
    {
      "entropy": 5.901585865020752,
      "epoch": 0.32928156748911463,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993571011572606,
      "loss": 5.6808,
      "mean_token_accuracy": 0.1630665123462677,
      "num_tokens": 7529910.0,
      "step": 3630
    },
    {
      "entropy": 5.800387811660767,
      "epoch": 0.32973512336719885,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993546545905053,
      "loss": 5.602,
      "mean_token_accuracy": 0.16489686965942382,
      "num_tokens": 7538966.0,
      "step": 3635
    },
    {
      "entropy": 5.775709533691407,
      "epoch": 0.330188679245283,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993522033840386,
      "loss": 5.6374,
      "mean_token_accuracy": 0.15789183378219604,
      "num_tokens": 7549683.0,
      "step": 3640
    },
    {
      "entropy": 5.700624227523804,
      "epoch": 0.3306422351233672,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499349747537911,
      "loss": 5.6191,
      "mean_token_accuracy": 0.1624668598175049,
      "num_tokens": 7560629.0,
      "step": 3645
    },
    {
      "entropy": 5.806013774871826,
      "epoch": 0.33109579100145137,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993472870521736,
      "loss": 5.5646,
      "mean_token_accuracy": 0.17270171344280244,
      "num_tokens": 7570613.0,
      "step": 3650
    },
    {
      "entropy": 5.79381422996521,
      "epoch": 0.3315493468795356,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993448219268771,
      "loss": 5.5499,
      "mean_token_accuracy": 0.16173658668994903,
      "num_tokens": 7580640.0,
      "step": 3655
    },
    {
      "entropy": 5.763698101043701,
      "epoch": 0.33200290275761973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993423521620726,
      "loss": 5.6381,
      "mean_token_accuracy": 0.16688979268074036,
      "num_tokens": 7590164.0,
      "step": 3660
    },
    {
      "entropy": 5.823714399337769,
      "epoch": 0.33245645863570394,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993398777578109,
      "loss": 5.6959,
      "mean_token_accuracy": 0.16008617281913756,
      "num_tokens": 7600173.0,
      "step": 3665
    },
    {
      "entropy": 5.71161642074585,
      "epoch": 0.3329100145137881,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993373987141435,
      "loss": 5.4841,
      "mean_token_accuracy": 0.17095217257738113,
      "num_tokens": 7609958.0,
      "step": 3670
    },
    {
      "entropy": 5.813071250915527,
      "epoch": 0.33336357039187225,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993349150311214,
      "loss": 5.6411,
      "mean_token_accuracy": 0.157687945663929,
      "num_tokens": 7620237.0,
      "step": 3675
    },
    {
      "entropy": 5.699023199081421,
      "epoch": 0.33381712626995647,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993324267087962,
      "loss": 5.5406,
      "mean_token_accuracy": 0.1682781383395195,
      "num_tokens": 7629886.0,
      "step": 3680
    },
    {
      "entropy": 5.728935241699219,
      "epoch": 0.3342706821480406,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993299337472192,
      "loss": 5.6095,
      "mean_token_accuracy": 0.16654768735170364,
      "num_tokens": 7640057.0,
      "step": 3685
    },
    {
      "entropy": 5.7182149410247805,
      "epoch": 0.33472423802612483,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499327436146442,
      "loss": 5.5532,
      "mean_token_accuracy": 0.16529571264982224,
      "num_tokens": 7649766.0,
      "step": 3690
    },
    {
      "entropy": 5.769032526016235,
      "epoch": 0.335177793904209,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993249339065162,
      "loss": 5.6399,
      "mean_token_accuracy": 0.16574999690055847,
      "num_tokens": 7660890.0,
      "step": 3695
    },
    {
      "entropy": 5.773209142684936,
      "epoch": 0.3356313497822932,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993224270274937,
      "loss": 5.6582,
      "mean_token_accuracy": 0.1607873558998108,
      "num_tokens": 7670737.0,
      "step": 3700
    },
    {
      "entropy": 5.7178552627563475,
      "epoch": 0.33608490566037735,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993199155094263,
      "loss": 5.6067,
      "mean_token_accuracy": 0.16417943835258483,
      "num_tokens": 7681407.0,
      "step": 3705
    },
    {
      "entropy": 5.768005990982056,
      "epoch": 0.33653846153846156,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993173993523658,
      "loss": 5.5356,
      "mean_token_accuracy": 0.17343078404664994,
      "num_tokens": 7691758.0,
      "step": 3710
    },
    {
      "entropy": 5.699445009231567,
      "epoch": 0.3369920174165457,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993148785563643,
      "loss": 5.5679,
      "mean_token_accuracy": 0.16385959088802338,
      "num_tokens": 7702934.0,
      "step": 3715
    },
    {
      "entropy": 5.71638708114624,
      "epoch": 0.3374455732946299,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499312353121474,
      "loss": 5.5705,
      "mean_token_accuracy": 0.16445922404527663,
      "num_tokens": 7713001.0,
      "step": 3720
    },
    {
      "entropy": 5.802838611602783,
      "epoch": 0.3378991291727141,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993098230477471,
      "loss": 5.6211,
      "mean_token_accuracy": 0.1572564959526062,
      "num_tokens": 7724005.0,
      "step": 3725
    },
    {
      "entropy": 5.785116243362427,
      "epoch": 0.33835268505079824,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993072883352358,
      "loss": 5.5802,
      "mean_token_accuracy": 0.16274530738592147,
      "num_tokens": 7733743.0,
      "step": 3730
    },
    {
      "entropy": 5.802304887771607,
      "epoch": 0.33880624092888245,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993047489839927,
      "loss": 5.7695,
      "mean_token_accuracy": 0.1579284265637398,
      "num_tokens": 7743762.0,
      "step": 3735
    },
    {
      "entropy": 5.850232982635498,
      "epoch": 0.3392597968069666,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004993022049940702,
      "loss": 5.655,
      "mean_token_accuracy": 0.16184027045965194,
      "num_tokens": 7753758.0,
      "step": 3740
    },
    {
      "entropy": 5.739925861358643,
      "epoch": 0.3397133526850508,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992996563655209,
      "loss": 5.6052,
      "mean_token_accuracy": 0.163077212870121,
      "num_tokens": 7764692.0,
      "step": 3745
    },
    {
      "entropy": 5.742031097412109,
      "epoch": 0.340166908563135,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992971030983975,
      "loss": 5.644,
      "mean_token_accuracy": 0.15631235092878343,
      "num_tokens": 7776892.0,
      "step": 3750
    },
    {
      "entropy": 5.875276947021485,
      "epoch": 0.3406204644412192,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992945451927531,
      "loss": 5.6769,
      "mean_token_accuracy": 0.16058121621608734,
      "num_tokens": 7787143.0,
      "step": 3755
    },
    {
      "entropy": 5.723575305938721,
      "epoch": 0.34107402031930334,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992919826486402,
      "loss": 5.5967,
      "mean_token_accuracy": 0.16272515058517456,
      "num_tokens": 7798537.0,
      "step": 3760
    },
    {
      "entropy": 5.713321161270142,
      "epoch": 0.3415275761973875,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992894154661118,
      "loss": 5.5709,
      "mean_token_accuracy": 0.16793359965085983,
      "num_tokens": 7808434.0,
      "step": 3765
    },
    {
      "entropy": 5.79578595161438,
      "epoch": 0.3419811320754717,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992868436452213,
      "loss": 5.6818,
      "mean_token_accuracy": 0.16078411191701888,
      "num_tokens": 7819592.0,
      "step": 3770
    },
    {
      "entropy": 5.809304237365723,
      "epoch": 0.34243468795355586,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992842671860217,
      "loss": 5.5643,
      "mean_token_accuracy": 0.16715734750032424,
      "num_tokens": 7830602.0,
      "step": 3775
    },
    {
      "entropy": 5.786276483535767,
      "epoch": 0.34288824383164007,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992816860885662,
      "loss": 5.7212,
      "mean_token_accuracy": 0.15924838036298752,
      "num_tokens": 7840850.0,
      "step": 3780
    },
    {
      "entropy": 5.884521865844727,
      "epoch": 0.3433417997097242,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992791003529084,
      "loss": 5.6558,
      "mean_token_accuracy": 0.15685309320688248,
      "num_tokens": 7852371.0,
      "step": 3785
    },
    {
      "entropy": 5.715097904205322,
      "epoch": 0.34379535558780844,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992765099791017,
      "loss": 5.6818,
      "mean_token_accuracy": 0.15666263401508332,
      "num_tokens": 7863290.0,
      "step": 3790
    },
    {
      "entropy": 5.71993088722229,
      "epoch": 0.3442489114658926,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992739149671995,
      "loss": 5.509,
      "mean_token_accuracy": 0.16685899645090102,
      "num_tokens": 7873927.0,
      "step": 3795
    },
    {
      "entropy": 5.733199405670166,
      "epoch": 0.3447024673439768,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992713153172556,
      "loss": 5.5502,
      "mean_token_accuracy": 0.16382715702056885,
      "num_tokens": 7884977.0,
      "step": 3800
    },
    {
      "entropy": 5.699240398406983,
      "epoch": 0.34515602322206096,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004992687110293238,
      "loss": 5.6304,
      "mean_token_accuracy": 0.16769205555319786,
      "num_tokens": 7894620.0,
      "step": 3805
    },
    {
      "entropy": 5.719654321670532,
      "epoch": 0.3456095791001451,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992661021034579,
      "loss": 5.5507,
      "mean_token_accuracy": 0.16906091570854187,
      "num_tokens": 7904541.0,
      "step": 3810
    },
    {
      "entropy": 5.726424789428711,
      "epoch": 0.3460631349782293,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004992634885397118,
      "loss": 5.5794,
      "mean_token_accuracy": 0.1659620225429535,
      "num_tokens": 7915435.0,
      "step": 3815
    },
    {
      "entropy": 5.722615718841553,
      "epoch": 0.3465166908563135,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992608703381396,
      "loss": 5.5616,
      "mean_token_accuracy": 0.16741435825824738,
      "num_tokens": 7925870.0,
      "step": 3820
    },
    {
      "entropy": 5.722995281219482,
      "epoch": 0.3469702467343977,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992582474987955,
      "loss": 5.5639,
      "mean_token_accuracy": 0.16274259686470033,
      "num_tokens": 7937046.0,
      "step": 3825
    },
    {
      "entropy": 5.75324854850769,
      "epoch": 0.34742380261248185,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004992556200217337,
      "loss": 5.5261,
      "mean_token_accuracy": 0.1641123980283737,
      "num_tokens": 7947150.0,
      "step": 3830
    },
    {
      "entropy": 5.795271539688111,
      "epoch": 0.34787735849056606,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992529879070085,
      "loss": 5.618,
      "mean_token_accuracy": 0.16771930009126662,
      "num_tokens": 7956815.0,
      "step": 3835
    },
    {
      "entropy": 5.789908409118652,
      "epoch": 0.3483309143686502,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992503511546745,
      "loss": 5.6657,
      "mean_token_accuracy": 0.15945106595754624,
      "num_tokens": 7968114.0,
      "step": 3840
    },
    {
      "entropy": 5.748790836334228,
      "epoch": 0.3487844702467344,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992477097647859,
      "loss": 5.5656,
      "mean_token_accuracy": 0.1627722293138504,
      "num_tokens": 7977840.0,
      "step": 3845
    },
    {
      "entropy": 5.814347887039185,
      "epoch": 0.3492380261248186,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992450637373976,
      "loss": 5.6907,
      "mean_token_accuracy": 0.16484753638505936,
      "num_tokens": 7988663.0,
      "step": 3850
    },
    {
      "entropy": 5.681796550750732,
      "epoch": 0.34969158200290273,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992424130725642,
      "loss": 5.4999,
      "mean_token_accuracy": 0.16489350348711013,
      "num_tokens": 7999693.0,
      "step": 3855
    },
    {
      "entropy": 5.772833013534546,
      "epoch": 0.35014513788098695,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004992397577703406,
      "loss": 5.6747,
      "mean_token_accuracy": 0.1532568469643593,
      "num_tokens": 8010154.0,
      "step": 3860
    },
    {
      "entropy": 5.825438737869263,
      "epoch": 0.3505986937590711,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992370978307816,
      "loss": 5.6536,
      "mean_token_accuracy": 0.1622571349143982,
      "num_tokens": 8020921.0,
      "step": 3865
    },
    {
      "entropy": 5.764519309997558,
      "epoch": 0.3510522496371553,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992344332539424,
      "loss": 5.5811,
      "mean_token_accuracy": 0.16872124671936034,
      "num_tokens": 8031404.0,
      "step": 3870
    },
    {
      "entropy": 5.684741640090943,
      "epoch": 0.35150580551523947,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992317640398779,
      "loss": 5.585,
      "mean_token_accuracy": 0.1686838299036026,
      "num_tokens": 8041310.0,
      "step": 3875
    },
    {
      "entropy": 5.710890054702759,
      "epoch": 0.3519593613933237,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992290901886434,
      "loss": 5.57,
      "mean_token_accuracy": 0.1684218779206276,
      "num_tokens": 8051054.0,
      "step": 3880
    },
    {
      "entropy": 5.744958877563477,
      "epoch": 0.35241291727140783,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499226411700294,
      "loss": 5.5578,
      "mean_token_accuracy": 0.16676690876483918,
      "num_tokens": 8061063.0,
      "step": 3885
    },
    {
      "entropy": 5.69763445854187,
      "epoch": 0.35286647314949204,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992237285748855,
      "loss": 5.5841,
      "mean_token_accuracy": 0.16478407829999925,
      "num_tokens": 8071715.0,
      "step": 3890
    },
    {
      "entropy": 5.80994873046875,
      "epoch": 0.3533200290275762,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992210408124731,
      "loss": 5.6496,
      "mean_token_accuracy": 0.15725049376487732,
      "num_tokens": 8082679.0,
      "step": 3895
    },
    {
      "entropy": 5.752315807342529,
      "epoch": 0.35377358490566035,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004992183484131123,
      "loss": 5.6808,
      "mean_token_accuracy": 0.16136541217565536,
      "num_tokens": 8093785.0,
      "step": 3900
    },
    {
      "entropy": 5.71539158821106,
      "epoch": 0.35422714078374457,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992156513768591,
      "loss": 5.5685,
      "mean_token_accuracy": 0.16779497861862183,
      "num_tokens": 8105822.0,
      "step": 3905
    },
    {
      "entropy": 5.709498691558838,
      "epoch": 0.3546806966618287,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499212949703769,
      "loss": 5.486,
      "mean_token_accuracy": 0.1679369956254959,
      "num_tokens": 8115521.0,
      "step": 3910
    },
    {
      "entropy": 5.726305246353149,
      "epoch": 0.35513425253991293,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992102433938981,
      "loss": 5.5783,
      "mean_token_accuracy": 0.16780869513750077,
      "num_tokens": 8125524.0,
      "step": 3915
    },
    {
      "entropy": 5.712926054000855,
      "epoch": 0.3555878084179971,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992075324473021,
      "loss": 5.482,
      "mean_token_accuracy": 0.17744722813367844,
      "num_tokens": 8135552.0,
      "step": 3920
    },
    {
      "entropy": 5.780481147766113,
      "epoch": 0.3560413642960813,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992048168640371,
      "loss": 5.6135,
      "mean_token_accuracy": 0.16508189737796783,
      "num_tokens": 8145617.0,
      "step": 3925
    },
    {
      "entropy": 5.725088024139405,
      "epoch": 0.35649492017416545,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004992020966441595,
      "loss": 5.6251,
      "mean_token_accuracy": 0.16034330129623414,
      "num_tokens": 8156729.0,
      "step": 3930
    },
    {
      "entropy": 5.764310026168824,
      "epoch": 0.35694847605224966,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991993717877254,
      "loss": 5.6076,
      "mean_token_accuracy": 0.16174564361572266,
      "num_tokens": 8167637.0,
      "step": 3935
    },
    {
      "entropy": 5.720431089401245,
      "epoch": 0.3574020319303338,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991966422947911,
      "loss": 5.5852,
      "mean_token_accuracy": 0.1702892929315567,
      "num_tokens": 8177354.0,
      "step": 3940
    },
    {
      "entropy": 5.718207550048828,
      "epoch": 0.357855587808418,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004991939081654132,
      "loss": 5.5502,
      "mean_token_accuracy": 0.1746809646487236,
      "num_tokens": 8187580.0,
      "step": 3945
    },
    {
      "entropy": 5.708486032485962,
      "epoch": 0.3583091436865022,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499191169399648,
      "loss": 5.6497,
      "mean_token_accuracy": 0.1604743108153343,
      "num_tokens": 8199382.0,
      "step": 3950
    },
    {
      "entropy": 5.712984323501587,
      "epoch": 0.35876269956458634,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991884259975525,
      "loss": 5.5404,
      "mean_token_accuracy": 0.16475948542356492,
      "num_tokens": 8208484.0,
      "step": 3955
    },
    {
      "entropy": 5.727371263504028,
      "epoch": 0.35921625544267055,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499185677959183,
      "loss": 5.5838,
      "mean_token_accuracy": 0.16461657136678695,
      "num_tokens": 8219056.0,
      "step": 3960
    },
    {
      "entropy": 5.731614685058593,
      "epoch": 0.3596698113207547,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991829252845966,
      "loss": 5.6047,
      "mean_token_accuracy": 0.16140197962522507,
      "num_tokens": 8230149.0,
      "step": 3965
    },
    {
      "entropy": 5.7445838928222654,
      "epoch": 0.3601233671988389,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991801679738502,
      "loss": 5.6652,
      "mean_token_accuracy": 0.15284715443849564,
      "num_tokens": 8240803.0,
      "step": 3970
    },
    {
      "entropy": 5.766950941085815,
      "epoch": 0.3605769230769231,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991774060270008,
      "loss": 5.6201,
      "mean_token_accuracy": 0.15810701847076417,
      "num_tokens": 8251563.0,
      "step": 3975
    },
    {
      "entropy": 5.701673126220703,
      "epoch": 0.3610304789550073,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991746394441055,
      "loss": 5.5522,
      "mean_token_accuracy": 0.16369420439004898,
      "num_tokens": 8261342.0,
      "step": 3980
    },
    {
      "entropy": 5.753665637969971,
      "epoch": 0.36148403483309144,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991718682252216,
      "loss": 5.6681,
      "mean_token_accuracy": 0.15975360572338104,
      "num_tokens": 8271540.0,
      "step": 3985
    },
    {
      "entropy": 5.782139301300049,
      "epoch": 0.3619375907111756,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991690923704063,
      "loss": 5.58,
      "mean_token_accuracy": 0.16486785262823106,
      "num_tokens": 8280593.0,
      "step": 3990
    },
    {
      "entropy": 5.742940139770508,
      "epoch": 0.3623911465892598,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991663118797171,
      "loss": 5.6104,
      "mean_token_accuracy": 0.16217703521251678,
      "num_tokens": 8290126.0,
      "step": 3995
    },
    {
      "entropy": 5.747898292541504,
      "epoch": 0.36284470246734396,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991635267532113,
      "loss": 5.5394,
      "mean_token_accuracy": 0.16359646618366241,
      "num_tokens": 8301416.0,
      "step": 4000
    },
    {
      "entropy": 5.775230884552002,
      "epoch": 0.36329825834542817,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991607369909468,
      "loss": 5.6135,
      "mean_token_accuracy": 0.1605329006910324,
      "num_tokens": 8311034.0,
      "step": 4005
    },
    {
      "entropy": 5.757994508743286,
      "epoch": 0.3637518142235123,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991579425929811,
      "loss": 5.6266,
      "mean_token_accuracy": 0.16281366795301438,
      "num_tokens": 8322203.0,
      "step": 4010
    },
    {
      "entropy": 5.7276918411254885,
      "epoch": 0.36420537010159654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499155143559372,
      "loss": 5.5117,
      "mean_token_accuracy": 0.16838518381118775,
      "num_tokens": 8330981.0,
      "step": 4015
    },
    {
      "entropy": 5.655632972717285,
      "epoch": 0.3646589259796807,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004991523398901773,
      "loss": 5.5437,
      "mean_token_accuracy": 0.16577242016792298,
      "num_tokens": 8341294.0,
      "step": 4020
    },
    {
      "entropy": 5.8728711128234865,
      "epoch": 0.3651124818577649,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991495315854553,
      "loss": 5.6921,
      "mean_token_accuracy": 0.15758731961250305,
      "num_tokens": 8351824.0,
      "step": 4025
    },
    {
      "entropy": 5.734429454803466,
      "epoch": 0.36556603773584906,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991467186452637,
      "loss": 5.5595,
      "mean_token_accuracy": 0.16328836232423782,
      "num_tokens": 8361975.0,
      "step": 4030
    },
    {
      "entropy": 5.646515369415283,
      "epoch": 0.3660195936139332,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991439010696609,
      "loss": 5.5053,
      "mean_token_accuracy": 0.16924880295991898,
      "num_tokens": 8372145.0,
      "step": 4035
    },
    {
      "entropy": 5.743934106826782,
      "epoch": 0.3664731494920174,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991410788587051,
      "loss": 5.6034,
      "mean_token_accuracy": 0.1616530880331993,
      "num_tokens": 8382542.0,
      "step": 4040
    },
    {
      "entropy": 5.771177625656128,
      "epoch": 0.3669267053701016,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991382520124547,
      "loss": 5.5942,
      "mean_token_accuracy": 0.16173005402088164,
      "num_tokens": 8393450.0,
      "step": 4045
    },
    {
      "entropy": 5.7502647876739506,
      "epoch": 0.3673802612481858,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004991354205309681,
      "loss": 5.5407,
      "mean_token_accuracy": 0.1637318655848503,
      "num_tokens": 8404929.0,
      "step": 4050
    },
    {
      "entropy": 5.762520360946655,
      "epoch": 0.36783381712626995,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499132584414304,
      "loss": 5.5531,
      "mean_token_accuracy": 0.1654103085398674,
      "num_tokens": 8415703.0,
      "step": 4055
    },
    {
      "entropy": 5.728594827651977,
      "epoch": 0.36828737300435416,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991297436625207,
      "loss": 5.6347,
      "mean_token_accuracy": 0.16382609456777572,
      "num_tokens": 8426220.0,
      "step": 4060
    },
    {
      "entropy": 5.665254354476929,
      "epoch": 0.3687409288824383,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991268982756774,
      "loss": 5.5135,
      "mean_token_accuracy": 0.16983363032341003,
      "num_tokens": 8436220.0,
      "step": 4065
    },
    {
      "entropy": 5.824220180511475,
      "epoch": 0.3691944847605225,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004991240482538327,
      "loss": 5.6576,
      "mean_token_accuracy": 0.1528759405016899,
      "num_tokens": 8446724.0,
      "step": 4070
    },
    {
      "entropy": 5.739574432373047,
      "epoch": 0.3696480406386067,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991211935970455,
      "loss": 5.5215,
      "mean_token_accuracy": 0.16571784317493438,
      "num_tokens": 8457270.0,
      "step": 4075
    },
    {
      "entropy": 5.698843717575073,
      "epoch": 0.37010159651669083,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991183343053748,
      "loss": 5.5808,
      "mean_token_accuracy": 0.16122987121343613,
      "num_tokens": 8468137.0,
      "step": 4080
    },
    {
      "entropy": 5.737043762207032,
      "epoch": 0.37055515239477504,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049911547037888,
      "loss": 5.5207,
      "mean_token_accuracy": 0.16846096962690355,
      "num_tokens": 8477574.0,
      "step": 4085
    },
    {
      "entropy": 5.718803071975708,
      "epoch": 0.3710087082728592,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991126018176202,
      "loss": 5.6211,
      "mean_token_accuracy": 0.1616816744208336,
      "num_tokens": 8487268.0,
      "step": 4090
    },
    {
      "entropy": 5.8145722389221195,
      "epoch": 0.3714622641509434,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991097286216545,
      "loss": 5.6662,
      "mean_token_accuracy": 0.168794247508049,
      "num_tokens": 8497785.0,
      "step": 4095
    },
    {
      "entropy": 5.753116083145142,
      "epoch": 0.37191582002902757,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991068507910427,
      "loss": 5.5533,
      "mean_token_accuracy": 0.16418525129556655,
      "num_tokens": 8508169.0,
      "step": 4100
    },
    {
      "entropy": 5.690785503387451,
      "epoch": 0.3723693759071118,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499103968325844,
      "loss": 5.522,
      "mean_token_accuracy": 0.16396052837371827,
      "num_tokens": 8519201.0,
      "step": 4105
    },
    {
      "entropy": 5.7371025562286375,
      "epoch": 0.37282293178519593,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991010812261181,
      "loss": 5.6384,
      "mean_token_accuracy": 0.1556764304637909,
      "num_tokens": 8531093.0,
      "step": 4110
    },
    {
      "entropy": 5.665953826904297,
      "epoch": 0.37327648766328014,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990981894919248,
      "loss": 5.4913,
      "mean_token_accuracy": 0.16958004534244536,
      "num_tokens": 8541730.0,
      "step": 4115
    },
    {
      "entropy": 5.693306016921997,
      "epoch": 0.3737300435413643,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990952931233238,
      "loss": 5.5646,
      "mean_token_accuracy": 0.16480782628059387,
      "num_tokens": 8552570.0,
      "step": 4120
    },
    {
      "entropy": 5.754336500167847,
      "epoch": 0.37418359941944845,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499092392120375,
      "loss": 5.4628,
      "mean_token_accuracy": 0.17125755399465561,
      "num_tokens": 8562340.0,
      "step": 4125
    },
    {
      "entropy": 5.735305070877075,
      "epoch": 0.37463715529753266,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004990894864831385,
      "loss": 5.6707,
      "mean_token_accuracy": 0.15796474665403365,
      "num_tokens": 8573240.0,
      "step": 4130
    },
    {
      "entropy": 5.769050550460816,
      "epoch": 0.3750907111756168,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990865762116743,
      "loss": 5.5299,
      "mean_token_accuracy": 0.16909209936857222,
      "num_tokens": 8583194.0,
      "step": 4135
    },
    {
      "entropy": 5.719068336486816,
      "epoch": 0.37554426705370103,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990836613060426,
      "loss": 5.5962,
      "mean_token_accuracy": 0.1610783204436302,
      "num_tokens": 8593603.0,
      "step": 4140
    },
    {
      "entropy": 5.664818382263183,
      "epoch": 0.3759978229317852,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990807417663036,
      "loss": 5.5998,
      "mean_token_accuracy": 0.16227242201566697,
      "num_tokens": 8603176.0,
      "step": 4145
    },
    {
      "entropy": 5.7971210956573485,
      "epoch": 0.3764513788098694,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004990778175925179,
      "loss": 5.6154,
      "mean_token_accuracy": 0.16660853177309037,
      "num_tokens": 8614801.0,
      "step": 4150
    },
    {
      "entropy": 5.768546104431152,
      "epoch": 0.37690493468795355,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004990748887847457,
      "loss": 5.6418,
      "mean_token_accuracy": 0.16194847822189332,
      "num_tokens": 8624999.0,
      "step": 4155
    },
    {
      "entropy": 5.667234706878662,
      "epoch": 0.37735849056603776,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990719553430477,
      "loss": 5.4948,
      "mean_token_accuracy": 0.17510903775691986,
      "num_tokens": 8635358.0,
      "step": 4160
    },
    {
      "entropy": 5.749764490127563,
      "epoch": 0.3778120464441219,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990690172674847,
      "loss": 5.5679,
      "mean_token_accuracy": 0.1612323611974716,
      "num_tokens": 8645430.0,
      "step": 4165
    },
    {
      "entropy": 5.726269006729126,
      "epoch": 0.3782656023222061,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990660745581172,
      "loss": 5.5234,
      "mean_token_accuracy": 0.17513987571001052,
      "num_tokens": 8655115.0,
      "step": 4170
    },
    {
      "entropy": 5.6731971263885494,
      "epoch": 0.3787191582002903,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004990631272150062,
      "loss": 5.5279,
      "mean_token_accuracy": 0.17234957814216614,
      "num_tokens": 8665391.0,
      "step": 4175
    },
    {
      "entropy": 5.764857435226441,
      "epoch": 0.37917271407837444,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004990601752382127,
      "loss": 5.6196,
      "mean_token_accuracy": 0.1626276418566704,
      "num_tokens": 8676819.0,
      "step": 4180
    },
    {
      "entropy": 5.78800048828125,
      "epoch": 0.37962626995645865,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004990572186277975,
      "loss": 5.5433,
      "mean_token_accuracy": 0.16920216530561447,
      "num_tokens": 8687052.0,
      "step": 4185
    },
    {
      "entropy": 5.716868305206299,
      "epoch": 0.3800798258345428,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990542573838221,
      "loss": 5.6207,
      "mean_token_accuracy": 0.16125022917985915,
      "num_tokens": 8697756.0,
      "step": 4190
    },
    {
      "entropy": 5.616197013854981,
      "epoch": 0.380533381712627,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990512915063475,
      "loss": 5.5242,
      "mean_token_accuracy": 0.17394591718912125,
      "num_tokens": 8708388.0,
      "step": 4195
    },
    {
      "entropy": 5.806944036483765,
      "epoch": 0.38098693759071117,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990483209954352,
      "loss": 5.6288,
      "mean_token_accuracy": 0.15857603251934052,
      "num_tokens": 8719475.0,
      "step": 4200
    },
    {
      "entropy": 5.71453013420105,
      "epoch": 0.3814404934687954,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004990453458511464,
      "loss": 5.477,
      "mean_token_accuracy": 0.17523704022169112,
      "num_tokens": 8729395.0,
      "step": 4205
    },
    {
      "entropy": 5.69320387840271,
      "epoch": 0.38189404934687954,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990423660735428,
      "loss": 5.5367,
      "mean_token_accuracy": 0.17120707631111146,
      "num_tokens": 8740351.0,
      "step": 4210
    },
    {
      "entropy": 5.747463369369507,
      "epoch": 0.3823476052249637,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499039381662686,
      "loss": 5.5519,
      "mean_token_accuracy": 0.16874152719974517,
      "num_tokens": 8749260.0,
      "step": 4215
    },
    {
      "entropy": 5.784005641937256,
      "epoch": 0.3828011611030479,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990363926186377,
      "loss": 5.6472,
      "mean_token_accuracy": 0.1575682356953621,
      "num_tokens": 8759604.0,
      "step": 4220
    },
    {
      "entropy": 5.778999853134155,
      "epoch": 0.38325471698113206,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990333989414597,
      "loss": 5.6582,
      "mean_token_accuracy": 0.1573771618306637,
      "num_tokens": 8769815.0,
      "step": 4225
    },
    {
      "entropy": 5.677911949157715,
      "epoch": 0.38370827285921627,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990304006312138,
      "loss": 5.5816,
      "mean_token_accuracy": 0.16669487804174424,
      "num_tokens": 8780557.0,
      "step": 4230
    },
    {
      "entropy": 5.731473064422607,
      "epoch": 0.3841618287373004,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990273976879622,
      "loss": 5.5429,
      "mean_token_accuracy": 0.17356031835079194,
      "num_tokens": 8790993.0,
      "step": 4235
    },
    {
      "entropy": 5.731623315811158,
      "epoch": 0.38461538461538464,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990243901117669,
      "loss": 5.6024,
      "mean_token_accuracy": 0.15951776504516602,
      "num_tokens": 8800657.0,
      "step": 4240
    },
    {
      "entropy": 5.657181215286255,
      "epoch": 0.3850689404934688,
      "grad_norm": 0.875,
      "learning_rate": 0.0004990213779026903,
      "loss": 5.5996,
      "mean_token_accuracy": 0.1628565177321434,
      "num_tokens": 8812983.0,
      "step": 4245
    },
    {
      "entropy": 5.771100807189941,
      "epoch": 0.385522496371553,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990183610607944,
      "loss": 5.571,
      "mean_token_accuracy": 0.1698968768119812,
      "num_tokens": 8822980.0,
      "step": 4250
    },
    {
      "entropy": 5.677217197418213,
      "epoch": 0.38597605224963716,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990153395861415,
      "loss": 5.5484,
      "mean_token_accuracy": 0.16861167252063752,
      "num_tokens": 8833821.0,
      "step": 4255
    },
    {
      "entropy": 5.625637340545654,
      "epoch": 0.3864296081277213,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004990123134787946,
      "loss": 5.4898,
      "mean_token_accuracy": 0.17129137367010117,
      "num_tokens": 8843929.0,
      "step": 4260
    },
    {
      "entropy": 5.754845666885376,
      "epoch": 0.3868831640058055,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990092827388159,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16929103881120683,
      "num_tokens": 8854180.0,
      "step": 4265
    },
    {
      "entropy": 5.630288219451904,
      "epoch": 0.3873367198838897,
      "grad_norm": 0.875,
      "learning_rate": 0.0004990062473662681,
      "loss": 5.4903,
      "mean_token_accuracy": 0.17342475801706314,
      "num_tokens": 8865018.0,
      "step": 4270
    },
    {
      "entropy": 5.752366018295288,
      "epoch": 0.3877902757619739,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499003207361214,
      "loss": 5.6753,
      "mean_token_accuracy": 0.16161389499902726,
      "num_tokens": 8875423.0,
      "step": 4275
    },
    {
      "entropy": 5.819116306304932,
      "epoch": 0.38824383164005805,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990001627237165,
      "loss": 5.6046,
      "mean_token_accuracy": 0.1682436063885689,
      "num_tokens": 8887850.0,
      "step": 4280
    },
    {
      "entropy": 5.644161605834961,
      "epoch": 0.38869738751814226,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989971134538386,
      "loss": 5.4972,
      "mean_token_accuracy": 0.17050439417362212,
      "num_tokens": 8897429.0,
      "step": 4285
    },
    {
      "entropy": 5.663009977340698,
      "epoch": 0.3891509433962264,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989940595516432,
      "loss": 5.5253,
      "mean_token_accuracy": 0.16778848469257354,
      "num_tokens": 8907048.0,
      "step": 4290
    },
    {
      "entropy": 5.702516841888428,
      "epoch": 0.3896044992743106,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989910010171938,
      "loss": 5.5746,
      "mean_token_accuracy": 0.16057664453983306,
      "num_tokens": 8917987.0,
      "step": 4295
    },
    {
      "entropy": 5.754789972305298,
      "epoch": 0.3900580551523948,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004989879378505532,
      "loss": 5.4806,
      "mean_token_accuracy": 0.16423581093549727,
      "num_tokens": 8929339.0,
      "step": 4300
    },
    {
      "entropy": 5.599805641174316,
      "epoch": 0.39051161103047893,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989848700517849,
      "loss": 5.4908,
      "mean_token_accuracy": 0.1674097627401352,
      "num_tokens": 8939473.0,
      "step": 4305
    },
    {
      "entropy": 5.650110673904419,
      "epoch": 0.39096516690856314,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989817976209526,
      "loss": 5.5306,
      "mean_token_accuracy": 0.16720281541347504,
      "num_tokens": 8949173.0,
      "step": 4310
    },
    {
      "entropy": 5.739528751373291,
      "epoch": 0.3914187227866473,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004989787205581196,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1708226829767227,
      "num_tokens": 8959879.0,
      "step": 4315
    },
    {
      "entropy": 5.707206296920776,
      "epoch": 0.3918722786647315,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989756388633496,
      "loss": 5.5819,
      "mean_token_accuracy": 0.16486465483903884,
      "num_tokens": 8970643.0,
      "step": 4320
    },
    {
      "entropy": 5.666028881072998,
      "epoch": 0.39232583454281567,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989725525367063,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16958049088716506,
      "num_tokens": 8979882.0,
      "step": 4325
    },
    {
      "entropy": 5.7163489818572994,
      "epoch": 0.3927793904208999,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989694615782534,
      "loss": 5.5912,
      "mean_token_accuracy": 0.1618393614888191,
      "num_tokens": 8990925.0,
      "step": 4330
    },
    {
      "entropy": 5.722356939315796,
      "epoch": 0.39323294629898403,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989663659880552,
      "loss": 5.6084,
      "mean_token_accuracy": 0.1621517851948738,
      "num_tokens": 9001365.0,
      "step": 4335
    },
    {
      "entropy": 5.79314546585083,
      "epoch": 0.3936865021770682,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989632657661754,
      "loss": 5.6,
      "mean_token_accuracy": 0.1683468908071518,
      "num_tokens": 9011942.0,
      "step": 4340
    },
    {
      "entropy": 5.737189674377442,
      "epoch": 0.3941400580551524,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000498960160912678,
      "loss": 5.6458,
      "mean_token_accuracy": 0.16355903297662736,
      "num_tokens": 9022610.0,
      "step": 4345
    },
    {
      "entropy": 5.703625679016113,
      "epoch": 0.39459361393323655,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989570514276276,
      "loss": 5.5296,
      "mean_token_accuracy": 0.16396254748106004,
      "num_tokens": 9033034.0,
      "step": 4350
    },
    {
      "entropy": 5.721978998184204,
      "epoch": 0.39504716981132076,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004989539373110883,
      "loss": 5.5728,
      "mean_token_accuracy": 0.16446860283613204,
      "num_tokens": 9043418.0,
      "step": 4355
    },
    {
      "entropy": 5.782990789413452,
      "epoch": 0.3955007256894049,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989508185631245,
      "loss": 5.6716,
      "mean_token_accuracy": 0.16027326881885529,
      "num_tokens": 9052708.0,
      "step": 4360
    },
    {
      "entropy": 5.617509841918945,
      "epoch": 0.39595428156748913,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989476951838006,
      "loss": 5.5176,
      "mean_token_accuracy": 0.16624224931001663,
      "num_tokens": 9062883.0,
      "step": 4365
    },
    {
      "entropy": 5.744745779037475,
      "epoch": 0.3964078374455733,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989445671731814,
      "loss": 5.4776,
      "mean_token_accuracy": 0.16907311379909515,
      "num_tokens": 9072830.0,
      "step": 4370
    },
    {
      "entropy": 5.634798336029053,
      "epoch": 0.3968613933236575,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989414345313315,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1632262200117111,
      "num_tokens": 9082749.0,
      "step": 4375
    },
    {
      "entropy": 5.70997200012207,
      "epoch": 0.39731494920174165,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004989382972583156,
      "loss": 5.5417,
      "mean_token_accuracy": 0.1624021738767624,
      "num_tokens": 9094188.0,
      "step": 4380
    },
    {
      "entropy": 5.6712054252624515,
      "epoch": 0.3977685050798258,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989351553541986,
      "loss": 5.5543,
      "mean_token_accuracy": 0.1630195915699005,
      "num_tokens": 9105023.0,
      "step": 4385
    },
    {
      "entropy": 5.616395616531372,
      "epoch": 0.39822206095791,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004989320088190456,
      "loss": 5.504,
      "mean_token_accuracy": 0.1688954085111618,
      "num_tokens": 9115563.0,
      "step": 4390
    },
    {
      "entropy": 5.713869190216064,
      "epoch": 0.3986756168359942,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989288576529216,
      "loss": 5.5355,
      "mean_token_accuracy": 0.16001897305250168,
      "num_tokens": 9125394.0,
      "step": 4395
    },
    {
      "entropy": 5.724222660064697,
      "epoch": 0.3991291727140784,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004989257018558917,
      "loss": 5.531,
      "mean_token_accuracy": 0.16804168224334717,
      "num_tokens": 9137061.0,
      "step": 4400
    },
    {
      "entropy": 5.685125684738159,
      "epoch": 0.39958272859216254,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004989225414280213,
      "loss": 5.4906,
      "mean_token_accuracy": 0.168438558280468,
      "num_tokens": 9148196.0,
      "step": 4405
    },
    {
      "entropy": 5.665059661865234,
      "epoch": 0.40003628447024675,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989193763693756,
      "loss": 5.5142,
      "mean_token_accuracy": 0.1706180453300476,
      "num_tokens": 9158013.0,
      "step": 4410
    },
    {
      "entropy": 5.651648664474488,
      "epoch": 0.4004898403483309,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004989162066800202,
      "loss": 5.4225,
      "mean_token_accuracy": 0.17173381000757218,
      "num_tokens": 9167813.0,
      "step": 4415
    },
    {
      "entropy": 5.637704706192016,
      "epoch": 0.4009433962264151,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989130323600205,
      "loss": 5.4884,
      "mean_token_accuracy": 0.16999162435531617,
      "num_tokens": 9177783.0,
      "step": 4420
    },
    {
      "entropy": 5.688581085205078,
      "epoch": 0.40139695210449927,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989098534094424,
      "loss": 5.5333,
      "mean_token_accuracy": 0.16801782250404357,
      "num_tokens": 9187556.0,
      "step": 4425
    },
    {
      "entropy": 5.703752279281616,
      "epoch": 0.4018505079825834,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989066698283512,
      "loss": 5.5988,
      "mean_token_accuracy": 0.16409932374954223,
      "num_tokens": 9198659.0,
      "step": 4430
    },
    {
      "entropy": 5.741377305984497,
      "epoch": 0.40230406386066764,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989034816168132,
      "loss": 5.5719,
      "mean_token_accuracy": 0.1686299681663513,
      "num_tokens": 9209546.0,
      "step": 4435
    },
    {
      "entropy": 5.628240251541138,
      "epoch": 0.4027576197387518,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004989002887748941,
      "loss": 5.4811,
      "mean_token_accuracy": 0.17123182117938995,
      "num_tokens": 9219339.0,
      "step": 4440
    },
    {
      "entropy": 5.702338361740113,
      "epoch": 0.403211175616836,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004988970913026601,
      "loss": 5.6741,
      "mean_token_accuracy": 0.1605461746454239,
      "num_tokens": 9230364.0,
      "step": 4445
    },
    {
      "entropy": 5.706683349609375,
      "epoch": 0.40366473149492016,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988938892001771,
      "loss": 5.5239,
      "mean_token_accuracy": 0.16557129025459288,
      "num_tokens": 9240732.0,
      "step": 4450
    },
    {
      "entropy": 5.59526686668396,
      "epoch": 0.40411828737300437,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988906824675113,
      "loss": 5.4616,
      "mean_token_accuracy": 0.17066627889871597,
      "num_tokens": 9251049.0,
      "step": 4455
    },
    {
      "entropy": 5.65702075958252,
      "epoch": 0.4045718432510885,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988874711047293,
      "loss": 5.5041,
      "mean_token_accuracy": 0.17010140866041185,
      "num_tokens": 9261660.0,
      "step": 4460
    },
    {
      "entropy": 5.707148694992066,
      "epoch": 0.40502539912917274,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004988842551118972,
      "loss": 5.5619,
      "mean_token_accuracy": 0.17123832255601884,
      "num_tokens": 9272114.0,
      "step": 4465
    },
    {
      "entropy": 5.657516384124756,
      "epoch": 0.4054789550072569,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988810344890818,
      "loss": 5.5274,
      "mean_token_accuracy": 0.17154074907302858,
      "num_tokens": 9282125.0,
      "step": 4470
    },
    {
      "entropy": 5.755674219131469,
      "epoch": 0.40593251088534105,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988778092363494,
      "loss": 5.5719,
      "mean_token_accuracy": 0.164210744202137,
      "num_tokens": 9294078.0,
      "step": 4475
    },
    {
      "entropy": 5.703670406341553,
      "epoch": 0.40638606676342526,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988745793537671,
      "loss": 5.5273,
      "mean_token_accuracy": 0.16619765460491182,
      "num_tokens": 9304026.0,
      "step": 4480
    },
    {
      "entropy": 5.6847145557403564,
      "epoch": 0.4068396226415094,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988713448414011,
      "loss": 5.5717,
      "mean_token_accuracy": 0.17452892959117888,
      "num_tokens": 9313933.0,
      "step": 4485
    },
    {
      "entropy": 5.682414197921753,
      "epoch": 0.4072931785195936,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000498868105699319,
      "loss": 5.6038,
      "mean_token_accuracy": 0.1653165489435196,
      "num_tokens": 9323932.0,
      "step": 4490
    },
    {
      "entropy": 5.62096529006958,
      "epoch": 0.4077467343976778,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988648619275872,
      "loss": 5.4957,
      "mean_token_accuracy": 0.17853589355945587,
      "num_tokens": 9333467.0,
      "step": 4495
    },
    {
      "entropy": 5.730397415161133,
      "epoch": 0.408200290275762,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004988616135262729,
      "loss": 5.5523,
      "mean_token_accuracy": 0.16236073523759842,
      "num_tokens": 9344909.0,
      "step": 4500
    },
    {
      "entropy": 5.778379106521607,
      "epoch": 0.40865384615384615,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004988583604954436,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1632135108113289,
      "num_tokens": 9355630.0,
      "step": 4505
    },
    {
      "entropy": 5.630618858337402,
      "epoch": 0.40910740203193036,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988551028351663,
      "loss": 5.539,
      "mean_token_accuracy": 0.16523584127426147,
      "num_tokens": 9365700.0,
      "step": 4510
    },
    {
      "entropy": 5.654221391677856,
      "epoch": 0.4095609579100145,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988518405455084,
      "loss": 5.6261,
      "mean_token_accuracy": 0.15864572525024415,
      "num_tokens": 9377088.0,
      "step": 4515
    },
    {
      "entropy": 5.703935956954956,
      "epoch": 0.41001451378809867,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004988485736265374,
      "loss": 5.4107,
      "mean_token_accuracy": 0.17571215331554413,
      "num_tokens": 9386855.0,
      "step": 4520
    },
    {
      "entropy": 5.696550416946411,
      "epoch": 0.4104680696661829,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498845302078321,
      "loss": 5.5307,
      "mean_token_accuracy": 0.16638017445802689,
      "num_tokens": 9398661.0,
      "step": 4525
    },
    {
      "entropy": 5.72752366065979,
      "epoch": 0.41092162554426703,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988420259009266,
      "loss": 5.5315,
      "mean_token_accuracy": 0.17306275367736818,
      "num_tokens": 9408670.0,
      "step": 4530
    },
    {
      "entropy": 5.688307809829712,
      "epoch": 0.41137518142235124,
      "grad_norm": 0.859375,
      "learning_rate": 0.000498838745094422,
      "loss": 5.5582,
      "mean_token_accuracy": 0.16831061094999314,
      "num_tokens": 9418740.0,
      "step": 4535
    },
    {
      "entropy": 5.6272059917449955,
      "epoch": 0.4118287373004354,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988354596588751,
      "loss": 5.5311,
      "mean_token_accuracy": 0.16975374519824982,
      "num_tokens": 9429339.0,
      "step": 4540
    },
    {
      "entropy": 5.678937482833862,
      "epoch": 0.4122822931785196,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004988321695943539,
      "loss": 5.4738,
      "mean_token_accuracy": 0.1705018848180771,
      "num_tokens": 9438996.0,
      "step": 4545
    },
    {
      "entropy": 5.698333978652954,
      "epoch": 0.41273584905660377,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988288749009264,
      "loss": 5.4912,
      "mean_token_accuracy": 0.1687777042388916,
      "num_tokens": 9448669.0,
      "step": 4550
    },
    {
      "entropy": 5.645137596130371,
      "epoch": 0.413189404934688,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988255755786608,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1714872419834137,
      "num_tokens": 9458857.0,
      "step": 4555
    },
    {
      "entropy": 5.650253391265869,
      "epoch": 0.41364296081277213,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988222716276251,
      "loss": 5.4782,
      "mean_token_accuracy": 0.17236211895942688,
      "num_tokens": 9470241.0,
      "step": 4560
    },
    {
      "entropy": 5.731791353225708,
      "epoch": 0.4140965166908563,
      "grad_norm": 0.90625,
      "learning_rate": 0.000498818963047888,
      "loss": 5.6346,
      "mean_token_accuracy": 0.16064894944429398,
      "num_tokens": 9480528.0,
      "step": 4565
    },
    {
      "entropy": 5.75153079032898,
      "epoch": 0.4145500725689405,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988156498395176,
      "loss": 5.5396,
      "mean_token_accuracy": 0.17293335497379303,
      "num_tokens": 9491000.0,
      "step": 4570
    },
    {
      "entropy": 5.723613452911377,
      "epoch": 0.41500362844702465,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004988123320025825,
      "loss": 5.5322,
      "mean_token_accuracy": 0.17224910855293274,
      "num_tokens": 9501612.0,
      "step": 4575
    },
    {
      "entropy": 5.692364740371704,
      "epoch": 0.41545718432510886,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988090095371514,
      "loss": 5.639,
      "mean_token_accuracy": 0.16442762464284896,
      "num_tokens": 9513999.0,
      "step": 4580
    },
    {
      "entropy": 5.704233598709107,
      "epoch": 0.415910740203193,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988056824432928,
      "loss": 5.5106,
      "mean_token_accuracy": 0.1647787868976593,
      "num_tokens": 9524392.0,
      "step": 4585
    },
    {
      "entropy": 5.709184408187866,
      "epoch": 0.41636429608127723,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988023507210758,
      "loss": 5.5108,
      "mean_token_accuracy": 0.1714200809597969,
      "num_tokens": 9534227.0,
      "step": 4590
    },
    {
      "entropy": 5.69327826499939,
      "epoch": 0.4168178519593614,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987990143705692,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1758728012442589,
      "num_tokens": 9544242.0,
      "step": 4595
    },
    {
      "entropy": 5.6579484939575195,
      "epoch": 0.4172714078374456,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987956733918418,
      "loss": 5.434,
      "mean_token_accuracy": 0.17048389017581939,
      "num_tokens": 9553921.0,
      "step": 4600
    },
    {
      "entropy": 5.562104272842407,
      "epoch": 0.41772496371552975,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004987923277849629,
      "loss": 5.4928,
      "mean_token_accuracy": 0.16425550878047943,
      "num_tokens": 9564768.0,
      "step": 4605
    },
    {
      "entropy": 5.7012450218200685,
      "epoch": 0.4181785195936139,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987889775500017,
      "loss": 5.5184,
      "mean_token_accuracy": 0.16336782574653624,
      "num_tokens": 9576360.0,
      "step": 4610
    },
    {
      "entropy": 5.6748738288879395,
      "epoch": 0.4186320754716981,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004987856226870272,
      "loss": 5.54,
      "mean_token_accuracy": 0.16737818121910095,
      "num_tokens": 9587623.0,
      "step": 4615
    },
    {
      "entropy": 5.636663770675659,
      "epoch": 0.4190856313497823,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987822631961092,
      "loss": 5.5691,
      "mean_token_accuracy": 0.1642339438199997,
      "num_tokens": 9597837.0,
      "step": 4620
    },
    {
      "entropy": 5.706202650070191,
      "epoch": 0.4195391872278665,
      "grad_norm": 0.875,
      "learning_rate": 0.000498778899077317,
      "loss": 5.5288,
      "mean_token_accuracy": 0.1656550332903862,
      "num_tokens": 9609509.0,
      "step": 4625
    },
    {
      "entropy": 5.667462539672852,
      "epoch": 0.41999274310595064,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049877553033072,
      "loss": 5.4957,
      "mean_token_accuracy": 0.17181909531354905,
      "num_tokens": 9619954.0,
      "step": 4630
    },
    {
      "entropy": 5.644245195388794,
      "epoch": 0.42044629898403485,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004987721569563882,
      "loss": 5.4973,
      "mean_token_accuracy": 0.16737744808197022,
      "num_tokens": 9629742.0,
      "step": 4635
    },
    {
      "entropy": 5.666360330581665,
      "epoch": 0.420899854862119,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000498768778954391,
      "loss": 5.515,
      "mean_token_accuracy": 0.1711300268769264,
      "num_tokens": 9641378.0,
      "step": 4640
    },
    {
      "entropy": 5.756295871734619,
      "epoch": 0.4213534107402032,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987653963247985,
      "loss": 5.5852,
      "mean_token_accuracy": 0.16515168696641921,
      "num_tokens": 9652279.0,
      "step": 4645
    },
    {
      "entropy": 5.715899276733398,
      "epoch": 0.42180696661828737,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004987620090676805,
      "loss": 5.5811,
      "mean_token_accuracy": 0.16966685950756072,
      "num_tokens": 9663416.0,
      "step": 4650
    },
    {
      "entropy": 5.65866527557373,
      "epoch": 0.4222605224963715,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004987586171831072,
      "loss": 5.5207,
      "mean_token_accuracy": 0.16795586943626403,
      "num_tokens": 9674647.0,
      "step": 4655
    },
    {
      "entropy": 5.646727991104126,
      "epoch": 0.42271407837445574,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987552206711487,
      "loss": 5.3947,
      "mean_token_accuracy": 0.17098441645503043,
      "num_tokens": 9685084.0,
      "step": 4660
    },
    {
      "entropy": 5.602826547622681,
      "epoch": 0.4231676342525399,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004987518195318752,
      "loss": 5.451,
      "mean_token_accuracy": 0.17014521807432176,
      "num_tokens": 9695661.0,
      "step": 4665
    },
    {
      "entropy": 5.699631309509277,
      "epoch": 0.4236211901306241,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498748413765357,
      "loss": 5.5111,
      "mean_token_accuracy": 0.1655079945921898,
      "num_tokens": 9705773.0,
      "step": 4670
    },
    {
      "entropy": 5.672311735153198,
      "epoch": 0.42407474600870826,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987450033716646,
      "loss": 5.5543,
      "mean_token_accuracy": 0.1671050012111664,
      "num_tokens": 9715799.0,
      "step": 4675
    },
    {
      "entropy": 5.70002236366272,
      "epoch": 0.42452830188679247,
      "grad_norm": 0.875,
      "learning_rate": 0.0004987415883508686,
      "loss": 5.5117,
      "mean_token_accuracy": 0.17100395262241364,
      "num_tokens": 9727012.0,
      "step": 4680
    },
    {
      "entropy": 5.699238252639771,
      "epoch": 0.4249818577648766,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987381687030394,
      "loss": 5.5134,
      "mean_token_accuracy": 0.16880285739898682,
      "num_tokens": 9737044.0,
      "step": 4685
    },
    {
      "entropy": 5.728292417526245,
      "epoch": 0.42543541364296084,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987347444282479,
      "loss": 5.5911,
      "mean_token_accuracy": 0.16349772363901138,
      "num_tokens": 9747648.0,
      "step": 4690
    },
    {
      "entropy": 5.686186790466309,
      "epoch": 0.425888969521045,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987313155265649,
      "loss": 5.5245,
      "mean_token_accuracy": 0.163296639919281,
      "num_tokens": 9757812.0,
      "step": 4695
    },
    {
      "entropy": 5.753760099411011,
      "epoch": 0.42634252539912915,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004987278819980613,
      "loss": 5.5178,
      "mean_token_accuracy": 0.174093596637249,
      "num_tokens": 9768337.0,
      "step": 4700
    },
    {
      "entropy": 5.703652429580688,
      "epoch": 0.42679608127721336,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987244438428081,
      "loss": 5.5634,
      "mean_token_accuracy": 0.16444121152162552,
      "num_tokens": 9779433.0,
      "step": 4705
    },
    {
      "entropy": 5.611905670166015,
      "epoch": 0.4272496371552975,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004987210010608764,
      "loss": 5.4602,
      "mean_token_accuracy": 0.1711221307516098,
      "num_tokens": 9790069.0,
      "step": 4710
    },
    {
      "entropy": 5.659437561035157,
      "epoch": 0.4277031930333817,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004987175536523373,
      "loss": 5.4316,
      "mean_token_accuracy": 0.17792649418115616,
      "num_tokens": 9799669.0,
      "step": 4715
    },
    {
      "entropy": 5.626423311233521,
      "epoch": 0.4281567489114659,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004987141016172622,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1750251442193985,
      "num_tokens": 9809615.0,
      "step": 4720
    },
    {
      "entropy": 5.633890104293823,
      "epoch": 0.4286103047895501,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987106449557225,
      "loss": 5.4379,
      "mean_token_accuracy": 0.17350732386112214,
      "num_tokens": 9819666.0,
      "step": 4725
    },
    {
      "entropy": 5.73306941986084,
      "epoch": 0.42906386066763424,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987071836677896,
      "loss": 5.6122,
      "mean_token_accuracy": 0.16756637394428253,
      "num_tokens": 9831015.0,
      "step": 4730
    },
    {
      "entropy": 5.635014009475708,
      "epoch": 0.42951741654571846,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987037177535353,
      "loss": 5.4679,
      "mean_token_accuracy": 0.1639992356300354,
      "num_tokens": 9841354.0,
      "step": 4735
    },
    {
      "entropy": 5.633705282211304,
      "epoch": 0.4299709724238026,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987002472130309,
      "loss": 5.4563,
      "mean_token_accuracy": 0.16753149628639222,
      "num_tokens": 9850467.0,
      "step": 4740
    },
    {
      "entropy": 5.770284414291382,
      "epoch": 0.43042452830188677,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004986967720463485,
      "loss": 5.5496,
      "mean_token_accuracy": 0.1623040944337845,
      "num_tokens": 9861115.0,
      "step": 4745
    },
    {
      "entropy": 5.7330851554870605,
      "epoch": 0.430878084179971,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004986932922535598,
      "loss": 5.5384,
      "mean_token_accuracy": 0.16642699092626573,
      "num_tokens": 9871172.0,
      "step": 4750
    },
    {
      "entropy": 5.652520322799683,
      "epoch": 0.43133164005805513,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986898078347368,
      "loss": 5.4515,
      "mean_token_accuracy": 0.17356225550174714,
      "num_tokens": 9881997.0,
      "step": 4755
    },
    {
      "entropy": 5.61650390625,
      "epoch": 0.43178519593613934,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004986863187899516,
      "loss": 5.4211,
      "mean_token_accuracy": 0.1723109319806099,
      "num_tokens": 9892310.0,
      "step": 4760
    },
    {
      "entropy": 5.6054548740386965,
      "epoch": 0.4322387518142235,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004986828251192764,
      "loss": 5.4786,
      "mean_token_accuracy": 0.17029564827680588,
      "num_tokens": 9902664.0,
      "step": 4765
    },
    {
      "entropy": 5.6452922344207765,
      "epoch": 0.4326923076923077,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986793268227833,
      "loss": 5.4662,
      "mean_token_accuracy": 0.17340250313282013,
      "num_tokens": 9912491.0,
      "step": 4770
    },
    {
      "entropy": 5.65220012664795,
      "epoch": 0.43314586357039186,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986758239005448,
      "loss": 5.439,
      "mean_token_accuracy": 0.17318663150072097,
      "num_tokens": 9922057.0,
      "step": 4775
    },
    {
      "entropy": 5.640111303329467,
      "epoch": 0.4335994194484761,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004986723163526331,
      "loss": 5.4976,
      "mean_token_accuracy": 0.1674135446548462,
      "num_tokens": 9933661.0,
      "step": 4780
    },
    {
      "entropy": 5.577649021148682,
      "epoch": 0.43405297532656023,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004986688041791209,
      "loss": 5.4532,
      "mean_token_accuracy": 0.16382990777492523,
      "num_tokens": 9945177.0,
      "step": 4785
    },
    {
      "entropy": 5.671372318267823,
      "epoch": 0.4345065312046444,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498665287380081,
      "loss": 5.479,
      "mean_token_accuracy": 0.1665464699268341,
      "num_tokens": 9956226.0,
      "step": 4790
    },
    {
      "entropy": 5.709822797775269,
      "epoch": 0.4349600870827286,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986617659555859,
      "loss": 5.5528,
      "mean_token_accuracy": 0.16944931149482728,
      "num_tokens": 9965877.0,
      "step": 4795
    },
    {
      "entropy": 5.627245140075684,
      "epoch": 0.43541364296081275,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986582399057085,
      "loss": 5.4958,
      "mean_token_accuracy": 0.17361783385276794,
      "num_tokens": 9975592.0,
      "step": 4800
    },
    {
      "entropy": 5.716971731185913,
      "epoch": 0.43586719883889696,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986547092305216,
      "loss": 5.5188,
      "mean_token_accuracy": 0.16906160414218901,
      "num_tokens": 9986094.0,
      "step": 4805
    },
    {
      "entropy": 5.747332000732422,
      "epoch": 0.4363207547169811,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986511739300984,
      "loss": 5.5563,
      "mean_token_accuracy": 0.16685436069965362,
      "num_tokens": 9997241.0,
      "step": 4810
    },
    {
      "entropy": 5.624169874191284,
      "epoch": 0.43677431059506533,
      "grad_norm": 0.875,
      "learning_rate": 0.0004986476340045119,
      "loss": 5.4805,
      "mean_token_accuracy": 0.1696499690413475,
      "num_tokens": 10008091.0,
      "step": 4815
    },
    {
      "entropy": 5.610916471481323,
      "epoch": 0.4372278664731495,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004986440894538354,
      "loss": 5.5358,
      "mean_token_accuracy": 0.16445003151893617,
      "num_tokens": 10018794.0,
      "step": 4820
    },
    {
      "entropy": 5.658578824996948,
      "epoch": 0.4376814223512337,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986405402781421,
      "loss": 5.4595,
      "mean_token_accuracy": 0.17329651415348052,
      "num_tokens": 10028414.0,
      "step": 4825
    },
    {
      "entropy": 5.683524417877197,
      "epoch": 0.43813497822931785,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004986369864775054,
      "loss": 5.5255,
      "mean_token_accuracy": 0.1668830245733261,
      "num_tokens": 10038749.0,
      "step": 4830
    },
    {
      "entropy": 5.680928516387939,
      "epoch": 0.438588534107402,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004986334280519989,
      "loss": 5.4706,
      "mean_token_accuracy": 0.16898296475410463,
      "num_tokens": 10049866.0,
      "step": 4835
    },
    {
      "entropy": 5.620717573165893,
      "epoch": 0.4390420899854862,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000498629865001696,
      "loss": 5.4971,
      "mean_token_accuracy": 0.176370669901371,
      "num_tokens": 10060680.0,
      "step": 4840
    },
    {
      "entropy": 5.6851057529449465,
      "epoch": 0.43949564586357037,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004986262973266706,
      "loss": 5.6094,
      "mean_token_accuracy": 0.16637811809778214,
      "num_tokens": 10070931.0,
      "step": 4845
    },
    {
      "entropy": 5.648931550979614,
      "epoch": 0.4399492017416546,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986227250269964,
      "loss": 5.5285,
      "mean_token_accuracy": 0.17198051065206527,
      "num_tokens": 10080189.0,
      "step": 4850
    },
    {
      "entropy": 5.683497285842895,
      "epoch": 0.44040275761973874,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986191481027472,
      "loss": 5.4349,
      "mean_token_accuracy": 0.17278559356927872,
      "num_tokens": 10088813.0,
      "step": 4855
    },
    {
      "entropy": 5.61211371421814,
      "epoch": 0.44085631349782295,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004986155665539971,
      "loss": 5.4644,
      "mean_token_accuracy": 0.17080731093883514,
      "num_tokens": 10098125.0,
      "step": 4860
    },
    {
      "entropy": 5.69177656173706,
      "epoch": 0.4413098693759071,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049861198038082,
      "loss": 5.506,
      "mean_token_accuracy": 0.17002110332250595,
      "num_tokens": 10109125.0,
      "step": 4865
    },
    {
      "entropy": 5.673552989959717,
      "epoch": 0.4417634252539913,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986083895832903,
      "loss": 5.4864,
      "mean_token_accuracy": 0.16711554676294327,
      "num_tokens": 10120182.0,
      "step": 4870
    },
    {
      "entropy": 5.619903326034546,
      "epoch": 0.44221698113207547,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986047941614821,
      "loss": 5.5446,
      "mean_token_accuracy": 0.1634003072977066,
      "num_tokens": 10130941.0,
      "step": 4875
    },
    {
      "entropy": 5.683604526519775,
      "epoch": 0.4426705370101596,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986011941154696,
      "loss": 5.5044,
      "mean_token_accuracy": 0.16631961166858672,
      "num_tokens": 10141708.0,
      "step": 4880
    },
    {
      "entropy": 5.637266683578491,
      "epoch": 0.44312409288824384,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985975894453275,
      "loss": 5.4431,
      "mean_token_accuracy": 0.16837728172540664,
      "num_tokens": 10151816.0,
      "step": 4885
    },
    {
      "entropy": 5.694721031188965,
      "epoch": 0.443577648766328,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985939801511304,
      "loss": 5.5662,
      "mean_token_accuracy": 0.16548764184117318,
      "num_tokens": 10162470.0,
      "step": 4890
    },
    {
      "entropy": 5.672745275497436,
      "epoch": 0.4440312046444122,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004985903662329527,
      "loss": 5.4515,
      "mean_token_accuracy": 0.17449744194746017,
      "num_tokens": 10173171.0,
      "step": 4895
    },
    {
      "entropy": 5.667294073104858,
      "epoch": 0.44448476052249636,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004985867476908693,
      "loss": 5.4873,
      "mean_token_accuracy": 0.17015801668167113,
      "num_tokens": 10184185.0,
      "step": 4900
    },
    {
      "entropy": 5.702019119262696,
      "epoch": 0.44493831640058057,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004985831245249549,
      "loss": 5.5464,
      "mean_token_accuracy": 0.169259525090456,
      "num_tokens": 10195516.0,
      "step": 4905
    },
    {
      "entropy": 5.6287211894989015,
      "epoch": 0.4453918722786647,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985794967352846,
      "loss": 5.4707,
      "mean_token_accuracy": 0.1748988300561905,
      "num_tokens": 10204936.0,
      "step": 4910
    },
    {
      "entropy": 5.658074283599854,
      "epoch": 0.44584542815674894,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004985758643219334,
      "loss": 5.4998,
      "mean_token_accuracy": 0.17041691839694978,
      "num_tokens": 10216218.0,
      "step": 4915
    },
    {
      "entropy": 5.744098043441772,
      "epoch": 0.4462989840348331,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985722272849762,
      "loss": 5.5099,
      "mean_token_accuracy": 0.16554411202669145,
      "num_tokens": 10226370.0,
      "step": 4920
    },
    {
      "entropy": 5.618660640716553,
      "epoch": 0.44675253991291725,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985685856244884,
      "loss": 5.4398,
      "mean_token_accuracy": 0.18081948906183243,
      "num_tokens": 10236170.0,
      "step": 4925
    },
    {
      "entropy": 5.666572141647339,
      "epoch": 0.44720609579100146,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985649393405453,
      "loss": 5.5025,
      "mean_token_accuracy": 0.16895841807127,
      "num_tokens": 10246119.0,
      "step": 4930
    },
    {
      "entropy": 5.671585464477539,
      "epoch": 0.4476596516690856,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004985612884332223,
      "loss": 5.5196,
      "mean_token_accuracy": 0.16341286897659302,
      "num_tokens": 10256762.0,
      "step": 4935
    },
    {
      "entropy": 5.617032051086426,
      "epoch": 0.4481132075471698,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985576329025949,
      "loss": 5.4652,
      "mean_token_accuracy": 0.17075106054544448,
      "num_tokens": 10267317.0,
      "step": 4940
    },
    {
      "entropy": 5.681619024276733,
      "epoch": 0.448566763425254,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985539727487385,
      "loss": 5.603,
      "mean_token_accuracy": 0.15708888173103333,
      "num_tokens": 10279111.0,
      "step": 4945
    },
    {
      "entropy": 5.6958630084991455,
      "epoch": 0.4490203193033382,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004985503079717292,
      "loss": 5.4674,
      "mean_token_accuracy": 0.1703536629676819,
      "num_tokens": 10289383.0,
      "step": 4950
    },
    {
      "entropy": 5.6642831325531,
      "epoch": 0.44947387518142234,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985466385716425,
      "loss": 5.4767,
      "mean_token_accuracy": 0.17260171324014664,
      "num_tokens": 10299806.0,
      "step": 4955
    },
    {
      "entropy": 5.578657579421997,
      "epoch": 0.44992743105950656,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004985429645485543,
      "loss": 5.5005,
      "mean_token_accuracy": 0.1660196453332901,
      "num_tokens": 10310348.0,
      "step": 4960
    },
    {
      "entropy": 5.674696922302246,
      "epoch": 0.4503809869375907,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985392859025407,
      "loss": 5.4875,
      "mean_token_accuracy": 0.17419226914644242,
      "num_tokens": 10320354.0,
      "step": 4965
    },
    {
      "entropy": 5.604598712921143,
      "epoch": 0.45083454281567487,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004985356026336776,
      "loss": 5.4148,
      "mean_token_accuracy": 0.16999066323041917,
      "num_tokens": 10330292.0,
      "step": 4970
    },
    {
      "entropy": 5.621414422988892,
      "epoch": 0.4512880986937591,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985319147420414,
      "loss": 5.3829,
      "mean_token_accuracy": 0.18479472696781157,
      "num_tokens": 10340449.0,
      "step": 4975
    },
    {
      "entropy": 5.649209356307983,
      "epoch": 0.45174165457184323,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985282222277082,
      "loss": 5.4569,
      "mean_token_accuracy": 0.16669374704360962,
      "num_tokens": 10350637.0,
      "step": 4980
    },
    {
      "entropy": 5.581578731536865,
      "epoch": 0.45219521044992744,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004985245250907544,
      "loss": 5.3903,
      "mean_token_accuracy": 0.17272900640964509,
      "num_tokens": 10361548.0,
      "step": 4985
    },
    {
      "entropy": 5.573017024993897,
      "epoch": 0.4526487663280116,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985208233312564,
      "loss": 5.4129,
      "mean_token_accuracy": 0.16686842143535613,
      "num_tokens": 10371765.0,
      "step": 4990
    },
    {
      "entropy": 5.696421766281128,
      "epoch": 0.4531023222060958,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004985171169492908,
      "loss": 5.5166,
      "mean_token_accuracy": 0.17461331635713578,
      "num_tokens": 10382108.0,
      "step": 4995
    },
    {
      "entropy": 5.630676031112671,
      "epoch": 0.45355587808417996,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985134059449344,
      "loss": 5.4666,
      "mean_token_accuracy": 0.18015743643045426,
      "num_tokens": 10392390.0,
      "step": 5000
    },
    {
      "entropy": 5.6141965866088865,
      "epoch": 0.4540094339622642,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985096903182637,
      "loss": 5.5127,
      "mean_token_accuracy": 0.17537071108818053,
      "num_tokens": 10402094.0,
      "step": 5005
    },
    {
      "entropy": 5.666981887817383,
      "epoch": 0.45446298984034833,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004985059700693557,
      "loss": 5.466,
      "mean_token_accuracy": 0.173527130484581,
      "num_tokens": 10413350.0,
      "step": 5010
    },
    {
      "entropy": 5.702916288375855,
      "epoch": 0.4549165457184325,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985022451982872,
      "loss": 5.5916,
      "mean_token_accuracy": 0.16218466758728028,
      "num_tokens": 10424292.0,
      "step": 5015
    },
    {
      "entropy": 5.756633567810058,
      "epoch": 0.4553701015965167,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984985157051354,
      "loss": 5.4481,
      "mean_token_accuracy": 0.1688245952129364,
      "num_tokens": 10433724.0,
      "step": 5020
    },
    {
      "entropy": 5.648002099990845,
      "epoch": 0.45582365747460085,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984947815899774,
      "loss": 5.5425,
      "mean_token_accuracy": 0.1708854228258133,
      "num_tokens": 10444724.0,
      "step": 5025
    },
    {
      "entropy": 5.65775728225708,
      "epoch": 0.45627721335268506,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984910428528902,
      "loss": 5.4623,
      "mean_token_accuracy": 0.17359500378370285,
      "num_tokens": 10455086.0,
      "step": 5030
    },
    {
      "entropy": 5.598028898239136,
      "epoch": 0.4567307692307692,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004984872994939514,
      "loss": 5.4155,
      "mean_token_accuracy": 0.1689280539751053,
      "num_tokens": 10465076.0,
      "step": 5035
    },
    {
      "entropy": 5.6523706912994385,
      "epoch": 0.45718432510885343,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984835515132382,
      "loss": 5.4321,
      "mean_token_accuracy": 0.17813173532485962,
      "num_tokens": 10474817.0,
      "step": 5040
    },
    {
      "entropy": 5.67584810256958,
      "epoch": 0.4576378809869376,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004984797989108282,
      "loss": 5.4584,
      "mean_token_accuracy": 0.1707393169403076,
      "num_tokens": 10485735.0,
      "step": 5045
    },
    {
      "entropy": 5.613812065124511,
      "epoch": 0.4580914368650218,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984760416867991,
      "loss": 5.5472,
      "mean_token_accuracy": 0.1682056277990341,
      "num_tokens": 10496731.0,
      "step": 5050
    },
    {
      "entropy": 5.568698740005493,
      "epoch": 0.45854499274310595,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004984722798412286,
      "loss": 5.4136,
      "mean_token_accuracy": 0.17073957175016402,
      "num_tokens": 10506787.0,
      "step": 5055
    },
    {
      "entropy": 5.659662580490112,
      "epoch": 0.4589985486211901,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984685133741941,
      "loss": 5.4739,
      "mean_token_accuracy": 0.16943865567445754,
      "num_tokens": 10517170.0,
      "step": 5060
    },
    {
      "entropy": 5.622515487670898,
      "epoch": 0.4594521044992743,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498464742285774,
      "loss": 5.4733,
      "mean_token_accuracy": 0.1717009276151657,
      "num_tokens": 10527342.0,
      "step": 5065
    },
    {
      "entropy": 5.642328071594238,
      "epoch": 0.45990566037735847,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000498460966576046,
      "loss": 5.4503,
      "mean_token_accuracy": 0.17293796241283416,
      "num_tokens": 10537253.0,
      "step": 5070
    },
    {
      "entropy": 5.698889923095703,
      "epoch": 0.4603592162554427,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984571862450884,
      "loss": 5.465,
      "mean_token_accuracy": 0.1680241197347641,
      "num_tokens": 10547249.0,
      "step": 5075
    },
    {
      "entropy": 5.7043322086334225,
      "epoch": 0.46081277213352684,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004984534012929791,
      "loss": 5.503,
      "mean_token_accuracy": 0.172734771668911,
      "num_tokens": 10558174.0,
      "step": 5080
    },
    {
      "entropy": 5.700099897384644,
      "epoch": 0.46126632801161105,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984496117197967,
      "loss": 5.4444,
      "mean_token_accuracy": 0.17787872403860092,
      "num_tokens": 10567610.0,
      "step": 5085
    },
    {
      "entropy": 5.602445316314697,
      "epoch": 0.4617198838896952,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004984458175256193,
      "loss": 5.462,
      "mean_token_accuracy": 0.16895331740379332,
      "num_tokens": 10578494.0,
      "step": 5090
    },
    {
      "entropy": 5.690131187438965,
      "epoch": 0.4621734397677794,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004984420187105253,
      "loss": 5.5668,
      "mean_token_accuracy": 0.16630938053131103,
      "num_tokens": 10589482.0,
      "step": 5095
    },
    {
      "entropy": 5.727915954589844,
      "epoch": 0.46262699564586357,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004984382152745936,
      "loss": 5.5203,
      "mean_token_accuracy": 0.1676805019378662,
      "num_tokens": 10599965.0,
      "step": 5100
    },
    {
      "entropy": 5.638349914550782,
      "epoch": 0.4630805515239477,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984344072179026,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1719616398215294,
      "num_tokens": 10610500.0,
      "step": 5105
    },
    {
      "entropy": 5.737101459503174,
      "epoch": 0.46353410740203194,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984305945405312,
      "loss": 5.5,
      "mean_token_accuracy": 0.17199551463127136,
      "num_tokens": 10619843.0,
      "step": 5110
    },
    {
      "entropy": 5.645010232925415,
      "epoch": 0.4639876632801161,
      "grad_norm": 0.921875,
      "learning_rate": 0.000498426777242558,
      "loss": 5.4291,
      "mean_token_accuracy": 0.1767846554517746,
      "num_tokens": 10631026.0,
      "step": 5115
    },
    {
      "entropy": 5.490650749206543,
      "epoch": 0.4644412191582003,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004984229553240623,
      "loss": 5.4348,
      "mean_token_accuracy": 0.18036371171474458,
      "num_tokens": 10641130.0,
      "step": 5120
    },
    {
      "entropy": 5.699585437774658,
      "epoch": 0.46489477503628446,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004984191287851228,
      "loss": 5.4958,
      "mean_token_accuracy": 0.17234579771757125,
      "num_tokens": 10652305.0,
      "step": 5125
    },
    {
      "entropy": 5.680755996704102,
      "epoch": 0.46534833091436867,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004984152976258188,
      "loss": 5.5328,
      "mean_token_accuracy": 0.17020701766014099,
      "num_tokens": 10663083.0,
      "step": 5130
    },
    {
      "entropy": 5.659653997421264,
      "epoch": 0.4658018867924528,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984114618462296,
      "loss": 5.5545,
      "mean_token_accuracy": 0.16575139313936232,
      "num_tokens": 10673602.0,
      "step": 5135
    },
    {
      "entropy": 5.619678211212158,
      "epoch": 0.46625544267053703,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004984076214464343,
      "loss": 5.3841,
      "mean_token_accuracy": 0.1807913675904274,
      "num_tokens": 10684622.0,
      "step": 5140
    },
    {
      "entropy": 5.624106788635254,
      "epoch": 0.4667089985486212,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984037764265126,
      "loss": 5.4969,
      "mean_token_accuracy": 0.17183390706777574,
      "num_tokens": 10695129.0,
      "step": 5145
    },
    {
      "entropy": 5.588241958618164,
      "epoch": 0.46716255442670535,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983999267865438,
      "loss": 5.3285,
      "mean_token_accuracy": 0.18249204456806184,
      "num_tokens": 10705236.0,
      "step": 5150
    },
    {
      "entropy": 5.593046188354492,
      "epoch": 0.46761611030478956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983960725266078,
      "loss": 5.4582,
      "mean_token_accuracy": 0.17504026740789413,
      "num_tokens": 10715748.0,
      "step": 5155
    },
    {
      "entropy": 5.624542713165283,
      "epoch": 0.4680696661828737,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004983922136467838,
      "loss": 5.3846,
      "mean_token_accuracy": 0.1776614785194397,
      "num_tokens": 10725799.0,
      "step": 5160
    },
    {
      "entropy": 5.6600791931152346,
      "epoch": 0.4685232220609579,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498388350147152,
      "loss": 5.5062,
      "mean_token_accuracy": 0.16867911964654922,
      "num_tokens": 10735668.0,
      "step": 5165
    },
    {
      "entropy": 5.662786149978638,
      "epoch": 0.4689767779390421,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983844820277922,
      "loss": 5.6079,
      "mean_token_accuracy": 0.167015041410923,
      "num_tokens": 10746466.0,
      "step": 5170
    },
    {
      "entropy": 5.752711057662964,
      "epoch": 0.4694303338171263,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983806092887843,
      "loss": 5.5063,
      "mean_token_accuracy": 0.1687478616833687,
      "num_tokens": 10756724.0,
      "step": 5175
    },
    {
      "entropy": 5.664383506774902,
      "epoch": 0.46988388969521044,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004983767319302086,
      "loss": 5.5499,
      "mean_token_accuracy": 0.16907021701335906,
      "num_tokens": 10766919.0,
      "step": 5180
    },
    {
      "entropy": 5.576836204528808,
      "epoch": 0.47033744557329465,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983728499521451,
      "loss": 5.3338,
      "mean_token_accuracy": 0.18249264657497405,
      "num_tokens": 10777550.0,
      "step": 5185
    },
    {
      "entropy": 5.655050230026245,
      "epoch": 0.4707910014513788,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004983689633546741,
      "loss": 5.4027,
      "mean_token_accuracy": 0.17136975079774858,
      "num_tokens": 10788456.0,
      "step": 5190
    },
    {
      "entropy": 5.663191461563111,
      "epoch": 0.47124455732946297,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004983650721378761,
      "loss": 5.4345,
      "mean_token_accuracy": 0.17229533791542054,
      "num_tokens": 10800049.0,
      "step": 5195
    },
    {
      "entropy": 5.655028867721557,
      "epoch": 0.4716981132075472,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004983611763018314,
      "loss": 5.4815,
      "mean_token_accuracy": 0.16865273416042328,
      "num_tokens": 10810743.0,
      "step": 5200
    },
    {
      "entropy": 5.753147888183594,
      "epoch": 0.47215166908563133,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983572758466208,
      "loss": 5.677,
      "mean_token_accuracy": 0.16416452825069427,
      "num_tokens": 10821897.0,
      "step": 5205
    },
    {
      "entropy": 5.7169407367706295,
      "epoch": 0.47260522496371554,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983533707723247,
      "loss": 5.527,
      "mean_token_accuracy": 0.16855133473873138,
      "num_tokens": 10831589.0,
      "step": 5210
    },
    {
      "entropy": 5.657151603698731,
      "epoch": 0.4730587808417997,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004983494610790241,
      "loss": 5.429,
      "mean_token_accuracy": 0.16840204745531082,
      "num_tokens": 10841404.0,
      "step": 5215
    },
    {
      "entropy": 5.6251708984375,
      "epoch": 0.4735123367198839,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983455467667996,
      "loss": 5.471,
      "mean_token_accuracy": 0.1659417137503624,
      "num_tokens": 10851577.0,
      "step": 5220
    },
    {
      "entropy": 5.615735912322998,
      "epoch": 0.47396589259796806,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983416278357322,
      "loss": 5.3906,
      "mean_token_accuracy": 0.1765459805727005,
      "num_tokens": 10861403.0,
      "step": 5225
    },
    {
      "entropy": 5.629169273376465,
      "epoch": 0.4744194484760523,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983377042859032,
      "loss": 5.4383,
      "mean_token_accuracy": 0.17094605714082717,
      "num_tokens": 10871933.0,
      "step": 5230
    },
    {
      "entropy": 5.5857017040252686,
      "epoch": 0.47487300435413643,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983337761173936,
      "loss": 5.5079,
      "mean_token_accuracy": 0.16988566964864732,
      "num_tokens": 10881800.0,
      "step": 5235
    },
    {
      "entropy": 5.733130645751953,
      "epoch": 0.4753265602322206,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004983298433302843,
      "loss": 5.4496,
      "mean_token_accuracy": 0.17222348898649215,
      "num_tokens": 10891830.0,
      "step": 5240
    },
    {
      "entropy": 5.616529273986816,
      "epoch": 0.4757801161103048,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983259059246572,
      "loss": 5.3783,
      "mean_token_accuracy": 0.18143776208162307,
      "num_tokens": 10901557.0,
      "step": 5245
    },
    {
      "entropy": 5.551515579223633,
      "epoch": 0.47623367198838895,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983219639005934,
      "loss": 5.5078,
      "mean_token_accuracy": 0.1646910712122917,
      "num_tokens": 10912155.0,
      "step": 5250
    },
    {
      "entropy": 5.639435863494873,
      "epoch": 0.47668722786647316,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983180172581745,
      "loss": 5.4085,
      "mean_token_accuracy": 0.17542287111282348,
      "num_tokens": 10922291.0,
      "step": 5255
    },
    {
      "entropy": 5.703779172897339,
      "epoch": 0.4771407837445573,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004983140659974819,
      "loss": 5.478,
      "mean_token_accuracy": 0.1703026846051216,
      "num_tokens": 10933176.0,
      "step": 5260
    },
    {
      "entropy": 5.653920316696167,
      "epoch": 0.47759433962264153,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004983101101185977,
      "loss": 5.5772,
      "mean_token_accuracy": 0.16286463141441346,
      "num_tokens": 10944049.0,
      "step": 5265
    },
    {
      "entropy": 5.619420862197876,
      "epoch": 0.4780478955007257,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004983061496216035,
      "loss": 5.4072,
      "mean_token_accuracy": 0.17815487831830978,
      "num_tokens": 10954347.0,
      "step": 5270
    },
    {
      "entropy": 5.614488506317139,
      "epoch": 0.4785014513788099,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004983021845065812,
      "loss": 5.4529,
      "mean_token_accuracy": 0.17562718093395233,
      "num_tokens": 10965123.0,
      "step": 5275
    },
    {
      "entropy": 5.580786514282226,
      "epoch": 0.47895500725689405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982982147736128,
      "loss": 5.4492,
      "mean_token_accuracy": 0.1764552414417267,
      "num_tokens": 10976395.0,
      "step": 5280
    },
    {
      "entropy": 5.654614782333374,
      "epoch": 0.4794085631349782,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004982942404227805,
      "loss": 5.437,
      "mean_token_accuracy": 0.16845346987247467,
      "num_tokens": 10986797.0,
      "step": 5285
    },
    {
      "entropy": 5.695258712768554,
      "epoch": 0.4798621190130624,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004982902614541662,
      "loss": 5.4915,
      "mean_token_accuracy": 0.16865674555301666,
      "num_tokens": 10997229.0,
      "step": 5290
    },
    {
      "entropy": 5.583453416824341,
      "epoch": 0.48031567489114657,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004982862778678526,
      "loss": 5.4231,
      "mean_token_accuracy": 0.1796005994081497,
      "num_tokens": 11008180.0,
      "step": 5295
    },
    {
      "entropy": 5.642146348953247,
      "epoch": 0.4807692307692308,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004982822896639218,
      "loss": 5.4431,
      "mean_token_accuracy": 0.17733724117279054,
      "num_tokens": 11019248.0,
      "step": 5300
    },
    {
      "entropy": 5.5944068908691404,
      "epoch": 0.48122278664731494,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004982782968424563,
      "loss": 5.3671,
      "mean_token_accuracy": 0.1745454654097557,
      "num_tokens": 11029760.0,
      "step": 5305
    },
    {
      "entropy": 5.60427680015564,
      "epoch": 0.48167634252539915,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982742994035388,
      "loss": 5.4575,
      "mean_token_accuracy": 0.17510539293289185,
      "num_tokens": 11039211.0,
      "step": 5310
    },
    {
      "entropy": 5.634851455688477,
      "epoch": 0.4821298984034833,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982702973472518,
      "loss": 5.4293,
      "mean_token_accuracy": 0.17235955595970154,
      "num_tokens": 11049892.0,
      "step": 5315
    },
    {
      "entropy": 5.697804927825928,
      "epoch": 0.4825834542815675,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004982662906736783,
      "loss": 5.4824,
      "mean_token_accuracy": 0.17303521037101746,
      "num_tokens": 11060823.0,
      "step": 5320
    },
    {
      "entropy": 5.578884744644165,
      "epoch": 0.48303701015965167,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004982622793829009,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1753349706530571,
      "num_tokens": 11072353.0,
      "step": 5325
    },
    {
      "entropy": 5.626900243759155,
      "epoch": 0.4834905660377358,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982582634750027,
      "loss": 5.4687,
      "mean_token_accuracy": 0.17250408083200455,
      "num_tokens": 11082655.0,
      "step": 5330
    },
    {
      "entropy": 5.67281928062439,
      "epoch": 0.48394412191582004,
      "grad_norm": 0.875,
      "learning_rate": 0.0004982542429500667,
      "loss": 5.4946,
      "mean_token_accuracy": 0.1740492969751358,
      "num_tokens": 11093315.0,
      "step": 5335
    },
    {
      "entropy": 5.53027229309082,
      "epoch": 0.4843976777939042,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982502178081761,
      "loss": 5.3898,
      "mean_token_accuracy": 0.17748946249485015,
      "num_tokens": 11103074.0,
      "step": 5340
    },
    {
      "entropy": 5.641853141784668,
      "epoch": 0.4848512336719884,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004982461880494141,
      "loss": 5.4088,
      "mean_token_accuracy": 0.17445126175880432,
      "num_tokens": 11113636.0,
      "step": 5345
    },
    {
      "entropy": 5.629180288314819,
      "epoch": 0.48530478955007256,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498242153673864,
      "loss": 5.3781,
      "mean_token_accuracy": 0.1742332860827446,
      "num_tokens": 11123509.0,
      "step": 5350
    },
    {
      "entropy": 5.594754266738891,
      "epoch": 0.48575834542815677,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982381146816094,
      "loss": 5.4325,
      "mean_token_accuracy": 0.170786514878273,
      "num_tokens": 11133405.0,
      "step": 5355
    },
    {
      "entropy": 5.694587469100952,
      "epoch": 0.4862119013062409,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004982340710727336,
      "loss": 5.5156,
      "mean_token_accuracy": 0.16179958879947662,
      "num_tokens": 11144725.0,
      "step": 5360
    },
    {
      "entropy": 5.644291639328003,
      "epoch": 0.48666545718432513,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982300228473203,
      "loss": 5.4028,
      "mean_token_accuracy": 0.17339923679828645,
      "num_tokens": 11155446.0,
      "step": 5365
    },
    {
      "entropy": 5.61758942604065,
      "epoch": 0.4871190130624093,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004982259700054533,
      "loss": 5.4547,
      "mean_token_accuracy": 0.18006922155618668,
      "num_tokens": 11165380.0,
      "step": 5370
    },
    {
      "entropy": 5.630903339385986,
      "epoch": 0.48757256894049344,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004982219125472163,
      "loss": 5.5089,
      "mean_token_accuracy": 0.1695322260260582,
      "num_tokens": 11175641.0,
      "step": 5375
    },
    {
      "entropy": 5.7712828636169435,
      "epoch": 0.48802612481857766,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004982178504726933,
      "loss": 5.6741,
      "mean_token_accuracy": 0.16982539594173432,
      "num_tokens": 11186257.0,
      "step": 5380
    },
    {
      "entropy": 5.703107261657715,
      "epoch": 0.4884796806966618,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004982137837819682,
      "loss": 5.4871,
      "mean_token_accuracy": 0.16374782472848892,
      "num_tokens": 11197284.0,
      "step": 5385
    },
    {
      "entropy": 5.680084371566773,
      "epoch": 0.488933236574746,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004982097124751252,
      "loss": 5.4638,
      "mean_token_accuracy": 0.17422798573970794,
      "num_tokens": 11208413.0,
      "step": 5390
    },
    {
      "entropy": 5.596026182174683,
      "epoch": 0.4893867924528302,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982056365522486,
      "loss": 5.465,
      "mean_token_accuracy": 0.17459869980812073,
      "num_tokens": 11219016.0,
      "step": 5395
    },
    {
      "entropy": 5.622074842453003,
      "epoch": 0.4898403483309144,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004982015560134225,
      "loss": 5.425,
      "mean_token_accuracy": 0.17910900115966796,
      "num_tokens": 11228934.0,
      "step": 5400
    },
    {
      "entropy": 5.652887344360352,
      "epoch": 0.49029390420899854,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981974708587312,
      "loss": 5.346,
      "mean_token_accuracy": 0.1806335538625717,
      "num_tokens": 11239596.0,
      "step": 5405
    },
    {
      "entropy": 5.567725992202758,
      "epoch": 0.49074746008708275,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004981933810882594,
      "loss": 5.342,
      "mean_token_accuracy": 0.17635241597890855,
      "num_tokens": 11249560.0,
      "step": 5410
    },
    {
      "entropy": 5.721574974060059,
      "epoch": 0.4912010159651669,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004981892867020917,
      "loss": 5.5447,
      "mean_token_accuracy": 0.1738561399281025,
      "num_tokens": 11259787.0,
      "step": 5415
    },
    {
      "entropy": 5.6990868091583256,
      "epoch": 0.49165457184325106,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981851877003125,
      "loss": 5.4694,
      "mean_token_accuracy": 0.17295820415019988,
      "num_tokens": 11269717.0,
      "step": 5420
    },
    {
      "entropy": 5.580885410308838,
      "epoch": 0.4921081277213353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981810840830068,
      "loss": 5.4482,
      "mean_token_accuracy": 0.18043429851531984,
      "num_tokens": 11279863.0,
      "step": 5425
    },
    {
      "entropy": 5.6650306224823,
      "epoch": 0.49256168359941943,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981769758502596,
      "loss": 5.5763,
      "mean_token_accuracy": 0.16078406423330308,
      "num_tokens": 11291161.0,
      "step": 5430
    },
    {
      "entropy": 5.6055395126342775,
      "epoch": 0.49301523947750364,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981728630021555,
      "loss": 5.3956,
      "mean_token_accuracy": 0.17572860419750214,
      "num_tokens": 11301151.0,
      "step": 5435
    },
    {
      "entropy": 5.575085067749024,
      "epoch": 0.4934687953555878,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004981687455387797,
      "loss": 5.3946,
      "mean_token_accuracy": 0.1716492608189583,
      "num_tokens": 11311466.0,
      "step": 5440
    },
    {
      "entropy": 5.647791528701783,
      "epoch": 0.493922351233672,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004981646234602173,
      "loss": 5.3629,
      "mean_token_accuracy": 0.1757534921169281,
      "num_tokens": 11321522.0,
      "step": 5445
    },
    {
      "entropy": 5.705031681060791,
      "epoch": 0.49437590711175616,
      "grad_norm": 0.875,
      "learning_rate": 0.0004981604967665539,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17398695945739745,
      "num_tokens": 11332317.0,
      "step": 5450
    },
    {
      "entropy": 5.611266326904297,
      "epoch": 0.4948294629898404,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004981563654578743,
      "loss": 5.4223,
      "mean_token_accuracy": 0.16848283857107163,
      "num_tokens": 11343024.0,
      "step": 5455
    },
    {
      "entropy": 5.585212087631225,
      "epoch": 0.49528301886792453,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981522295342643,
      "loss": 5.381,
      "mean_token_accuracy": 0.17383768558502197,
      "num_tokens": 11353114.0,
      "step": 5460
    },
    {
      "entropy": 5.634729814529419,
      "epoch": 0.4957365747460087,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004981480889958092,
      "loss": 5.4311,
      "mean_token_accuracy": 0.17489740252494812,
      "num_tokens": 11363446.0,
      "step": 5465
    },
    {
      "entropy": 5.55680570602417,
      "epoch": 0.4961901306240929,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981439438425949,
      "loss": 5.4316,
      "mean_token_accuracy": 0.17906904518604277,
      "num_tokens": 11373413.0,
      "step": 5470
    },
    {
      "entropy": 5.541783714294434,
      "epoch": 0.49664368650217705,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498139794074707,
      "loss": 5.3189,
      "mean_token_accuracy": 0.180606409907341,
      "num_tokens": 11383142.0,
      "step": 5475
    },
    {
      "entropy": 5.681115627288818,
      "epoch": 0.49709724238026126,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981356396922313,
      "loss": 5.4639,
      "mean_token_accuracy": 0.17151203900575637,
      "num_tokens": 11394635.0,
      "step": 5480
    },
    {
      "entropy": 5.54714994430542,
      "epoch": 0.4975507982583454,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004981314806952535,
      "loss": 5.3762,
      "mean_token_accuracy": 0.17595487385988234,
      "num_tokens": 11404442.0,
      "step": 5485
    },
    {
      "entropy": 5.550731897354126,
      "epoch": 0.49800435413642963,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049812731708386,
      "loss": 5.3204,
      "mean_token_accuracy": 0.18081480413675308,
      "num_tokens": 11414217.0,
      "step": 5490
    },
    {
      "entropy": 5.680741786956787,
      "epoch": 0.4984579100145138,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981231488581367,
      "loss": 5.456,
      "mean_token_accuracy": 0.17206876575946808,
      "num_tokens": 11423834.0,
      "step": 5495
    },
    {
      "entropy": 5.532857179641724,
      "epoch": 0.498911465892598,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004981189760181699,
      "loss": 5.3341,
      "mean_token_accuracy": 0.17510698288679122,
      "num_tokens": 11433613.0,
      "step": 5500
    },
    {
      "entropy": 5.559586668014527,
      "epoch": 0.49936502177068215,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004981147985640456,
      "loss": 5.412,
      "mean_token_accuracy": 0.17702824473381043,
      "num_tokens": 11443915.0,
      "step": 5505
    },
    {
      "entropy": 5.699520635604858,
      "epoch": 0.4998185776487663,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004981106164958506,
      "loss": 5.4326,
      "mean_token_accuracy": 0.17155378311872482,
      "num_tokens": 11454588.0,
      "step": 5510
    },
    {
      "entropy": 5.580562543869019,
      "epoch": 0.5002721335268505,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004981064298136712,
      "loss": 5.4901,
      "mean_token_accuracy": 0.17163256406784058,
      "num_tokens": 11465533.0,
      "step": 5515
    },
    {
      "entropy": 5.621740531921387,
      "epoch": 0.5007256894049347,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004981022385175939,
      "loss": 5.4445,
      "mean_token_accuracy": 0.1629154846072197,
      "num_tokens": 11475776.0,
      "step": 5520
    },
    {
      "entropy": 5.647769451141357,
      "epoch": 0.5011792452830188,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980980426077054,
      "loss": 5.417,
      "mean_token_accuracy": 0.18146403580904008,
      "num_tokens": 11486200.0,
      "step": 5525
    },
    {
      "entropy": 5.586569929122925,
      "epoch": 0.5016328011611031,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004980938420840926,
      "loss": 5.4449,
      "mean_token_accuracy": 0.16791134476661682,
      "num_tokens": 11496768.0,
      "step": 5530
    },
    {
      "entropy": 5.628451347351074,
      "epoch": 0.5020863570391872,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980896369468423,
      "loss": 5.4208,
      "mean_token_accuracy": 0.17588136047124864,
      "num_tokens": 11507096.0,
      "step": 5535
    },
    {
      "entropy": 5.637207078933716,
      "epoch": 0.5025399129172714,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004980854271960415,
      "loss": 5.4631,
      "mean_token_accuracy": 0.17342848032712938,
      "num_tokens": 11517301.0,
      "step": 5540
    },
    {
      "entropy": 5.666760635375977,
      "epoch": 0.5029934687953556,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004980812128317772,
      "loss": 5.5259,
      "mean_token_accuracy": 0.1743517428636551,
      "num_tokens": 11528616.0,
      "step": 5545
    },
    {
      "entropy": 5.676811027526855,
      "epoch": 0.5034470246734397,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004980769938541364,
      "loss": 5.4487,
      "mean_token_accuracy": 0.17927418798208236,
      "num_tokens": 11539344.0,
      "step": 5550
    },
    {
      "entropy": 5.635175704956055,
      "epoch": 0.503900580551524,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980727702632066,
      "loss": 5.4897,
      "mean_token_accuracy": 0.17068908959627152,
      "num_tokens": 11549511.0,
      "step": 5555
    },
    {
      "entropy": 5.6430299282073975,
      "epoch": 0.5043541364296081,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980685420590752,
      "loss": 5.4369,
      "mean_token_accuracy": 0.17500991076231004,
      "num_tokens": 11559425.0,
      "step": 5560
    },
    {
      "entropy": 5.707560014724732,
      "epoch": 0.5048076923076923,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004980643092418294,
      "loss": 5.5405,
      "mean_token_accuracy": 0.16627538055181504,
      "num_tokens": 11570406.0,
      "step": 5565
    },
    {
      "entropy": 5.634152126312256,
      "epoch": 0.5052612481857764,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004980600718115568,
      "loss": 5.4295,
      "mean_token_accuracy": 0.175694976747036,
      "num_tokens": 11580509.0,
      "step": 5570
    },
    {
      "entropy": 5.635722541809082,
      "epoch": 0.5057148040638607,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004980558297683452,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1730503410100937,
      "num_tokens": 11590860.0,
      "step": 5575
    },
    {
      "entropy": 5.6468902111053465,
      "epoch": 0.5061683599419449,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004980515831122822,
      "loss": 5.4468,
      "mean_token_accuracy": 0.17353652268648148,
      "num_tokens": 11601954.0,
      "step": 5580
    },
    {
      "entropy": 5.652958011627197,
      "epoch": 0.506621915820029,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004980473318434555,
      "loss": 5.4462,
      "mean_token_accuracy": 0.17210652977228164,
      "num_tokens": 11612581.0,
      "step": 5585
    },
    {
      "entropy": 5.573605346679687,
      "epoch": 0.5070754716981132,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980430759619532,
      "loss": 5.3652,
      "mean_token_accuracy": 0.17937352061271666,
      "num_tokens": 11622599.0,
      "step": 5590
    },
    {
      "entropy": 5.543113279342651,
      "epoch": 0.5075290275761973,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980388154678633,
      "loss": 5.3908,
      "mean_token_accuracy": 0.17439862638711928,
      "num_tokens": 11633611.0,
      "step": 5595
    },
    {
      "entropy": 5.585327100753784,
      "epoch": 0.5079825834542816,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004980345503612738,
      "loss": 5.365,
      "mean_token_accuracy": 0.17676355540752411,
      "num_tokens": 11643159.0,
      "step": 5600
    },
    {
      "entropy": 5.5928394317626955,
      "epoch": 0.5084361393323658,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980302806422729,
      "loss": 5.3449,
      "mean_token_accuracy": 0.177506685256958,
      "num_tokens": 11653174.0,
      "step": 5605
    },
    {
      "entropy": 5.6118120670318605,
      "epoch": 0.5088896952104499,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980260063109491,
      "loss": 5.4038,
      "mean_token_accuracy": 0.17569431215524672,
      "num_tokens": 11662621.0,
      "step": 5610
    },
    {
      "entropy": 5.528201961517334,
      "epoch": 0.5093432510885341,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004980217273673906,
      "loss": 5.3599,
      "mean_token_accuracy": 0.17619221061468124,
      "num_tokens": 11673538.0,
      "step": 5615
    },
    {
      "entropy": 5.644331693649292,
      "epoch": 0.5097968069666183,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980174438116859,
      "loss": 5.3752,
      "mean_token_accuracy": 0.1794475644826889,
      "num_tokens": 11683203.0,
      "step": 5620
    },
    {
      "entropy": 5.552001285552978,
      "epoch": 0.5102503628447025,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980131556439236,
      "loss": 5.3636,
      "mean_token_accuracy": 0.18283542692661287,
      "num_tokens": 11693313.0,
      "step": 5625
    },
    {
      "entropy": 5.612477016448975,
      "epoch": 0.5107039187227866,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004980088628641925,
      "loss": 5.4534,
      "mean_token_accuracy": 0.17441582083702087,
      "num_tokens": 11704194.0,
      "step": 5630
    },
    {
      "entropy": 5.56378436088562,
      "epoch": 0.5111574746008708,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004980045654725812,
      "loss": 5.3146,
      "mean_token_accuracy": 0.17968470156192778,
      "num_tokens": 11715109.0,
      "step": 5635
    },
    {
      "entropy": 5.534660005569458,
      "epoch": 0.511611030478955,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980002634691787,
      "loss": 5.3489,
      "mean_token_accuracy": 0.18387647569179535,
      "num_tokens": 11724961.0,
      "step": 5640
    },
    {
      "entropy": 5.55992374420166,
      "epoch": 0.5120645863570392,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497995956854074,
      "loss": 5.3785,
      "mean_token_accuracy": 0.1765351787209511,
      "num_tokens": 11735334.0,
      "step": 5645
    },
    {
      "entropy": 5.569534349441528,
      "epoch": 0.5125181422351234,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004979916456273561,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1730036035180092,
      "num_tokens": 11746858.0,
      "step": 5650
    },
    {
      "entropy": 5.661505699157715,
      "epoch": 0.5129716981132075,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979873297891141,
      "loss": 5.4903,
      "mean_token_accuracy": 0.17127009034156798,
      "num_tokens": 11757819.0,
      "step": 5655
    },
    {
      "entropy": 5.618195486068726,
      "epoch": 0.5134252539912917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979830093394372,
      "loss": 5.4053,
      "mean_token_accuracy": 0.17758016139268876,
      "num_tokens": 11767470.0,
      "step": 5660
    },
    {
      "entropy": 5.549928283691406,
      "epoch": 0.513878809869376,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004979786842784149,
      "loss": 5.3273,
      "mean_token_accuracy": 0.1853337362408638,
      "num_tokens": 11777060.0,
      "step": 5665
    },
    {
      "entropy": 5.577957963943481,
      "epoch": 0.5143323657474601,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979743546061366,
      "loss": 5.3962,
      "mean_token_accuracy": 0.17992057055234909,
      "num_tokens": 11787176.0,
      "step": 5670
    },
    {
      "entropy": 5.539264678955078,
      "epoch": 0.5147859216255443,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979700203226918,
      "loss": 5.2699,
      "mean_token_accuracy": 0.18505061715841292,
      "num_tokens": 11796419.0,
      "step": 5675
    },
    {
      "entropy": 5.661307001113892,
      "epoch": 0.5152394775036284,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004979656814281702,
      "loss": 5.4999,
      "mean_token_accuracy": 0.17273307591676712,
      "num_tokens": 11807279.0,
      "step": 5680
    },
    {
      "entropy": 5.561037874221801,
      "epoch": 0.5156930333817126,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979613379226615,
      "loss": 5.3961,
      "mean_token_accuracy": 0.17279176861047746,
      "num_tokens": 11817669.0,
      "step": 5685
    },
    {
      "entropy": 5.53184027671814,
      "epoch": 0.5161465892597968,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979569898062554,
      "loss": 5.3938,
      "mean_token_accuracy": 0.18203373849391938,
      "num_tokens": 11827247.0,
      "step": 5690
    },
    {
      "entropy": 5.618873500823975,
      "epoch": 0.516600145137881,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497952637079042,
      "loss": 5.3803,
      "mean_token_accuracy": 0.17191717773675919,
      "num_tokens": 11838201.0,
      "step": 5695
    },
    {
      "entropy": 5.568843269348145,
      "epoch": 0.5170537010159652,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979482797411114,
      "loss": 5.322,
      "mean_token_accuracy": 0.18213216811418534,
      "num_tokens": 11847561.0,
      "step": 5700
    },
    {
      "entropy": 5.6255693435668945,
      "epoch": 0.5175072568940493,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979439177925534,
      "loss": 5.4481,
      "mean_token_accuracy": 0.180612213909626,
      "num_tokens": 11858272.0,
      "step": 5705
    },
    {
      "entropy": 5.612488603591919,
      "epoch": 0.5179608127721336,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004979395512334584,
      "loss": 5.4823,
      "mean_token_accuracy": 0.17200103104114534,
      "num_tokens": 11869660.0,
      "step": 5710
    },
    {
      "entropy": 5.6795634746551515,
      "epoch": 0.5184143686502177,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004979351800639166,
      "loss": 5.5096,
      "mean_token_accuracy": 0.1676321730017662,
      "num_tokens": 11880239.0,
      "step": 5715
    },
    {
      "entropy": 5.712689352035523,
      "epoch": 0.5188679245283019,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979308042840186,
      "loss": 5.4816,
      "mean_token_accuracy": 0.16780400574207305,
      "num_tokens": 11890630.0,
      "step": 5720
    },
    {
      "entropy": 5.669008064270019,
      "epoch": 0.519321480406386,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004979264238938547,
      "loss": 5.5112,
      "mean_token_accuracy": 0.1703075349330902,
      "num_tokens": 11900948.0,
      "step": 5725
    },
    {
      "entropy": 5.541024541854858,
      "epoch": 0.5197750362844702,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004979220388935155,
      "loss": 5.3925,
      "mean_token_accuracy": 0.17539578080177307,
      "num_tokens": 11912528.0,
      "step": 5730
    },
    {
      "entropy": 5.5685014724731445,
      "epoch": 0.5202285921625545,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004979176492830918,
      "loss": 5.353,
      "mean_token_accuracy": 0.17433730214834214,
      "num_tokens": 11922211.0,
      "step": 5735
    },
    {
      "entropy": 5.598578691482544,
      "epoch": 0.5206821480406386,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979132550626742,
      "loss": 5.4084,
      "mean_token_accuracy": 0.17142235189676286,
      "num_tokens": 11932520.0,
      "step": 5740
    },
    {
      "entropy": 5.59015007019043,
      "epoch": 0.5211357039187228,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979088562323538,
      "loss": 5.3221,
      "mean_token_accuracy": 0.18398308157920837,
      "num_tokens": 11943312.0,
      "step": 5745
    },
    {
      "entropy": 5.559282016754151,
      "epoch": 0.5215892597968069,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979044527922214,
      "loss": 5.4661,
      "mean_token_accuracy": 0.17238451689481735,
      "num_tokens": 11954455.0,
      "step": 5750
    },
    {
      "entropy": 5.626161718368531,
      "epoch": 0.5220428156748912,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979000447423681,
      "loss": 5.3677,
      "mean_token_accuracy": 0.18028335571289061,
      "num_tokens": 11964650.0,
      "step": 5755
    },
    {
      "entropy": 5.579872035980225,
      "epoch": 0.5224963715529753,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004978956320828851,
      "loss": 5.4013,
      "mean_token_accuracy": 0.17466039061546326,
      "num_tokens": 11975134.0,
      "step": 5760
    },
    {
      "entropy": 5.607253074645996,
      "epoch": 0.5229499274310595,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004978912148138637,
      "loss": 5.45,
      "mean_token_accuracy": 0.17419570237398146,
      "num_tokens": 11986486.0,
      "step": 5765
    },
    {
      "entropy": 5.655194902420044,
      "epoch": 0.5234034833091437,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497886792935395,
      "loss": 5.4488,
      "mean_token_accuracy": 0.1717308521270752,
      "num_tokens": 11998045.0,
      "step": 5770
    },
    {
      "entropy": 5.663045930862427,
      "epoch": 0.5238570391872278,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004978823664475707,
      "loss": 5.4513,
      "mean_token_accuracy": 0.17971017062664033,
      "num_tokens": 12008130.0,
      "step": 5775
    },
    {
      "entropy": 5.596103620529175,
      "epoch": 0.5243105950653121,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004978779353504824,
      "loss": 5.5171,
      "mean_token_accuracy": 0.16514042466878892,
      "num_tokens": 12018753.0,
      "step": 5780
    },
    {
      "entropy": 5.621715688705445,
      "epoch": 0.5247641509433962,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004978734996442214,
      "loss": 5.3943,
      "mean_token_accuracy": 0.18264525085687638,
      "num_tokens": 12028202.0,
      "step": 5785
    },
    {
      "entropy": 5.640986490249634,
      "epoch": 0.5252177068214804,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004978690593288798,
      "loss": 5.4546,
      "mean_token_accuracy": 0.17733746469020845,
      "num_tokens": 12039652.0,
      "step": 5790
    },
    {
      "entropy": 5.637750244140625,
      "epoch": 0.5256712626995645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978646144045491,
      "loss": 5.3854,
      "mean_token_accuracy": 0.17197972387075425,
      "num_tokens": 12051269.0,
      "step": 5795
    },
    {
      "entropy": 5.482382678985596,
      "epoch": 0.5261248185776488,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978601648713215,
      "loss": 5.2795,
      "mean_token_accuracy": 0.17774615287780762,
      "num_tokens": 12060647.0,
      "step": 5800
    },
    {
      "entropy": 5.569826936721801,
      "epoch": 0.526578374455733,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978557107292889,
      "loss": 5.408,
      "mean_token_accuracy": 0.17919009327888488,
      "num_tokens": 12071669.0,
      "step": 5805
    },
    {
      "entropy": 5.547264099121094,
      "epoch": 0.5270319303338171,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004978512519785433,
      "loss": 5.282,
      "mean_token_accuracy": 0.18562869876623153,
      "num_tokens": 12081971.0,
      "step": 5810
    },
    {
      "entropy": 5.5526330947875975,
      "epoch": 0.5274854862119013,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000497846788619177,
      "loss": 5.4195,
      "mean_token_accuracy": 0.17687285095453262,
      "num_tokens": 12092465.0,
      "step": 5815
    },
    {
      "entropy": 5.617220735549926,
      "epoch": 0.5279390420899854,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004978423206512825,
      "loss": 5.3735,
      "mean_token_accuracy": 0.17665890902280806,
      "num_tokens": 12101752.0,
      "step": 5820
    },
    {
      "entropy": 5.661663913726807,
      "epoch": 0.5283925979680697,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497837848074952,
      "loss": 5.456,
      "mean_token_accuracy": 0.17533332258462905,
      "num_tokens": 12111805.0,
      "step": 5825
    },
    {
      "entropy": 5.665722894668579,
      "epoch": 0.5288461538461539,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000497833370890278,
      "loss": 5.4739,
      "mean_token_accuracy": 0.17663151323795317,
      "num_tokens": 12122511.0,
      "step": 5830
    },
    {
      "entropy": 5.596258163452148,
      "epoch": 0.529299709724238,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004978288890973529,
      "loss": 5.3516,
      "mean_token_accuracy": 0.17615063041448592,
      "num_tokens": 12132510.0,
      "step": 5835
    },
    {
      "entropy": 5.578837966918945,
      "epoch": 0.5297532656023222,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978244026962698,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1739246740937233,
      "num_tokens": 12142419.0,
      "step": 5840
    },
    {
      "entropy": 5.594649934768677,
      "epoch": 0.5302068214804064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978199116871212,
      "loss": 5.4149,
      "mean_token_accuracy": 0.17248719334602355,
      "num_tokens": 12152759.0,
      "step": 5845
    },
    {
      "entropy": 5.554829692840576,
      "epoch": 0.5306603773584906,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00049781541607,
      "loss": 5.3653,
      "mean_token_accuracy": 0.17687786668539046,
      "num_tokens": 12162758.0,
      "step": 5850
    },
    {
      "entropy": 5.583818531036377,
      "epoch": 0.5311139332365747,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004978109158449993,
      "loss": 5.4229,
      "mean_token_accuracy": 0.17397976219654082,
      "num_tokens": 12172664.0,
      "step": 5855
    },
    {
      "entropy": 5.605308771133423,
      "epoch": 0.5315674891146589,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004978064110122121,
      "loss": 5.3847,
      "mean_token_accuracy": 0.17937863022089004,
      "num_tokens": 12182028.0,
      "step": 5860
    },
    {
      "entropy": 5.624365615844726,
      "epoch": 0.532021044992743,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978019015717314,
      "loss": 5.4322,
      "mean_token_accuracy": 0.17571227252483368,
      "num_tokens": 12192168.0,
      "step": 5865
    },
    {
      "entropy": 5.526874589920044,
      "epoch": 0.5324746008708273,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004977973875236508,
      "loss": 5.3325,
      "mean_token_accuracy": 0.17963843196630477,
      "num_tokens": 12202889.0,
      "step": 5870
    },
    {
      "entropy": 5.54812593460083,
      "epoch": 0.5329281567489115,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004977928688680634,
      "loss": 5.3732,
      "mean_token_accuracy": 0.18322223126888276,
      "num_tokens": 12212514.0,
      "step": 5875
    },
    {
      "entropy": 5.584365177154541,
      "epoch": 0.5333817126269956,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004977883456050626,
      "loss": 5.247,
      "mean_token_accuracy": 0.1861517012119293,
      "num_tokens": 12222353.0,
      "step": 5880
    },
    {
      "entropy": 5.6489053726196286,
      "epoch": 0.5338352685050798,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000497783817734742,
      "loss": 5.5219,
      "mean_token_accuracy": 0.16939357966184615,
      "num_tokens": 12232845.0,
      "step": 5885
    },
    {
      "entropy": 5.61934814453125,
      "epoch": 0.534288824383164,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004977792852571954,
      "loss": 5.45,
      "mean_token_accuracy": 0.16973866671323776,
      "num_tokens": 12242950.0,
      "step": 5890
    },
    {
      "entropy": 5.633284950256348,
      "epoch": 0.5347423802612482,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004977747481725164,
      "loss": 5.4321,
      "mean_token_accuracy": 0.1735300526022911,
      "num_tokens": 12253132.0,
      "step": 5895
    },
    {
      "entropy": 5.599405288696289,
      "epoch": 0.5351959361393324,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977702064807987,
      "loss": 5.3785,
      "mean_token_accuracy": 0.17767370939254762,
      "num_tokens": 12263214.0,
      "step": 5900
    },
    {
      "entropy": 5.526389789581299,
      "epoch": 0.5356494920174165,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977656601821366,
      "loss": 5.3095,
      "mean_token_accuracy": 0.18471042364835738,
      "num_tokens": 12273250.0,
      "step": 5905
    },
    {
      "entropy": 5.716156530380249,
      "epoch": 0.5361030478955007,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004977611092766238,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1808107778429985,
      "num_tokens": 12283767.0,
      "step": 5910
    },
    {
      "entropy": 5.523726463317871,
      "epoch": 0.5365566037735849,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004977565537643544,
      "loss": 5.3421,
      "mean_token_accuracy": 0.17266201227903366,
      "num_tokens": 12294612.0,
      "step": 5915
    },
    {
      "entropy": 5.554253149032593,
      "epoch": 0.5370101596516691,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004977519936454227,
      "loss": 5.4225,
      "mean_token_accuracy": 0.18037798404693603,
      "num_tokens": 12305614.0,
      "step": 5920
    },
    {
      "entropy": 5.550849533081054,
      "epoch": 0.5374637155297532,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977474289199231,
      "loss": 5.2966,
      "mean_token_accuracy": 0.17565731406211854,
      "num_tokens": 12316152.0,
      "step": 5925
    },
    {
      "entropy": 5.522589588165284,
      "epoch": 0.5379172714078374,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977428595879499,
      "loss": 5.3751,
      "mean_token_accuracy": 0.17792995423078536,
      "num_tokens": 12325940.0,
      "step": 5930
    },
    {
      "entropy": 5.570098352432251,
      "epoch": 0.5383708272859217,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977382856495975,
      "loss": 5.3606,
      "mean_token_accuracy": 0.17704337388277053,
      "num_tokens": 12335482.0,
      "step": 5935
    },
    {
      "entropy": 5.6550994396209715,
      "epoch": 0.5388243831640058,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977337071049606,
      "loss": 5.4564,
      "mean_token_accuracy": 0.17251523584127426,
      "num_tokens": 12346386.0,
      "step": 5940
    },
    {
      "entropy": 5.668819808959961,
      "epoch": 0.53927793904209,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497729123954134,
      "loss": 5.4197,
      "mean_token_accuracy": 0.1771680533885956,
      "num_tokens": 12356282.0,
      "step": 5945
    },
    {
      "entropy": 5.580939483642578,
      "epoch": 0.5397314949201741,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004977245361972123,
      "loss": 5.3956,
      "mean_token_accuracy": 0.17961778491735458,
      "num_tokens": 12366994.0,
      "step": 5950
    },
    {
      "entropy": 5.550912141799927,
      "epoch": 0.5401850507982583,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004977199438342904,
      "loss": 5.2834,
      "mean_token_accuracy": 0.1829832002520561,
      "num_tokens": 12377524.0,
      "step": 5955
    },
    {
      "entropy": 5.6025049686431885,
      "epoch": 0.5406386066763426,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004977153468654634,
      "loss": 5.4502,
      "mean_token_accuracy": 0.18007041662931442,
      "num_tokens": 12387593.0,
      "step": 5960
    },
    {
      "entropy": 5.677999258041382,
      "epoch": 0.5410921625544267,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004977107452908262,
      "loss": 5.4079,
      "mean_token_accuracy": 0.18105369359254836,
      "num_tokens": 12398732.0,
      "step": 5965
    },
    {
      "entropy": 5.5877197265625,
      "epoch": 0.5415457184325109,
      "grad_norm": 1.0,
      "learning_rate": 0.000497706139110474,
      "loss": 5.405,
      "mean_token_accuracy": 0.18087859451770782,
      "num_tokens": 12409821.0,
      "step": 5970
    },
    {
      "entropy": 5.5568990230560305,
      "epoch": 0.541999274310595,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977015283245021,
      "loss": 5.4217,
      "mean_token_accuracy": 0.17138040512800218,
      "num_tokens": 12420044.0,
      "step": 5975
    },
    {
      "entropy": 5.519027042388916,
      "epoch": 0.5424528301886793,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004976969129330059,
      "loss": 5.3454,
      "mean_token_accuracy": 0.1738667070865631,
      "num_tokens": 12430980.0,
      "step": 5980
    },
    {
      "entropy": 5.6030182361602785,
      "epoch": 0.5429063860667634,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004976922929360808,
      "loss": 5.3948,
      "mean_token_accuracy": 0.18232907652854918,
      "num_tokens": 12441999.0,
      "step": 5985
    },
    {
      "entropy": 5.621866083145141,
      "epoch": 0.5433599419448476,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004976876683338222,
      "loss": 5.4362,
      "mean_token_accuracy": 0.17455211877822877,
      "num_tokens": 12452837.0,
      "step": 5990
    },
    {
      "entropy": 5.572230195999145,
      "epoch": 0.5438134978229318,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000497683039126326,
      "loss": 5.3547,
      "mean_token_accuracy": 0.18271825313568116,
      "num_tokens": 12462142.0,
      "step": 5995
    },
    {
      "entropy": 5.600415372848511,
      "epoch": 0.5442670537010159,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004976784053136878,
      "loss": 5.4043,
      "mean_token_accuracy": 0.17925617694854737,
      "num_tokens": 12472756.0,
      "step": 6000
    },
    {
      "epoch": 0.5442670537010159,
      "eval_entropy": 5.379152549122254,
      "eval_loss": 5.41491174697876,
      "eval_mean_token_accuracy": 0.18314881672433733,
      "eval_num_tokens": 12472756.0,
      "eval_runtime": 20.6957,
      "eval_samples_per_second": 1708.571,
      "eval_steps_per_second": 213.571,
      "step": 6000
    },
    {
      "entropy": 5.644152069091797,
      "epoch": 0.5447206095791002,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004976737668960034,
      "loss": 5.5029,
      "mean_token_accuracy": 0.17155803442001344,
      "num_tokens": 12483115.0,
      "step": 6005
    },
    {
      "entropy": 5.652310562133789,
      "epoch": 0.5451741654571843,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004976691238733688,
      "loss": 5.3646,
      "mean_token_accuracy": 0.17381614148616792,
      "num_tokens": 12493280.0,
      "step": 6010
    },
    {
      "entropy": 5.596336126327515,
      "epoch": 0.5456277213352685,
      "grad_norm": 0.953125,
      "learning_rate": 0.00049766447624588,
      "loss": 5.5069,
      "mean_token_accuracy": 0.1748632788658142,
      "num_tokens": 12503665.0,
      "step": 6015
    },
    {
      "entropy": 5.5472839832305905,
      "epoch": 0.5460812772133526,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497659824013633,
      "loss": 5.4058,
      "mean_token_accuracy": 0.17200566828250885,
      "num_tokens": 12513381.0,
      "step": 6020
    },
    {
      "entropy": 5.680693578720093,
      "epoch": 0.5465348330914369,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004976551671767241,
      "loss": 5.4393,
      "mean_token_accuracy": 0.17311064749956132,
      "num_tokens": 12524121.0,
      "step": 6025
    },
    {
      "entropy": 5.601163005828857,
      "epoch": 0.5469883889695211,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004976505057352496,
      "loss": 5.4147,
      "mean_token_accuracy": 0.18124686181545258,
      "num_tokens": 12533972.0,
      "step": 6030
    },
    {
      "entropy": 5.644373226165771,
      "epoch": 0.5474419448476052,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004976458396893059,
      "loss": 5.5213,
      "mean_token_accuracy": 0.16715265363454818,
      "num_tokens": 12544110.0,
      "step": 6035
    },
    {
      "entropy": 5.5955469608306885,
      "epoch": 0.5478955007256894,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004976411690389896,
      "loss": 5.3403,
      "mean_token_accuracy": 0.17307349890470505,
      "num_tokens": 12554273.0,
      "step": 6040
    },
    {
      "entropy": 5.661379146575928,
      "epoch": 0.5483490566037735,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497636493784397,
      "loss": 5.4215,
      "mean_token_accuracy": 0.17500090450048447,
      "num_tokens": 12564633.0,
      "step": 6045
    },
    {
      "entropy": 5.586629152297974,
      "epoch": 0.5488026124818578,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976318139256251,
      "loss": 5.3537,
      "mean_token_accuracy": 0.178970268368721,
      "num_tokens": 12573790.0,
      "step": 6050
    },
    {
      "entropy": 5.6261004447937015,
      "epoch": 0.549256168359942,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004976271294627706,
      "loss": 5.4317,
      "mean_token_accuracy": 0.17064266204833983,
      "num_tokens": 12584265.0,
      "step": 6055
    },
    {
      "entropy": 5.570170545578003,
      "epoch": 0.5497097242380261,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004976224403959303,
      "loss": 5.2914,
      "mean_token_accuracy": 0.1787312373518944,
      "num_tokens": 12594413.0,
      "step": 6060
    },
    {
      "entropy": 5.510761976242065,
      "epoch": 0.5501632801161103,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976177467252012,
      "loss": 5.3683,
      "mean_token_accuracy": 0.18546044677495957,
      "num_tokens": 12604637.0,
      "step": 6065
    },
    {
      "entropy": 5.589938545227051,
      "epoch": 0.5506168359941945,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004976130484506804,
      "loss": 5.4391,
      "mean_token_accuracy": 0.1787260204553604,
      "num_tokens": 12614873.0,
      "step": 6070
    },
    {
      "entropy": 5.601373863220215,
      "epoch": 0.5510703918722787,
      "grad_norm": 0.875,
      "learning_rate": 0.0004976083455724649,
      "loss": 5.3712,
      "mean_token_accuracy": 0.1769331619143486,
      "num_tokens": 12626182.0,
      "step": 6075
    },
    {
      "entropy": 5.540160465240478,
      "epoch": 0.5515239477503628,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976036380906523,
      "loss": 5.3062,
      "mean_token_accuracy": 0.1818635180592537,
      "num_tokens": 12635418.0,
      "step": 6080
    },
    {
      "entropy": 5.5531182289123535,
      "epoch": 0.551977503628447,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004975989260053395,
      "loss": 5.4516,
      "mean_token_accuracy": 0.16748463958501816,
      "num_tokens": 12645942.0,
      "step": 6085
    },
    {
      "entropy": 5.6305876731872555,
      "epoch": 0.5524310595065312,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004975942093166242,
      "loss": 5.3756,
      "mean_token_accuracy": 0.17721232026815414,
      "num_tokens": 12656460.0,
      "step": 6090
    },
    {
      "entropy": 5.610075330734253,
      "epoch": 0.5528846153846154,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975894880246041,
      "loss": 5.3241,
      "mean_token_accuracy": 0.17855470925569533,
      "num_tokens": 12666916.0,
      "step": 6095
    },
    {
      "entropy": 5.539130449295044,
      "epoch": 0.5533381712626996,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004975847621293766,
      "loss": 5.3682,
      "mean_token_accuracy": 0.18464813530445098,
      "num_tokens": 12677562.0,
      "step": 6100
    },
    {
      "entropy": 5.5453942775726315,
      "epoch": 0.5537917271407837,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004975800316310394,
      "loss": 5.3443,
      "mean_token_accuracy": 0.1776389956474304,
      "num_tokens": 12687349.0,
      "step": 6105
    },
    {
      "entropy": 5.601655769348144,
      "epoch": 0.5542452830188679,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004975752965296906,
      "loss": 5.4005,
      "mean_token_accuracy": 0.17570678889751434,
      "num_tokens": 12697850.0,
      "step": 6110
    },
    {
      "entropy": 5.50960373878479,
      "epoch": 0.5546988388969522,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975705568254278,
      "loss": 5.2554,
      "mean_token_accuracy": 0.1816887453198433,
      "num_tokens": 12708283.0,
      "step": 6115
    },
    {
      "entropy": 5.553288125991822,
      "epoch": 0.5551523947750363,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004975658125183493,
      "loss": 5.3604,
      "mean_token_accuracy": 0.1766258791089058,
      "num_tokens": 12718201.0,
      "step": 6120
    },
    {
      "entropy": 5.5428098201751705,
      "epoch": 0.5556059506531205,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497561063608553,
      "loss": 5.3093,
      "mean_token_accuracy": 0.18314361572265625,
      "num_tokens": 12728860.0,
      "step": 6125
    },
    {
      "entropy": 5.565022611618042,
      "epoch": 0.5560595065312046,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004975563100961372,
      "loss": 5.3704,
      "mean_token_accuracy": 0.17591018676757814,
      "num_tokens": 12739007.0,
      "step": 6130
    },
    {
      "entropy": 5.546350717544556,
      "epoch": 0.5565130624092888,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004975515519812002,
      "loss": 5.2309,
      "mean_token_accuracy": 0.1811644896864891,
      "num_tokens": 12749004.0,
      "step": 6135
    },
    {
      "entropy": 5.555451583862305,
      "epoch": 0.556966618287373,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004975467892638402,
      "loss": 5.3761,
      "mean_token_accuracy": 0.1802150845527649,
      "num_tokens": 12760085.0,
      "step": 6140
    },
    {
      "entropy": 5.568343782424927,
      "epoch": 0.5574201741654572,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004975420219441561,
      "loss": 5.3294,
      "mean_token_accuracy": 0.18599051386117935,
      "num_tokens": 12770303.0,
      "step": 6145
    },
    {
      "entropy": 5.574814367294311,
      "epoch": 0.5578737300435413,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975372500222462,
      "loss": 5.3613,
      "mean_token_accuracy": 0.18204200863838196,
      "num_tokens": 12780334.0,
      "step": 6150
    },
    {
      "entropy": 5.568981695175171,
      "epoch": 0.5583272859216255,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975324734982094,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1729448139667511,
      "num_tokens": 12790980.0,
      "step": 6155
    },
    {
      "entropy": 5.61474723815918,
      "epoch": 0.5587808417997098,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975276923721442,
      "loss": 5.4477,
      "mean_token_accuracy": 0.17725956439971924,
      "num_tokens": 12801713.0,
      "step": 6160
    },
    {
      "entropy": 5.613516807556152,
      "epoch": 0.5592343976777939,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004975229066441496,
      "loss": 5.4632,
      "mean_token_accuracy": 0.17113297879695893,
      "num_tokens": 12813040.0,
      "step": 6165
    },
    {
      "entropy": 5.557818269729614,
      "epoch": 0.5596879535558781,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004975181163143247,
      "loss": 5.3897,
      "mean_token_accuracy": 0.17339551895856858,
      "num_tokens": 12824048.0,
      "step": 6170
    },
    {
      "entropy": 5.607717227935791,
      "epoch": 0.5601415094339622,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004975133213827684,
      "loss": 5.3425,
      "mean_token_accuracy": 0.17274201065301895,
      "num_tokens": 12834297.0,
      "step": 6175
    },
    {
      "entropy": 5.6809532165527346,
      "epoch": 0.5605950653120464,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004975085218495798,
      "loss": 5.4172,
      "mean_token_accuracy": 0.17106309533119202,
      "num_tokens": 12844574.0,
      "step": 6180
    },
    {
      "entropy": 5.524085903167725,
      "epoch": 0.5610486211901307,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004975037177148585,
      "loss": 5.3589,
      "mean_token_accuracy": 0.1753137394785881,
      "num_tokens": 12854670.0,
      "step": 6185
    },
    {
      "entropy": 5.51950306892395,
      "epoch": 0.5615021770682148,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004974989089787035,
      "loss": 5.3713,
      "mean_token_accuracy": 0.18194560408592225,
      "num_tokens": 12865281.0,
      "step": 6190
    },
    {
      "entropy": 5.6557981967926025,
      "epoch": 0.561955732946299,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004974940956412143,
      "loss": 5.3725,
      "mean_token_accuracy": 0.18128142803907393,
      "num_tokens": 12875222.0,
      "step": 6195
    },
    {
      "entropy": 5.714055490493775,
      "epoch": 0.5624092888243831,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004974892777024905,
      "loss": 5.4478,
      "mean_token_accuracy": 0.1658184453845024,
      "num_tokens": 12885079.0,
      "step": 6200
    },
    {
      "entropy": 5.553147268295288,
      "epoch": 0.5628628447024674,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004974844551626318,
      "loss": 5.3775,
      "mean_token_accuracy": 0.18091828525066375,
      "num_tokens": 12896402.0,
      "step": 6205
    },
    {
      "entropy": 5.556194925308228,
      "epoch": 0.5633164005805515,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004974796280217379,
      "loss": 5.3496,
      "mean_token_accuracy": 0.18283513188362122,
      "num_tokens": 12906403.0,
      "step": 6210
    },
    {
      "entropy": 5.598826265335083,
      "epoch": 0.5637699564586357,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004974747962799086,
      "loss": 5.3687,
      "mean_token_accuracy": 0.1756142258644104,
      "num_tokens": 12917061.0,
      "step": 6215
    },
    {
      "entropy": 5.568496322631836,
      "epoch": 0.5642235123367199,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004974699599372437,
      "loss": 5.3911,
      "mean_token_accuracy": 0.1795428067445755,
      "num_tokens": 12927182.0,
      "step": 6220
    },
    {
      "entropy": 5.529246139526367,
      "epoch": 0.564677068214804,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004974651189938434,
      "loss": 5.361,
      "mean_token_accuracy": 0.17926921993494033,
      "num_tokens": 12937371.0,
      "step": 6225
    },
    {
      "entropy": 5.573128747940063,
      "epoch": 0.5651306240928883,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004974602734498079,
      "loss": 5.4119,
      "mean_token_accuracy": 0.1759198173880577,
      "num_tokens": 12948569.0,
      "step": 6230
    },
    {
      "entropy": 5.610108995437622,
      "epoch": 0.5655841799709724,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000497455423305237,
      "loss": 5.3696,
      "mean_token_accuracy": 0.1703425258398056,
      "num_tokens": 12958503.0,
      "step": 6235
    },
    {
      "entropy": 5.612152004241944,
      "epoch": 0.5660377358490566,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004974505685602314,
      "loss": 5.3857,
      "mean_token_accuracy": 0.1823065236210823,
      "num_tokens": 12968714.0,
      "step": 6240
    },
    {
      "entropy": 5.555819892883301,
      "epoch": 0.5664912917271407,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004974457092148912,
      "loss": 5.4102,
      "mean_token_accuracy": 0.17348227500915528,
      "num_tokens": 12978857.0,
      "step": 6245
    },
    {
      "entropy": 5.670070648193359,
      "epoch": 0.566944847605225,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004974408452693172,
      "loss": 5.5099,
      "mean_token_accuracy": 0.17125375419855118,
      "num_tokens": 12989863.0,
      "step": 6250
    },
    {
      "entropy": 5.5948718070983885,
      "epoch": 0.5673984034833092,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974359767236099,
      "loss": 5.3118,
      "mean_token_accuracy": 0.17757333517074586,
      "num_tokens": 13000307.0,
      "step": 6255
    },
    {
      "entropy": 5.592511749267578,
      "epoch": 0.5678519593613933,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004974311035778698,
      "loss": 5.3883,
      "mean_token_accuracy": 0.18470816314220428,
      "num_tokens": 13009053.0,
      "step": 6260
    },
    {
      "entropy": 5.578414249420166,
      "epoch": 0.5683055152394775,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004974262258321979,
      "loss": 5.4205,
      "mean_token_accuracy": 0.172482830286026,
      "num_tokens": 13019136.0,
      "step": 6265
    },
    {
      "entropy": 5.6188208103179935,
      "epoch": 0.5687590711175616,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004974213434866949,
      "loss": 5.4601,
      "mean_token_accuracy": 0.17041391432285308,
      "num_tokens": 13030356.0,
      "step": 6270
    },
    {
      "entropy": 5.613703393936158,
      "epoch": 0.5692126269956459,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974164565414617,
      "loss": 5.4759,
      "mean_token_accuracy": 0.17600406408309938,
      "num_tokens": 13040477.0,
      "step": 6275
    },
    {
      "entropy": 5.604294204711914,
      "epoch": 0.56966618287373,
      "grad_norm": 0.875,
      "learning_rate": 0.0004974115649965998,
      "loss": 5.3422,
      "mean_token_accuracy": 0.18165747225284576,
      "num_tokens": 13051755.0,
      "step": 6280
    },
    {
      "entropy": 5.5737049102783205,
      "epoch": 0.5701197387518142,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004974066688522098,
      "loss": 5.3627,
      "mean_token_accuracy": 0.1862332984805107,
      "num_tokens": 13061670.0,
      "step": 6285
    },
    {
      "entropy": 5.556718826293945,
      "epoch": 0.5705732946298984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004974017681083934,
      "loss": 5.3469,
      "mean_token_accuracy": 0.18105981051921843,
      "num_tokens": 13070917.0,
      "step": 6290
    },
    {
      "entropy": 5.554030179977417,
      "epoch": 0.5710268505079826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004973968627652516,
      "loss": 5.3933,
      "mean_token_accuracy": 0.17703206092119217,
      "num_tokens": 13080577.0,
      "step": 6295
    },
    {
      "entropy": 5.614200782775879,
      "epoch": 0.5714804063860668,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497391952822886,
      "loss": 5.4314,
      "mean_token_accuracy": 0.17582101076841355,
      "num_tokens": 13091446.0,
      "step": 6300
    },
    {
      "entropy": 5.623776531219482,
      "epoch": 0.5719339622641509,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004973870382813983,
      "loss": 5.4603,
      "mean_token_accuracy": 0.17539089173078537,
      "num_tokens": 13102752.0,
      "step": 6305
    },
    {
      "entropy": 5.665589141845703,
      "epoch": 0.5723875181422351,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004973821191408899,
      "loss": 5.465,
      "mean_token_accuracy": 0.17193345874547958,
      "num_tokens": 13113944.0,
      "step": 6310
    },
    {
      "entropy": 5.614248418807984,
      "epoch": 0.5728410740203193,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004973771954014626,
      "loss": 5.399,
      "mean_token_accuracy": 0.17831117212772368,
      "num_tokens": 13124316.0,
      "step": 6315
    },
    {
      "entropy": 5.65424370765686,
      "epoch": 0.5732946298984035,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004973722670632182,
      "loss": 5.4797,
      "mean_token_accuracy": 0.17098383456468583,
      "num_tokens": 13133567.0,
      "step": 6320
    },
    {
      "entropy": 5.5662015914917,
      "epoch": 0.5737481857764877,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004973673341262587,
      "loss": 5.3917,
      "mean_token_accuracy": 0.1759439915418625,
      "num_tokens": 13144279.0,
      "step": 6325
    },
    {
      "entropy": 5.604485321044922,
      "epoch": 0.5742017416545718,
      "grad_norm": 0.953125,
      "learning_rate": 0.000497362396590686,
      "loss": 5.4699,
      "mean_token_accuracy": 0.17275095134973525,
      "num_tokens": 13154463.0,
      "step": 6330
    },
    {
      "entropy": 5.647492170333862,
      "epoch": 0.574655297532656,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004973574544566024,
      "loss": 5.4281,
      "mean_token_accuracy": 0.16383579671382903,
      "num_tokens": 13165565.0,
      "step": 6335
    },
    {
      "entropy": 5.578436040878296,
      "epoch": 0.5751088534107403,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973525077241098,
      "loss": 5.3417,
      "mean_token_accuracy": 0.18116174638271332,
      "num_tokens": 13175839.0,
      "step": 6340
    },
    {
      "entropy": 5.509803342819214,
      "epoch": 0.5755624092888244,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973475563933109,
      "loss": 5.3268,
      "mean_token_accuracy": 0.17926134169101715,
      "num_tokens": 13186372.0,
      "step": 6345
    },
    {
      "entropy": 5.569689559936523,
      "epoch": 0.5760159651669086,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004973426004643077,
      "loss": 5.3894,
      "mean_token_accuracy": 0.1798923596739769,
      "num_tokens": 13196259.0,
      "step": 6350
    },
    {
      "entropy": 5.578006315231323,
      "epoch": 0.5764695210449927,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973376399372029,
      "loss": 5.3612,
      "mean_token_accuracy": 0.1823232352733612,
      "num_tokens": 13206497.0,
      "step": 6355
    },
    {
      "entropy": 5.582531070709228,
      "epoch": 0.5769230769230769,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004973326748120991,
      "loss": 5.4179,
      "mean_token_accuracy": 0.17883315533399582,
      "num_tokens": 13217188.0,
      "step": 6360
    },
    {
      "entropy": 5.578738641738892,
      "epoch": 0.5773766328011611,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004973277050890988,
      "loss": 5.4042,
      "mean_token_accuracy": 0.17636267542839051,
      "num_tokens": 13226544.0,
      "step": 6365
    },
    {
      "entropy": 5.636392402648926,
      "epoch": 0.5778301886792453,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973227307683049,
      "loss": 5.4464,
      "mean_token_accuracy": 0.17221278697252274,
      "num_tokens": 13236572.0,
      "step": 6370
    },
    {
      "entropy": 5.705385446548462,
      "epoch": 0.5782837445573294,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973177518498204,
      "loss": 5.4333,
      "mean_token_accuracy": 0.1773147389292717,
      "num_tokens": 13247958.0,
      "step": 6375
    },
    {
      "entropy": 5.586960983276367,
      "epoch": 0.5787373004354136,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973127683337482,
      "loss": 5.342,
      "mean_token_accuracy": 0.17918977737426758,
      "num_tokens": 13258255.0,
      "step": 6380
    },
    {
      "entropy": 5.52049298286438,
      "epoch": 0.5791908563134979,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004973077802201911,
      "loss": 5.3718,
      "mean_token_accuracy": 0.17743289172649385,
      "num_tokens": 13268416.0,
      "step": 6385
    },
    {
      "entropy": 5.5609814643859865,
      "epoch": 0.579644412191582,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973027875092526,
      "loss": 5.346,
      "mean_token_accuracy": 0.17577312141656876,
      "num_tokens": 13278087.0,
      "step": 6390
    },
    {
      "entropy": 5.551668643951416,
      "epoch": 0.5800979680696662,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004972977902010357,
      "loss": 5.2501,
      "mean_token_accuracy": 0.18471986949443817,
      "num_tokens": 13288121.0,
      "step": 6395
    },
    {
      "entropy": 5.666141223907471,
      "epoch": 0.5805515239477503,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972927882956439,
      "loss": 5.4785,
      "mean_token_accuracy": 0.16452504992485045,
      "num_tokens": 13298620.0,
      "step": 6400
    },
    {
      "entropy": 5.551081323623658,
      "epoch": 0.5810050798258345,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004972877817931807,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17660645842552186,
      "num_tokens": 13308413.0,
      "step": 6405
    },
    {
      "entropy": 5.522010469436646,
      "epoch": 0.5814586357039188,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004972827706937493,
      "loss": 5.3758,
      "mean_token_accuracy": 0.17947073578834533,
      "num_tokens": 13319387.0,
      "step": 6410
    },
    {
      "entropy": 5.620809888839721,
      "epoch": 0.5819121915820029,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004972777549974537,
      "loss": 5.3266,
      "mean_token_accuracy": 0.17694490402936935,
      "num_tokens": 13330494.0,
      "step": 6415
    },
    {
      "entropy": 5.556739234924317,
      "epoch": 0.5823657474600871,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004972727347043975,
      "loss": 5.3374,
      "mean_token_accuracy": 0.17858588695526123,
      "num_tokens": 13341009.0,
      "step": 6420
    },
    {
      "entropy": 5.526559066772461,
      "epoch": 0.5828193033381712,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004972677098146845,
      "loss": 5.2915,
      "mean_token_accuracy": 0.18030916303396224,
      "num_tokens": 13351269.0,
      "step": 6425
    },
    {
      "entropy": 5.536026811599731,
      "epoch": 0.5832728592162555,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004972626803284185,
      "loss": 5.4053,
      "mean_token_accuracy": 0.17801292687654496,
      "num_tokens": 13362160.0,
      "step": 6430
    },
    {
      "entropy": 5.641278648376465,
      "epoch": 0.5837264150943396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972576462457038,
      "loss": 5.4702,
      "mean_token_accuracy": 0.17617461383342742,
      "num_tokens": 13372497.0,
      "step": 6435
    },
    {
      "entropy": 5.673011207580567,
      "epoch": 0.5841799709724238,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004972526075666442,
      "loss": 5.3986,
      "mean_token_accuracy": 0.1806288555264473,
      "num_tokens": 13381665.0,
      "step": 6440
    },
    {
      "entropy": 5.523349046707153,
      "epoch": 0.584633526850508,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004972475642913442,
      "loss": 5.3026,
      "mean_token_accuracy": 0.1852041155099869,
      "num_tokens": 13392507.0,
      "step": 6445
    },
    {
      "entropy": 5.5107386112213135,
      "epoch": 0.5850870827285921,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972425164199078,
      "loss": 5.2899,
      "mean_token_accuracy": 0.18328872472047805,
      "num_tokens": 13402171.0,
      "step": 6450
    },
    {
      "entropy": 5.651026678085327,
      "epoch": 0.5855406386066764,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004972374639524396,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1725245714187622,
      "num_tokens": 13412933.0,
      "step": 6455
    },
    {
      "entropy": 5.612742567062378,
      "epoch": 0.5859941944847605,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000497232406889044,
      "loss": 5.3512,
      "mean_token_accuracy": 0.1728028178215027,
      "num_tokens": 13423563.0,
      "step": 6460
    },
    {
      "entropy": 5.638546085357666,
      "epoch": 0.5864477503628447,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004972273452298255,
      "loss": 5.36,
      "mean_token_accuracy": 0.18153707385063172,
      "num_tokens": 13433998.0,
      "step": 6465
    },
    {
      "entropy": 5.532230949401855,
      "epoch": 0.5869013062409288,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004972222789748888,
      "loss": 5.3105,
      "mean_token_accuracy": 0.18037805259227752,
      "num_tokens": 13443878.0,
      "step": 6470
    },
    {
      "entropy": 5.536468267440796,
      "epoch": 0.5873548621190131,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004972172081243389,
      "loss": 5.3923,
      "mean_token_accuracy": 0.17271317690610885,
      "num_tokens": 13456051.0,
      "step": 6475
    },
    {
      "entropy": 5.576761245727539,
      "epoch": 0.5878084179970973,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004972121326782806,
      "loss": 5.3829,
      "mean_token_accuracy": 0.18032484352588654,
      "num_tokens": 13467276.0,
      "step": 6480
    },
    {
      "entropy": 5.621661853790283,
      "epoch": 0.5882619738751814,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004972070526368186,
      "loss": 5.3517,
      "mean_token_accuracy": 0.17507115453481675,
      "num_tokens": 13478331.0,
      "step": 6485
    },
    {
      "entropy": 5.605046653747559,
      "epoch": 0.5887155297532656,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004972019680000582,
      "loss": 5.3101,
      "mean_token_accuracy": 0.18173345625400544,
      "num_tokens": 13489879.0,
      "step": 6490
    },
    {
      "entropy": 5.548620986938476,
      "epoch": 0.5891690856313497,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004971968787681044,
      "loss": 5.4208,
      "mean_token_accuracy": 0.1835034653544426,
      "num_tokens": 13500879.0,
      "step": 6495
    },
    {
      "entropy": 5.66513090133667,
      "epoch": 0.589622641509434,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004971917849410627,
      "loss": 5.3983,
      "mean_token_accuracy": 0.1771375998854637,
      "num_tokens": 13511719.0,
      "step": 6500
    },
    {
      "entropy": 5.545218276977539,
      "epoch": 0.5900761973875182,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004971866865190381,
      "loss": 5.3364,
      "mean_token_accuracy": 0.18135370761156083,
      "num_tokens": 13521850.0,
      "step": 6505
    },
    {
      "entropy": 5.597626638412476,
      "epoch": 0.5905297532656023,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004971815835021362,
      "loss": 5.3715,
      "mean_token_accuracy": 0.1825823336839676,
      "num_tokens": 13532095.0,
      "step": 6510
    },
    {
      "entropy": 5.5470359325408936,
      "epoch": 0.5909833091436865,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004971764758904626,
      "loss": 5.3513,
      "mean_token_accuracy": 0.17569283097982408,
      "num_tokens": 13543133.0,
      "step": 6515
    },
    {
      "entropy": 5.5723918914794925,
      "epoch": 0.5914368650217707,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004971713636841229,
      "loss": 5.4301,
      "mean_token_accuracy": 0.1692360073328018,
      "num_tokens": 13553771.0,
      "step": 6520
    },
    {
      "entropy": 5.536196327209472,
      "epoch": 0.5918904208998549,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004971662468832228,
      "loss": 5.3591,
      "mean_token_accuracy": 0.18100160360336304,
      "num_tokens": 13564548.0,
      "step": 6525
    },
    {
      "entropy": 5.500099849700928,
      "epoch": 0.592343976777939,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497161125487868,
      "loss": 5.343,
      "mean_token_accuracy": 0.17980441451072693,
      "num_tokens": 13574567.0,
      "step": 6530
    },
    {
      "entropy": 5.531034755706787,
      "epoch": 0.5927975326560232,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004971559994981645,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1790984570980072,
      "num_tokens": 13585302.0,
      "step": 6535
    },
    {
      "entropy": 5.616375589370728,
      "epoch": 0.5932510885341074,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004971508689142184,
      "loss": 5.4028,
      "mean_token_accuracy": 0.1724355548620224,
      "num_tokens": 13594754.0,
      "step": 6540
    },
    {
      "entropy": 5.548051548004151,
      "epoch": 0.5937046444121916,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971457337361356,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17818880081176758,
      "num_tokens": 13604691.0,
      "step": 6545
    },
    {
      "entropy": 5.455195903778076,
      "epoch": 0.5941582002902758,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004971405939640225,
      "loss": 5.2268,
      "mean_token_accuracy": 0.18700708597898483,
      "num_tokens": 13613911.0,
      "step": 6550
    },
    {
      "entropy": 5.672217941284179,
      "epoch": 0.5946117561683599,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004971354495979854,
      "loss": 5.456,
      "mean_token_accuracy": 0.17892757952213287,
      "num_tokens": 13624285.0,
      "step": 6555
    },
    {
      "entropy": 5.586663722991943,
      "epoch": 0.5950653120464441,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971303006381305,
      "loss": 5.3054,
      "mean_token_accuracy": 0.1812276229262352,
      "num_tokens": 13635434.0,
      "step": 6560
    },
    {
      "entropy": 5.5121111392974855,
      "epoch": 0.5955188679245284,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004971251470845643,
      "loss": 5.361,
      "mean_token_accuracy": 0.17556140720844268,
      "num_tokens": 13645582.0,
      "step": 6565
    },
    {
      "entropy": 5.546522235870361,
      "epoch": 0.5959724238026125,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004971199889373935,
      "loss": 5.3174,
      "mean_token_accuracy": 0.17810338735580444,
      "num_tokens": 13655029.0,
      "step": 6570
    },
    {
      "entropy": 5.550983810424805,
      "epoch": 0.5964259796806967,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004971148261967249,
      "loss": 5.309,
      "mean_token_accuracy": 0.1805609330534935,
      "num_tokens": 13665204.0,
      "step": 6575
    },
    {
      "entropy": 5.571804332733154,
      "epoch": 0.5968795355587808,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004971096588626649,
      "loss": 5.4215,
      "mean_token_accuracy": 0.17430134862661362,
      "num_tokens": 13676281.0,
      "step": 6580
    },
    {
      "entropy": 5.537905597686768,
      "epoch": 0.597333091436865,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004971044869353206,
      "loss": 5.3613,
      "mean_token_accuracy": 0.18184176832437515,
      "num_tokens": 13687144.0,
      "step": 6585
    },
    {
      "entropy": 5.562548398971558,
      "epoch": 0.5977866473149492,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497099310414799,
      "loss": 5.3553,
      "mean_token_accuracy": 0.18140233010053636,
      "num_tokens": 13696371.0,
      "step": 6590
    },
    {
      "entropy": 5.626462841033936,
      "epoch": 0.5982402031930334,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970941293012069,
      "loss": 5.3001,
      "mean_token_accuracy": 0.1769891545176506,
      "num_tokens": 13706900.0,
      "step": 6595
    },
    {
      "entropy": 5.530804586410523,
      "epoch": 0.5986937590711175,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004970889435946518,
      "loss": 5.3671,
      "mean_token_accuracy": 0.1814924567937851,
      "num_tokens": 13717874.0,
      "step": 6600
    },
    {
      "entropy": 5.539168167114258,
      "epoch": 0.5991473149492017,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004970837532952406,
      "loss": 5.3369,
      "mean_token_accuracy": 0.18162619471549987,
      "num_tokens": 13728020.0,
      "step": 6605
    },
    {
      "entropy": 5.58751859664917,
      "epoch": 0.599600870827286,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004970785584030808,
      "loss": 5.3918,
      "mean_token_accuracy": 0.17638065218925475,
      "num_tokens": 13738007.0,
      "step": 6610
    },
    {
      "entropy": 5.566587495803833,
      "epoch": 0.6000544267053701,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004970733589182799,
      "loss": 5.3877,
      "mean_token_accuracy": 0.17494157403707505,
      "num_tokens": 13749429.0,
      "step": 6615
    },
    {
      "entropy": 5.558034038543701,
      "epoch": 0.6005079825834543,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970681548409452,
      "loss": 5.262,
      "mean_token_accuracy": 0.18570734858512877,
      "num_tokens": 13760279.0,
      "step": 6620
    },
    {
      "entropy": 5.600715589523316,
      "epoch": 0.6009615384615384,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004970629461711846,
      "loss": 5.4823,
      "mean_token_accuracy": 0.17011196166276932,
      "num_tokens": 13770575.0,
      "step": 6625
    },
    {
      "entropy": 5.498355531692505,
      "epoch": 0.6014150943396226,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004970577329091057,
      "loss": 5.3063,
      "mean_token_accuracy": 0.1857804089784622,
      "num_tokens": 13781068.0,
      "step": 6630
    },
    {
      "entropy": 5.5711182117462155,
      "epoch": 0.6018686502177069,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970525150548162,
      "loss": 5.3957,
      "mean_token_accuracy": 0.18031402826309204,
      "num_tokens": 13790745.0,
      "step": 6635
    },
    {
      "entropy": 5.585378217697143,
      "epoch": 0.602322206095791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004970472926084242,
      "loss": 5.3303,
      "mean_token_accuracy": 0.17876851856708526,
      "num_tokens": 13800196.0,
      "step": 6640
    },
    {
      "entropy": 5.521693229675293,
      "epoch": 0.6027757619738752,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004970420655700375,
      "loss": 5.2459,
      "mean_token_accuracy": 0.1828948974609375,
      "num_tokens": 13809954.0,
      "step": 6645
    },
    {
      "entropy": 5.5161940574646,
      "epoch": 0.6032293178519593,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004970368339397643,
      "loss": 5.4225,
      "mean_token_accuracy": 0.1718314230442047,
      "num_tokens": 13820281.0,
      "step": 6650
    },
    {
      "entropy": 5.634420919418335,
      "epoch": 0.6036828737300436,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004970315977177129,
      "loss": 5.3772,
      "mean_token_accuracy": 0.17908216267824173,
      "num_tokens": 13830614.0,
      "step": 6655
    },
    {
      "entropy": 5.5539734840393065,
      "epoch": 0.6041364296081277,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004970263569039914,
      "loss": 5.3185,
      "mean_token_accuracy": 0.18047489523887633,
      "num_tokens": 13840551.0,
      "step": 6660
    },
    {
      "entropy": 5.62788462638855,
      "epoch": 0.6045899854862119,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004970211114987083,
      "loss": 5.4994,
      "mean_token_accuracy": 0.1773001953959465,
      "num_tokens": 13851830.0,
      "step": 6665
    },
    {
      "entropy": 5.5775432109832765,
      "epoch": 0.6050435413642961,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004970158615019719,
      "loss": 5.2978,
      "mean_token_accuracy": 0.18300541788339614,
      "num_tokens": 13862299.0,
      "step": 6670
    },
    {
      "entropy": 5.607674980163575,
      "epoch": 0.6054970972423802,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004970106069138912,
      "loss": 5.4325,
      "mean_token_accuracy": 0.1705495685338974,
      "num_tokens": 13872852.0,
      "step": 6675
    },
    {
      "entropy": 5.530221462249756,
      "epoch": 0.6059506531204645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004970053477345743,
      "loss": 5.3784,
      "mean_token_accuracy": 0.18458971232175828,
      "num_tokens": 13883601.0,
      "step": 6680
    },
    {
      "entropy": 5.578758192062378,
      "epoch": 0.6064042089985486,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004970000839641304,
      "loss": 5.3524,
      "mean_token_accuracy": 0.18701042830944062,
      "num_tokens": 13894246.0,
      "step": 6685
    },
    {
      "entropy": 5.521432018280029,
      "epoch": 0.6068577648766328,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004969948156026681,
      "loss": 5.3868,
      "mean_token_accuracy": 0.1802896812558174,
      "num_tokens": 13904823.0,
      "step": 6690
    },
    {
      "entropy": 5.647825717926025,
      "epoch": 0.6073113207547169,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969895426502965,
      "loss": 5.403,
      "mean_token_accuracy": 0.1710766464471817,
      "num_tokens": 13915338.0,
      "step": 6695
    },
    {
      "entropy": 5.546239614486694,
      "epoch": 0.6077648766328012,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004969842651071246,
      "loss": 5.3611,
      "mean_token_accuracy": 0.17356563061475755,
      "num_tokens": 13925838.0,
      "step": 6700
    },
    {
      "entropy": 5.606841611862182,
      "epoch": 0.6082184325108854,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004969789829732615,
      "loss": 5.3653,
      "mean_token_accuracy": 0.17559958398342132,
      "num_tokens": 13936341.0,
      "step": 6705
    },
    {
      "entropy": 5.611944770812988,
      "epoch": 0.6086719883889695,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004969736962488164,
      "loss": 5.3672,
      "mean_token_accuracy": 0.1796737715601921,
      "num_tokens": 13945438.0,
      "step": 6710
    },
    {
      "entropy": 5.558967113494873,
      "epoch": 0.6091255442670537,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004969684049338987,
      "loss": 5.2584,
      "mean_token_accuracy": 0.18215785771608353,
      "num_tokens": 13956029.0,
      "step": 6715
    },
    {
      "entropy": 5.596481418609619,
      "epoch": 0.6095791001451378,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000496963109028618,
      "loss": 5.3783,
      "mean_token_accuracy": 0.1746341571211815,
      "num_tokens": 13966498.0,
      "step": 6720
    },
    {
      "entropy": 5.53505163192749,
      "epoch": 0.6100326560232221,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969578085330834,
      "loss": 5.3298,
      "mean_token_accuracy": 0.18196677714586257,
      "num_tokens": 13976997.0,
      "step": 6725
    },
    {
      "entropy": 5.615462064743042,
      "epoch": 0.6104862119013063,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004969525034474049,
      "loss": 5.3307,
      "mean_token_accuracy": 0.17902466505765915,
      "num_tokens": 13986664.0,
      "step": 6730
    },
    {
      "entropy": 5.498200178146362,
      "epoch": 0.6109397677793904,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000496947193771692,
      "loss": 5.3209,
      "mean_token_accuracy": 0.18532713502645493,
      "num_tokens": 13997828.0,
      "step": 6735
    },
    {
      "entropy": 5.508232593536377,
      "epoch": 0.6113933236574746,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004969418795060548,
      "loss": 5.2761,
      "mean_token_accuracy": 0.17970932871103287,
      "num_tokens": 14008293.0,
      "step": 6740
    },
    {
      "entropy": 5.507388973236084,
      "epoch": 0.6118468795355588,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004969365606506027,
      "loss": 5.2646,
      "mean_token_accuracy": 0.18298047930002212,
      "num_tokens": 14018031.0,
      "step": 6745
    },
    {
      "entropy": 5.491404008865357,
      "epoch": 0.612300435413643,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004969312372054461,
      "loss": 5.2387,
      "mean_token_accuracy": 0.1883564054965973,
      "num_tokens": 14028013.0,
      "step": 6750
    },
    {
      "entropy": 5.524714374542237,
      "epoch": 0.6127539912917271,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004969259091706948,
      "loss": 5.3723,
      "mean_token_accuracy": 0.17930554449558259,
      "num_tokens": 14039787.0,
      "step": 6755
    },
    {
      "entropy": 5.5531907081604,
      "epoch": 0.6132075471698113,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004969205765464593,
      "loss": 5.2734,
      "mean_token_accuracy": 0.18047384172677994,
      "num_tokens": 14049223.0,
      "step": 6760
    },
    {
      "entropy": 5.597138452529907,
      "epoch": 0.6136611030478955,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004969152393328497,
      "loss": 5.3133,
      "mean_token_accuracy": 0.1850910186767578,
      "num_tokens": 14059004.0,
      "step": 6765
    },
    {
      "entropy": 5.561625194549561,
      "epoch": 0.6141146589259797,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969098975299764,
      "loss": 5.4108,
      "mean_token_accuracy": 0.1744755983352661,
      "num_tokens": 14069615.0,
      "step": 6770
    },
    {
      "entropy": 5.550989723205566,
      "epoch": 0.6145682148040639,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004969045511379499,
      "loss": 5.3455,
      "mean_token_accuracy": 0.17937582731246948,
      "num_tokens": 14078886.0,
      "step": 6775
    },
    {
      "entropy": 5.646713638305664,
      "epoch": 0.615021770682148,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004968992001568806,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1722983568906784,
      "num_tokens": 14089021.0,
      "step": 6780
    },
    {
      "entropy": 5.55102949142456,
      "epoch": 0.6154753265602322,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004968938445868795,
      "loss": 5.235,
      "mean_token_accuracy": 0.19036040008068084,
      "num_tokens": 14098316.0,
      "step": 6785
    },
    {
      "entropy": 5.502194881439209,
      "epoch": 0.6159288824383164,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004968884844280569,
      "loss": 5.3202,
      "mean_token_accuracy": 0.1710915148258209,
      "num_tokens": 14108874.0,
      "step": 6790
    },
    {
      "entropy": 5.5451150894165036,
      "epoch": 0.6163824383164006,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004968831196805242,
      "loss": 5.3731,
      "mean_token_accuracy": 0.1814247488975525,
      "num_tokens": 14119613.0,
      "step": 6795
    },
    {
      "entropy": 5.577469968795777,
      "epoch": 0.6168359941944848,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004968777503443918,
      "loss": 5.3335,
      "mean_token_accuracy": 0.1749727502465248,
      "num_tokens": 14129085.0,
      "step": 6800
    },
    {
      "entropy": 5.4897855758667,
      "epoch": 0.6172895500725689,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004968723764197711,
      "loss": 5.2292,
      "mean_token_accuracy": 0.19681656509637832,
      "num_tokens": 14138461.0,
      "step": 6805
    },
    {
      "entropy": 5.5140174388885494,
      "epoch": 0.6177431059506531,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000496866997906773,
      "loss": 5.2926,
      "mean_token_accuracy": 0.18786997944116593,
      "num_tokens": 14149126.0,
      "step": 6810
    },
    {
      "entropy": 5.499987840652466,
      "epoch": 0.6181966618287373,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004968616148055089,
      "loss": 5.306,
      "mean_token_accuracy": 0.1850553944706917,
      "num_tokens": 14160068.0,
      "step": 6815
    },
    {
      "entropy": 5.559629440307617,
      "epoch": 0.6186502177068215,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004968562271160901,
      "loss": 5.3981,
      "mean_token_accuracy": 0.18361826986074448,
      "num_tokens": 14170288.0,
      "step": 6820
    },
    {
      "entropy": 5.63637866973877,
      "epoch": 0.6191037735849056,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000496850834838628,
      "loss": 5.3922,
      "mean_token_accuracy": 0.17993116676807402,
      "num_tokens": 14181231.0,
      "step": 6825
    },
    {
      "entropy": 5.592602682113648,
      "epoch": 0.6195573294629898,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004968454379732342,
      "loss": 5.294,
      "mean_token_accuracy": 0.18232411295175552,
      "num_tokens": 14191715.0,
      "step": 6830
    },
    {
      "entropy": 5.52168231010437,
      "epoch": 0.6200108853410741,
      "grad_norm": 0.921875,
      "learning_rate": 0.00049684003652002,
      "loss": 5.2998,
      "mean_token_accuracy": 0.17813656479120255,
      "num_tokens": 14203416.0,
      "step": 6835
    },
    {
      "entropy": 5.446109342575073,
      "epoch": 0.6204644412191582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004968346304790973,
      "loss": 5.2524,
      "mean_token_accuracy": 0.19115934818983077,
      "num_tokens": 14213658.0,
      "step": 6840
    },
    {
      "entropy": 5.593048906326294,
      "epoch": 0.6209179970972424,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004968292198505781,
      "loss": 5.4235,
      "mean_token_accuracy": 0.17608387172222137,
      "num_tokens": 14223956.0,
      "step": 6845
    },
    {
      "entropy": 5.618917846679688,
      "epoch": 0.6213715529753265,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000496823804634574,
      "loss": 5.4129,
      "mean_token_accuracy": 0.17932555377483367,
      "num_tokens": 14236801.0,
      "step": 6850
    },
    {
      "entropy": 5.529294729232788,
      "epoch": 0.6218251088534107,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968183848311972,
      "loss": 5.2253,
      "mean_token_accuracy": 0.1876459464430809,
      "num_tokens": 14247382.0,
      "step": 6855
    },
    {
      "entropy": 5.560374402999878,
      "epoch": 0.622278664731495,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004968129604405596,
      "loss": 5.3169,
      "mean_token_accuracy": 0.18302041888237,
      "num_tokens": 14258627.0,
      "step": 6860
    },
    {
      "entropy": 5.464026880264282,
      "epoch": 0.6227322206095791,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004968075314627734,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1861542820930481,
      "num_tokens": 14269013.0,
      "step": 6865
    },
    {
      "entropy": 5.536890077590942,
      "epoch": 0.6231857764876633,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496802097897951,
      "loss": 5.3071,
      "mean_token_accuracy": 0.18000025004148484,
      "num_tokens": 14279620.0,
      "step": 6870
    },
    {
      "entropy": 5.5682155132293705,
      "epoch": 0.6236393323657474,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004967966597462046,
      "loss": 5.2997,
      "mean_token_accuracy": 0.18479396849870683,
      "num_tokens": 14289448.0,
      "step": 6875
    },
    {
      "entropy": 5.561263513565064,
      "epoch": 0.6240928882438317,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004967912170076469,
      "loss": 5.2893,
      "mean_token_accuracy": 0.17907041907310486,
      "num_tokens": 14299999.0,
      "step": 6880
    },
    {
      "entropy": 5.501129531860352,
      "epoch": 0.6245464441219158,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004967857696823901,
      "loss": 5.2876,
      "mean_token_accuracy": 0.18401148319244384,
      "num_tokens": 14310384.0,
      "step": 6885
    },
    {
      "entropy": 5.550896310806275,
      "epoch": 0.625,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004967803177705472,
      "loss": 5.4017,
      "mean_token_accuracy": 0.17491451203823088,
      "num_tokens": 14320644.0,
      "step": 6890
    },
    {
      "entropy": 5.607715702056884,
      "epoch": 0.6254535558780842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004967748612722308,
      "loss": 5.3716,
      "mean_token_accuracy": 0.18048068434000014,
      "num_tokens": 14330077.0,
      "step": 6895
    },
    {
      "entropy": 5.613873863220215,
      "epoch": 0.6259071117561683,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004967694001875538,
      "loss": 5.4525,
      "mean_token_accuracy": 0.17149625718593597,
      "num_tokens": 14341549.0,
      "step": 6900
    },
    {
      "entropy": 5.585393762588501,
      "epoch": 0.6263606676342526,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000496763934516629,
      "loss": 5.3118,
      "mean_token_accuracy": 0.18692418187856674,
      "num_tokens": 14352146.0,
      "step": 6905
    },
    {
      "entropy": 5.495177698135376,
      "epoch": 0.6268142235123367,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004967584642595696,
      "loss": 5.3372,
      "mean_token_accuracy": 0.1760642945766449,
      "num_tokens": 14362521.0,
      "step": 6910
    },
    {
      "entropy": 5.564804315567017,
      "epoch": 0.6272677793904209,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004967529894164887,
      "loss": 5.3685,
      "mean_token_accuracy": 0.1765128716826439,
      "num_tokens": 14373216.0,
      "step": 6915
    },
    {
      "entropy": 5.5402006149292,
      "epoch": 0.627721335268505,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004967475099874992,
      "loss": 5.2748,
      "mean_token_accuracy": 0.18698276132345198,
      "num_tokens": 14383104.0,
      "step": 6920
    },
    {
      "entropy": 5.488488435745239,
      "epoch": 0.6281748911465893,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000496742025972715,
      "loss": 5.31,
      "mean_token_accuracy": 0.1918911427259445,
      "num_tokens": 14393854.0,
      "step": 6925
    },
    {
      "entropy": 5.564506006240845,
      "epoch": 0.6286284470246735,
      "grad_norm": 1.046875,
      "learning_rate": 0.000496736537372249,
      "loss": 5.2489,
      "mean_token_accuracy": 0.18805742710828782,
      "num_tokens": 14403213.0,
      "step": 6930
    },
    {
      "entropy": 5.641831588745117,
      "epoch": 0.6290820029027576,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004967310441862151,
      "loss": 5.478,
      "mean_token_accuracy": 0.17123783379793167,
      "num_tokens": 14412792.0,
      "step": 6935
    },
    {
      "entropy": 5.584077024459839,
      "epoch": 0.6295355587808418,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004967255464147265,
      "loss": 5.3367,
      "mean_token_accuracy": 0.18502951860427858,
      "num_tokens": 14423061.0,
      "step": 6940
    },
    {
      "entropy": 5.563900518417358,
      "epoch": 0.6299891146589259,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004967200440578972,
      "loss": 5.2726,
      "mean_token_accuracy": 0.1876469150185585,
      "num_tokens": 14433338.0,
      "step": 6945
    },
    {
      "entropy": 5.481202793121338,
      "epoch": 0.6304426705370102,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004967145371158408,
      "loss": 5.2585,
      "mean_token_accuracy": 0.19061126410961152,
      "num_tokens": 14444124.0,
      "step": 6950
    },
    {
      "entropy": 5.514369153976441,
      "epoch": 0.6308962264150944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967090255886714,
      "loss": 5.3887,
      "mean_token_accuracy": 0.18010223805904388,
      "num_tokens": 14453868.0,
      "step": 6955
    },
    {
      "entropy": 5.5535437107086185,
      "epoch": 0.6313497822931785,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004967035094765028,
      "loss": 5.2781,
      "mean_token_accuracy": 0.17700921893119811,
      "num_tokens": 14465015.0,
      "step": 6960
    },
    {
      "entropy": 5.504481458663941,
      "epoch": 0.6318033381712627,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004966979887794491,
      "loss": 5.2372,
      "mean_token_accuracy": 0.19093358367681504,
      "num_tokens": 14475523.0,
      "step": 6965
    },
    {
      "entropy": 5.507241201400757,
      "epoch": 0.6322568940493469,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966924634976245,
      "loss": 5.3185,
      "mean_token_accuracy": 0.18237321078777313,
      "num_tokens": 14485523.0,
      "step": 6970
    },
    {
      "entropy": 5.608642530441284,
      "epoch": 0.6327104499274311,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004966869336311433,
      "loss": 5.371,
      "mean_token_accuracy": 0.1829627588391304,
      "num_tokens": 14495237.0,
      "step": 6975
    },
    {
      "entropy": 5.584758949279785,
      "epoch": 0.6331640058055152,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966813991801197,
      "loss": 5.5108,
      "mean_token_accuracy": 0.17335500121116637,
      "num_tokens": 14506431.0,
      "step": 6980
    },
    {
      "entropy": 5.615596199035645,
      "epoch": 0.6336175616835994,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004966758601446684,
      "loss": 5.4274,
      "mean_token_accuracy": 0.17262813448905945,
      "num_tokens": 14517445.0,
      "step": 6985
    },
    {
      "entropy": 5.597373199462891,
      "epoch": 0.6340711175616836,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004966703165249038,
      "loss": 5.3624,
      "mean_token_accuracy": 0.17349915504455565,
      "num_tokens": 14527818.0,
      "step": 6990
    },
    {
      "entropy": 5.553152656555175,
      "epoch": 0.6345246734397678,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004966647683209406,
      "loss": 5.4443,
      "mean_token_accuracy": 0.16868529468774796,
      "num_tokens": 14538235.0,
      "step": 6995
    },
    {
      "entropy": 5.693959903717041,
      "epoch": 0.634978229317852,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004966592155328935,
      "loss": 5.4539,
      "mean_token_accuracy": 0.1710901975631714,
      "num_tokens": 14549784.0,
      "step": 7000
    },
    {
      "entropy": 5.5767427444458,
      "epoch": 0.6354317851959361,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004966536581608771,
      "loss": 5.4413,
      "mean_token_accuracy": 0.16910024285316466,
      "num_tokens": 14561057.0,
      "step": 7005
    },
    {
      "entropy": 5.555035877227783,
      "epoch": 0.6358853410740203,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004966480962050069,
      "loss": 5.3529,
      "mean_token_accuracy": 0.1803895801305771,
      "num_tokens": 14570859.0,
      "step": 7010
    },
    {
      "entropy": 5.552107143402099,
      "epoch": 0.6363388969521045,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004966425296653974,
      "loss": 5.353,
      "mean_token_accuracy": 0.1830982819199562,
      "num_tokens": 14581066.0,
      "step": 7015
    },
    {
      "entropy": 5.559674692153931,
      "epoch": 0.6367924528301887,
      "grad_norm": 0.90625,
      "learning_rate": 0.000496636958542164,
      "loss": 5.3239,
      "mean_token_accuracy": 0.17896756231784822,
      "num_tokens": 14590709.0,
      "step": 7020
    },
    {
      "entropy": 5.556455469131469,
      "epoch": 0.6372460087082729,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004966313828354218,
      "loss": 5.3089,
      "mean_token_accuracy": 0.18295130729675294,
      "num_tokens": 14600849.0,
      "step": 7025
    },
    {
      "entropy": 5.481048727035523,
      "epoch": 0.637699564586357,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496625802545286,
      "loss": 5.1954,
      "mean_token_accuracy": 0.18557194024324417,
      "num_tokens": 14611427.0,
      "step": 7030
    },
    {
      "entropy": 5.520011472702026,
      "epoch": 0.6381531204644412,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004966202176718723,
      "loss": 5.3651,
      "mean_token_accuracy": 0.18566489666700364,
      "num_tokens": 14623860.0,
      "step": 7035
    },
    {
      "entropy": 5.556991815567017,
      "epoch": 0.6386066763425254,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004966146282152959,
      "loss": 5.2855,
      "mean_token_accuracy": 0.1784844383597374,
      "num_tokens": 14633481.0,
      "step": 7040
    },
    {
      "entropy": 5.5632165431976315,
      "epoch": 0.6390602322206096,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966090341756726,
      "loss": 5.2627,
      "mean_token_accuracy": 0.19155942499637604,
      "num_tokens": 14643519.0,
      "step": 7045
    },
    {
      "entropy": 5.432686424255371,
      "epoch": 0.6395137880986937,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496603435553118,
      "loss": 5.2478,
      "mean_token_accuracy": 0.1885633021593094,
      "num_tokens": 14654570.0,
      "step": 7050
    },
    {
      "entropy": 5.48898401260376,
      "epoch": 0.6399673439767779,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004965978323477478,
      "loss": 5.3028,
      "mean_token_accuracy": 0.1765861302614212,
      "num_tokens": 14665596.0,
      "step": 7055
    },
    {
      "entropy": 5.582192134857178,
      "epoch": 0.6404208998548622,
      "grad_norm": 1.03125,
      "learning_rate": 0.000496592224559678,
      "loss": 5.2517,
      "mean_token_accuracy": 0.18795465528964997,
      "num_tokens": 14675660.0,
      "step": 7060
    },
    {
      "entropy": 5.563787317276001,
      "epoch": 0.6408744557329463,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004965866121890245,
      "loss": 5.4095,
      "mean_token_accuracy": 0.18029894530773163,
      "num_tokens": 14685229.0,
      "step": 7065
    },
    {
      "entropy": 5.557628488540649,
      "epoch": 0.6413280116110305,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965809952359034,
      "loss": 5.3122,
      "mean_token_accuracy": 0.17769250571727752,
      "num_tokens": 14695672.0,
      "step": 7070
    },
    {
      "entropy": 5.514695119857788,
      "epoch": 0.6417815674891146,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004965753737004309,
      "loss": 5.2615,
      "mean_token_accuracy": 0.18418668806552888,
      "num_tokens": 14705160.0,
      "step": 7075
    },
    {
      "entropy": 5.52557144165039,
      "epoch": 0.6422351233671988,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004965697475827231,
      "loss": 5.3097,
      "mean_token_accuracy": 0.18284096121788024,
      "num_tokens": 14715487.0,
      "step": 7080
    },
    {
      "entropy": 5.577736806869507,
      "epoch": 0.6426886792452831,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004965641168828965,
      "loss": 5.3544,
      "mean_token_accuracy": 0.18357906490564346,
      "num_tokens": 14726987.0,
      "step": 7085
    },
    {
      "entropy": 5.4941871643066404,
      "epoch": 0.6431422351233672,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004965584816010676,
      "loss": 5.3026,
      "mean_token_accuracy": 0.18404889553785325,
      "num_tokens": 14737810.0,
      "step": 7090
    },
    {
      "entropy": 5.579785442352295,
      "epoch": 0.6435957910014514,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004965528417373528,
      "loss": 5.3193,
      "mean_token_accuracy": 0.18094680607318878,
      "num_tokens": 14748997.0,
      "step": 7095
    },
    {
      "entropy": 5.553784418106079,
      "epoch": 0.6440493468795355,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004965471972918688,
      "loss": 5.3318,
      "mean_token_accuracy": 0.17817866504192353,
      "num_tokens": 14760439.0,
      "step": 7100
    },
    {
      "entropy": 5.445217418670654,
      "epoch": 0.6445029027576198,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004965415482647322,
      "loss": 5.2362,
      "mean_token_accuracy": 0.18102772533893585,
      "num_tokens": 14770828.0,
      "step": 7105
    },
    {
      "entropy": 5.490187788009644,
      "epoch": 0.6449564586357039,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049653589465606,
      "loss": 5.3809,
      "mean_token_accuracy": 0.18126174360513686,
      "num_tokens": 14781753.0,
      "step": 7110
    },
    {
      "entropy": 5.619928598403931,
      "epoch": 0.6454100145137881,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000496530236465969,
      "loss": 5.3987,
      "mean_token_accuracy": 0.1711141899228096,
      "num_tokens": 14792241.0,
      "step": 7115
    },
    {
      "entropy": 5.504394292831421,
      "epoch": 0.6458635703918723,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004965245736945763,
      "loss": 5.2172,
      "mean_token_accuracy": 0.1904732972383499,
      "num_tokens": 14802580.0,
      "step": 7120
    },
    {
      "entropy": 5.5241124629974365,
      "epoch": 0.6463171262699564,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004965189063419989,
      "loss": 5.4283,
      "mean_token_accuracy": 0.17418646961450576,
      "num_tokens": 14813103.0,
      "step": 7125
    },
    {
      "entropy": 5.571353006362915,
      "epoch": 0.6467706821480407,
      "grad_norm": 1.0,
      "learning_rate": 0.0004965132344083541,
      "loss": 5.3258,
      "mean_token_accuracy": 0.17736517637968063,
      "num_tokens": 14823334.0,
      "step": 7130
    },
    {
      "entropy": 5.56951904296875,
      "epoch": 0.6472242380261248,
      "grad_norm": 0.875,
      "learning_rate": 0.0004965075578937593,
      "loss": 5.336,
      "mean_token_accuracy": 0.17852421551942826,
      "num_tokens": 14834850.0,
      "step": 7135
    },
    {
      "entropy": 5.485752820968628,
      "epoch": 0.647677793904209,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004965018767983316,
      "loss": 5.3493,
      "mean_token_accuracy": 0.18051422238349915,
      "num_tokens": 14844276.0,
      "step": 7140
    },
    {
      "entropy": 5.440044736862182,
      "epoch": 0.6481313497822931,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004964961911221886,
      "loss": 5.1867,
      "mean_token_accuracy": 0.19364167600870133,
      "num_tokens": 14854352.0,
      "step": 7145
    },
    {
      "entropy": 5.579373216629028,
      "epoch": 0.6485849056603774,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000496490500865448,
      "loss": 5.3523,
      "mean_token_accuracy": 0.18060466051101684,
      "num_tokens": 14864395.0,
      "step": 7150
    },
    {
      "entropy": 5.513666248321533,
      "epoch": 0.6490384615384616,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004964848060282275,
      "loss": 5.2513,
      "mean_token_accuracy": 0.18520902544260026,
      "num_tokens": 14874373.0,
      "step": 7155
    },
    {
      "entropy": 5.515260982513428,
      "epoch": 0.6494920174165457,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964791066106447,
      "loss": 5.2496,
      "mean_token_accuracy": 0.17809188216924668,
      "num_tokens": 14884932.0,
      "step": 7160
    },
    {
      "entropy": 5.572841691970825,
      "epoch": 0.6499455732946299,
      "grad_norm": 0.875,
      "learning_rate": 0.0004964734026128175,
      "loss": 5.3991,
      "mean_token_accuracy": 0.17553634494543074,
      "num_tokens": 14896254.0,
      "step": 7165
    },
    {
      "entropy": 5.601190614700317,
      "epoch": 0.650399129172714,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964676940348638,
      "loss": 5.3319,
      "mean_token_accuracy": 0.17750634700059892,
      "num_tokens": 14907392.0,
      "step": 7170
    },
    {
      "entropy": 5.573651218414307,
      "epoch": 0.6508526850507983,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000496461980876902,
      "loss": 5.4395,
      "mean_token_accuracy": 0.17827751934528352,
      "num_tokens": 14917676.0,
      "step": 7175
    },
    {
      "entropy": 5.549517297744751,
      "epoch": 0.6513062409288825,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004964562631390498,
      "loss": 5.2987,
      "mean_token_accuracy": 0.18335486352443695,
      "num_tokens": 14928500.0,
      "step": 7180
    },
    {
      "entropy": 5.538741636276245,
      "epoch": 0.6517597968069666,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004964505408214257,
      "loss": 5.2492,
      "mean_token_accuracy": 0.18961041122674943,
      "num_tokens": 14938970.0,
      "step": 7185
    },
    {
      "entropy": 5.511117887496948,
      "epoch": 0.6522133526850508,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496444813924148,
      "loss": 5.311,
      "mean_token_accuracy": 0.17935390323400496,
      "num_tokens": 14949017.0,
      "step": 7190
    },
    {
      "entropy": 5.51752781867981,
      "epoch": 0.652666908563135,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004964390824473351,
      "loss": 5.2404,
      "mean_token_accuracy": 0.18981554210186005,
      "num_tokens": 14958529.0,
      "step": 7195
    },
    {
      "entropy": 5.509324216842652,
      "epoch": 0.6531204644412192,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004964333463911056,
      "loss": 5.3481,
      "mean_token_accuracy": 0.1777355194091797,
      "num_tokens": 14968044.0,
      "step": 7200
    },
    {
      "entropy": 5.569104433059692,
      "epoch": 0.6535740203193033,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004964276057555779,
      "loss": 5.365,
      "mean_token_accuracy": 0.17823416888713836,
      "num_tokens": 14977867.0,
      "step": 7205
    },
    {
      "entropy": 5.6319859504699705,
      "epoch": 0.6540275761973875,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496421860540871,
      "loss": 5.3719,
      "mean_token_accuracy": 0.1820246011018753,
      "num_tokens": 14986936.0,
      "step": 7210
    },
    {
      "entropy": 5.530765676498413,
      "epoch": 0.6544811320754716,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004964161107471036,
      "loss": 5.331,
      "mean_token_accuracy": 0.18487488329410554,
      "num_tokens": 14996681.0,
      "step": 7215
    },
    {
      "entropy": 5.552242612838745,
      "epoch": 0.6549346879535559,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004964103563743946,
      "loss": 5.3224,
      "mean_token_accuracy": 0.1786444827914238,
      "num_tokens": 15007100.0,
      "step": 7220
    },
    {
      "entropy": 5.599308681488037,
      "epoch": 0.6553882438316401,
      "grad_norm": 1.0,
      "learning_rate": 0.000496404597422863,
      "loss": 5.3806,
      "mean_token_accuracy": 0.18596629649400712,
      "num_tokens": 15017452.0,
      "step": 7225
    },
    {
      "entropy": 5.516591310501099,
      "epoch": 0.6558417997097242,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000496398833892628,
      "loss": 5.2531,
      "mean_token_accuracy": 0.18542855978012085,
      "num_tokens": 15028362.0,
      "step": 7230
    },
    {
      "entropy": 5.581466817855835,
      "epoch": 0.6562953555878084,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004963930657838086,
      "loss": 5.3469,
      "mean_token_accuracy": 0.17714816331863403,
      "num_tokens": 15038691.0,
      "step": 7235
    },
    {
      "entropy": 5.552061080932617,
      "epoch": 0.6567489114658926,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004963872930965242,
      "loss": 5.3117,
      "mean_token_accuracy": 0.1878270536661148,
      "num_tokens": 15050037.0,
      "step": 7240
    },
    {
      "entropy": 5.552956056594849,
      "epoch": 0.6572024673439768,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963815158308941,
      "loss": 5.2747,
      "mean_token_accuracy": 0.18535771667957307,
      "num_tokens": 15059793.0,
      "step": 7245
    },
    {
      "entropy": 5.5456095218658445,
      "epoch": 0.657656023222061,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004963757339870379,
      "loss": 5.3749,
      "mean_token_accuracy": 0.17560529112815856,
      "num_tokens": 15070500.0,
      "step": 7250
    },
    {
      "entropy": 5.579104375839234,
      "epoch": 0.6581095791001451,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004963699475650751,
      "loss": 5.4274,
      "mean_token_accuracy": 0.1789287731051445,
      "num_tokens": 15081109.0,
      "step": 7255
    },
    {
      "entropy": 5.514655733108521,
      "epoch": 0.6585631349782293,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963641565651255,
      "loss": 5.3029,
      "mean_token_accuracy": 0.18567312955856324,
      "num_tokens": 15092315.0,
      "step": 7260
    },
    {
      "entropy": 5.513403844833374,
      "epoch": 0.6590166908563135,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004963583609873084,
      "loss": 5.3694,
      "mean_token_accuracy": 0.18358769416809081,
      "num_tokens": 15103957.0,
      "step": 7265
    },
    {
      "entropy": 5.562368488311767,
      "epoch": 0.6594702467343977,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004963525608317443,
      "loss": 5.362,
      "mean_token_accuracy": 0.1775441661477089,
      "num_tokens": 15114996.0,
      "step": 7270
    },
    {
      "entropy": 5.581370878219604,
      "epoch": 0.6599238026124818,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004963467560985527,
      "loss": 5.2938,
      "mean_token_accuracy": 0.17815622687339783,
      "num_tokens": 15125098.0,
      "step": 7275
    },
    {
      "entropy": 5.480411434173584,
      "epoch": 0.660377358490566,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004963409467878538,
      "loss": 5.2984,
      "mean_token_accuracy": 0.1813468173146248,
      "num_tokens": 15134746.0,
      "step": 7280
    },
    {
      "entropy": 5.59424524307251,
      "epoch": 0.6608309143686503,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004963351328997678,
      "loss": 5.3253,
      "mean_token_accuracy": 0.17637452483177185,
      "num_tokens": 15145698.0,
      "step": 7285
    },
    {
      "entropy": 5.577039194107056,
      "epoch": 0.6612844702467344,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963293144344148,
      "loss": 5.3688,
      "mean_token_accuracy": 0.17920732200145723,
      "num_tokens": 15156026.0,
      "step": 7290
    },
    {
      "entropy": 5.529924106597901,
      "epoch": 0.6617380261248186,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004963234913919151,
      "loss": 5.3084,
      "mean_token_accuracy": 0.17793872058391572,
      "num_tokens": 15165900.0,
      "step": 7295
    },
    {
      "entropy": 5.5461030960083,
      "epoch": 0.6621915820029027,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004963176637723893,
      "loss": 5.3744,
      "mean_token_accuracy": 0.182407346367836,
      "num_tokens": 15176693.0,
      "step": 7300
    },
    {
      "entropy": 5.591224098205567,
      "epoch": 0.6626451378809869,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004963118315759578,
      "loss": 5.3901,
      "mean_token_accuracy": 0.16922723799943923,
      "num_tokens": 15186535.0,
      "step": 7305
    },
    {
      "entropy": 5.577974796295166,
      "epoch": 0.6630986937590712,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004963059948027411,
      "loss": 5.3013,
      "mean_token_accuracy": 0.182105216383934,
      "num_tokens": 15196019.0,
      "step": 7310
    },
    {
      "entropy": 5.474636697769165,
      "epoch": 0.6635522496371553,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004963001534528601,
      "loss": 5.1884,
      "mean_token_accuracy": 0.192403119802475,
      "num_tokens": 15205273.0,
      "step": 7315
    },
    {
      "entropy": 5.537912750244141,
      "epoch": 0.6640058055152395,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004962943075264355,
      "loss": 5.3325,
      "mean_token_accuracy": 0.1802896723151207,
      "num_tokens": 15216716.0,
      "step": 7320
    },
    {
      "entropy": 5.575633764266968,
      "epoch": 0.6644593613933236,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004962884570235884,
      "loss": 5.3761,
      "mean_token_accuracy": 0.17417128533124923,
      "num_tokens": 15227306.0,
      "step": 7325
    },
    {
      "entropy": 5.511094474792481,
      "epoch": 0.6649129172714079,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004962826019444394,
      "loss": 5.2652,
      "mean_token_accuracy": 0.1794412538409233,
      "num_tokens": 15237302.0,
      "step": 7330
    },
    {
      "entropy": 5.517401790618896,
      "epoch": 0.665366473149492,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004962767422891099,
      "loss": 5.2692,
      "mean_token_accuracy": 0.18371451795101165,
      "num_tokens": 15247064.0,
      "step": 7335
    },
    {
      "entropy": 5.537144470214844,
      "epoch": 0.6658200290275762,
      "grad_norm": 0.890625,
      "learning_rate": 0.000496270878057721,
      "loss": 5.3427,
      "mean_token_accuracy": 0.18440829068422318,
      "num_tokens": 15257006.0,
      "step": 7340
    },
    {
      "entropy": 5.58497314453125,
      "epoch": 0.6662735849056604,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004962650092503939,
      "loss": 5.3172,
      "mean_token_accuracy": 0.1943359598517418,
      "num_tokens": 15266720.0,
      "step": 7345
    },
    {
      "entropy": 5.543534851074218,
      "epoch": 0.6667271407837445,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004962591358672502,
      "loss": 5.3427,
      "mean_token_accuracy": 0.18019769638776778,
      "num_tokens": 15277330.0,
      "step": 7350
    },
    {
      "entropy": 5.5373796939849855,
      "epoch": 0.6671806966618288,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000496253257908411,
      "loss": 5.271,
      "mean_token_accuracy": 0.1839846909046173,
      "num_tokens": 15287401.0,
      "step": 7355
    },
    {
      "entropy": 5.489133691787719,
      "epoch": 0.6676342525399129,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004962473753739981,
      "loss": 5.2292,
      "mean_token_accuracy": 0.1890961855649948,
      "num_tokens": 15297426.0,
      "step": 7360
    },
    {
      "entropy": 5.556583213806152,
      "epoch": 0.6680878084179971,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000496241488264133,
      "loss": 5.2272,
      "mean_token_accuracy": 0.1845378175377846,
      "num_tokens": 15308412.0,
      "step": 7365
    },
    {
      "entropy": 5.527732086181641,
      "epoch": 0.6685413642960812,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004962355965789377,
      "loss": 5.2293,
      "mean_token_accuracy": 0.18310726881027223,
      "num_tokens": 15318821.0,
      "step": 7370
    },
    {
      "entropy": 5.476294469833374,
      "epoch": 0.6689949201741655,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004962297003185338,
      "loss": 5.2916,
      "mean_token_accuracy": 0.1801626518368721,
      "num_tokens": 15328737.0,
      "step": 7375
    },
    {
      "entropy": 5.622493934631348,
      "epoch": 0.6694484760522497,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962237994830433,
      "loss": 5.393,
      "mean_token_accuracy": 0.17233246713876724,
      "num_tokens": 15339695.0,
      "step": 7380
    },
    {
      "entropy": 5.615177869796753,
      "epoch": 0.6699020319303338,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004962178940725883,
      "loss": 5.364,
      "mean_token_accuracy": 0.18082518577575685,
      "num_tokens": 15350270.0,
      "step": 7385
    },
    {
      "entropy": 5.520081806182861,
      "epoch": 0.670355587808418,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004962119840872908,
      "loss": 5.3439,
      "mean_token_accuracy": 0.18389475494623184,
      "num_tokens": 15360986.0,
      "step": 7390
    },
    {
      "entropy": 5.516814851760865,
      "epoch": 0.6708091436865021,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004962060695272733,
      "loss": 5.3391,
      "mean_token_accuracy": 0.18298070579767228,
      "num_tokens": 15371786.0,
      "step": 7395
    },
    {
      "entropy": 5.55599799156189,
      "epoch": 0.6712626995645864,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004962001503926577,
      "loss": 5.2993,
      "mean_token_accuracy": 0.1882503256201744,
      "num_tokens": 15382960.0,
      "step": 7400
    },
    {
      "entropy": 5.526637887954712,
      "epoch": 0.6717162554426706,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961942266835667,
      "loss": 5.287,
      "mean_token_accuracy": 0.18534947335720062,
      "num_tokens": 15392807.0,
      "step": 7405
    },
    {
      "entropy": 5.559031534194946,
      "epoch": 0.6721698113207547,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004961882984001227,
      "loss": 5.3307,
      "mean_token_accuracy": 0.1767762467265129,
      "num_tokens": 15403712.0,
      "step": 7410
    },
    {
      "entropy": 5.584656190872193,
      "epoch": 0.6726233671988389,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004961823655424484,
      "loss": 5.3916,
      "mean_token_accuracy": 0.17293067425489425,
      "num_tokens": 15414839.0,
      "step": 7415
    },
    {
      "entropy": 5.518201398849487,
      "epoch": 0.6730769230769231,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004961764281106664,
      "loss": 5.2838,
      "mean_token_accuracy": 0.1870151072740555,
      "num_tokens": 15425320.0,
      "step": 7420
    },
    {
      "entropy": 5.517269277572632,
      "epoch": 0.6735304789550073,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961704861048995,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17715439349412918,
      "num_tokens": 15435476.0,
      "step": 7425
    },
    {
      "entropy": 5.53749532699585,
      "epoch": 0.6739840348330914,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961645395252705,
      "loss": 5.3947,
      "mean_token_accuracy": 0.17859402149915696,
      "num_tokens": 15445786.0,
      "step": 7430
    },
    {
      "entropy": 5.562483930587769,
      "epoch": 0.6744375907111756,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961585883719025,
      "loss": 5.3762,
      "mean_token_accuracy": 0.18556168675422668,
      "num_tokens": 15455915.0,
      "step": 7435
    },
    {
      "entropy": 5.612591075897217,
      "epoch": 0.6748911465892597,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961526326449186,
      "loss": 5.3614,
      "mean_token_accuracy": 0.1754249468445778,
      "num_tokens": 15466523.0,
      "step": 7440
    },
    {
      "entropy": 5.561124181747436,
      "epoch": 0.675344702467344,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961466723444419,
      "loss": 5.3425,
      "mean_token_accuracy": 0.18115431517362596,
      "num_tokens": 15476821.0,
      "step": 7445
    },
    {
      "entropy": 5.506720113754272,
      "epoch": 0.6757982583454282,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961407074705955,
      "loss": 5.2759,
      "mean_token_accuracy": 0.1748933583498001,
      "num_tokens": 15487182.0,
      "step": 7450
    },
    {
      "entropy": 5.5438395023345945,
      "epoch": 0.6762518142235123,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961347380235031,
      "loss": 5.267,
      "mean_token_accuracy": 0.18452272266149522,
      "num_tokens": 15497600.0,
      "step": 7455
    },
    {
      "entropy": 5.560304641723633,
      "epoch": 0.6767053701015965,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961287640032879,
      "loss": 5.41,
      "mean_token_accuracy": 0.17247417122125625,
      "num_tokens": 15508278.0,
      "step": 7460
    },
    {
      "entropy": 5.6194220066070555,
      "epoch": 0.6771589259796807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961227854100734,
      "loss": 5.4261,
      "mean_token_accuracy": 0.17710963487625123,
      "num_tokens": 15518458.0,
      "step": 7465
    },
    {
      "entropy": 5.551078796386719,
      "epoch": 0.6776124818577649,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961168022439835,
      "loss": 5.2497,
      "mean_token_accuracy": 0.1808410331606865,
      "num_tokens": 15528738.0,
      "step": 7470
    },
    {
      "entropy": 5.536972951889038,
      "epoch": 0.6780660377358491,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961108145051417,
      "loss": 5.2783,
      "mean_token_accuracy": 0.1874865859746933,
      "num_tokens": 15538656.0,
      "step": 7475
    },
    {
      "entropy": 5.481096410751343,
      "epoch": 0.6785195936139332,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496104822193672,
      "loss": 5.2837,
      "mean_token_accuracy": 0.18348010629415512,
      "num_tokens": 15549285.0,
      "step": 7480
    },
    {
      "entropy": 5.595572376251221,
      "epoch": 0.6789731494920174,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004960988253096982,
      "loss": 5.3666,
      "mean_token_accuracy": 0.18583495169878006,
      "num_tokens": 15559766.0,
      "step": 7485
    },
    {
      "entropy": 5.505124187469482,
      "epoch": 0.6794267053701016,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004960928238533443,
      "loss": 5.2299,
      "mean_token_accuracy": 0.19369809031486512,
      "num_tokens": 15570873.0,
      "step": 7490
    },
    {
      "entropy": 5.5578302383422855,
      "epoch": 0.6798802612481858,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004960868178247344,
      "loss": 5.4746,
      "mean_token_accuracy": 0.17800090312957764,
      "num_tokens": 15581229.0,
      "step": 7495
    },
    {
      "entropy": 5.530143356323242,
      "epoch": 0.68033381712627,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960808072239929,
      "loss": 5.3422,
      "mean_token_accuracy": 0.17350580245256425,
      "num_tokens": 15591686.0,
      "step": 7500
    },
    {
      "entropy": 5.646744251251221,
      "epoch": 0.6807873730043541,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960747920512439,
      "loss": 5.3575,
      "mean_token_accuracy": 0.17898468375205995,
      "num_tokens": 15601869.0,
      "step": 7505
    },
    {
      "entropy": 5.519121694564819,
      "epoch": 0.6812409288824384,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000496068772306612,
      "loss": 5.3459,
      "mean_token_accuracy": 0.17619211375713348,
      "num_tokens": 15612946.0,
      "step": 7510
    },
    {
      "entropy": 5.554516172409057,
      "epoch": 0.6816944847605225,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004960627479902214,
      "loss": 5.3769,
      "mean_token_accuracy": 0.17832256257534027,
      "num_tokens": 15623761.0,
      "step": 7515
    },
    {
      "entropy": 5.549900197982788,
      "epoch": 0.6821480406386067,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000496056719102197,
      "loss": 5.369,
      "mean_token_accuracy": 0.17480377554893495,
      "num_tokens": 15634500.0,
      "step": 7520
    },
    {
      "entropy": 5.474475002288818,
      "epoch": 0.6826015965166908,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004960506856426631,
      "loss": 5.2228,
      "mean_token_accuracy": 0.18139137476682662,
      "num_tokens": 15644896.0,
      "step": 7525
    },
    {
      "entropy": 5.5556963920593265,
      "epoch": 0.683055152394775,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004960446476117448,
      "loss": 5.3439,
      "mean_token_accuracy": 0.18209579437971116,
      "num_tokens": 15655008.0,
      "step": 7530
    },
    {
      "entropy": 5.526656913757324,
      "epoch": 0.6835087082728593,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004960386050095668,
      "loss": 5.265,
      "mean_token_accuracy": 0.17771433889865876,
      "num_tokens": 15665944.0,
      "step": 7535
    },
    {
      "entropy": 5.55998010635376,
      "epoch": 0.6839622641509434,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004960325578362543,
      "loss": 5.3639,
      "mean_token_accuracy": 0.18044693470001222,
      "num_tokens": 15677099.0,
      "step": 7540
    },
    {
      "entropy": 5.593668031692505,
      "epoch": 0.6844158200290276,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496026506091932,
      "loss": 5.3198,
      "mean_token_accuracy": 0.1848937004804611,
      "num_tokens": 15687531.0,
      "step": 7545
    },
    {
      "entropy": 5.586200761795044,
      "epoch": 0.6848693759071117,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004960204497767253,
      "loss": 5.3321,
      "mean_token_accuracy": 0.17955249547958374,
      "num_tokens": 15697002.0,
      "step": 7550
    },
    {
      "entropy": 5.43728461265564,
      "epoch": 0.685322931785196,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960143888907594,
      "loss": 5.2166,
      "mean_token_accuracy": 0.18282410353422165,
      "num_tokens": 15707249.0,
      "step": 7555
    },
    {
      "entropy": 5.5209822177886965,
      "epoch": 0.6857764876632801,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004960083234341595,
      "loss": 5.2848,
      "mean_token_accuracy": 0.18320000618696214,
      "num_tokens": 15718764.0,
      "step": 7560
    },
    {
      "entropy": 5.546800947189331,
      "epoch": 0.6862300435413643,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004960022534070513,
      "loss": 5.2361,
      "mean_token_accuracy": 0.1883462518453598,
      "num_tokens": 15728854.0,
      "step": 7565
    },
    {
      "entropy": 5.43729796409607,
      "epoch": 0.6866835994194485,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049599617880956,
      "loss": 5.2635,
      "mean_token_accuracy": 0.18559020906686782,
      "num_tokens": 15739682.0,
      "step": 7570
    },
    {
      "entropy": 5.473732614517212,
      "epoch": 0.6871371552975326,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004959900996418115,
      "loss": 5.2534,
      "mean_token_accuracy": 0.18659544587135315,
      "num_tokens": 15749628.0,
      "step": 7575
    },
    {
      "entropy": 5.554725790023804,
      "epoch": 0.6875907111756169,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959840159039315,
      "loss": 5.3221,
      "mean_token_accuracy": 0.1806378498673439,
      "num_tokens": 15760137.0,
      "step": 7580
    },
    {
      "entropy": 5.579864168167115,
      "epoch": 0.688044267053701,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004959779275960456,
      "loss": 5.2711,
      "mean_token_accuracy": 0.18420590311288834,
      "num_tokens": 15769889.0,
      "step": 7585
    },
    {
      "entropy": 5.514402961730957,
      "epoch": 0.6884978229317852,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00049597183471828,
      "loss": 5.3859,
      "mean_token_accuracy": 0.18321892023086547,
      "num_tokens": 15780912.0,
      "step": 7590
    },
    {
      "entropy": 5.531727409362793,
      "epoch": 0.6889513788098693,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004959657372707604,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18849852085113525,
      "num_tokens": 15791719.0,
      "step": 7595
    },
    {
      "entropy": 5.572351932525635,
      "epoch": 0.6894049346879536,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959596352536132,
      "loss": 5.2987,
      "mean_token_accuracy": 0.18423290997743608,
      "num_tokens": 15802551.0,
      "step": 7600
    },
    {
      "entropy": 5.585467529296875,
      "epoch": 0.6898584905660378,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004959535286669643,
      "loss": 5.3539,
      "mean_token_accuracy": 0.1799274355173111,
      "num_tokens": 15813551.0,
      "step": 7605
    },
    {
      "entropy": 5.518627882003784,
      "epoch": 0.6903120464441219,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004959474175109402,
      "loss": 5.3222,
      "mean_token_accuracy": 0.1847722724080086,
      "num_tokens": 15824262.0,
      "step": 7610
    },
    {
      "entropy": 5.595053720474243,
      "epoch": 0.6907656023222061,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004959413017856672,
      "loss": 5.3735,
      "mean_token_accuracy": 0.18128857016563416,
      "num_tokens": 15835493.0,
      "step": 7615
    },
    {
      "entropy": 5.448233413696289,
      "epoch": 0.6912191582002902,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004959351814912716,
      "loss": 5.2606,
      "mean_token_accuracy": 0.18287449926137925,
      "num_tokens": 15846224.0,
      "step": 7620
    },
    {
      "entropy": 5.5843593120574955,
      "epoch": 0.6916727140783745,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959290566278803,
      "loss": 5.3062,
      "mean_token_accuracy": 0.1838727205991745,
      "num_tokens": 15855622.0,
      "step": 7625
    },
    {
      "entropy": 5.5864767074584964,
      "epoch": 0.6921262699564587,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004959229271956199,
      "loss": 5.3493,
      "mean_token_accuracy": 0.18438090682029723,
      "num_tokens": 15865835.0,
      "step": 7630
    },
    {
      "entropy": 5.511325550079346,
      "epoch": 0.6925798258345428,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004959167931946169,
      "loss": 5.2916,
      "mean_token_accuracy": 0.18278788179159164,
      "num_tokens": 15875915.0,
      "step": 7635
    },
    {
      "entropy": 5.570894479751587,
      "epoch": 0.693033381712627,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004959106546249984,
      "loss": 5.301,
      "mean_token_accuracy": 0.18494617342948913,
      "num_tokens": 15887111.0,
      "step": 7640
    },
    {
      "entropy": 5.589294576644898,
      "epoch": 0.6934869375907112,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004959045114868913,
      "loss": 5.4105,
      "mean_token_accuracy": 0.17258675992488862,
      "num_tokens": 15899528.0,
      "step": 7645
    },
    {
      "entropy": 5.547265100479126,
      "epoch": 0.6939404934687954,
      "grad_norm": 0.875,
      "learning_rate": 0.0004958983637804225,
      "loss": 5.3194,
      "mean_token_accuracy": 0.18939780741930007,
      "num_tokens": 15910881.0,
      "step": 7650
    },
    {
      "entropy": 5.6097711563110355,
      "epoch": 0.6943940493468795,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004958922115057194,
      "loss": 5.3506,
      "mean_token_accuracy": 0.17834327965974808,
      "num_tokens": 15921189.0,
      "step": 7655
    },
    {
      "entropy": 5.611061525344849,
      "epoch": 0.6948476052249637,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004958860546629089,
      "loss": 5.3497,
      "mean_token_accuracy": 0.18507591933012008,
      "num_tokens": 15931276.0,
      "step": 7660
    },
    {
      "entropy": 5.58590931892395,
      "epoch": 0.6953011611030478,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004958798932521186,
      "loss": 5.429,
      "mean_token_accuracy": 0.1739620342850685,
      "num_tokens": 15941506.0,
      "step": 7665
    },
    {
      "entropy": 5.493170309066772,
      "epoch": 0.6957547169811321,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004958737272734759,
      "loss": 5.2574,
      "mean_token_accuracy": 0.18339498937129975,
      "num_tokens": 15951158.0,
      "step": 7670
    },
    {
      "entropy": 5.588201665878296,
      "epoch": 0.6962082728592163,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004958675567271081,
      "loss": 5.2997,
      "mean_token_accuracy": 0.18696109801530839,
      "num_tokens": 15960852.0,
      "step": 7675
    },
    {
      "entropy": 5.544498586654663,
      "epoch": 0.6966618287373004,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000495861381613143,
      "loss": 5.2936,
      "mean_token_accuracy": 0.17795197516679764,
      "num_tokens": 15971291.0,
      "step": 7680
    },
    {
      "entropy": 5.560691261291504,
      "epoch": 0.6971153846153846,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958552019317082,
      "loss": 5.3665,
      "mean_token_accuracy": 0.18023389130830764,
      "num_tokens": 15982180.0,
      "step": 7685
    },
    {
      "entropy": 5.486464738845825,
      "epoch": 0.6975689404934688,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004958490176829316,
      "loss": 5.2506,
      "mean_token_accuracy": 0.18482301533222198,
      "num_tokens": 15992363.0,
      "step": 7690
    },
    {
      "entropy": 5.53675217628479,
      "epoch": 0.698022496371553,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000495842828866941,
      "loss": 5.3368,
      "mean_token_accuracy": 0.1778530851006508,
      "num_tokens": 16003477.0,
      "step": 7695
    },
    {
      "entropy": 5.585207319259643,
      "epoch": 0.6984760522496372,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004958366354838645,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1820820912718773,
      "num_tokens": 16013747.0,
      "step": 7700
    },
    {
      "entropy": 5.542598628997803,
      "epoch": 0.6989296081277213,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00049583043753383,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1875500798225403,
      "num_tokens": 16025662.0,
      "step": 7705
    },
    {
      "entropy": 5.572959136962891,
      "epoch": 0.6993831640058055,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004958242350169658,
      "loss": 5.3992,
      "mean_token_accuracy": 0.1829013928771019,
      "num_tokens": 16037315.0,
      "step": 7710
    },
    {
      "entropy": 5.502584838867188,
      "epoch": 0.6998367198838897,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004958180279334001,
      "loss": 5.2262,
      "mean_token_accuracy": 0.1828617572784424,
      "num_tokens": 16047297.0,
      "step": 7715
    },
    {
      "entropy": 5.482805871963501,
      "epoch": 0.7002902757619739,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004958118162832614,
      "loss": 5.2115,
      "mean_token_accuracy": 0.18582056909799577,
      "num_tokens": 16057573.0,
      "step": 7720
    },
    {
      "entropy": 5.51785831451416,
      "epoch": 0.700743831640058,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004958056000666781,
      "loss": 5.3708,
      "mean_token_accuracy": 0.17752822488546371,
      "num_tokens": 16067926.0,
      "step": 7725
    },
    {
      "entropy": 5.560018396377563,
      "epoch": 0.7011973875181422,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004957993792837786,
      "loss": 5.3089,
      "mean_token_accuracy": 0.18352362066507338,
      "num_tokens": 16077938.0,
      "step": 7730
    },
    {
      "entropy": 5.509559869766235,
      "epoch": 0.7016509433962265,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004957931539346917,
      "loss": 5.1887,
      "mean_token_accuracy": 0.18910111486911774,
      "num_tokens": 16088163.0,
      "step": 7735
    },
    {
      "entropy": 5.6044848442077635,
      "epoch": 0.7021044992743106,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004957869240195462,
      "loss": 5.3964,
      "mean_token_accuracy": 0.17759514451026917,
      "num_tokens": 16100640.0,
      "step": 7740
    },
    {
      "entropy": 5.525788402557373,
      "epoch": 0.7025580551523948,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004957806895384707,
      "loss": 5.22,
      "mean_token_accuracy": 0.1873020723462105,
      "num_tokens": 16111018.0,
      "step": 7745
    },
    {
      "entropy": 5.509130525588989,
      "epoch": 0.7030116110304789,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004957744504915944,
      "loss": 5.2634,
      "mean_token_accuracy": 0.17722718566656112,
      "num_tokens": 16120873.0,
      "step": 7750
    },
    {
      "entropy": 5.563656759262085,
      "epoch": 0.7034651669085631,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957682068790463,
      "loss": 5.2845,
      "mean_token_accuracy": 0.18420600146055222,
      "num_tokens": 16131178.0,
      "step": 7755
    },
    {
      "entropy": 5.503865575790405,
      "epoch": 0.7039187227866474,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004957619587009553,
      "loss": 5.2961,
      "mean_token_accuracy": 0.18365350663661956,
      "num_tokens": 16141156.0,
      "step": 7760
    },
    {
      "entropy": 5.544031953811645,
      "epoch": 0.7043722786647315,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004957557059574507,
      "loss": 5.397,
      "mean_token_accuracy": 0.18116726726293564,
      "num_tokens": 16150077.0,
      "step": 7765
    },
    {
      "entropy": 5.521406173706055,
      "epoch": 0.7048258345428157,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004957494486486618,
      "loss": 5.307,
      "mean_token_accuracy": 0.1804828479886055,
      "num_tokens": 16160514.0,
      "step": 7770
    },
    {
      "entropy": 5.567109632492065,
      "epoch": 0.7052793904208998,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004957431867747182,
      "loss": 5.3432,
      "mean_token_accuracy": 0.187213496863842,
      "num_tokens": 16170811.0,
      "step": 7775
    },
    {
      "entropy": 5.480603742599487,
      "epoch": 0.7057329462989841,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004957369203357492,
      "loss": 5.2402,
      "mean_token_accuracy": 0.18310723453760147,
      "num_tokens": 16181553.0,
      "step": 7780
    },
    {
      "entropy": 5.523513269424439,
      "epoch": 0.7061865021770682,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004957306493318844,
      "loss": 5.2696,
      "mean_token_accuracy": 0.18861202895641327,
      "num_tokens": 16191308.0,
      "step": 7785
    },
    {
      "entropy": 5.526185083389282,
      "epoch": 0.7066400580551524,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004957243737632535,
      "loss": 5.2552,
      "mean_token_accuracy": 0.18830116391181945,
      "num_tokens": 16201311.0,
      "step": 7790
    },
    {
      "entropy": 5.584646558761596,
      "epoch": 0.7070936139332366,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004957180936299863,
      "loss": 5.2822,
      "mean_token_accuracy": 0.19124689400196077,
      "num_tokens": 16211105.0,
      "step": 7795
    },
    {
      "entropy": 5.554888391494751,
      "epoch": 0.7075471698113207,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004957118089322127,
      "loss": 5.3012,
      "mean_token_accuracy": 0.1761303499341011,
      "num_tokens": 16220836.0,
      "step": 7800
    },
    {
      "entropy": 5.484191131591797,
      "epoch": 0.708000725689405,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004957055196700628,
      "loss": 5.2146,
      "mean_token_accuracy": 0.18652159571647645,
      "num_tokens": 16231630.0,
      "step": 7805
    },
    {
      "entropy": 5.544210433959961,
      "epoch": 0.7084542815674891,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004956992258436663,
      "loss": 5.304,
      "mean_token_accuracy": 0.18482640385627747,
      "num_tokens": 16242272.0,
      "step": 7810
    },
    {
      "entropy": 5.429587078094483,
      "epoch": 0.7089078374455733,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004956929274531536,
      "loss": 5.2099,
      "mean_token_accuracy": 0.19073296040296556,
      "num_tokens": 16252555.0,
      "step": 7815
    },
    {
      "entropy": 5.518241500854492,
      "epoch": 0.7093613933236574,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000495686624498655,
      "loss": 5.331,
      "mean_token_accuracy": 0.187027183175087,
      "num_tokens": 16261964.0,
      "step": 7820
    },
    {
      "entropy": 5.617752361297607,
      "epoch": 0.7098149492017417,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004956803169803007,
      "loss": 5.3859,
      "mean_token_accuracy": 0.17545270025730134,
      "num_tokens": 16274543.0,
      "step": 7825
    },
    {
      "entropy": 5.507696151733398,
      "epoch": 0.7102685050798259,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004956740048982212,
      "loss": 5.2941,
      "mean_token_accuracy": 0.1758976697921753,
      "num_tokens": 16285124.0,
      "step": 7830
    },
    {
      "entropy": 5.528030920028686,
      "epoch": 0.71072206095791,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956676882525472,
      "loss": 5.3634,
      "mean_token_accuracy": 0.18425541818141938,
      "num_tokens": 16295727.0,
      "step": 7835
    },
    {
      "entropy": 5.628829622268677,
      "epoch": 0.7111756168359942,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000495661367043409,
      "loss": 5.2686,
      "mean_token_accuracy": 0.18808830976486207,
      "num_tokens": 16305496.0,
      "step": 7840
    },
    {
      "entropy": 5.535170650482177,
      "epoch": 0.7116291727140783,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956550412709376,
      "loss": 5.2455,
      "mean_token_accuracy": 0.19257238060235976,
      "num_tokens": 16315112.0,
      "step": 7845
    },
    {
      "entropy": 5.514770650863648,
      "epoch": 0.7120827285921626,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004956487109352637,
      "loss": 5.2948,
      "mean_token_accuracy": 0.18559942841529847,
      "num_tokens": 16325613.0,
      "step": 7850
    },
    {
      "entropy": 5.510515832901001,
      "epoch": 0.7125362844702468,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004956423760365181,
      "loss": 5.2236,
      "mean_token_accuracy": 0.18841205388307572,
      "num_tokens": 16335317.0,
      "step": 7855
    },
    {
      "entropy": 5.594292879104614,
      "epoch": 0.7129898403483309,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956360365748321,
      "loss": 5.3684,
      "mean_token_accuracy": 0.17692358642816544,
      "num_tokens": 16345003.0,
      "step": 7860
    },
    {
      "entropy": 5.5583514213562015,
      "epoch": 0.7134433962264151,
      "grad_norm": 0.875,
      "learning_rate": 0.0004956296925503367,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1829149454832077,
      "num_tokens": 16356108.0,
      "step": 7865
    },
    {
      "entropy": 5.483655261993408,
      "epoch": 0.7138969521044993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004956233439631628,
      "loss": 5.1666,
      "mean_token_accuracy": 0.19129244983196259,
      "num_tokens": 16364733.0,
      "step": 7870
    },
    {
      "entropy": 5.558326196670532,
      "epoch": 0.7143505079825835,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004956169908134421,
      "loss": 5.273,
      "mean_token_accuracy": 0.18959595412015914,
      "num_tokens": 16375229.0,
      "step": 7875
    },
    {
      "entropy": 5.519115591049195,
      "epoch": 0.7148040638606676,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004956106331013057,
      "loss": 5.3594,
      "mean_token_accuracy": 0.18587884157896042,
      "num_tokens": 16386366.0,
      "step": 7880
    },
    {
      "entropy": 5.490304279327392,
      "epoch": 0.7152576197387518,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004956042708268853,
      "loss": 5.2427,
      "mean_token_accuracy": 0.1905361145734787,
      "num_tokens": 16397067.0,
      "step": 7885
    },
    {
      "entropy": 5.452323913574219,
      "epoch": 0.715711175616836,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004955979039903123,
      "loss": 5.1938,
      "mean_token_accuracy": 0.18583076745271682,
      "num_tokens": 16407256.0,
      "step": 7890
    },
    {
      "entropy": 5.501726055145264,
      "epoch": 0.7161647314949202,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004955915325917183,
      "loss": 5.2573,
      "mean_token_accuracy": 0.18402386009693145,
      "num_tokens": 16418701.0,
      "step": 7895
    },
    {
      "entropy": 5.472748422622681,
      "epoch": 0.7166182873730044,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004955851566312354,
      "loss": 5.1796,
      "mean_token_accuracy": 0.19416148364543914,
      "num_tokens": 16428019.0,
      "step": 7900
    },
    {
      "entropy": 5.482213354110717,
      "epoch": 0.7170718432510885,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004955787761089952,
      "loss": 5.2658,
      "mean_token_accuracy": 0.179888778924942,
      "num_tokens": 16440122.0,
      "step": 7905
    },
    {
      "entropy": 5.606745100021362,
      "epoch": 0.7175253991291727,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955723910251296,
      "loss": 5.3425,
      "mean_token_accuracy": 0.1781800538301468,
      "num_tokens": 16449801.0,
      "step": 7910
    },
    {
      "entropy": 5.498953914642334,
      "epoch": 0.717978955007257,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004955660013797709,
      "loss": 5.2049,
      "mean_token_accuracy": 0.18978928178548812,
      "num_tokens": 16460155.0,
      "step": 7915
    },
    {
      "entropy": 5.5611310482025145,
      "epoch": 0.7184325108853411,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495559607173051,
      "loss": 5.4094,
      "mean_token_accuracy": 0.17007175385951995,
      "num_tokens": 16470656.0,
      "step": 7920
    },
    {
      "entropy": 5.565587759017944,
      "epoch": 0.7188860667634253,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004955532084051022,
      "loss": 5.3028,
      "mean_token_accuracy": 0.18778245747089387,
      "num_tokens": 16481449.0,
      "step": 7925
    },
    {
      "entropy": 5.544949960708618,
      "epoch": 0.7193396226415094,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004955468050760568,
      "loss": 5.3213,
      "mean_token_accuracy": 0.18307759761810302,
      "num_tokens": 16492032.0,
      "step": 7930
    },
    {
      "entropy": 5.559630060195923,
      "epoch": 0.7197931785195936,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004955403971860474,
      "loss": 5.2449,
      "mean_token_accuracy": 0.18228569626808167,
      "num_tokens": 16502221.0,
      "step": 7935
    },
    {
      "entropy": 5.516655302047729,
      "epoch": 0.7202467343976778,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004955339847352063,
      "loss": 5.2255,
      "mean_token_accuracy": 0.18727567791938782,
      "num_tokens": 16512021.0,
      "step": 7940
    },
    {
      "entropy": 5.526650905609131,
      "epoch": 0.720700290275762,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004955275677236662,
      "loss": 5.2781,
      "mean_token_accuracy": 0.18725835978984834,
      "num_tokens": 16522060.0,
      "step": 7945
    },
    {
      "entropy": 5.522590780258179,
      "epoch": 0.7211538461538461,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004955211461515599,
      "loss": 5.2851,
      "mean_token_accuracy": 0.18848953992128373,
      "num_tokens": 16532827.0,
      "step": 7950
    },
    {
      "entropy": 5.5282035827636715,
      "epoch": 0.7216074020319303,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00049551472001902,
      "loss": 5.3291,
      "mean_token_accuracy": 0.18570098727941514,
      "num_tokens": 16542838.0,
      "step": 7955
    },
    {
      "entropy": 5.49766263961792,
      "epoch": 0.7220609579100146,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004955082893261796,
      "loss": 5.2574,
      "mean_token_accuracy": 0.18782561421394348,
      "num_tokens": 16552615.0,
      "step": 7960
    },
    {
      "entropy": 5.5685553550720215,
      "epoch": 0.7225145137880987,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955018540731714,
      "loss": 5.3358,
      "mean_token_accuracy": 0.18197497874498367,
      "num_tokens": 16561777.0,
      "step": 7965
    },
    {
      "entropy": 5.561930131912232,
      "epoch": 0.7229680696661829,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004954954142601288,
      "loss": 5.3046,
      "mean_token_accuracy": 0.1859089732170105,
      "num_tokens": 16572111.0,
      "step": 7970
    },
    {
      "entropy": 5.553031778335571,
      "epoch": 0.723421625544267,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004954889698871849,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1823531538248062,
      "num_tokens": 16581674.0,
      "step": 7975
    },
    {
      "entropy": 5.5707708358764645,
      "epoch": 0.7238751814223512,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004954825209544728,
      "loss": 5.3265,
      "mean_token_accuracy": 0.18853056877851487,
      "num_tokens": 16591802.0,
      "step": 7980
    },
    {
      "entropy": 5.4827385425567625,
      "epoch": 0.7243287373004355,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495476067462126,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1914340078830719,
      "num_tokens": 16602772.0,
      "step": 7985
    },
    {
      "entropy": 5.56507248878479,
      "epoch": 0.7247822931785196,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495469609410278,
      "loss": 5.2697,
      "mean_token_accuracy": 0.18687217831611633,
      "num_tokens": 16613526.0,
      "step": 7990
    },
    {
      "entropy": 5.475165891647339,
      "epoch": 0.7252358490566038,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004954631467990622,
      "loss": 5.2738,
      "mean_token_accuracy": 0.1871074303984642,
      "num_tokens": 16623583.0,
      "step": 7995
    },
    {
      "entropy": 5.543488788604736,
      "epoch": 0.7256894049346879,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004954566796286124,
      "loss": 5.3322,
      "mean_token_accuracy": 0.1807237446308136,
      "num_tokens": 16635068.0,
      "step": 8000
    },
    {
      "entropy": 5.588771057128906,
      "epoch": 0.7261429608127722,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004954502078990623,
      "loss": 5.2332,
      "mean_token_accuracy": 0.18706244379281997,
      "num_tokens": 16645021.0,
      "step": 8005
    },
    {
      "entropy": 5.526928997039795,
      "epoch": 0.7265965166908563,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004954437316105457,
      "loss": 5.3204,
      "mean_token_accuracy": 0.17738464921712876,
      "num_tokens": 16656021.0,
      "step": 8010
    },
    {
      "entropy": 5.476295518875122,
      "epoch": 0.7270500725689405,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954372507631965,
      "loss": 5.2173,
      "mean_token_accuracy": 0.19314096868038177,
      "num_tokens": 16665236.0,
      "step": 8015
    },
    {
      "entropy": 5.545669603347778,
      "epoch": 0.7275036284470247,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004954307653571489,
      "loss": 5.4112,
      "mean_token_accuracy": 0.17193434983491898,
      "num_tokens": 16676690.0,
      "step": 8020
    },
    {
      "entropy": 5.553662109375,
      "epoch": 0.7279571843251088,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954242753925367,
      "loss": 5.2901,
      "mean_token_accuracy": 0.17670878767967224,
      "num_tokens": 16687288.0,
      "step": 8025
    },
    {
      "entropy": 5.5189173221588135,
      "epoch": 0.7284107402031931,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004954177808694946,
      "loss": 5.3052,
      "mean_token_accuracy": 0.1775049388408661,
      "num_tokens": 16697873.0,
      "step": 8030
    },
    {
      "entropy": 5.551232862472534,
      "epoch": 0.7288642960812772,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004954112817881565,
      "loss": 5.3207,
      "mean_token_accuracy": 0.18077222257852554,
      "num_tokens": 16709002.0,
      "step": 8035
    },
    {
      "entropy": 5.493309879302979,
      "epoch": 0.7293178519593614,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004954047781486568,
      "loss": 5.278,
      "mean_token_accuracy": 0.18425150364637374,
      "num_tokens": 16718989.0,
      "step": 8040
    },
    {
      "entropy": 5.620672035217285,
      "epoch": 0.7297714078374455,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004953982699511303,
      "loss": 5.3692,
      "mean_token_accuracy": 0.18326299786567687,
      "num_tokens": 16730425.0,
      "step": 8045
    },
    {
      "entropy": 5.562078189849854,
      "epoch": 0.7302249637155298,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004953917571957114,
      "loss": 5.2638,
      "mean_token_accuracy": 0.18367500156164168,
      "num_tokens": 16740216.0,
      "step": 8050
    },
    {
      "entropy": 5.507948780059815,
      "epoch": 0.730678519593614,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004953852398825346,
      "loss": 5.2517,
      "mean_token_accuracy": 0.18657175153493882,
      "num_tokens": 16752436.0,
      "step": 8055
    },
    {
      "entropy": 5.471053600311279,
      "epoch": 0.7311320754716981,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495378718011735,
      "loss": 5.2782,
      "mean_token_accuracy": 0.17842504680156707,
      "num_tokens": 16762631.0,
      "step": 8060
    },
    {
      "entropy": 5.580714368820191,
      "epoch": 0.7315856313497823,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004953721915834473,
      "loss": 5.2878,
      "mean_token_accuracy": 0.18535203933715821,
      "num_tokens": 16772613.0,
      "step": 8065
    },
    {
      "entropy": 5.585743761062622,
      "epoch": 0.7320391872278664,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004953656605978067,
      "loss": 5.2995,
      "mean_token_accuracy": 0.17919224500656128,
      "num_tokens": 16782963.0,
      "step": 8070
    },
    {
      "entropy": 5.497876787185669,
      "epoch": 0.7324927431059507,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000495359125054948,
      "loss": 5.2248,
      "mean_token_accuracy": 0.18719250857830047,
      "num_tokens": 16793892.0,
      "step": 8075
    },
    {
      "entropy": 5.433376407623291,
      "epoch": 0.7329462989840348,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953525849550063,
      "loss": 5.2243,
      "mean_token_accuracy": 0.1793593570590019,
      "num_tokens": 16803388.0,
      "step": 8080
    },
    {
      "entropy": 5.505691719055176,
      "epoch": 0.733399854862119,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004953460402981172,
      "loss": 5.1801,
      "mean_token_accuracy": 0.19310829937458038,
      "num_tokens": 16813201.0,
      "step": 8085
    },
    {
      "entropy": 5.602481555938721,
      "epoch": 0.7338534107402032,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004953394910844157,
      "loss": 5.4435,
      "mean_token_accuracy": 0.18122223764657974,
      "num_tokens": 16823265.0,
      "step": 8090
    },
    {
      "entropy": 5.53153166770935,
      "epoch": 0.7343069666182874,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953329373140374,
      "loss": 5.2311,
      "mean_token_accuracy": 0.18478136509656906,
      "num_tokens": 16833057.0,
      "step": 8095
    },
    {
      "entropy": 5.486098194122315,
      "epoch": 0.7347605224963716,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004953263789871179,
      "loss": 5.1915,
      "mean_token_accuracy": 0.18385787308216095,
      "num_tokens": 16843488.0,
      "step": 8100
    },
    {
      "entropy": 5.463052749633789,
      "epoch": 0.7352140783744557,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004953198161037925,
      "loss": 5.2102,
      "mean_token_accuracy": 0.19679825603961945,
      "num_tokens": 16852448.0,
      "step": 8105
    },
    {
      "entropy": 5.431389713287354,
      "epoch": 0.7356676342525399,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004953132486641975,
      "loss": 5.2512,
      "mean_token_accuracy": 0.1843101128935814,
      "num_tokens": 16863944.0,
      "step": 8110
    },
    {
      "entropy": 5.52769889831543,
      "epoch": 0.736121190130624,
      "grad_norm": 1.046875,
      "learning_rate": 0.000495306676668468,
      "loss": 5.1792,
      "mean_token_accuracy": 0.18727706223726273,
      "num_tokens": 16873923.0,
      "step": 8115
    },
    {
      "entropy": 5.547335052490235,
      "epoch": 0.7365747460087083,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004953001001167406,
      "loss": 5.2912,
      "mean_token_accuracy": 0.18075185418128967,
      "num_tokens": 16884373.0,
      "step": 8120
    },
    {
      "entropy": 5.570522165298462,
      "epoch": 0.7370283018867925,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004952935190091508,
      "loss": 5.4223,
      "mean_token_accuracy": 0.17607153505086898,
      "num_tokens": 16895869.0,
      "step": 8125
    },
    {
      "entropy": 5.455164957046509,
      "epoch": 0.7374818577648766,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952869333458349,
      "loss": 5.266,
      "mean_token_accuracy": 0.18525791615247728,
      "num_tokens": 16906737.0,
      "step": 8130
    },
    {
      "entropy": 5.48662314414978,
      "epoch": 0.7379354136429608,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004952803431269292,
      "loss": 5.1975,
      "mean_token_accuracy": 0.19422420412302016,
      "num_tokens": 16916083.0,
      "step": 8135
    },
    {
      "entropy": 5.597974872589111,
      "epoch": 0.738388969521045,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004952737483525699,
      "loss": 5.3612,
      "mean_token_accuracy": 0.18378081768751145,
      "num_tokens": 16926419.0,
      "step": 8140
    },
    {
      "entropy": 5.5718834400177,
      "epoch": 0.7388425253991292,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004952671490228932,
      "loss": 5.2942,
      "mean_token_accuracy": 0.1853596895933151,
      "num_tokens": 16936067.0,
      "step": 8145
    },
    {
      "entropy": 5.585398387908936,
      "epoch": 0.7392960812772134,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004952605451380357,
      "loss": 5.2489,
      "mean_token_accuracy": 0.18981460332870484,
      "num_tokens": 16946571.0,
      "step": 8150
    },
    {
      "entropy": 5.453392887115479,
      "epoch": 0.7397496371552975,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004952539366981342,
      "loss": 5.2966,
      "mean_token_accuracy": 0.18083768635988234,
      "num_tokens": 16956764.0,
      "step": 8155
    },
    {
      "entropy": 5.519579267501831,
      "epoch": 0.7402031930333817,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000495247323703325,
      "loss": 5.2249,
      "mean_token_accuracy": 0.1882965698838234,
      "num_tokens": 16966002.0,
      "step": 8160
    },
    {
      "entropy": 5.572782278060913,
      "epoch": 0.7406567489114659,
      "grad_norm": 1.03125,
      "learning_rate": 0.000495240706153745,
      "loss": 5.3257,
      "mean_token_accuracy": 0.18548445403575897,
      "num_tokens": 16975705.0,
      "step": 8165
    },
    {
      "entropy": 5.522975587844849,
      "epoch": 0.7411103047895501,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952340840495312,
      "loss": 5.2265,
      "mean_token_accuracy": 0.1855018049478531,
      "num_tokens": 16985945.0,
      "step": 8170
    },
    {
      "entropy": 5.4940248966217045,
      "epoch": 0.7415638606676342,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952274573908203,
      "loss": 5.2005,
      "mean_token_accuracy": 0.18602324575185775,
      "num_tokens": 16995475.0,
      "step": 8175
    },
    {
      "entropy": 5.510499334335327,
      "epoch": 0.7420174165457184,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004952208261777496,
      "loss": 5.2237,
      "mean_token_accuracy": 0.18946883529424668,
      "num_tokens": 17005028.0,
      "step": 8180
    },
    {
      "entropy": 5.531285905838013,
      "epoch": 0.7424709724238027,
      "grad_norm": 0.953125,
      "learning_rate": 0.000495214190410456,
      "loss": 5.2916,
      "mean_token_accuracy": 0.18801498115062715,
      "num_tokens": 17014786.0,
      "step": 8185
    },
    {
      "entropy": 5.563917732238769,
      "epoch": 0.7429245283018868,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004952075500890768,
      "loss": 5.2945,
      "mean_token_accuracy": 0.18754068166017532,
      "num_tokens": 17024431.0,
      "step": 8190
    },
    {
      "entropy": 5.541070652008057,
      "epoch": 0.743378084179971,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004952009052137493,
      "loss": 5.2722,
      "mean_token_accuracy": 0.18842438012361526,
      "num_tokens": 17034368.0,
      "step": 8195
    },
    {
      "entropy": 5.53280725479126,
      "epoch": 0.7438316400580551,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004951942557846112,
      "loss": 5.3039,
      "mean_token_accuracy": 0.182603819668293,
      "num_tokens": 17043818.0,
      "step": 8200
    },
    {
      "entropy": 5.568868350982666,
      "epoch": 0.7442851959361393,
      "grad_norm": 1.0,
      "learning_rate": 0.0004951876018017997,
      "loss": 5.3101,
      "mean_token_accuracy": 0.18415388315916062,
      "num_tokens": 17054843.0,
      "step": 8205
    },
    {
      "entropy": 5.5368658065795895,
      "epoch": 0.7447387518142236,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004951809432654524,
      "loss": 5.3268,
      "mean_token_accuracy": 0.1834358587861061,
      "num_tokens": 17065539.0,
      "step": 8210
    },
    {
      "entropy": 5.491271352767944,
      "epoch": 0.7451923076923077,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000495174280175707,
      "loss": 5.2211,
      "mean_token_accuracy": 0.18677894920110702,
      "num_tokens": 17075443.0,
      "step": 8215
    },
    {
      "entropy": 5.592065238952637,
      "epoch": 0.7456458635703919,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004951676125327015,
      "loss": 5.3549,
      "mean_token_accuracy": 0.18309056311845778,
      "num_tokens": 17085490.0,
      "step": 8220
    },
    {
      "entropy": 5.606427907943726,
      "epoch": 0.746099419448476,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004951609403365737,
      "loss": 5.3441,
      "mean_token_accuracy": 0.1845713585615158,
      "num_tokens": 17094803.0,
      "step": 8225
    },
    {
      "entropy": 5.5416947364807125,
      "epoch": 0.7465529753265603,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004951542635874615,
      "loss": 5.2032,
      "mean_token_accuracy": 0.1835956171154976,
      "num_tokens": 17104373.0,
      "step": 8230
    },
    {
      "entropy": 5.5435961246490475,
      "epoch": 0.7470065312046444,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004951475822855029,
      "loss": 5.2741,
      "mean_token_accuracy": 0.18503042459487914,
      "num_tokens": 17115469.0,
      "step": 8235
    },
    {
      "entropy": 5.504471731185913,
      "epoch": 0.7474600870827286,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004951408964308364,
      "loss": 5.2491,
      "mean_token_accuracy": 0.1830831825733185,
      "num_tokens": 17125443.0,
      "step": 8240
    },
    {
      "entropy": 5.5065467834472654,
      "epoch": 0.7479136429608128,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004951342060236,
      "loss": 5.2229,
      "mean_token_accuracy": 0.19070644974708556,
      "num_tokens": 17136253.0,
      "step": 8245
    },
    {
      "entropy": 5.5766904830932615,
      "epoch": 0.7483671988388969,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951275110639321,
      "loss": 5.2796,
      "mean_token_accuracy": 0.18655992895364762,
      "num_tokens": 17145544.0,
      "step": 8250
    },
    {
      "entropy": 5.52774977684021,
      "epoch": 0.7488207547169812,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004951208115519712,
      "loss": 5.2905,
      "mean_token_accuracy": 0.18376002162694932,
      "num_tokens": 17155043.0,
      "step": 8255
    },
    {
      "entropy": 5.510969734191894,
      "epoch": 0.7492743105950653,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951141074878558,
      "loss": 5.3221,
      "mean_token_accuracy": 0.183181294798851,
      "num_tokens": 17164885.0,
      "step": 8260
    },
    {
      "entropy": 5.609808874130249,
      "epoch": 0.7497278664731495,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004951073988717246,
      "loss": 5.3253,
      "mean_token_accuracy": 0.18268342316150665,
      "num_tokens": 17175681.0,
      "step": 8265
    },
    {
      "entropy": 5.468179035186767,
      "epoch": 0.7501814223512336,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004951006857037163,
      "loss": 5.2009,
      "mean_token_accuracy": 0.18825145214796066,
      "num_tokens": 17186002.0,
      "step": 8270
    },
    {
      "entropy": 5.501715898513794,
      "epoch": 0.7506349782293179,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004950939679839696,
      "loss": 5.2434,
      "mean_token_accuracy": 0.18338658064603805,
      "num_tokens": 17195499.0,
      "step": 8275
    },
    {
      "entropy": 5.538295269012451,
      "epoch": 0.7510885341074021,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004950872457126237,
      "loss": 5.2498,
      "mean_token_accuracy": 0.18770517855882646,
      "num_tokens": 17204684.0,
      "step": 8280
    },
    {
      "entropy": 5.519716119766235,
      "epoch": 0.7515420899854862,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950805188898175,
      "loss": 5.3005,
      "mean_token_accuracy": 0.18001003712415695,
      "num_tokens": 17213987.0,
      "step": 8285
    },
    {
      "entropy": 5.542646598815918,
      "epoch": 0.7519956458635704,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049507378751569,
      "loss": 5.3083,
      "mean_token_accuracy": 0.18153341710567475,
      "num_tokens": 17224512.0,
      "step": 8290
    },
    {
      "entropy": 5.592179584503174,
      "epoch": 0.7524492017416545,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004950670515903805,
      "loss": 5.3213,
      "mean_token_accuracy": 0.18630932569503783,
      "num_tokens": 17234362.0,
      "step": 8295
    },
    {
      "entropy": 5.5631856441497805,
      "epoch": 0.7529027576197388,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004950603111140283,
      "loss": 5.3047,
      "mean_token_accuracy": 0.18705005645751954,
      "num_tokens": 17244241.0,
      "step": 8300
    },
    {
      "entropy": 5.497237014770508,
      "epoch": 0.753356313497823,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950535660867728,
      "loss": 5.2429,
      "mean_token_accuracy": 0.1937210038304329,
      "num_tokens": 17254689.0,
      "step": 8305
    },
    {
      "entropy": 5.5569257736206055,
      "epoch": 0.7538098693759071,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004950468165087535,
      "loss": 5.3237,
      "mean_token_accuracy": 0.1863400772213936,
      "num_tokens": 17265506.0,
      "step": 8310
    },
    {
      "entropy": 5.533684968948364,
      "epoch": 0.7542634252539913,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004950400623801098,
      "loss": 5.3973,
      "mean_token_accuracy": 0.18226426541805268,
      "num_tokens": 17276376.0,
      "step": 8315
    },
    {
      "entropy": 5.59690899848938,
      "epoch": 0.7547169811320755,
      "grad_norm": 0.875,
      "learning_rate": 0.0004950333037009818,
      "loss": 5.2811,
      "mean_token_accuracy": 0.17492688000202178,
      "num_tokens": 17287005.0,
      "step": 8320
    },
    {
      "entropy": 5.592009592056274,
      "epoch": 0.7551705370101597,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004950265404715089,
      "loss": 5.2924,
      "mean_token_accuracy": 0.1817374274134636,
      "num_tokens": 17297595.0,
      "step": 8325
    },
    {
      "entropy": 5.455151844024658,
      "epoch": 0.7556240928882438,
      "grad_norm": 0.953125,
      "learning_rate": 0.000495019772691831,
      "loss": 5.1694,
      "mean_token_accuracy": 0.1883062943816185,
      "num_tokens": 17307319.0,
      "step": 8330
    },
    {
      "entropy": 5.569844818115234,
      "epoch": 0.756077648766328,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950130003620882,
      "loss": 5.3959,
      "mean_token_accuracy": 0.18379779458045958,
      "num_tokens": 17317978.0,
      "step": 8335
    },
    {
      "entropy": 5.532192945480347,
      "epoch": 0.7565312046444121,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004950062234824204,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1818419873714447,
      "num_tokens": 17328339.0,
      "step": 8340
    },
    {
      "entropy": 5.53981204032898,
      "epoch": 0.7569847605224964,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004949994420529679,
      "loss": 5.3248,
      "mean_token_accuracy": 0.18118658065795898,
      "num_tokens": 17338668.0,
      "step": 8345
    },
    {
      "entropy": 5.4830098152160645,
      "epoch": 0.7574383164005806,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004949926560738709,
      "loss": 5.2692,
      "mean_token_accuracy": 0.17884783148765565,
      "num_tokens": 17348609.0,
      "step": 8350
    },
    {
      "entropy": 5.543556213378906,
      "epoch": 0.7578918722786647,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004949858655452697,
      "loss": 5.2878,
      "mean_token_accuracy": 0.18408870846033096,
      "num_tokens": 17358014.0,
      "step": 8355
    },
    {
      "entropy": 5.600489568710327,
      "epoch": 0.7583454281567489,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004949790704673048,
      "loss": 5.2905,
      "mean_token_accuracy": 0.18661998957395554,
      "num_tokens": 17367975.0,
      "step": 8360
    },
    {
      "entropy": 5.554596185684204,
      "epoch": 0.7587989840348331,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004949722708401166,
      "loss": 5.2372,
      "mean_token_accuracy": 0.18600034564733506,
      "num_tokens": 17377960.0,
      "step": 8365
    },
    {
      "entropy": 5.555735635757446,
      "epoch": 0.7592525399129173,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004949654666638457,
      "loss": 5.3269,
      "mean_token_accuracy": 0.18860286325216294,
      "num_tokens": 17387334.0,
      "step": 8370
    },
    {
      "entropy": 5.5340049266815186,
      "epoch": 0.7597060957910015,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494958657938633,
      "loss": 5.2708,
      "mean_token_accuracy": 0.18800106942653655,
      "num_tokens": 17396521.0,
      "step": 8375
    },
    {
      "entropy": 5.478679418563843,
      "epoch": 0.7601596516690856,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949518446646192,
      "loss": 5.2504,
      "mean_token_accuracy": 0.18262986093759537,
      "num_tokens": 17407217.0,
      "step": 8380
    },
    {
      "entropy": 5.634729862213135,
      "epoch": 0.7606132075471698,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004949450268419451,
      "loss": 5.3708,
      "mean_token_accuracy": 0.17887072116136551,
      "num_tokens": 17416982.0,
      "step": 8385
    },
    {
      "entropy": 5.479625225067139,
      "epoch": 0.761066763425254,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000494938204470752,
      "loss": 5.2013,
      "mean_token_accuracy": 0.18697880059480668,
      "num_tokens": 17427636.0,
      "step": 8390
    },
    {
      "entropy": 5.509683084487915,
      "epoch": 0.7615203193033382,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004949313775511806,
      "loss": 5.3342,
      "mean_token_accuracy": 0.179373337328434,
      "num_tokens": 17437488.0,
      "step": 8395
    },
    {
      "entropy": 5.50303807258606,
      "epoch": 0.7619738751814223,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004949245460833723,
      "loss": 5.1156,
      "mean_token_accuracy": 0.1915082797408104,
      "num_tokens": 17447838.0,
      "step": 8400
    },
    {
      "entropy": 5.572708368301392,
      "epoch": 0.7624274310595065,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949177100674684,
      "loss": 5.34,
      "mean_token_accuracy": 0.18229371309280396,
      "num_tokens": 17457257.0,
      "step": 8405
    },
    {
      "entropy": 5.531822872161865,
      "epoch": 0.7628809869375908,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004949108695036102,
      "loss": 5.2602,
      "mean_token_accuracy": 0.18731117099523545,
      "num_tokens": 17467729.0,
      "step": 8410
    },
    {
      "entropy": 5.531580114364624,
      "epoch": 0.7633345428156749,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949040243919392,
      "loss": 5.2793,
      "mean_token_accuracy": 0.18056346476078033,
      "num_tokens": 17477692.0,
      "step": 8415
    },
    {
      "entropy": 5.491760206222534,
      "epoch": 0.7637880986937591,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004948971747325968,
      "loss": 5.2807,
      "mean_token_accuracy": 0.18631262183189393,
      "num_tokens": 17488730.0,
      "step": 8420
    },
    {
      "entropy": 5.567406415939331,
      "epoch": 0.7642416545718432,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000494890320525725,
      "loss": 5.3518,
      "mean_token_accuracy": 0.18223876208066941,
      "num_tokens": 17498374.0,
      "step": 8425
    },
    {
      "entropy": 5.554366874694824,
      "epoch": 0.7646952104499274,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948834617714652,
      "loss": 5.2025,
      "mean_token_accuracy": 0.1882804214954376,
      "num_tokens": 17508601.0,
      "step": 8430
    },
    {
      "entropy": 5.512381458282471,
      "epoch": 0.7651487663280117,
      "grad_norm": 1.0,
      "learning_rate": 0.0004948765984699595,
      "loss": 5.2487,
      "mean_token_accuracy": 0.18550052344799042,
      "num_tokens": 17518164.0,
      "step": 8435
    },
    {
      "entropy": 5.567288446426391,
      "epoch": 0.7656023222060958,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004948697306213494,
      "loss": 5.2225,
      "mean_token_accuracy": 0.19012609124183655,
      "num_tokens": 17527244.0,
      "step": 8440
    },
    {
      "entropy": 5.549527502059936,
      "epoch": 0.76605587808418,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004948628582257775,
      "loss": 5.2897,
      "mean_token_accuracy": 0.1844577297568321,
      "num_tokens": 17538045.0,
      "step": 8445
    },
    {
      "entropy": 5.501225328445434,
      "epoch": 0.7665094339622641,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948559812833857,
      "loss": 5.2492,
      "mean_token_accuracy": 0.18562017679214476,
      "num_tokens": 17547584.0,
      "step": 8450
    },
    {
      "entropy": 5.550030612945557,
      "epoch": 0.7669629898403484,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004948490997943162,
      "loss": 5.3011,
      "mean_token_accuracy": 0.18173202872276306,
      "num_tokens": 17558925.0,
      "step": 8455
    },
    {
      "entropy": 5.558056306838989,
      "epoch": 0.7674165457184325,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948422137587111,
      "loss": 5.3712,
      "mean_token_accuracy": 0.17508765757083894,
      "num_tokens": 17569355.0,
      "step": 8460
    },
    {
      "entropy": 5.5319421768188475,
      "epoch": 0.7678701015965167,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004948353231767131,
      "loss": 5.2348,
      "mean_token_accuracy": 0.18913586884737016,
      "num_tokens": 17579782.0,
      "step": 8465
    },
    {
      "entropy": 5.539266347885132,
      "epoch": 0.7683236574746009,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004948284280484645,
      "loss": 5.2996,
      "mean_token_accuracy": 0.19015204161405563,
      "num_tokens": 17589772.0,
      "step": 8470
    },
    {
      "entropy": 5.481799364089966,
      "epoch": 0.768777213352685,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004948215283741081,
      "loss": 5.2051,
      "mean_token_accuracy": 0.18939271271228791,
      "num_tokens": 17599613.0,
      "step": 8475
    },
    {
      "entropy": 5.554580450057983,
      "epoch": 0.7692307692307693,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948146241537866,
      "loss": 5.2984,
      "mean_token_accuracy": 0.180326946079731,
      "num_tokens": 17609056.0,
      "step": 8480
    },
    {
      "entropy": 5.499348592758179,
      "epoch": 0.7696843251088534,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004948077153876425,
      "loss": 5.2124,
      "mean_token_accuracy": 0.1857810825109482,
      "num_tokens": 17618620.0,
      "step": 8485
    },
    {
      "entropy": 5.500620794296265,
      "epoch": 0.7701378809869376,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948008020758188,
      "loss": 5.1845,
      "mean_token_accuracy": 0.19046938568353652,
      "num_tokens": 17627630.0,
      "step": 8490
    },
    {
      "entropy": 5.409573554992676,
      "epoch": 0.7705914368650217,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004947938842184585,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1935472920536995,
      "num_tokens": 17638903.0,
      "step": 8495
    },
    {
      "entropy": 5.469343042373657,
      "epoch": 0.771044992743106,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004947869618157048,
      "loss": 5.2378,
      "mean_token_accuracy": 0.18261487334966658,
      "num_tokens": 17649257.0,
      "step": 8500
    },
    {
      "entropy": 5.481932640075684,
      "epoch": 0.7714985486211902,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004947800348677006,
      "loss": 5.2129,
      "mean_token_accuracy": 0.19850903004407883,
      "num_tokens": 17659495.0,
      "step": 8505
    },
    {
      "entropy": 5.464404773712158,
      "epoch": 0.7719521044992743,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947731033745893,
      "loss": 5.1901,
      "mean_token_accuracy": 0.18522430658340455,
      "num_tokens": 17669671.0,
      "step": 8510
    },
    {
      "entropy": 5.5648661136627195,
      "epoch": 0.7724056603773585,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004947661673365142,
      "loss": 5.3616,
      "mean_token_accuracy": 0.18332356959581375,
      "num_tokens": 17680927.0,
      "step": 8515
    },
    {
      "entropy": 5.506349372863769,
      "epoch": 0.7728592162554426,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004947592267536188,
      "loss": 5.2245,
      "mean_token_accuracy": 0.19083321392536162,
      "num_tokens": 17691378.0,
      "step": 8520
    },
    {
      "entropy": 5.522361421585083,
      "epoch": 0.7733127721335269,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004947522816260466,
      "loss": 5.3236,
      "mean_token_accuracy": 0.18290676027536393,
      "num_tokens": 17700942.0,
      "step": 8525
    },
    {
      "entropy": 5.499392032623291,
      "epoch": 0.773766328011611,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947453319539411,
      "loss": 5.2144,
      "mean_token_accuracy": 0.18726398050785065,
      "num_tokens": 17710076.0,
      "step": 8530
    },
    {
      "entropy": 5.495902681350708,
      "epoch": 0.7742198838896952,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004947383777374462,
      "loss": 5.2539,
      "mean_token_accuracy": 0.1863068625330925,
      "num_tokens": 17721009.0,
      "step": 8535
    },
    {
      "entropy": 5.479369497299194,
      "epoch": 0.7746734397677794,
      "grad_norm": 0.875,
      "learning_rate": 0.0004947314189767058,
      "loss": 5.2782,
      "mean_token_accuracy": 0.18749626576900483,
      "num_tokens": 17731821.0,
      "step": 8540
    },
    {
      "entropy": 5.466885328292847,
      "epoch": 0.7751269956458636,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004947244556718635,
      "loss": 5.1684,
      "mean_token_accuracy": 0.19538922607898712,
      "num_tokens": 17741425.0,
      "step": 8545
    },
    {
      "entropy": 5.487680578231812,
      "epoch": 0.7755805515239478,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004947174878230635,
      "loss": 5.2167,
      "mean_token_accuracy": 0.19239004105329513,
      "num_tokens": 17752381.0,
      "step": 8550
    },
    {
      "entropy": 5.448737573623657,
      "epoch": 0.7760341074020319,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004947105154304498,
      "loss": 5.2126,
      "mean_token_accuracy": 0.18385005593299866,
      "num_tokens": 17762768.0,
      "step": 8555
    },
    {
      "entropy": 5.496819543838501,
      "epoch": 0.7764876632801161,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004947035384941667,
      "loss": 5.2306,
      "mean_token_accuracy": 0.1835770294070244,
      "num_tokens": 17774032.0,
      "step": 8560
    },
    {
      "entropy": 5.567213773727417,
      "epoch": 0.7769412191582002,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004946965570143584,
      "loss": 5.3774,
      "mean_token_accuracy": 0.17739971727132797,
      "num_tokens": 17785098.0,
      "step": 8565
    },
    {
      "entropy": 5.56190505027771,
      "epoch": 0.7773947750362845,
      "grad_norm": 1.0,
      "learning_rate": 0.0004946895709911695,
      "loss": 5.2412,
      "mean_token_accuracy": 0.1863827884197235,
      "num_tokens": 17795487.0,
      "step": 8570
    },
    {
      "entropy": 5.525628662109375,
      "epoch": 0.7778483309143687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004946825804247442,
      "loss": 5.3538,
      "mean_token_accuracy": 0.17812175750732423,
      "num_tokens": 17806045.0,
      "step": 8575
    },
    {
      "entropy": 5.567047452926635,
      "epoch": 0.7783018867924528,
      "grad_norm": 0.890625,
      "learning_rate": 0.000494675585315227,
      "loss": 5.314,
      "mean_token_accuracy": 0.18278959393501282,
      "num_tokens": 17816948.0,
      "step": 8580
    },
    {
      "entropy": 5.596051406860352,
      "epoch": 0.778755442670537,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004946685856627628,
      "loss": 5.2593,
      "mean_token_accuracy": 0.18712812513113022,
      "num_tokens": 17826929.0,
      "step": 8585
    },
    {
      "entropy": 5.555044555664063,
      "epoch": 0.7792089985486212,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946615814674963,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1852679133415222,
      "num_tokens": 17837780.0,
      "step": 8590
    },
    {
      "entropy": 5.394699335098267,
      "epoch": 0.7796625544267054,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004946545727295724,
      "loss": 5.1144,
      "mean_token_accuracy": 0.19915736168622972,
      "num_tokens": 17847835.0,
      "step": 8595
    },
    {
      "entropy": 5.485675191879272,
      "epoch": 0.7801161103047896,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946475594491359,
      "loss": 5.2043,
      "mean_token_accuracy": 0.19405315816402435,
      "num_tokens": 17858324.0,
      "step": 8600
    },
    {
      "entropy": 5.496988296508789,
      "epoch": 0.7805696661828737,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946405416263319,
      "loss": 5.2818,
      "mean_token_accuracy": 0.17678985744714737,
      "num_tokens": 17868455.0,
      "step": 8605
    },
    {
      "entropy": 5.545461845397949,
      "epoch": 0.7810232220609579,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946335192613054,
      "loss": 5.2465,
      "mean_token_accuracy": 0.18767861723899842,
      "num_tokens": 17879038.0,
      "step": 8610
    },
    {
      "entropy": 5.460289812088012,
      "epoch": 0.7814767779390421,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004946264923542019,
      "loss": 5.1285,
      "mean_token_accuracy": 0.19217866361141206,
      "num_tokens": 17888153.0,
      "step": 8615
    },
    {
      "entropy": 5.4844536781311035,
      "epoch": 0.7819303338171263,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004946194609051664,
      "loss": 5.2519,
      "mean_token_accuracy": 0.19185152649879456,
      "num_tokens": 17898289.0,
      "step": 8620
    },
    {
      "entropy": 5.501482677459717,
      "epoch": 0.7823838896952104,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004946124249143447,
      "loss": 5.2547,
      "mean_token_accuracy": 0.18857731074094772,
      "num_tokens": 17909192.0,
      "step": 8625
    },
    {
      "entropy": 5.4674238681793215,
      "epoch": 0.7828374455732946,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000494605384381882,
      "loss": 5.2012,
      "mean_token_accuracy": 0.18121700286865233,
      "num_tokens": 17918937.0,
      "step": 8630
    },
    {
      "entropy": 5.441970586776733,
      "epoch": 0.7832910014513788,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004945983393079239,
      "loss": 5.2272,
      "mean_token_accuracy": 0.1953953266143799,
      "num_tokens": 17929346.0,
      "step": 8635
    },
    {
      "entropy": 5.503742122650147,
      "epoch": 0.783744557329463,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004945912896926163,
      "loss": 5.2471,
      "mean_token_accuracy": 0.1864580824971199,
      "num_tokens": 17939479.0,
      "step": 8640
    },
    {
      "entropy": 5.592639398574829,
      "epoch": 0.7841981132075472,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004945842355361049,
      "loss": 5.3252,
      "mean_token_accuracy": 0.18374345749616622,
      "num_tokens": 17949841.0,
      "step": 8645
    },
    {
      "entropy": 5.471678686141968,
      "epoch": 0.7846516690856313,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945771768385355,
      "loss": 5.2526,
      "mean_token_accuracy": 0.19229422211647035,
      "num_tokens": 17960334.0,
      "step": 8650
    },
    {
      "entropy": 5.478883457183838,
      "epoch": 0.7851052249637155,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000494570113600054,
      "loss": 5.1961,
      "mean_token_accuracy": 0.19275954067707063,
      "num_tokens": 17970595.0,
      "step": 8655
    },
    {
      "entropy": 5.50038251876831,
      "epoch": 0.7855587808417998,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004945630458208068,
      "loss": 5.3421,
      "mean_token_accuracy": 0.17421732395887374,
      "num_tokens": 17981301.0,
      "step": 8660
    },
    {
      "entropy": 5.457895803451538,
      "epoch": 0.7860123367198839,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004945559735009397,
      "loss": 5.1658,
      "mean_token_accuracy": 0.191603122651577,
      "num_tokens": 17991262.0,
      "step": 8665
    },
    {
      "entropy": 5.568162202835083,
      "epoch": 0.7864658925979681,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004945488966405993,
      "loss": 5.2024,
      "mean_token_accuracy": 0.18861326873302459,
      "num_tokens": 18001212.0,
      "step": 8670
    },
    {
      "entropy": 5.501408624649048,
      "epoch": 0.7869194484760522,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004945418152399317,
      "loss": 5.2229,
      "mean_token_accuracy": 0.19007271379232407,
      "num_tokens": 18010432.0,
      "step": 8675
    },
    {
      "entropy": 5.5106878757476805,
      "epoch": 0.7873730043541364,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004945347292990833,
      "loss": 5.2655,
      "mean_token_accuracy": 0.18418308347463608,
      "num_tokens": 18020902.0,
      "step": 8680
    },
    {
      "entropy": 5.512370157241821,
      "epoch": 0.7878265602322206,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004945276388182008,
      "loss": 5.132,
      "mean_token_accuracy": 0.1858739361166954,
      "num_tokens": 18031133.0,
      "step": 8685
    },
    {
      "entropy": 5.567989873886108,
      "epoch": 0.7882801161103048,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004945205437974309,
      "loss": 5.2472,
      "mean_token_accuracy": 0.18936766386032106,
      "num_tokens": 18042050.0,
      "step": 8690
    },
    {
      "entropy": 5.428337478637696,
      "epoch": 0.788733671988389,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004945134442369201,
      "loss": 5.214,
      "mean_token_accuracy": 0.19088352620601653,
      "num_tokens": 18051642.0,
      "step": 8695
    },
    {
      "entropy": 5.552801275253296,
      "epoch": 0.7891872278664731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945063401368154,
      "loss": 5.2664,
      "mean_token_accuracy": 0.1811126798391342,
      "num_tokens": 18062180.0,
      "step": 8700
    },
    {
      "entropy": 5.50035719871521,
      "epoch": 0.7896407837445574,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004944992314972637,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18759729713201523,
      "num_tokens": 18072302.0,
      "step": 8705
    },
    {
      "entropy": 5.486184597015381,
      "epoch": 0.7900943396226415,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004944921183184118,
      "loss": 5.2179,
      "mean_token_accuracy": 0.18937012553215027,
      "num_tokens": 18081824.0,
      "step": 8710
    },
    {
      "entropy": 5.53223443031311,
      "epoch": 0.7905478955007257,
      "grad_norm": 0.9375,
      "learning_rate": 0.000494485000600407,
      "loss": 5.2632,
      "mean_token_accuracy": 0.1890762209892273,
      "num_tokens": 18092696.0,
      "step": 8715
    },
    {
      "entropy": 5.482760906219482,
      "epoch": 0.7910014513788098,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944778783433965,
      "loss": 5.1892,
      "mean_token_accuracy": 0.1936694249510765,
      "num_tokens": 18101790.0,
      "step": 8720
    },
    {
      "entropy": 5.550623178482056,
      "epoch": 0.791455007256894,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004944707515475276,
      "loss": 5.3132,
      "mean_token_accuracy": 0.1843470126390457,
      "num_tokens": 18111572.0,
      "step": 8725
    },
    {
      "entropy": 5.5540313720703125,
      "epoch": 0.7919085631349783,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004944636202129475,
      "loss": 5.2942,
      "mean_token_accuracy": 0.1926032617688179,
      "num_tokens": 18120650.0,
      "step": 8730
    },
    {
      "entropy": 5.47811222076416,
      "epoch": 0.7923621190130624,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004944564843398039,
      "loss": 5.1675,
      "mean_token_accuracy": 0.1985896944999695,
      "num_tokens": 18131055.0,
      "step": 8735
    },
    {
      "entropy": 5.573776483535767,
      "epoch": 0.7928156748911466,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004944493439282443,
      "loss": 5.2739,
      "mean_token_accuracy": 0.18772944211959838,
      "num_tokens": 18141423.0,
      "step": 8740
    },
    {
      "entropy": 5.4522857666015625,
      "epoch": 0.7932692307692307,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004944421989784162,
      "loss": 5.1932,
      "mean_token_accuracy": 0.18533381223678588,
      "num_tokens": 18152015.0,
      "step": 8745
    },
    {
      "entropy": 5.539948177337647,
      "epoch": 0.793722786647315,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004944350494904676,
      "loss": 5.2892,
      "mean_token_accuracy": 0.18949988037347792,
      "num_tokens": 18161934.0,
      "step": 8750
    },
    {
      "entropy": 5.496873092651367,
      "epoch": 0.7941763425253991,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004944278954645463,
      "loss": 5.2086,
      "mean_token_accuracy": 0.18725937306880952,
      "num_tokens": 18172287.0,
      "step": 8755
    },
    {
      "entropy": 5.482796335220337,
      "epoch": 0.7946298984034833,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004944207369008002,
      "loss": 5.2505,
      "mean_token_accuracy": 0.19790160655975342,
      "num_tokens": 18181015.0,
      "step": 8760
    },
    {
      "entropy": 5.577034568786621,
      "epoch": 0.7950834542815675,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004944135737993773,
      "loss": 5.3461,
      "mean_token_accuracy": 0.18178513646125793,
      "num_tokens": 18192382.0,
      "step": 8765
    },
    {
      "entropy": 5.488105583190918,
      "epoch": 0.7955370101596516,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004944064061604258,
      "loss": 5.2109,
      "mean_token_accuracy": 0.19367620944976807,
      "num_tokens": 18202597.0,
      "step": 8770
    },
    {
      "entropy": 5.523237991333008,
      "epoch": 0.7959905660377359,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004943992339840939,
      "loss": 5.2774,
      "mean_token_accuracy": 0.18589625358581544,
      "num_tokens": 18213756.0,
      "step": 8775
    },
    {
      "entropy": 5.434082651138306,
      "epoch": 0.79644412191582,
      "grad_norm": 0.921875,
      "learning_rate": 0.00049439205727053,
      "loss": 5.1884,
      "mean_token_accuracy": 0.18429725021123886,
      "num_tokens": 18223669.0,
      "step": 8780
    },
    {
      "entropy": 5.497088479995727,
      "epoch": 0.7968976777939042,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004943848760198823,
      "loss": 5.2104,
      "mean_token_accuracy": 0.19213344007730485,
      "num_tokens": 18234884.0,
      "step": 8785
    },
    {
      "entropy": 5.466268301010132,
      "epoch": 0.7973512336719883,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004943776902322995,
      "loss": 5.1506,
      "mean_token_accuracy": 0.1955432653427124,
      "num_tokens": 18245155.0,
      "step": 8790
    },
    {
      "entropy": 5.37814736366272,
      "epoch": 0.7978047895500726,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004943704999079302,
      "loss": 5.0843,
      "mean_token_accuracy": 0.19864733517169952,
      "num_tokens": 18254686.0,
      "step": 8795
    },
    {
      "entropy": 5.471942377090454,
      "epoch": 0.7982583454281568,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004943633050469231,
      "loss": 5.121,
      "mean_token_accuracy": 0.19447069764137268,
      "num_tokens": 18265473.0,
      "step": 8800
    },
    {
      "entropy": 5.53497724533081,
      "epoch": 0.7987119013062409,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004943561056494269,
      "loss": 5.3336,
      "mean_token_accuracy": 0.18408775627613067,
      "num_tokens": 18276024.0,
      "step": 8805
    },
    {
      "entropy": 5.5106282234191895,
      "epoch": 0.7991654571843251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004943489017155906,
      "loss": 5.2888,
      "mean_token_accuracy": 0.1785351738333702,
      "num_tokens": 18287261.0,
      "step": 8810
    },
    {
      "entropy": 5.560506153106689,
      "epoch": 0.7996190130624092,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004943416932455631,
      "loss": 5.305,
      "mean_token_accuracy": 0.18382198810577394,
      "num_tokens": 18299374.0,
      "step": 8815
    },
    {
      "entropy": 5.472782135009766,
      "epoch": 0.8000725689404935,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004943344802394935,
      "loss": 5.2679,
      "mean_token_accuracy": 0.1808692380785942,
      "num_tokens": 18309344.0,
      "step": 8820
    },
    {
      "entropy": 5.548999357223511,
      "epoch": 0.8005261248185777,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004943272626975309,
      "loss": 5.2525,
      "mean_token_accuracy": 0.18682713359594344,
      "num_tokens": 18319778.0,
      "step": 8825
    },
    {
      "entropy": 5.542529487609864,
      "epoch": 0.8009796806966618,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004943200406198246,
      "loss": 5.2544,
      "mean_token_accuracy": 0.18820199817419053,
      "num_tokens": 18329787.0,
      "step": 8830
    },
    {
      "entropy": 5.463485479354858,
      "epoch": 0.801433236574746,
      "grad_norm": 0.921875,
      "learning_rate": 0.000494312814006524,
      "loss": 5.2068,
      "mean_token_accuracy": 0.19309862405061723,
      "num_tokens": 18340670.0,
      "step": 8835
    },
    {
      "entropy": 5.48480076789856,
      "epoch": 0.8018867924528302,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004943055828577784,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18602313101291656,
      "num_tokens": 18352484.0,
      "step": 8840
    },
    {
      "entropy": 5.469893169403076,
      "epoch": 0.8023403483309144,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004942983471737376,
      "loss": 5.1777,
      "mean_token_accuracy": 0.19207196831703185,
      "num_tokens": 18363103.0,
      "step": 8845
    },
    {
      "entropy": 5.4201044082641605,
      "epoch": 0.8027939042089985,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004942911069545511,
      "loss": 5.2278,
      "mean_token_accuracy": 0.19178254008293152,
      "num_tokens": 18373925.0,
      "step": 8850
    },
    {
      "entropy": 5.5190746784210205,
      "epoch": 0.8032474600870827,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004942838622003686,
      "loss": 5.288,
      "mean_token_accuracy": 0.18306922763586045,
      "num_tokens": 18384081.0,
      "step": 8855
    },
    {
      "entropy": 5.582727241516113,
      "epoch": 0.8037010159651669,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942766129113399,
      "loss": 5.1672,
      "mean_token_accuracy": 0.19331018328666688,
      "num_tokens": 18392510.0,
      "step": 8860
    },
    {
      "entropy": 5.567781639099121,
      "epoch": 0.8041545718432511,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004942693590876151,
      "loss": 5.3186,
      "mean_token_accuracy": 0.182000532746315,
      "num_tokens": 18402809.0,
      "step": 8865
    },
    {
      "entropy": 5.5828429698944095,
      "epoch": 0.8046081277213353,
      "grad_norm": 1.078125,
      "learning_rate": 0.000494262100729344,
      "loss": 5.3322,
      "mean_token_accuracy": 0.18444409668445588,
      "num_tokens": 18413342.0,
      "step": 8870
    },
    {
      "entropy": 5.521259212493897,
      "epoch": 0.8050616835994194,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004942548378366768,
      "loss": 5.3185,
      "mean_token_accuracy": 0.1856078937649727,
      "num_tokens": 18424735.0,
      "step": 8875
    },
    {
      "entropy": 5.5181769847869875,
      "epoch": 0.8055152394775036,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004942475704097635,
      "loss": 5.2072,
      "mean_token_accuracy": 0.18817635774612426,
      "num_tokens": 18434797.0,
      "step": 8880
    },
    {
      "entropy": 5.514129638671875,
      "epoch": 0.8059687953555879,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004942402984487548,
      "loss": 5.2634,
      "mean_token_accuracy": 0.18100712597370147,
      "num_tokens": 18445456.0,
      "step": 8885
    },
    {
      "entropy": 5.485908126831054,
      "epoch": 0.806422351233672,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004942330219538008,
      "loss": 5.2398,
      "mean_token_accuracy": 0.18712347149848937,
      "num_tokens": 18456728.0,
      "step": 8890
    },
    {
      "entropy": 5.682804965972901,
      "epoch": 0.8068759071117562,
      "grad_norm": 1.125,
      "learning_rate": 0.000494225740925052,
      "loss": 5.3729,
      "mean_token_accuracy": 0.17928425073623658,
      "num_tokens": 18468813.0,
      "step": 8895
    },
    {
      "entropy": 5.553705883026123,
      "epoch": 0.8073294629898403,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004942184553626591,
      "loss": 5.2608,
      "mean_token_accuracy": 0.185865418612957,
      "num_tokens": 18477912.0,
      "step": 8900
    },
    {
      "entropy": 5.464331388473511,
      "epoch": 0.8077830188679245,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004942111652667726,
      "loss": 5.2155,
      "mean_token_accuracy": 0.19424942433834075,
      "num_tokens": 18488115.0,
      "step": 8905
    },
    {
      "entropy": 5.492030239105224,
      "epoch": 0.8082365747460087,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004942038706375433,
      "loss": 5.2499,
      "mean_token_accuracy": 0.18399515151977539,
      "num_tokens": 18498853.0,
      "step": 8910
    },
    {
      "entropy": 5.5277331352233885,
      "epoch": 0.8086901306240929,
      "grad_norm": 0.875,
      "learning_rate": 0.0004941965714751222,
      "loss": 5.2947,
      "mean_token_accuracy": 0.18574539422988892,
      "num_tokens": 18510047.0,
      "step": 8915
    },
    {
      "entropy": 5.511319303512574,
      "epoch": 0.809143686502177,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941892677796601,
      "loss": 5.2691,
      "mean_token_accuracy": 0.18705060333013535,
      "num_tokens": 18519673.0,
      "step": 8920
    },
    {
      "entropy": 5.497614240646362,
      "epoch": 0.8095972423802612,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004941819595513081,
      "loss": 5.2878,
      "mean_token_accuracy": 0.18286352902650832,
      "num_tokens": 18530090.0,
      "step": 8925
    },
    {
      "entropy": 5.597482538223266,
      "epoch": 0.8100507982583455,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004941746467902173,
      "loss": 5.3243,
      "mean_token_accuracy": 0.1817130759358406,
      "num_tokens": 18541862.0,
      "step": 8930
    },
    {
      "entropy": 5.573345136642456,
      "epoch": 0.8105043541364296,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000494167329496539,
      "loss": 5.3369,
      "mean_token_accuracy": 0.17775724679231644,
      "num_tokens": 18551624.0,
      "step": 8935
    },
    {
      "entropy": 5.514851140975952,
      "epoch": 0.8109579100145138,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941600076704244,
      "loss": 5.2497,
      "mean_token_accuracy": 0.18574765473604202,
      "num_tokens": 18561784.0,
      "step": 8940
    },
    {
      "entropy": 5.55006742477417,
      "epoch": 0.8114114658925979,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004941526813120251,
      "loss": 5.318,
      "mean_token_accuracy": 0.1817800968885422,
      "num_tokens": 18572110.0,
      "step": 8945
    },
    {
      "entropy": 5.550695466995239,
      "epoch": 0.8118650217706821,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941453504214924,
      "loss": 5.2702,
      "mean_token_accuracy": 0.1783347800374031,
      "num_tokens": 18582012.0,
      "step": 8950
    },
    {
      "entropy": 5.4474639892578125,
      "epoch": 0.8123185776487664,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004941380149989781,
      "loss": 5.1115,
      "mean_token_accuracy": 0.1958514556288719,
      "num_tokens": 18592850.0,
      "step": 8955
    },
    {
      "entropy": 5.4674022674560545,
      "epoch": 0.8127721335268505,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004941306750446336,
      "loss": 5.2945,
      "mean_token_accuracy": 0.1863408148288727,
      "num_tokens": 18602590.0,
      "step": 8960
    },
    {
      "entropy": 5.463942813873291,
      "epoch": 0.8132256894049347,
      "grad_norm": 1.015625,
      "learning_rate": 0.000494123330558611,
      "loss": 5.1962,
      "mean_token_accuracy": 0.19270357340574265,
      "num_tokens": 18612607.0,
      "step": 8965
    },
    {
      "entropy": 5.4990180969238285,
      "epoch": 0.8136792452830188,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004941159815410622,
      "loss": 5.1769,
      "mean_token_accuracy": 0.19223012775182724,
      "num_tokens": 18622936.0,
      "step": 8970
    },
    {
      "entropy": 5.516388368606568,
      "epoch": 0.8141328011611031,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000494108627992139,
      "loss": 5.3316,
      "mean_token_accuracy": 0.18003688007593155,
      "num_tokens": 18633997.0,
      "step": 8975
    },
    {
      "entropy": 5.481217002868652,
      "epoch": 0.8145863570391872,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004941012699119936,
      "loss": 5.1633,
      "mean_token_accuracy": 0.19657521694898605,
      "num_tokens": 18644347.0,
      "step": 8980
    },
    {
      "entropy": 5.528944206237793,
      "epoch": 0.8150399129172714,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004940939073007779,
      "loss": 5.2399,
      "mean_token_accuracy": 0.18880648016929627,
      "num_tokens": 18654989.0,
      "step": 8985
    },
    {
      "entropy": 5.4825615882873535,
      "epoch": 0.8154934687953556,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004940865401586446,
      "loss": 5.2116,
      "mean_token_accuracy": 0.19149848967790603,
      "num_tokens": 18665526.0,
      "step": 8990
    },
    {
      "entropy": 5.465846824645996,
      "epoch": 0.8159470246734397,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004940791684857458,
      "loss": 5.2131,
      "mean_token_accuracy": 0.19184787720441818,
      "num_tokens": 18676226.0,
      "step": 8995
    },
    {
      "entropy": 5.5053871154785154,
      "epoch": 0.816400580551524,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004940717922822339,
      "loss": 5.1511,
      "mean_token_accuracy": 0.1933094546198845,
      "num_tokens": 18686499.0,
      "step": 9000
    },
    {
      "epoch": 0.816400580551524,
      "eval_entropy": 5.358133128541627,
      "eval_loss": 5.252007961273193,
      "eval_mean_token_accuracy": 0.19481117964819142,
      "eval_num_tokens": 18686499.0,
      "eval_runtime": 20.7388,
      "eval_samples_per_second": 1705.014,
      "eval_steps_per_second": 213.127,
      "step": 9000
    },
    {
      "entropy": 5.569316291809082,
      "epoch": 0.8168541364296081,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004940644115482615,
      "loss": 5.2286,
      "mean_token_accuracy": 0.18744052350521087,
      "num_tokens": 18696543.0,
      "step": 9005
    },
    {
      "entropy": 5.459391164779663,
      "epoch": 0.8173076923076923,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004940570262839812,
      "loss": 5.2701,
      "mean_token_accuracy": 0.18510979413986206,
      "num_tokens": 18707029.0,
      "step": 9010
    },
    {
      "entropy": 5.494412612915039,
      "epoch": 0.8177612481857764,
      "grad_norm": 1.0,
      "learning_rate": 0.0004940496364895457,
      "loss": 5.2571,
      "mean_token_accuracy": 0.18970807492733002,
      "num_tokens": 18716937.0,
      "step": 9015
    },
    {
      "entropy": 5.612600326538086,
      "epoch": 0.8182148040638607,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004940422421651082,
      "loss": 5.2543,
      "mean_token_accuracy": 0.1858663633465767,
      "num_tokens": 18727478.0,
      "step": 9020
    },
    {
      "entropy": 5.491196346282959,
      "epoch": 0.8186683599419449,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004940348433108211,
      "loss": 5.1684,
      "mean_token_accuracy": 0.18780348002910613,
      "num_tokens": 18737339.0,
      "step": 9025
    },
    {
      "entropy": 5.48309326171875,
      "epoch": 0.819121915820029,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004940274399268377,
      "loss": 5.276,
      "mean_token_accuracy": 0.18714603930711746,
      "num_tokens": 18747810.0,
      "step": 9030
    },
    {
      "entropy": 5.493037223815918,
      "epoch": 0.8195754716981132,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004940200320133109,
      "loss": 5.2579,
      "mean_token_accuracy": 0.18732280135154725,
      "num_tokens": 18757564.0,
      "step": 9035
    },
    {
      "entropy": 5.595683336257935,
      "epoch": 0.8200290275761973,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004940126195703941,
      "loss": 5.2798,
      "mean_token_accuracy": 0.18251087218523027,
      "num_tokens": 18768196.0,
      "step": 9040
    },
    {
      "entropy": 5.569827604293823,
      "epoch": 0.8204825834542816,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004940052025982405,
      "loss": 5.2491,
      "mean_token_accuracy": 0.19636135995388032,
      "num_tokens": 18778350.0,
      "step": 9045
    },
    {
      "entropy": 5.475778579711914,
      "epoch": 0.8209361393323658,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004939977810970035,
      "loss": 5.2048,
      "mean_token_accuracy": 0.18468899726867677,
      "num_tokens": 18789825.0,
      "step": 9050
    },
    {
      "entropy": 5.49319748878479,
      "epoch": 0.8213896952104499,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939903550668366,
      "loss": 5.2059,
      "mean_token_accuracy": 0.18663357943296432,
      "num_tokens": 18799793.0,
      "step": 9055
    },
    {
      "entropy": 5.511374473571777,
      "epoch": 0.8218432510885341,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004939829245078934,
      "loss": 5.2397,
      "mean_token_accuracy": 0.19193035662174224,
      "num_tokens": 18809771.0,
      "step": 9060
    },
    {
      "entropy": 5.4916544437408445,
      "epoch": 0.8222968069666183,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004939754894203275,
      "loss": 5.2606,
      "mean_token_accuracy": 0.18900091648101808,
      "num_tokens": 18820529.0,
      "step": 9065
    },
    {
      "entropy": 5.425688123703003,
      "epoch": 0.8227503628447025,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939680498042926,
      "loss": 5.1706,
      "mean_token_accuracy": 0.19591638892889024,
      "num_tokens": 18830797.0,
      "step": 9070
    },
    {
      "entropy": 5.465984773635864,
      "epoch": 0.8232039187227866,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004939606056599428,
      "loss": 5.1989,
      "mean_token_accuracy": 0.1867137596011162,
      "num_tokens": 18840526.0,
      "step": 9075
    },
    {
      "entropy": 5.483081436157226,
      "epoch": 0.8236574746008708,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004939531569874316,
      "loss": 5.2148,
      "mean_token_accuracy": 0.1837407886981964,
      "num_tokens": 18851201.0,
      "step": 9080
    },
    {
      "entropy": 5.4127593517303465,
      "epoch": 0.824111030478955,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939457037869135,
      "loss": 5.0298,
      "mean_token_accuracy": 0.20760098844766617,
      "num_tokens": 18861676.0,
      "step": 9085
    },
    {
      "entropy": 5.430079507827759,
      "epoch": 0.8245645863570392,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004939382460585424,
      "loss": 5.178,
      "mean_token_accuracy": 0.184494811296463,
      "num_tokens": 18872669.0,
      "step": 9090
    },
    {
      "entropy": 5.434750080108643,
      "epoch": 0.8250181422351234,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004939307838024726,
      "loss": 5.2201,
      "mean_token_accuracy": 0.1835554853081703,
      "num_tokens": 18882130.0,
      "step": 9095
    },
    {
      "entropy": 5.456328392028809,
      "epoch": 0.8254716981132075,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004939233170188585,
      "loss": 5.2109,
      "mean_token_accuracy": 0.1889977216720581,
      "num_tokens": 18892457.0,
      "step": 9100
    },
    {
      "entropy": 5.537774133682251,
      "epoch": 0.8259252539912917,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004939158457078543,
      "loss": 5.2316,
      "mean_token_accuracy": 0.18953631073236465,
      "num_tokens": 18903152.0,
      "step": 9105
    },
    {
      "entropy": 5.513712215423584,
      "epoch": 0.826378809869376,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004939083698696146,
      "loss": 5.1924,
      "mean_token_accuracy": 0.18583756536245347,
      "num_tokens": 18913740.0,
      "step": 9110
    },
    {
      "entropy": 5.487645053863526,
      "epoch": 0.8268323657474601,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000493900889504294,
      "loss": 5.2403,
      "mean_token_accuracy": 0.18907872438430787,
      "num_tokens": 18923501.0,
      "step": 9115
    },
    {
      "entropy": 5.49286789894104,
      "epoch": 0.8272859216255443,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004938934046120473,
      "loss": 5.2371,
      "mean_token_accuracy": 0.18701096922159194,
      "num_tokens": 18933255.0,
      "step": 9120
    },
    {
      "entropy": 5.486018419265747,
      "epoch": 0.8277394775036284,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004938859151930293,
      "loss": 5.1943,
      "mean_token_accuracy": 0.19105144888162612,
      "num_tokens": 18943649.0,
      "step": 9125
    },
    {
      "entropy": 5.471429109573364,
      "epoch": 0.8281930333817126,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938784212473947,
      "loss": 5.0977,
      "mean_token_accuracy": 0.19930363148450853,
      "num_tokens": 18953187.0,
      "step": 9130
    },
    {
      "entropy": 5.544489240646362,
      "epoch": 0.8286465892597968,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004938709227752985,
      "loss": 5.2041,
      "mean_token_accuracy": 0.1965079352259636,
      "num_tokens": 18963015.0,
      "step": 9135
    },
    {
      "entropy": 5.482017135620117,
      "epoch": 0.829100145137881,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000493863419776896,
      "loss": 5.2664,
      "mean_token_accuracy": 0.1819402039051056,
      "num_tokens": 18974184.0,
      "step": 9140
    },
    {
      "entropy": 5.475565242767334,
      "epoch": 0.8295537010159652,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004938559122523421,
      "loss": 5.2073,
      "mean_token_accuracy": 0.1881326138973236,
      "num_tokens": 18985036.0,
      "step": 9145
    },
    {
      "entropy": 5.559447145462036,
      "epoch": 0.8300072568940493,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004938484002017921,
      "loss": 5.3234,
      "mean_token_accuracy": 0.1855930745601654,
      "num_tokens": 18995237.0,
      "step": 9150
    },
    {
      "entropy": 5.519651746749878,
      "epoch": 0.8304608127721336,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004938408836254015,
      "loss": 5.3199,
      "mean_token_accuracy": 0.17952737510204314,
      "num_tokens": 19006272.0,
      "step": 9155
    },
    {
      "entropy": 5.461530971527099,
      "epoch": 0.8309143686502177,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004938333625233258,
      "loss": 5.2331,
      "mean_token_accuracy": 0.18802201896905898,
      "num_tokens": 19016728.0,
      "step": 9160
    },
    {
      "entropy": 5.531713771820068,
      "epoch": 0.8313679245283019,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004938258368957202,
      "loss": 5.2856,
      "mean_token_accuracy": 0.17781373262405395,
      "num_tokens": 19028024.0,
      "step": 9165
    },
    {
      "entropy": 5.528930377960205,
      "epoch": 0.831821480406386,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938183067427406,
      "loss": 5.2252,
      "mean_token_accuracy": 0.1947782501578331,
      "num_tokens": 19038218.0,
      "step": 9170
    },
    {
      "entropy": 5.461911010742187,
      "epoch": 0.8322750362844702,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004938107720645426,
      "loss": 5.254,
      "mean_token_accuracy": 0.18719436675310136,
      "num_tokens": 19048574.0,
      "step": 9175
    },
    {
      "entropy": 5.498357486724854,
      "epoch": 0.8327285921625545,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004938032328612821,
      "loss": 5.2288,
      "mean_token_accuracy": 0.18767004311084748,
      "num_tokens": 19057721.0,
      "step": 9180
    },
    {
      "entropy": 5.501523447036743,
      "epoch": 0.8331821480406386,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000493795689133115,
      "loss": 5.2002,
      "mean_token_accuracy": 0.1808902621269226,
      "num_tokens": 19067582.0,
      "step": 9185
    },
    {
      "entropy": 5.48468132019043,
      "epoch": 0.8336357039187228,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004937881408801974,
      "loss": 5.1627,
      "mean_token_accuracy": 0.19163392037153243,
      "num_tokens": 19077896.0,
      "step": 9190
    },
    {
      "entropy": 5.507964372634888,
      "epoch": 0.8340892597968069,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000493780588102685,
      "loss": 5.2563,
      "mean_token_accuracy": 0.192536298930645,
      "num_tokens": 19088082.0,
      "step": 9195
    },
    {
      "entropy": 5.55622878074646,
      "epoch": 0.8345428156748912,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004937730308007346,
      "loss": 5.2814,
      "mean_token_accuracy": 0.18960726112127305,
      "num_tokens": 19097176.0,
      "step": 9200
    },
    {
      "entropy": 5.497657394409179,
      "epoch": 0.8349963715529753,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004937654689745021,
      "loss": 5.1633,
      "mean_token_accuracy": 0.19420606195926665,
      "num_tokens": 19107835.0,
      "step": 9205
    },
    {
      "entropy": 5.424288654327393,
      "epoch": 0.8354499274310595,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004937579026241439,
      "loss": 5.1904,
      "mean_token_accuracy": 0.19160815477371215,
      "num_tokens": 19118028.0,
      "step": 9210
    },
    {
      "entropy": 5.48676643371582,
      "epoch": 0.8359034833091437,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004937503317498165,
      "loss": 5.2278,
      "mean_token_accuracy": 0.19056888222694396,
      "num_tokens": 19129223.0,
      "step": 9215
    },
    {
      "entropy": 5.492285490036011,
      "epoch": 0.8363570391872278,
      "grad_norm": 0.875,
      "learning_rate": 0.0004937427563516765,
      "loss": 5.3239,
      "mean_token_accuracy": 0.17992374151945115,
      "num_tokens": 19140048.0,
      "step": 9220
    },
    {
      "entropy": 5.564370107650757,
      "epoch": 0.8368105950653121,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004937351764298805,
      "loss": 5.287,
      "mean_token_accuracy": 0.178375805914402,
      "num_tokens": 19150853.0,
      "step": 9225
    },
    {
      "entropy": 5.543662738800049,
      "epoch": 0.8372641509433962,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004937275919845855,
      "loss": 5.2718,
      "mean_token_accuracy": 0.18463384956121445,
      "num_tokens": 19161645.0,
      "step": 9230
    },
    {
      "entropy": 5.514537143707275,
      "epoch": 0.8377177068214804,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004937200030159481,
      "loss": 5.3572,
      "mean_token_accuracy": 0.18064510822296143,
      "num_tokens": 19173276.0,
      "step": 9235
    },
    {
      "entropy": 5.551925563812256,
      "epoch": 0.8381712626995645,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004937124095241252,
      "loss": 5.2393,
      "mean_token_accuracy": 0.19420024901628494,
      "num_tokens": 19182949.0,
      "step": 9240
    },
    {
      "entropy": 5.564868927001953,
      "epoch": 0.8386248185776488,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004937048115092739,
      "loss": 5.2393,
      "mean_token_accuracy": 0.1857432618737221,
      "num_tokens": 19193504.0,
      "step": 9245
    },
    {
      "entropy": 5.488748836517334,
      "epoch": 0.839078374455733,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004936972089715515,
      "loss": 5.3068,
      "mean_token_accuracy": 0.1808287188410759,
      "num_tokens": 19203406.0,
      "step": 9250
    },
    {
      "entropy": 5.481642866134644,
      "epoch": 0.8395319303338171,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000493689601911115,
      "loss": 5.2549,
      "mean_token_accuracy": 0.18709076792001725,
      "num_tokens": 19214325.0,
      "step": 9255
    },
    {
      "entropy": 5.523412084579467,
      "epoch": 0.8399854862119013,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004936819903281219,
      "loss": 5.3033,
      "mean_token_accuracy": 0.17142430394887925,
      "num_tokens": 19226466.0,
      "step": 9260
    },
    {
      "entropy": 5.605291700363159,
      "epoch": 0.8404390420899854,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004936743742227295,
      "loss": 5.3239,
      "mean_token_accuracy": 0.18541250377893448,
      "num_tokens": 19237032.0,
      "step": 9265
    },
    {
      "entropy": 5.553051137924195,
      "epoch": 0.8408925979680697,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004936667535950952,
      "loss": 5.3008,
      "mean_token_accuracy": 0.18086010366678237,
      "num_tokens": 19248009.0,
      "step": 9270
    },
    {
      "entropy": 5.570973443984985,
      "epoch": 0.8413461538461539,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004936591284453768,
      "loss": 5.3585,
      "mean_token_accuracy": 0.18207215815782546,
      "num_tokens": 19259363.0,
      "step": 9275
    },
    {
      "entropy": 5.544232416152954,
      "epoch": 0.841799709724238,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004936514987737319,
      "loss": 5.2609,
      "mean_token_accuracy": 0.18669627606868744,
      "num_tokens": 19269191.0,
      "step": 9280
    },
    {
      "entropy": 5.55557804107666,
      "epoch": 0.8422532656023222,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004936438645803182,
      "loss": 5.2303,
      "mean_token_accuracy": 0.18535475730895995,
      "num_tokens": 19280278.0,
      "step": 9285
    },
    {
      "entropy": 5.522267198562622,
      "epoch": 0.8427068214804064,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004936362258652938,
      "loss": 5.2129,
      "mean_token_accuracy": 0.1926440954208374,
      "num_tokens": 19290130.0,
      "step": 9290
    },
    {
      "entropy": 5.5636334896087645,
      "epoch": 0.8431603773584906,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004936285826288165,
      "loss": 5.3232,
      "mean_token_accuracy": 0.19174562841653825,
      "num_tokens": 19301669.0,
      "step": 9295
    },
    {
      "entropy": 5.453507995605468,
      "epoch": 0.8436139332365747,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004936209348710444,
      "loss": 5.1815,
      "mean_token_accuracy": 0.195254348218441,
      "num_tokens": 19312490.0,
      "step": 9300
    },
    {
      "entropy": 5.478431653976441,
      "epoch": 0.8440674891146589,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004936132825921355,
      "loss": 5.2246,
      "mean_token_accuracy": 0.1872256428003311,
      "num_tokens": 19321743.0,
      "step": 9305
    },
    {
      "entropy": 5.496931743621826,
      "epoch": 0.844521044992743,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004936056257922484,
      "loss": 5.1403,
      "mean_token_accuracy": 0.193279492855072,
      "num_tokens": 19332149.0,
      "step": 9310
    },
    {
      "entropy": 5.508009672164917,
      "epoch": 0.8449746008708273,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004935979644715412,
      "loss": 5.2505,
      "mean_token_accuracy": 0.18201816231012344,
      "num_tokens": 19341655.0,
      "step": 9315
    },
    {
      "entropy": 5.538322639465332,
      "epoch": 0.8454281567489115,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004935902986301724,
      "loss": 5.282,
      "mean_token_accuracy": 0.1822471797466278,
      "num_tokens": 19351404.0,
      "step": 9320
    },
    {
      "entropy": 5.544085264205933,
      "epoch": 0.8458817126269956,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004935826282683005,
      "loss": 5.3194,
      "mean_token_accuracy": 0.18208804428577424,
      "num_tokens": 19361788.0,
      "step": 9325
    },
    {
      "entropy": 5.524391937255859,
      "epoch": 0.8463352685050798,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004935749533860842,
      "loss": 5.2728,
      "mean_token_accuracy": 0.18818075209856033,
      "num_tokens": 19372902.0,
      "step": 9330
    },
    {
      "entropy": 5.45745267868042,
      "epoch": 0.846788824383164,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000493567273983682,
      "loss": 5.1709,
      "mean_token_accuracy": 0.1850125476717949,
      "num_tokens": 19382181.0,
      "step": 9335
    },
    {
      "entropy": 5.514938163757324,
      "epoch": 0.8472423802612482,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000493559590061253,
      "loss": 5.229,
      "mean_token_accuracy": 0.18603635281324388,
      "num_tokens": 19393692.0,
      "step": 9340
    },
    {
      "entropy": 5.539198207855224,
      "epoch": 0.8476959361393324,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004935519016189559,
      "loss": 5.2581,
      "mean_token_accuracy": 0.19449927061796188,
      "num_tokens": 19404587.0,
      "step": 9345
    },
    {
      "entropy": 5.51031002998352,
      "epoch": 0.8481494920174165,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004935442086569498,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1873893231153488,
      "num_tokens": 19415296.0,
      "step": 9350
    },
    {
      "entropy": 5.568622016906739,
      "epoch": 0.8486030478955007,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004935365111753939,
      "loss": 5.2907,
      "mean_token_accuracy": 0.18625998497009277,
      "num_tokens": 19426119.0,
      "step": 9355
    },
    {
      "entropy": 5.55921778678894,
      "epoch": 0.8490566037735849,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004935288091744472,
      "loss": 5.2751,
      "mean_token_accuracy": 0.183457387983799,
      "num_tokens": 19436828.0,
      "step": 9360
    },
    {
      "entropy": 5.553300952911377,
      "epoch": 0.8495101596516691,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493521102654269,
      "loss": 5.2551,
      "mean_token_accuracy": 0.1870663121342659,
      "num_tokens": 19446583.0,
      "step": 9365
    },
    {
      "entropy": 5.452865266799927,
      "epoch": 0.8499637155297532,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004935133916150187,
      "loss": 5.1336,
      "mean_token_accuracy": 0.19529592543840407,
      "num_tokens": 19456757.0,
      "step": 9370
    },
    {
      "entropy": 5.455078649520874,
      "epoch": 0.8504172714078374,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004935056760568558,
      "loss": 5.2172,
      "mean_token_accuracy": 0.18394441306591033,
      "num_tokens": 19467624.0,
      "step": 9375
    },
    {
      "entropy": 5.49309778213501,
      "epoch": 0.8508708272859217,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004934979559799399,
      "loss": 5.1752,
      "mean_token_accuracy": 0.19133689105510712,
      "num_tokens": 19477759.0,
      "step": 9380
    },
    {
      "entropy": 5.515308380126953,
      "epoch": 0.8513243831640058,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004934902313844306,
      "loss": 5.1779,
      "mean_token_accuracy": 0.19061967581510544,
      "num_tokens": 19488877.0,
      "step": 9385
    },
    {
      "entropy": 5.406462621688843,
      "epoch": 0.85177793904209,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004934825022704875,
      "loss": 5.1733,
      "mean_token_accuracy": 0.19388926029205322,
      "num_tokens": 19499170.0,
      "step": 9390
    },
    {
      "entropy": 5.4668279647827145,
      "epoch": 0.8522314949201741,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004934747686382707,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18104320466518403,
      "num_tokens": 19509944.0,
      "step": 9395
    },
    {
      "entropy": 5.513626003265381,
      "epoch": 0.8526850507982583,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004934670304879398,
      "loss": 5.1753,
      "mean_token_accuracy": 0.19153966307640075,
      "num_tokens": 19519014.0,
      "step": 9400
    },
    {
      "entropy": 5.448259162902832,
      "epoch": 0.8531386066763426,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934592878196553,
      "loss": 5.187,
      "mean_token_accuracy": 0.19212067872285843,
      "num_tokens": 19528586.0,
      "step": 9405
    },
    {
      "entropy": 5.49515585899353,
      "epoch": 0.8535921625544267,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000493451540633577,
      "loss": 5.3225,
      "mean_token_accuracy": 0.18525902032852173,
      "num_tokens": 19539912.0,
      "step": 9410
    },
    {
      "entropy": 5.518271636962891,
      "epoch": 0.8540457184325109,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004934437889298651,
      "loss": 5.2805,
      "mean_token_accuracy": 0.1793683409690857,
      "num_tokens": 19549860.0,
      "step": 9415
    },
    {
      "entropy": 5.540545892715454,
      "epoch": 0.854499274310595,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004934360327086799,
      "loss": 5.2327,
      "mean_token_accuracy": 0.18420504480600358,
      "num_tokens": 19559446.0,
      "step": 9420
    },
    {
      "entropy": 5.539361572265625,
      "epoch": 0.8549528301886793,
      "grad_norm": 0.921875,
      "learning_rate": 0.000493428271970182,
      "loss": 5.2417,
      "mean_token_accuracy": 0.1881583198904991,
      "num_tokens": 19570282.0,
      "step": 9425
    },
    {
      "entropy": 5.510746669769287,
      "epoch": 0.8554063860667634,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004934205067145318,
      "loss": 5.1678,
      "mean_token_accuracy": 0.18670809119939805,
      "num_tokens": 19581087.0,
      "step": 9430
    },
    {
      "entropy": 5.492790555953979,
      "epoch": 0.8558599419448476,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004934127369418898,
      "loss": 5.2616,
      "mean_token_accuracy": 0.18760718405246735,
      "num_tokens": 19592100.0,
      "step": 9435
    },
    {
      "entropy": 5.487104320526123,
      "epoch": 0.8563134978229318,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004934049626524168,
      "loss": 5.2409,
      "mean_token_accuracy": 0.18586532920598983,
      "num_tokens": 19602557.0,
      "step": 9440
    },
    {
      "entropy": 5.5392913818359375,
      "epoch": 0.8567670537010159,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933971838462734,
      "loss": 5.2899,
      "mean_token_accuracy": 0.18158393800258638,
      "num_tokens": 19613643.0,
      "step": 9445
    },
    {
      "entropy": 5.4967609405517575,
      "epoch": 0.8572206095791002,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004933894005236207,
      "loss": 5.2001,
      "mean_token_accuracy": 0.1887534514069557,
      "num_tokens": 19624180.0,
      "step": 9450
    },
    {
      "entropy": 5.516611337661743,
      "epoch": 0.8576741654571843,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004933816126846195,
      "loss": 5.2246,
      "mean_token_accuracy": 0.18624820560216904,
      "num_tokens": 19634799.0,
      "step": 9455
    },
    {
      "entropy": 5.485091066360473,
      "epoch": 0.8581277213352685,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004933738203294309,
      "loss": 5.1903,
      "mean_token_accuracy": 0.19355441778898239,
      "num_tokens": 19644618.0,
      "step": 9460
    },
    {
      "entropy": 5.47304892539978,
      "epoch": 0.8585812772133526,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000493366023458216,
      "loss": 5.1999,
      "mean_token_accuracy": 0.18136429488658906,
      "num_tokens": 19655550.0,
      "step": 9465
    },
    {
      "entropy": 5.517861652374267,
      "epoch": 0.8590348330914369,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004933582220711362,
      "loss": 5.2673,
      "mean_token_accuracy": 0.1912809804081917,
      "num_tokens": 19665734.0,
      "step": 9470
    },
    {
      "entropy": 5.552363920211792,
      "epoch": 0.8594883889695211,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004933504161683526,
      "loss": 5.2249,
      "mean_token_accuracy": 0.18164515048265456,
      "num_tokens": 19677130.0,
      "step": 9475
    },
    {
      "entropy": 5.463605499267578,
      "epoch": 0.8599419448476052,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933426057500269,
      "loss": 5.1415,
      "mean_token_accuracy": 0.19647871553897858,
      "num_tokens": 19687374.0,
      "step": 9480
    },
    {
      "entropy": 5.548071193695068,
      "epoch": 0.8603955007256894,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004933347908163203,
      "loss": 5.2359,
      "mean_token_accuracy": 0.1858404666185379,
      "num_tokens": 19698182.0,
      "step": 9485
    },
    {
      "entropy": 5.489673233032226,
      "epoch": 0.8608490566037735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933269713673947,
      "loss": 5.209,
      "mean_token_accuracy": 0.18983528316020964,
      "num_tokens": 19707664.0,
      "step": 9490
    },
    {
      "entropy": 5.418010711669922,
      "epoch": 0.8613026124818578,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004933191474034116,
      "loss": 5.2464,
      "mean_token_accuracy": 0.18493668735027313,
      "num_tokens": 19717647.0,
      "step": 9495
    },
    {
      "entropy": 5.481851577758789,
      "epoch": 0.861756168359942,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000493311318924533,
      "loss": 5.2824,
      "mean_token_accuracy": 0.18406570553779603,
      "num_tokens": 19727444.0,
      "step": 9500
    },
    {
      "entropy": 5.547173070907593,
      "epoch": 0.8622097242380261,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004933034859309208,
      "loss": 5.3011,
      "mean_token_accuracy": 0.1775568827986717,
      "num_tokens": 19738719.0,
      "step": 9505
    },
    {
      "entropy": 5.583591938018799,
      "epoch": 0.8626632801161103,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004932956484227367,
      "loss": 5.2774,
      "mean_token_accuracy": 0.19096940159797668,
      "num_tokens": 19749786.0,
      "step": 9510
    },
    {
      "entropy": 5.515065097808838,
      "epoch": 0.8631168359941945,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000493287806400143,
      "loss": 5.2094,
      "mean_token_accuracy": 0.19534773230552674,
      "num_tokens": 19759623.0,
      "step": 9515
    },
    {
      "entropy": 5.536817026138306,
      "epoch": 0.8635703918722787,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004932799598633018,
      "loss": 5.2199,
      "mean_token_accuracy": 0.18721261024475097,
      "num_tokens": 19770117.0,
      "step": 9520
    },
    {
      "entropy": 5.51656551361084,
      "epoch": 0.8640239477503628,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932721088123753,
      "loss": 5.3074,
      "mean_token_accuracy": 0.18952126950025558,
      "num_tokens": 19780748.0,
      "step": 9525
    },
    {
      "entropy": 5.547321510314942,
      "epoch": 0.864477503628447,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004932642532475262,
      "loss": 5.2737,
      "mean_token_accuracy": 0.18902790993452073,
      "num_tokens": 19791816.0,
      "step": 9530
    },
    {
      "entropy": 5.57507176399231,
      "epoch": 0.8649310595065312,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004932563931689168,
      "loss": 5.3212,
      "mean_token_accuracy": 0.18597906678915024,
      "num_tokens": 19801096.0,
      "step": 9535
    },
    {
      "entropy": 5.51464467048645,
      "epoch": 0.8653846153846154,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004932485285767093,
      "loss": 5.1896,
      "mean_token_accuracy": 0.19145326763391496,
      "num_tokens": 19811258.0,
      "step": 9540
    },
    {
      "entropy": 5.531577491760254,
      "epoch": 0.8658381712626996,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004932406594710667,
      "loss": 5.2594,
      "mean_token_accuracy": 0.19362047761678697,
      "num_tokens": 19821765.0,
      "step": 9545
    },
    {
      "entropy": 5.4913736343383786,
      "epoch": 0.8662917271407837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004932327858521517,
      "loss": 5.1323,
      "mean_token_accuracy": 0.19672255367040634,
      "num_tokens": 19831537.0,
      "step": 9550
    },
    {
      "entropy": 5.447400045394898,
      "epoch": 0.8667452830188679,
      "grad_norm": 0.984375,
      "learning_rate": 0.000493224907720127,
      "loss": 5.1531,
      "mean_token_accuracy": 0.19029980450868605,
      "num_tokens": 19841830.0,
      "step": 9555
    },
    {
      "entropy": 5.512387990951538,
      "epoch": 0.8671988388969522,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004932170250751556,
      "loss": 5.256,
      "mean_token_accuracy": 0.18012520521879197,
      "num_tokens": 19851896.0,
      "step": 9560
    },
    {
      "entropy": 5.537576293945312,
      "epoch": 0.8676523947750363,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004932091379174006,
      "loss": 5.289,
      "mean_token_accuracy": 0.1798422083258629,
      "num_tokens": 19861735.0,
      "step": 9565
    },
    {
      "entropy": 5.497203254699707,
      "epoch": 0.8681059506531205,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000493201246247025,
      "loss": 5.1747,
      "mean_token_accuracy": 0.19459885060787202,
      "num_tokens": 19872333.0,
      "step": 9570
    },
    {
      "entropy": 5.521367979049683,
      "epoch": 0.8685595065312046,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000493193350064192,
      "loss": 5.1918,
      "mean_token_accuracy": 0.18699753284454346,
      "num_tokens": 19882896.0,
      "step": 9575
    },
    {
      "entropy": 5.4799529075622555,
      "epoch": 0.8690130624092888,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004931854493690649,
      "loss": 5.1517,
      "mean_token_accuracy": 0.189018677175045,
      "num_tokens": 19893767.0,
      "step": 9580
    },
    {
      "entropy": 5.41310601234436,
      "epoch": 0.869466618287373,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004931775441618071,
      "loss": 5.1264,
      "mean_token_accuracy": 0.19416602104902267,
      "num_tokens": 19904242.0,
      "step": 9585
    },
    {
      "entropy": 5.478859567642212,
      "epoch": 0.8699201741654572,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004931696344425821,
      "loss": 5.2633,
      "mean_token_accuracy": 0.188524429500103,
      "num_tokens": 19913898.0,
      "step": 9590
    },
    {
      "entropy": 5.614615726470947,
      "epoch": 0.8703737300435413,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004931617202115535,
      "loss": 5.2541,
      "mean_token_accuracy": 0.19075826853513717,
      "num_tokens": 19923756.0,
      "step": 9595
    },
    {
      "entropy": 5.524111700057984,
      "epoch": 0.8708272859216255,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004931538014688849,
      "loss": 5.1939,
      "mean_token_accuracy": 0.1916771948337555,
      "num_tokens": 19932720.0,
      "step": 9600
    },
    {
      "entropy": 5.558842515945434,
      "epoch": 0.8712808417997098,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004931458782147401,
      "loss": 5.4063,
      "mean_token_accuracy": 0.17672424167394638,
      "num_tokens": 19942962.0,
      "step": 9605
    },
    {
      "entropy": 5.560279273986817,
      "epoch": 0.8717343976777939,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000493137950449283,
      "loss": 5.276,
      "mean_token_accuracy": 0.19128557592630385,
      "num_tokens": 19953735.0,
      "step": 9610
    },
    {
      "entropy": 5.505196857452392,
      "epoch": 0.8721879535558781,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004931300181726775,
      "loss": 5.2476,
      "mean_token_accuracy": 0.18588921427726746,
      "num_tokens": 19964416.0,
      "step": 9615
    },
    {
      "entropy": 5.517534017562866,
      "epoch": 0.8726415094339622,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004931220813850875,
      "loss": 5.2602,
      "mean_token_accuracy": 0.18303596675395967,
      "num_tokens": 19974531.0,
      "step": 9620
    },
    {
      "entropy": 5.540431547164917,
      "epoch": 0.8730950653120464,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004931141400866773,
      "loss": 5.2387,
      "mean_token_accuracy": 0.19180816859006883,
      "num_tokens": 19984964.0,
      "step": 9625
    },
    {
      "entropy": 5.482637929916382,
      "epoch": 0.8735486211901307,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004931061942776112,
      "loss": 5.1951,
      "mean_token_accuracy": 0.19486953020095826,
      "num_tokens": 19995568.0,
      "step": 9630
    },
    {
      "entropy": 5.608856678009033,
      "epoch": 0.8740021770682148,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004930982439580534,
      "loss": 5.3692,
      "mean_token_accuracy": 0.17587941586971284,
      "num_tokens": 20005856.0,
      "step": 9635
    },
    {
      "entropy": 5.535829830169678,
      "epoch": 0.874455732946299,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004930902891281683,
      "loss": 5.2749,
      "mean_token_accuracy": 0.18487472534179689,
      "num_tokens": 20015683.0,
      "step": 9640
    },
    {
      "entropy": 5.421100807189942,
      "epoch": 0.8749092888243831,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004930823297881205,
      "loss": 5.0891,
      "mean_token_accuracy": 0.1941447973251343,
      "num_tokens": 20025998.0,
      "step": 9645
    },
    {
      "entropy": 5.542952680587769,
      "epoch": 0.8753628447024674,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004930743659380745,
      "loss": 5.2547,
      "mean_token_accuracy": 0.18763321340084077,
      "num_tokens": 20036486.0,
      "step": 9650
    },
    {
      "entropy": 5.459162616729737,
      "epoch": 0.8758164005805515,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000493066397578195,
      "loss": 5.1449,
      "mean_token_accuracy": 0.19946982264518737,
      "num_tokens": 20046784.0,
      "step": 9655
    },
    {
      "entropy": 5.504218053817749,
      "epoch": 0.8762699564586357,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004930584247086468,
      "loss": 5.2238,
      "mean_token_accuracy": 0.18850288093090056,
      "num_tokens": 20056999.0,
      "step": 9660
    },
    {
      "entropy": 5.464111852645874,
      "epoch": 0.8767235123367199,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000493050447329595,
      "loss": 5.1184,
      "mean_token_accuracy": 0.19502724707126617,
      "num_tokens": 20067612.0,
      "step": 9665
    },
    {
      "entropy": 5.520261335372925,
      "epoch": 0.877177068214804,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004930424654412043,
      "loss": 5.2666,
      "mean_token_accuracy": 0.1904868870973587,
      "num_tokens": 20077330.0,
      "step": 9670
    },
    {
      "entropy": 5.511299180984497,
      "epoch": 0.8776306240928883,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004930344790436398,
      "loss": 5.26,
      "mean_token_accuracy": 0.18457795083522796,
      "num_tokens": 20086801.0,
      "step": 9675
    },
    {
      "entropy": 5.532819318771362,
      "epoch": 0.8780841799709724,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004930264881370666,
      "loss": 5.2115,
      "mean_token_accuracy": 0.18478081822395326,
      "num_tokens": 20097047.0,
      "step": 9680
    },
    {
      "entropy": 5.5098706722259525,
      "epoch": 0.8785377358490566,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930184927216504,
      "loss": 5.2484,
      "mean_token_accuracy": 0.19049520045518875,
      "num_tokens": 20107185.0,
      "step": 9685
    },
    {
      "entropy": 5.523552942276001,
      "epoch": 0.8789912917271407,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004930104927975559,
      "loss": 5.186,
      "mean_token_accuracy": 0.1903187081217766,
      "num_tokens": 20118493.0,
      "step": 9690
    },
    {
      "entropy": 5.510832071304321,
      "epoch": 0.879444847605225,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004930024883649489,
      "loss": 5.2423,
      "mean_token_accuracy": 0.1870877966284752,
      "num_tokens": 20129294.0,
      "step": 9695
    },
    {
      "entropy": 5.439377927780152,
      "epoch": 0.8798984034833092,
      "grad_norm": 1.0,
      "learning_rate": 0.000492994479423995,
      "loss": 5.2038,
      "mean_token_accuracy": 0.19338049739599228,
      "num_tokens": 20139194.0,
      "step": 9700
    },
    {
      "entropy": 5.4923159122467045,
      "epoch": 0.8803519593613933,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004929864659748597,
      "loss": 5.2241,
      "mean_token_accuracy": 0.18959469497203826,
      "num_tokens": 20149681.0,
      "step": 9705
    },
    {
      "entropy": 5.547377967834473,
      "epoch": 0.8808055152394775,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004929784480177087,
      "loss": 5.3143,
      "mean_token_accuracy": 0.18003763556480407,
      "num_tokens": 20160557.0,
      "step": 9710
    },
    {
      "entropy": 5.4973267078399655,
      "epoch": 0.8812590711175616,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004929704255527078,
      "loss": 5.1858,
      "mean_token_accuracy": 0.18966007977724075,
      "num_tokens": 20171047.0,
      "step": 9715
    },
    {
      "entropy": 5.519015216827393,
      "epoch": 0.8817126269956459,
      "grad_norm": 0.90625,
      "learning_rate": 0.000492962398580023,
      "loss": 5.3549,
      "mean_token_accuracy": 0.18190275877714157,
      "num_tokens": 20182148.0,
      "step": 9720
    },
    {
      "entropy": 5.434145259857178,
      "epoch": 0.88216618287373,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004929543670998203,
      "loss": 5.107,
      "mean_token_accuracy": 0.19520194828510284,
      "num_tokens": 20192131.0,
      "step": 9725
    },
    {
      "entropy": 5.524809503555298,
      "epoch": 0.8826197387518142,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004929463311122657,
      "loss": 5.228,
      "mean_token_accuracy": 0.1856295123696327,
      "num_tokens": 20202805.0,
      "step": 9730
    },
    {
      "entropy": 5.528243207931519,
      "epoch": 0.8830732946298984,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004929382906175253,
      "loss": 5.1684,
      "mean_token_accuracy": 0.19349892288446427,
      "num_tokens": 20212702.0,
      "step": 9735
    },
    {
      "entropy": 5.468289422988891,
      "epoch": 0.8835268505079826,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004929302456157657,
      "loss": 5.1914,
      "mean_token_accuracy": 0.1951130896806717,
      "num_tokens": 20222661.0,
      "step": 9740
    },
    {
      "entropy": 5.48810658454895,
      "epoch": 0.8839804063860668,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929221961071531,
      "loss": 5.239,
      "mean_token_accuracy": 0.1886950433254242,
      "num_tokens": 20233012.0,
      "step": 9745
    },
    {
      "entropy": 5.531930875778198,
      "epoch": 0.8844339622641509,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004929141420918538,
      "loss": 5.2469,
      "mean_token_accuracy": 0.1919715628027916,
      "num_tokens": 20242918.0,
      "step": 9750
    },
    {
      "entropy": 5.4830118179321286,
      "epoch": 0.8848875181422351,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004929060835700347,
      "loss": 5.1766,
      "mean_token_accuracy": 0.19638995528221131,
      "num_tokens": 20253254.0,
      "step": 9755
    },
    {
      "entropy": 5.4783000469207765,
      "epoch": 0.8853410740203193,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492898020541862,
      "loss": 5.2378,
      "mean_token_accuracy": 0.18468834161758424,
      "num_tokens": 20263355.0,
      "step": 9760
    },
    {
      "entropy": 5.460058784484863,
      "epoch": 0.8857946298984035,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004928899530075029,
      "loss": 5.1601,
      "mean_token_accuracy": 0.19061171710491182,
      "num_tokens": 20274716.0,
      "step": 9765
    },
    {
      "entropy": 5.433362579345703,
      "epoch": 0.8862481857764877,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004928818809671241,
      "loss": 5.1507,
      "mean_token_accuracy": 0.19014403820037842,
      "num_tokens": 20285631.0,
      "step": 9770
    },
    {
      "entropy": 5.545337724685669,
      "epoch": 0.8867017416545718,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928738044208923,
      "loss": 5.2384,
      "mean_token_accuracy": 0.19255442023277283,
      "num_tokens": 20295717.0,
      "step": 9775
    },
    {
      "entropy": 5.456331539154053,
      "epoch": 0.887155297532656,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004928657233689748,
      "loss": 5.2041,
      "mean_token_accuracy": 0.1925482451915741,
      "num_tokens": 20306637.0,
      "step": 9780
    },
    {
      "entropy": 5.484044456481934,
      "epoch": 0.8876088534107403,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004928576378115387,
      "loss": 5.1831,
      "mean_token_accuracy": 0.1889451801776886,
      "num_tokens": 20316629.0,
      "step": 9785
    },
    {
      "entropy": 5.5872455596923825,
      "epoch": 0.8880624092888244,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004928495477487509,
      "loss": 5.355,
      "mean_token_accuracy": 0.17307652533054352,
      "num_tokens": 20327935.0,
      "step": 9790
    },
    {
      "entropy": 5.530547428131103,
      "epoch": 0.8885159651669086,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004928414531807792,
      "loss": 5.1728,
      "mean_token_accuracy": 0.1830437198281288,
      "num_tokens": 20339142.0,
      "step": 9795
    },
    {
      "entropy": 5.4817322254180905,
      "epoch": 0.8889695210449927,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004928333541077906,
      "loss": 5.172,
      "mean_token_accuracy": 0.2016322910785675,
      "num_tokens": 20348474.0,
      "step": 9800
    },
    {
      "entropy": 5.479598665237427,
      "epoch": 0.8894230769230769,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004928252505299527,
      "loss": 5.2418,
      "mean_token_accuracy": 0.19005548357963561,
      "num_tokens": 20359440.0,
      "step": 9805
    },
    {
      "entropy": 5.569894742965698,
      "epoch": 0.8898766328011611,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004928171424474332,
      "loss": 5.2767,
      "mean_token_accuracy": 0.1847678691148758,
      "num_tokens": 20370355.0,
      "step": 9810
    },
    {
      "entropy": 5.4881508350372314,
      "epoch": 0.8903301886792453,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004928090298603996,
      "loss": 5.1705,
      "mean_token_accuracy": 0.1922898545861244,
      "num_tokens": 20380214.0,
      "step": 9815
    },
    {
      "entropy": 5.526920175552368,
      "epoch": 0.8907837445573294,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004928009127690199,
      "loss": 5.1952,
      "mean_token_accuracy": 0.18694787621498107,
      "num_tokens": 20389190.0,
      "step": 9820
    },
    {
      "entropy": 5.458229827880859,
      "epoch": 0.8912373004354136,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927927911734617,
      "loss": 5.232,
      "mean_token_accuracy": 0.18914854824542998,
      "num_tokens": 20399531.0,
      "step": 9825
    },
    {
      "entropy": 5.465860223770141,
      "epoch": 0.8916908563134979,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004927846650738931,
      "loss": 5.2215,
      "mean_token_accuracy": 0.18792601376771928,
      "num_tokens": 20410201.0,
      "step": 9830
    },
    {
      "entropy": 5.48523736000061,
      "epoch": 0.892144412191582,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004927765344704821,
      "loss": 5.1693,
      "mean_token_accuracy": 0.19596148431301116,
      "num_tokens": 20421579.0,
      "step": 9835
    },
    {
      "entropy": 5.474275684356689,
      "epoch": 0.8925979680696662,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004927683993633969,
      "loss": 5.2427,
      "mean_token_accuracy": 0.19045363068580629,
      "num_tokens": 20432048.0,
      "step": 9840
    },
    {
      "entropy": 5.553377866744995,
      "epoch": 0.8930515239477503,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004927602597528057,
      "loss": 5.2663,
      "mean_token_accuracy": 0.18120875656604768,
      "num_tokens": 20441752.0,
      "step": 9845
    },
    {
      "entropy": 5.456312799453736,
      "epoch": 0.8935050798258345,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004927521156388769,
      "loss": 5.13,
      "mean_token_accuracy": 0.1914264976978302,
      "num_tokens": 20451693.0,
      "step": 9850
    },
    {
      "entropy": 5.502097320556641,
      "epoch": 0.8939586357039188,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004927439670217788,
      "loss": 5.2738,
      "mean_token_accuracy": 0.18670168668031692,
      "num_tokens": 20462312.0,
      "step": 9855
    },
    {
      "entropy": 5.483501243591308,
      "epoch": 0.8944121915820029,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049273581390168,
      "loss": 5.1804,
      "mean_token_accuracy": 0.18460638374090194,
      "num_tokens": 20472546.0,
      "step": 9860
    },
    {
      "entropy": 5.506111812591553,
      "epoch": 0.8948657474600871,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000492727656278749,
      "loss": 5.2123,
      "mean_token_accuracy": 0.1930461436510086,
      "num_tokens": 20483266.0,
      "step": 9865
    },
    {
      "entropy": 5.464397954940796,
      "epoch": 0.8953193033381712,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004927194941531547,
      "loss": 5.1512,
      "mean_token_accuracy": 0.18788621425628663,
      "num_tokens": 20492490.0,
      "step": 9870
    },
    {
      "entropy": 5.485383033752441,
      "epoch": 0.8957728592162555,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004927113275250657,
      "loss": 5.1922,
      "mean_token_accuracy": 0.19176370054483413,
      "num_tokens": 20503077.0,
      "step": 9875
    },
    {
      "entropy": 5.464278745651245,
      "epoch": 0.8962264150943396,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004927031563946511,
      "loss": 5.2078,
      "mean_token_accuracy": 0.18573808223009108,
      "num_tokens": 20513771.0,
      "step": 9880
    },
    {
      "entropy": 5.532713842391968,
      "epoch": 0.8966799709724238,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004926949807620796,
      "loss": 5.3044,
      "mean_token_accuracy": 0.1865783840417862,
      "num_tokens": 20523616.0,
      "step": 9885
    },
    {
      "entropy": 5.4971997261047365,
      "epoch": 0.897133526850508,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004926868006275204,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19763742238283158,
      "num_tokens": 20533613.0,
      "step": 9890
    },
    {
      "entropy": 5.4906604290008545,
      "epoch": 0.8975870827285921,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004926786159911427,
      "loss": 5.2075,
      "mean_token_accuracy": 0.1880497470498085,
      "num_tokens": 20543812.0,
      "step": 9895
    },
    {
      "entropy": 5.460688543319702,
      "epoch": 0.8980406386066764,
      "grad_norm": 1.0,
      "learning_rate": 0.000492670426853116,
      "loss": 5.1363,
      "mean_token_accuracy": 0.192536161839962,
      "num_tokens": 20553785.0,
      "step": 9900
    },
    {
      "entropy": 5.562257194519043,
      "epoch": 0.8984941944847605,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004926622332136092,
      "loss": 5.2559,
      "mean_token_accuracy": 0.19221415668725966,
      "num_tokens": 20563326.0,
      "step": 9905
    },
    {
      "entropy": 5.436643981933594,
      "epoch": 0.8989477503628447,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004926540350727918,
      "loss": 5.1005,
      "mean_token_accuracy": 0.19912851303815843,
      "num_tokens": 20573643.0,
      "step": 9910
    },
    {
      "entropy": 5.426002407073975,
      "epoch": 0.8994013062409288,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004926458324308337,
      "loss": 5.1318,
      "mean_token_accuracy": 0.19306659698486328,
      "num_tokens": 20583916.0,
      "step": 9915
    },
    {
      "entropy": 5.516725540161133,
      "epoch": 0.8998548621190131,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004926376252879043,
      "loss": 5.2894,
      "mean_token_accuracy": 0.19082354307174682,
      "num_tokens": 20593984.0,
      "step": 9920
    },
    {
      "entropy": 5.518711185455322,
      "epoch": 0.9003084179970973,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004926294136441733,
      "loss": 5.1265,
      "mean_token_accuracy": 0.1930823028087616,
      "num_tokens": 20603927.0,
      "step": 9925
    },
    {
      "entropy": 5.496563529968261,
      "epoch": 0.9007619738751814,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004926211974998106,
      "loss": 5.2612,
      "mean_token_accuracy": 0.1923177346587181,
      "num_tokens": 20614147.0,
      "step": 9930
    },
    {
      "entropy": 5.523381233215332,
      "epoch": 0.9012155297532656,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000492612976854986,
      "loss": 5.2238,
      "mean_token_accuracy": 0.1915097326040268,
      "num_tokens": 20623631.0,
      "step": 9935
    },
    {
      "entropy": 5.465784215927124,
      "epoch": 0.9016690856313497,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004926047517098697,
      "loss": 5.1087,
      "mean_token_accuracy": 0.19526478499174119,
      "num_tokens": 20632566.0,
      "step": 9940
    },
    {
      "entropy": 5.512724876403809,
      "epoch": 0.902122641509434,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004925965220646315,
      "loss": 5.2151,
      "mean_token_accuracy": 0.18857671916484833,
      "num_tokens": 20643246.0,
      "step": 9945
    },
    {
      "entropy": 5.507947826385498,
      "epoch": 0.9025761973875182,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004925882879194419,
      "loss": 5.296,
      "mean_token_accuracy": 0.18535405844449998,
      "num_tokens": 20653372.0,
      "step": 9950
    },
    {
      "entropy": 5.5078880310058596,
      "epoch": 0.9030297532656023,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004925800492744712,
      "loss": 5.2362,
      "mean_token_accuracy": 0.18510301411151886,
      "num_tokens": 20662687.0,
      "step": 9955
    },
    {
      "entropy": 5.463442754745484,
      "epoch": 0.9034833091436865,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004925718061298895,
      "loss": 5.1509,
      "mean_token_accuracy": 0.196966715157032,
      "num_tokens": 20673168.0,
      "step": 9960
    },
    {
      "entropy": 5.560179567337036,
      "epoch": 0.9039368650217707,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004925635584858675,
      "loss": 5.267,
      "mean_token_accuracy": 0.1841796949505806,
      "num_tokens": 20683100.0,
      "step": 9965
    },
    {
      "entropy": 5.587424993515015,
      "epoch": 0.9043904208998549,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004925553063425755,
      "loss": 5.2772,
      "mean_token_accuracy": 0.186713108420372,
      "num_tokens": 20694097.0,
      "step": 9970
    },
    {
      "entropy": 5.5363075733184814,
      "epoch": 0.904843976777939,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004925470497001844,
      "loss": 5.2393,
      "mean_token_accuracy": 0.19200791716575621,
      "num_tokens": 20704640.0,
      "step": 9975
    },
    {
      "entropy": 5.471827363967895,
      "epoch": 0.9052975326560232,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004925387885588648,
      "loss": 5.1802,
      "mean_token_accuracy": 0.18870692253112792,
      "num_tokens": 20714419.0,
      "step": 9980
    },
    {
      "entropy": 5.544799852371216,
      "epoch": 0.9057510885341074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004925305229187877,
      "loss": 5.2693,
      "mean_token_accuracy": 0.18884519636631011,
      "num_tokens": 20724705.0,
      "step": 9985
    },
    {
      "entropy": 5.597160959243775,
      "epoch": 0.9062046444121916,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492522252780124,
      "loss": 5.3261,
      "mean_token_accuracy": 0.18726316094398499,
      "num_tokens": 20734753.0,
      "step": 9990
    },
    {
      "entropy": 5.465805244445801,
      "epoch": 0.9066582002902758,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004925139781430446,
      "loss": 5.1156,
      "mean_token_accuracy": 0.1963346555829048,
      "num_tokens": 20745389.0,
      "step": 9995
    },
    {
      "entropy": 5.480003547668457,
      "epoch": 0.9071117561683599,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004925056990077207,
      "loss": 5.1943,
      "mean_token_accuracy": 0.1931140884757042,
      "num_tokens": 20755702.0,
      "step": 10000
    },
    {
      "entropy": 5.498392629623413,
      "epoch": 0.9075653120464441,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004924974153743235,
      "loss": 5.0984,
      "mean_token_accuracy": 0.19833627194166184,
      "num_tokens": 20766046.0,
      "step": 10005
    },
    {
      "entropy": 5.50446720123291,
      "epoch": 0.9080188679245284,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004924891272430244,
      "loss": 5.2141,
      "mean_token_accuracy": 0.19061006605625153,
      "num_tokens": 20776654.0,
      "step": 10010
    },
    {
      "entropy": 5.5594000816345215,
      "epoch": 0.9084724238026125,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004924808346139948,
      "loss": 5.2713,
      "mean_token_accuracy": 0.19167041033506393,
      "num_tokens": 20787066.0,
      "step": 10015
    },
    {
      "entropy": 5.47709379196167,
      "epoch": 0.9089259796806967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924725374874059,
      "loss": 5.1736,
      "mean_token_accuracy": 0.1896710604429245,
      "num_tokens": 20796822.0,
      "step": 10020
    },
    {
      "entropy": 5.495741128921509,
      "epoch": 0.9093795355587808,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924642358634297,
      "loss": 5.2297,
      "mean_token_accuracy": 0.18855800926685334,
      "num_tokens": 20807602.0,
      "step": 10025
    },
    {
      "entropy": 5.5386439800262455,
      "epoch": 0.909833091436865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004924559297422376,
      "loss": 5.1864,
      "mean_token_accuracy": 0.19395318776369094,
      "num_tokens": 20819046.0,
      "step": 10030
    },
    {
      "entropy": 5.51975998878479,
      "epoch": 0.9102866473149492,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004924476191240014,
      "loss": 5.2153,
      "mean_token_accuracy": 0.19031708985567092,
      "num_tokens": 20828989.0,
      "step": 10035
    },
    {
      "entropy": 5.446819305419922,
      "epoch": 0.9107402031930334,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000492439304008893,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18754609823226928,
      "num_tokens": 20839366.0,
      "step": 10040
    },
    {
      "entropy": 5.519721603393554,
      "epoch": 0.9111937590711175,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004924309843970847,
      "loss": 5.2112,
      "mean_token_accuracy": 0.19083768427371978,
      "num_tokens": 20849513.0,
      "step": 10045
    },
    {
      "entropy": 5.559159517288208,
      "epoch": 0.9116473149492017,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004924226602887479,
      "loss": 5.2674,
      "mean_token_accuracy": 0.18467978686094283,
      "num_tokens": 20861406.0,
      "step": 10050
    },
    {
      "entropy": 5.47953314781189,
      "epoch": 0.912100870827286,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004924143316840552,
      "loss": 5.1645,
      "mean_token_accuracy": 0.19939935952425003,
      "num_tokens": 20871190.0,
      "step": 10055
    },
    {
      "entropy": 5.531355714797973,
      "epoch": 0.9125544267053701,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004924059985831787,
      "loss": 5.3149,
      "mean_token_accuracy": 0.1820805475115776,
      "num_tokens": 20881589.0,
      "step": 10060
    },
    {
      "entropy": 5.5043120861053465,
      "epoch": 0.9130079825834543,
      "grad_norm": 1.0,
      "learning_rate": 0.0004923976609862908,
      "loss": 5.249,
      "mean_token_accuracy": 0.17866823971271514,
      "num_tokens": 20892072.0,
      "step": 10065
    },
    {
      "entropy": 5.541967296600342,
      "epoch": 0.9134615384615384,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004923893188935638,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18548781424760818,
      "num_tokens": 20902616.0,
      "step": 10070
    },
    {
      "entropy": 5.578791618347168,
      "epoch": 0.9139150943396226,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004923809723051702,
      "loss": 5.231,
      "mean_token_accuracy": 0.1868339642882347,
      "num_tokens": 20913832.0,
      "step": 10075
    },
    {
      "entropy": 5.451777458190918,
      "epoch": 0.9143686502177069,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004923726212212829,
      "loss": 5.1258,
      "mean_token_accuracy": 0.1962735816836357,
      "num_tokens": 20923322.0,
      "step": 10080
    },
    {
      "entropy": 5.44694151878357,
      "epoch": 0.914822206095791,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923642656420742,
      "loss": 5.2078,
      "mean_token_accuracy": 0.19073516130447388,
      "num_tokens": 20933334.0,
      "step": 10085
    },
    {
      "entropy": 5.462447452545166,
      "epoch": 0.9152757619738752,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004923559055677173,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1864377275109291,
      "num_tokens": 20944171.0,
      "step": 10090
    },
    {
      "entropy": 5.524150848388672,
      "epoch": 0.9157293178519593,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004923475409983848,
      "loss": 5.215,
      "mean_token_accuracy": 0.18702605217695237,
      "num_tokens": 20954200.0,
      "step": 10095
    },
    {
      "entropy": 5.490021753311157,
      "epoch": 0.9161828737300436,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004923391719342497,
      "loss": 5.2339,
      "mean_token_accuracy": 0.1857544004917145,
      "num_tokens": 20964935.0,
      "step": 10100
    },
    {
      "entropy": 5.470995235443115,
      "epoch": 0.9166364296081277,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004923307983754852,
      "loss": 5.2232,
      "mean_token_accuracy": 0.18978999555110931,
      "num_tokens": 20975499.0,
      "step": 10105
    },
    {
      "entropy": 5.406945657730103,
      "epoch": 0.9170899854862119,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004923224203222645,
      "loss": 5.1461,
      "mean_token_accuracy": 0.19023018777370454,
      "num_tokens": 20986199.0,
      "step": 10110
    },
    {
      "entropy": 5.485232019424439,
      "epoch": 0.9175435413642961,
      "grad_norm": 1.0,
      "learning_rate": 0.0004923140377747606,
      "loss": 5.2263,
      "mean_token_accuracy": 0.18715826719999312,
      "num_tokens": 20996313.0,
      "step": 10115
    },
    {
      "entropy": 5.515059518814087,
      "epoch": 0.9179970972423802,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004923056507331471,
      "loss": 5.2129,
      "mean_token_accuracy": 0.18567863702774048,
      "num_tokens": 21006133.0,
      "step": 10120
    },
    {
      "entropy": 5.515254068374634,
      "epoch": 0.9184506531204645,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004922972591975974,
      "loss": 5.1735,
      "mean_token_accuracy": 0.18915437906980515,
      "num_tokens": 21016448.0,
      "step": 10125
    },
    {
      "entropy": 5.455716228485107,
      "epoch": 0.9189042089985486,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004922888631682849,
      "loss": 5.1148,
      "mean_token_accuracy": 0.18959181159734725,
      "num_tokens": 21026458.0,
      "step": 10130
    },
    {
      "entropy": 5.464995288848877,
      "epoch": 0.9193577648766328,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004922804626453834,
      "loss": 5.1567,
      "mean_token_accuracy": 0.19512831717729567,
      "num_tokens": 21037111.0,
      "step": 10135
    },
    {
      "entropy": 5.4613899230957035,
      "epoch": 0.9198113207547169,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004922720576290665,
      "loss": 5.2107,
      "mean_token_accuracy": 0.19339829832315444,
      "num_tokens": 21048041.0,
      "step": 10140
    },
    {
      "entropy": 5.438198232650757,
      "epoch": 0.9202648766328012,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004922636481195082,
      "loss": 5.0888,
      "mean_token_accuracy": 0.20385636240243912,
      "num_tokens": 21058099.0,
      "step": 10145
    },
    {
      "entropy": 5.448825025558472,
      "epoch": 0.9207184325108854,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004922552341168823,
      "loss": 5.2224,
      "mean_token_accuracy": 0.18477773368358613,
      "num_tokens": 21068137.0,
      "step": 10150
    },
    {
      "entropy": 5.417740678787231,
      "epoch": 0.9211719883889695,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004922468156213627,
      "loss": 5.115,
      "mean_token_accuracy": 0.19422985017299652,
      "num_tokens": 21077556.0,
      "step": 10155
    },
    {
      "entropy": 5.478008985519409,
      "epoch": 0.9216255442670537,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004922383926331236,
      "loss": 5.1405,
      "mean_token_accuracy": 0.19280444234609603,
      "num_tokens": 21087524.0,
      "step": 10160
    },
    {
      "entropy": 5.425194454193115,
      "epoch": 0.9220791001451378,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004922299651523392,
      "loss": 5.2124,
      "mean_token_accuracy": 0.18724745512008667,
      "num_tokens": 21098446.0,
      "step": 10165
    },
    {
      "entropy": 5.459459257125855,
      "epoch": 0.9225326560232221,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004922215331791837,
      "loss": 5.2302,
      "mean_token_accuracy": 0.18599504977464676,
      "num_tokens": 21109291.0,
      "step": 10170
    },
    {
      "entropy": 5.501635313034058,
      "epoch": 0.9229862119013063,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004922130967138316,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19838399291038514,
      "num_tokens": 21120043.0,
      "step": 10175
    },
    {
      "entropy": 5.518260431289673,
      "epoch": 0.9234397677793904,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004922046557564574,
      "loss": 5.2318,
      "mean_token_accuracy": 0.18464371412992478,
      "num_tokens": 21130339.0,
      "step": 10180
    },
    {
      "entropy": 5.4879247665405275,
      "epoch": 0.9238933236574746,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004921962103072355,
      "loss": 5.2982,
      "mean_token_accuracy": 0.18253571540117264,
      "num_tokens": 21142307.0,
      "step": 10185
    },
    {
      "entropy": 5.516984462738037,
      "epoch": 0.9243468795355588,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004921877603663406,
      "loss": 5.1916,
      "mean_token_accuracy": 0.19168877601623535,
      "num_tokens": 21151843.0,
      "step": 10190
    },
    {
      "entropy": 5.52185640335083,
      "epoch": 0.924800435413643,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004921793059339474,
      "loss": 5.1917,
      "mean_token_accuracy": 0.1947432205080986,
      "num_tokens": 21162574.0,
      "step": 10195
    },
    {
      "entropy": 5.514074325561523,
      "epoch": 0.9252539912917271,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004921708470102309,
      "loss": 5.1502,
      "mean_token_accuracy": 0.19119355827569962,
      "num_tokens": 21172637.0,
      "step": 10200
    },
    {
      "entropy": 5.465252828598023,
      "epoch": 0.9257075471698113,
      "grad_norm": 0.9375,
      "learning_rate": 0.000492162383595366,
      "loss": 5.1695,
      "mean_token_accuracy": 0.19286154508590697,
      "num_tokens": 21182272.0,
      "step": 10205
    },
    {
      "entropy": 5.5177301406860355,
      "epoch": 0.9261611030478955,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004921539156895275,
      "loss": 5.287,
      "mean_token_accuracy": 0.18583095222711563,
      "num_tokens": 21193712.0,
      "step": 10210
    },
    {
      "entropy": 5.565571260452271,
      "epoch": 0.9266146589259797,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004921454432928908,
      "loss": 5.3146,
      "mean_token_accuracy": 0.18066612184047698,
      "num_tokens": 21205104.0,
      "step": 10215
    },
    {
      "entropy": 5.5436365604400635,
      "epoch": 0.9270682148040639,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000492136966405631,
      "loss": 5.1809,
      "mean_token_accuracy": 0.1893284574151039,
      "num_tokens": 21215495.0,
      "step": 10220
    },
    {
      "entropy": 5.496471214294433,
      "epoch": 0.927521770682148,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004921284850279233,
      "loss": 5.2052,
      "mean_token_accuracy": 0.18886639475822448,
      "num_tokens": 21226047.0,
      "step": 10225
    },
    {
      "entropy": 5.502618789672852,
      "epoch": 0.9279753265602322,
      "grad_norm": 1.0,
      "learning_rate": 0.0004921199991599433,
      "loss": 5.2528,
      "mean_token_accuracy": 0.18845335245132447,
      "num_tokens": 21235784.0,
      "step": 10230
    },
    {
      "entropy": 5.557482576370239,
      "epoch": 0.9284288824383164,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004921115088018665,
      "loss": 5.2097,
      "mean_token_accuracy": 0.19081354588270188,
      "num_tokens": 21245493.0,
      "step": 10235
    },
    {
      "entropy": 5.612878608703613,
      "epoch": 0.9288824383164006,
      "grad_norm": 0.96875,
      "learning_rate": 0.000492103013953868,
      "loss": 5.2906,
      "mean_token_accuracy": 0.1868404284119606,
      "num_tokens": 21255541.0,
      "step": 10240
    },
    {
      "entropy": 5.442387437820434,
      "epoch": 0.9293359941944848,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004920945146161242,
      "loss": 5.1463,
      "mean_token_accuracy": 0.19286169111728668,
      "num_tokens": 21265427.0,
      "step": 10245
    },
    {
      "entropy": 5.464549398422241,
      "epoch": 0.9297895500725689,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004920860107888105,
      "loss": 5.1543,
      "mean_token_accuracy": 0.1940089598298073,
      "num_tokens": 21276597.0,
      "step": 10250
    },
    {
      "entropy": 5.460808277130127,
      "epoch": 0.9302431059506531,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004920775024721026,
      "loss": 5.1745,
      "mean_token_accuracy": 0.18666350245475768,
      "num_tokens": 21286416.0,
      "step": 10255
    },
    {
      "entropy": 5.482029771804809,
      "epoch": 0.9306966618287373,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004920689896661767,
      "loss": 5.1954,
      "mean_token_accuracy": 0.19428568035364152,
      "num_tokens": 21296064.0,
      "step": 10260
    },
    {
      "entropy": 5.459726476669312,
      "epoch": 0.9311502177068215,
      "grad_norm": 1.0,
      "learning_rate": 0.0004920604723712088,
      "loss": 5.2188,
      "mean_token_accuracy": 0.19196224957704544,
      "num_tokens": 21306836.0,
      "step": 10265
    },
    {
      "entropy": 5.48441743850708,
      "epoch": 0.9316037735849056,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004920519505873751,
      "loss": 5.1828,
      "mean_token_accuracy": 0.19369848519563676,
      "num_tokens": 21316936.0,
      "step": 10270
    },
    {
      "entropy": 5.464742708206177,
      "epoch": 0.9320573294629898,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004920434243148515,
      "loss": 5.1155,
      "mean_token_accuracy": 0.20192406624555587,
      "num_tokens": 21326892.0,
      "step": 10275
    },
    {
      "entropy": 5.472868013381958,
      "epoch": 0.9325108853410741,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004920348935538147,
      "loss": 5.1444,
      "mean_token_accuracy": 0.1884632021188736,
      "num_tokens": 21337270.0,
      "step": 10280
    },
    {
      "entropy": 5.532982063293457,
      "epoch": 0.9329644412191582,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492026358304441,
      "loss": 5.2328,
      "mean_token_accuracy": 0.19010204523801805,
      "num_tokens": 21347968.0,
      "step": 10285
    },
    {
      "entropy": 5.47656831741333,
      "epoch": 0.9334179970972424,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000492017818566907,
      "loss": 5.1746,
      "mean_token_accuracy": 0.19259967803955078,
      "num_tokens": 21358312.0,
      "step": 10290
    },
    {
      "entropy": 5.470859527587891,
      "epoch": 0.9338715529753265,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004920092743413892,
      "loss": 5.142,
      "mean_token_accuracy": 0.19385140240192414,
      "num_tokens": 21368794.0,
      "step": 10295
    },
    {
      "entropy": 5.508439350128174,
      "epoch": 0.9343251088534107,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004920007256280642,
      "loss": 5.2078,
      "mean_token_accuracy": 0.1947756752371788,
      "num_tokens": 21380080.0,
      "step": 10300
    },
    {
      "entropy": 5.497937631607056,
      "epoch": 0.934778664731495,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000491992172427109,
      "loss": 5.1906,
      "mean_token_accuracy": 0.19269357919692992,
      "num_tokens": 21389961.0,
      "step": 10305
    },
    {
      "entropy": 5.450817489624024,
      "epoch": 0.9352322206095791,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004919836147387002,
      "loss": 5.2161,
      "mean_token_accuracy": 0.1923079416155815,
      "num_tokens": 21399869.0,
      "step": 10310
    },
    {
      "entropy": 5.519410276412964,
      "epoch": 0.9356857764876633,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004919750525630151,
      "loss": 5.2044,
      "mean_token_accuracy": 0.1905330315232277,
      "num_tokens": 21410343.0,
      "step": 10315
    },
    {
      "entropy": 5.512972068786621,
      "epoch": 0.9361393323657474,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004919664859002306,
      "loss": 5.1847,
      "mean_token_accuracy": 0.18935477286577224,
      "num_tokens": 21420915.0,
      "step": 10320
    },
    {
      "entropy": 5.480722951889038,
      "epoch": 0.9365928882438317,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004919579147505239,
      "loss": 5.138,
      "mean_token_accuracy": 0.19782340973615647,
      "num_tokens": 21430879.0,
      "step": 10325
    },
    {
      "entropy": 5.5278362274169925,
      "epoch": 0.9370464441219158,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004919493391140721,
      "loss": 5.2364,
      "mean_token_accuracy": 0.186504028737545,
      "num_tokens": 21440843.0,
      "step": 10330
    },
    {
      "entropy": 5.567490196228027,
      "epoch": 0.9375,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004919407589910529,
      "loss": 5.4008,
      "mean_token_accuracy": 0.18015343099832534,
      "num_tokens": 21452132.0,
      "step": 10335
    },
    {
      "entropy": 5.494717454910278,
      "epoch": 0.9379535558780842,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004919321743816434,
      "loss": 5.185,
      "mean_token_accuracy": 0.1896796151995659,
      "num_tokens": 21461861.0,
      "step": 10340
    },
    {
      "entropy": 5.5109964370727536,
      "epoch": 0.9384071117561683,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004919235852860213,
      "loss": 5.1841,
      "mean_token_accuracy": 0.1865747943520546,
      "num_tokens": 21472244.0,
      "step": 10345
    },
    {
      "entropy": 5.528714323043824,
      "epoch": 0.9388606676342526,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004919149917043641,
      "loss": 5.2286,
      "mean_token_accuracy": 0.18693966567516326,
      "num_tokens": 21482394.0,
      "step": 10350
    },
    {
      "entropy": 5.504456090927124,
      "epoch": 0.9393142235123367,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004919063936368496,
      "loss": 5.2679,
      "mean_token_accuracy": 0.18453455120325088,
      "num_tokens": 21493334.0,
      "step": 10355
    },
    {
      "entropy": 5.5536915302276615,
      "epoch": 0.9397677793904209,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004918977910836557,
      "loss": 5.2333,
      "mean_token_accuracy": 0.18519501984119416,
      "num_tokens": 21504443.0,
      "step": 10360
    },
    {
      "entropy": 5.539934301376343,
      "epoch": 0.940221335268505,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004918891840449601,
      "loss": 5.2497,
      "mean_token_accuracy": 0.18833032846450806,
      "num_tokens": 21513908.0,
      "step": 10365
    },
    {
      "entropy": 5.542459392547608,
      "epoch": 0.9406748911465893,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000491880572520941,
      "loss": 5.2724,
      "mean_token_accuracy": 0.18035230338573455,
      "num_tokens": 21524349.0,
      "step": 10370
    },
    {
      "entropy": 5.6526354312896725,
      "epoch": 0.9411284470246735,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004918719565117763,
      "loss": 5.3279,
      "mean_token_accuracy": 0.1903665080666542,
      "num_tokens": 21535003.0,
      "step": 10375
    },
    {
      "entropy": 5.616309881210327,
      "epoch": 0.9415820029027576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004918633360176444,
      "loss": 5.3643,
      "mean_token_accuracy": 0.1793342188000679,
      "num_tokens": 21545045.0,
      "step": 10380
    },
    {
      "entropy": 5.4643491268157955,
      "epoch": 0.9420355587808418,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004918547110387234,
      "loss": 5.182,
      "mean_token_accuracy": 0.19717971086502076,
      "num_tokens": 21554830.0,
      "step": 10385
    },
    {
      "entropy": 5.5400168895721436,
      "epoch": 0.9424891146589259,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004918460815751916,
      "loss": 5.1437,
      "mean_token_accuracy": 0.19479870349168776,
      "num_tokens": 21564693.0,
      "step": 10390
    },
    {
      "entropy": 5.491382455825805,
      "epoch": 0.9429426705370102,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004918374476272277,
      "loss": 5.167,
      "mean_token_accuracy": 0.187149715423584,
      "num_tokens": 21575475.0,
      "step": 10395
    },
    {
      "entropy": 5.4525188446044925,
      "epoch": 0.9433962264150944,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004918288091950101,
      "loss": 5.1887,
      "mean_token_accuracy": 0.18847269862890242,
      "num_tokens": 21586944.0,
      "step": 10400
    },
    {
      "entropy": 5.544979858398437,
      "epoch": 0.9438497822931785,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004918201662787174,
      "loss": 5.2002,
      "mean_token_accuracy": 0.1872720867395401,
      "num_tokens": 21598170.0,
      "step": 10405
    },
    {
      "entropy": 5.511740016937256,
      "epoch": 0.9443033381712627,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004918115188785285,
      "loss": 5.2188,
      "mean_token_accuracy": 0.1880117490887642,
      "num_tokens": 21609170.0,
      "step": 10410
    },
    {
      "entropy": 5.515832614898682,
      "epoch": 0.9447568940493469,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004918028669946221,
      "loss": 5.2956,
      "mean_token_accuracy": 0.18484372794628143,
      "num_tokens": 21620171.0,
      "step": 10415
    },
    {
      "entropy": 5.486926412582397,
      "epoch": 0.9452104499274311,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917942106271771,
      "loss": 5.2104,
      "mean_token_accuracy": 0.19340168982744216,
      "num_tokens": 21631072.0,
      "step": 10420
    },
    {
      "entropy": 5.4726005554199215,
      "epoch": 0.9456640058055152,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004917855497763727,
      "loss": 5.0846,
      "mean_token_accuracy": 0.1976454183459282,
      "num_tokens": 21641543.0,
      "step": 10425
    },
    {
      "entropy": 5.4695212841033936,
      "epoch": 0.9461175616835994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004917768844423878,
      "loss": 5.2123,
      "mean_token_accuracy": 0.19243092685937882,
      "num_tokens": 21651342.0,
      "step": 10430
    },
    {
      "entropy": 5.524942016601562,
      "epoch": 0.9465711175616836,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004917682146254017,
      "loss": 5.1712,
      "mean_token_accuracy": 0.19580309987068176,
      "num_tokens": 21661043.0,
      "step": 10435
    },
    {
      "entropy": 5.4407624244689945,
      "epoch": 0.9470246734397678,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004917595403255938,
      "loss": 5.0551,
      "mean_token_accuracy": 0.1993831992149353,
      "num_tokens": 21671197.0,
      "step": 10440
    },
    {
      "entropy": 5.421433401107788,
      "epoch": 0.947478229317852,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004917508615431431,
      "loss": 5.1253,
      "mean_token_accuracy": 0.19686731249094008,
      "num_tokens": 21682066.0,
      "step": 10445
    },
    {
      "entropy": 5.554642391204834,
      "epoch": 0.9479317851959361,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004917421782782296,
      "loss": 5.1912,
      "mean_token_accuracy": 0.19019131511449813,
      "num_tokens": 21692117.0,
      "step": 10450
    },
    {
      "entropy": 5.5294716358184814,
      "epoch": 0.9483853410740203,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004917334905310325,
      "loss": 5.2939,
      "mean_token_accuracy": 0.18246906250715256,
      "num_tokens": 21703544.0,
      "step": 10455
    },
    {
      "entropy": 5.445235729217529,
      "epoch": 0.9488388969521045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004917247983017317,
      "loss": 5.1233,
      "mean_token_accuracy": 0.19221487492322922,
      "num_tokens": 21712850.0,
      "step": 10460
    },
    {
      "entropy": 5.474435758590698,
      "epoch": 0.9492924528301887,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004917161015905067,
      "loss": 5.2636,
      "mean_token_accuracy": 0.18826578557491302,
      "num_tokens": 21723015.0,
      "step": 10465
    },
    {
      "entropy": 5.513061094284057,
      "epoch": 0.9497460087082729,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004917074003975376,
      "loss": 5.2182,
      "mean_token_accuracy": 0.1881333664059639,
      "num_tokens": 21732453.0,
      "step": 10470
    },
    {
      "entropy": 5.551966810226441,
      "epoch": 0.950199564586357,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004916986947230042,
      "loss": 5.2908,
      "mean_token_accuracy": 0.18900395184755325,
      "num_tokens": 21743251.0,
      "step": 10475
    },
    {
      "entropy": 5.515570878982544,
      "epoch": 0.9506531204644412,
      "grad_norm": 1.0,
      "learning_rate": 0.0004916899845670866,
      "loss": 5.2076,
      "mean_token_accuracy": 0.18819339275360109,
      "num_tokens": 21752891.0,
      "step": 10480
    },
    {
      "entropy": 5.543520164489746,
      "epoch": 0.9511066763425254,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004916812699299649,
      "loss": 5.2014,
      "mean_token_accuracy": 0.19229596108198166,
      "num_tokens": 21763230.0,
      "step": 10485
    },
    {
      "entropy": 5.49464168548584,
      "epoch": 0.9515602322206096,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004916725508118193,
      "loss": 5.1474,
      "mean_token_accuracy": 0.1914653703570366,
      "num_tokens": 21773663.0,
      "step": 10490
    },
    {
      "entropy": 5.490184593200683,
      "epoch": 0.9520137880986937,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004916638272128301,
      "loss": 5.216,
      "mean_token_accuracy": 0.19228923469781875,
      "num_tokens": 21784265.0,
      "step": 10495
    },
    {
      "entropy": 5.456570482254028,
      "epoch": 0.9524673439767779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004916550991331777,
      "loss": 5.1794,
      "mean_token_accuracy": 0.19147132486104965,
      "num_tokens": 21795056.0,
      "step": 10500
    },
    {
      "entropy": 5.484953880310059,
      "epoch": 0.9529208998548622,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004916463665730427,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1929536059498787,
      "num_tokens": 21806204.0,
      "step": 10505
    },
    {
      "entropy": 5.4476096630096436,
      "epoch": 0.9533744557329463,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004916376295326057,
      "loss": 5.2107,
      "mean_token_accuracy": 0.19114798158407212,
      "num_tokens": 21816993.0,
      "step": 10510
    },
    {
      "entropy": 5.462368154525757,
      "epoch": 0.9538280116110305,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004916288880120472,
      "loss": 5.0671,
      "mean_token_accuracy": 0.19932386875152588,
      "num_tokens": 21827457.0,
      "step": 10515
    },
    {
      "entropy": 5.480139207839966,
      "epoch": 0.9542815674891146,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004916201420115482,
      "loss": 5.1847,
      "mean_token_accuracy": 0.18778882622718812,
      "num_tokens": 21837555.0,
      "step": 10520
    },
    {
      "entropy": 5.487935304641724,
      "epoch": 0.9547351233671988,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004916113915312895,
      "loss": 5.2163,
      "mean_token_accuracy": 0.19396420568227768,
      "num_tokens": 21847074.0,
      "step": 10525
    },
    {
      "entropy": 5.557778024673462,
      "epoch": 0.9551886792452831,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004916026365714519,
      "loss": 5.2066,
      "mean_token_accuracy": 0.18701234608888626,
      "num_tokens": 21858333.0,
      "step": 10530
    },
    {
      "entropy": 5.46713604927063,
      "epoch": 0.9556422351233672,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004915938771322165,
      "loss": 5.1986,
      "mean_token_accuracy": 0.19187063127756118,
      "num_tokens": 21869132.0,
      "step": 10535
    },
    {
      "entropy": 5.460751390457153,
      "epoch": 0.9560957910014514,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915851132137646,
      "loss": 5.2371,
      "mean_token_accuracy": 0.189180788397789,
      "num_tokens": 21879153.0,
      "step": 10540
    },
    {
      "entropy": 5.506651306152344,
      "epoch": 0.9565493468795355,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004915763448162774,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19192200899124146,
      "num_tokens": 21889063.0,
      "step": 10545
    },
    {
      "entropy": 5.5077903270721436,
      "epoch": 0.9570029027576198,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004915675719399362,
      "loss": 5.2454,
      "mean_token_accuracy": 0.1865275591611862,
      "num_tokens": 21899005.0,
      "step": 10550
    },
    {
      "entropy": 5.494836759567261,
      "epoch": 0.9574564586357039,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004915587945849225,
      "loss": 5.1099,
      "mean_token_accuracy": 0.2017269641160965,
      "num_tokens": 21908391.0,
      "step": 10555
    },
    {
      "entropy": 5.501866388320923,
      "epoch": 0.9579100145137881,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004915500127514176,
      "loss": 5.1823,
      "mean_token_accuracy": 0.1938887879252434,
      "num_tokens": 21917655.0,
      "step": 10560
    },
    {
      "entropy": 5.5588977336883545,
      "epoch": 0.9583635703918723,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004915412264396034,
      "loss": 5.2381,
      "mean_token_accuracy": 0.19208064824342727,
      "num_tokens": 21928038.0,
      "step": 10565
    },
    {
      "entropy": 5.502170753479004,
      "epoch": 0.9588171262699564,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004915324356496613,
      "loss": 5.1422,
      "mean_token_accuracy": 0.19868719577789307,
      "num_tokens": 21938106.0,
      "step": 10570
    },
    {
      "entropy": 5.459872579574585,
      "epoch": 0.9592706821480407,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004915236403817733,
      "loss": 5.1307,
      "mean_token_accuracy": 0.19554639905691146,
      "num_tokens": 21947484.0,
      "step": 10575
    },
    {
      "entropy": 5.483737945556641,
      "epoch": 0.9597242380261248,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004915148406361213,
      "loss": 5.2229,
      "mean_token_accuracy": 0.18886553943157197,
      "num_tokens": 21957139.0,
      "step": 10580
    },
    {
      "entropy": 5.475294065475464,
      "epoch": 0.960177793904209,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004915060364128872,
      "loss": 5.219,
      "mean_token_accuracy": 0.18840531557798385,
      "num_tokens": 21967030.0,
      "step": 10585
    },
    {
      "entropy": 5.461275815963745,
      "epoch": 0.9606313497822931,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004914972277122532,
      "loss": 5.1727,
      "mean_token_accuracy": 0.19003562033176422,
      "num_tokens": 21976418.0,
      "step": 10590
    },
    {
      "entropy": 5.482760334014893,
      "epoch": 0.9610849056603774,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004914884145344012,
      "loss": 5.1779,
      "mean_token_accuracy": 0.19301039576530457,
      "num_tokens": 21987284.0,
      "step": 10595
    },
    {
      "entropy": 5.4282470703125,
      "epoch": 0.9615384615384616,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004914795968795137,
      "loss": 5.0948,
      "mean_token_accuracy": 0.19979993253946304,
      "num_tokens": 21997106.0,
      "step": 10600
    },
    {
      "entropy": 5.441180753707886,
      "epoch": 0.9619920174165457,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004914707747477728,
      "loss": 5.0567,
      "mean_token_accuracy": 0.19796952307224275,
      "num_tokens": 22007589.0,
      "step": 10605
    },
    {
      "entropy": 5.470518398284912,
      "epoch": 0.9624455732946299,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004914619481393613,
      "loss": 5.2181,
      "mean_token_accuracy": 0.1860770180821419,
      "num_tokens": 22017669.0,
      "step": 10610
    },
    {
      "entropy": 5.617669343948364,
      "epoch": 0.962899129172714,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004914531170544615,
      "loss": 5.3642,
      "mean_token_accuracy": 0.18315273523330688,
      "num_tokens": 22029330.0,
      "step": 10615
    },
    {
      "entropy": 5.45698356628418,
      "epoch": 0.9633526850507983,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004914442814932561,
      "loss": 5.1359,
      "mean_token_accuracy": 0.19763439446687697,
      "num_tokens": 22039877.0,
      "step": 10620
    },
    {
      "entropy": 5.481981801986694,
      "epoch": 0.9638062409288825,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004914354414559278,
      "loss": 5.1859,
      "mean_token_accuracy": 0.19522749185562133,
      "num_tokens": 22049579.0,
      "step": 10625
    },
    {
      "entropy": 5.393762636184692,
      "epoch": 0.9642597968069666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004914265969426594,
      "loss": 5.0752,
      "mean_token_accuracy": 0.20582508593797683,
      "num_tokens": 22059608.0,
      "step": 10630
    },
    {
      "entropy": 5.472809171676635,
      "epoch": 0.9647133526850508,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004914177479536338,
      "loss": 5.2292,
      "mean_token_accuracy": 0.19040220081806183,
      "num_tokens": 22070036.0,
      "step": 10635
    },
    {
      "entropy": 5.493038988113403,
      "epoch": 0.965166908563135,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000491408894489034,
      "loss": 5.2222,
      "mean_token_accuracy": 0.18583654314279557,
      "num_tokens": 22080679.0,
      "step": 10640
    },
    {
      "entropy": 5.500444984436035,
      "epoch": 0.9656204644412192,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004914000365490432,
      "loss": 5.0437,
      "mean_token_accuracy": 0.20014223903417588,
      "num_tokens": 22092192.0,
      "step": 10645
    },
    {
      "entropy": 5.490348196029663,
      "epoch": 0.9660740203193033,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004913911741338445,
      "loss": 5.2413,
      "mean_token_accuracy": 0.1852397084236145,
      "num_tokens": 22102722.0,
      "step": 10650
    },
    {
      "entropy": 5.406291580200195,
      "epoch": 0.9665275761973875,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004913823072436211,
      "loss": 5.1155,
      "mean_token_accuracy": 0.19892334789037705,
      "num_tokens": 22113521.0,
      "step": 10655
    },
    {
      "entropy": 5.501695013046264,
      "epoch": 0.9669811320754716,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004913734358785565,
      "loss": 5.1965,
      "mean_token_accuracy": 0.19285878390073777,
      "num_tokens": 22124891.0,
      "step": 10660
    },
    {
      "entropy": 5.486810874938965,
      "epoch": 0.9674346879535559,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004913645600388341,
      "loss": 5.2107,
      "mean_token_accuracy": 0.18844253867864608,
      "num_tokens": 22135972.0,
      "step": 10665
    },
    {
      "entropy": 5.540691900253296,
      "epoch": 0.9678882438316401,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004913556797246374,
      "loss": 5.2649,
      "mean_token_accuracy": 0.18179991245269775,
      "num_tokens": 22146623.0,
      "step": 10670
    },
    {
      "entropy": 5.5169305324554445,
      "epoch": 0.9683417997097242,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004913467949361503,
      "loss": 5.1767,
      "mean_token_accuracy": 0.18900265991687776,
      "num_tokens": 22157619.0,
      "step": 10675
    },
    {
      "entropy": 5.546804666519165,
      "epoch": 0.9687953555878084,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004913379056735561,
      "loss": 5.1826,
      "mean_token_accuracy": 0.18539999574422836,
      "num_tokens": 22168655.0,
      "step": 10680
    },
    {
      "entropy": 5.47217059135437,
      "epoch": 0.9692489114658926,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000491329011937039,
      "loss": 5.2184,
      "mean_token_accuracy": 0.18761395961046218,
      "num_tokens": 22178500.0,
      "step": 10685
    },
    {
      "entropy": 5.515644979476929,
      "epoch": 0.9697024673439768,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004913201137267828,
      "loss": 5.224,
      "mean_token_accuracy": 0.18653703182935716,
      "num_tokens": 22188873.0,
      "step": 10690
    },
    {
      "entropy": 5.454254674911499,
      "epoch": 0.970156023222061,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004913112110429715,
      "loss": 5.1422,
      "mean_token_accuracy": 0.19080322682857515,
      "num_tokens": 22199743.0,
      "step": 10695
    },
    {
      "entropy": 5.465599822998047,
      "epoch": 0.9706095791001451,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004913023038857892,
      "loss": 5.0907,
      "mean_token_accuracy": 0.2004644140601158,
      "num_tokens": 22210301.0,
      "step": 10700
    },
    {
      "entropy": 5.4484978199005125,
      "epoch": 0.9710631349782293,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004912933922554203,
      "loss": 5.1006,
      "mean_token_accuracy": 0.1963333547115326,
      "num_tokens": 22220771.0,
      "step": 10705
    },
    {
      "entropy": 5.459470510482788,
      "epoch": 0.9715166908563135,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004912844761520487,
      "loss": 5.2071,
      "mean_token_accuracy": 0.1855422705411911,
      "num_tokens": 22231100.0,
      "step": 10710
    },
    {
      "entropy": 5.4956214427948,
      "epoch": 0.9719702467343977,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004912755555758592,
      "loss": 5.1826,
      "mean_token_accuracy": 0.19422222226858138,
      "num_tokens": 22240997.0,
      "step": 10715
    },
    {
      "entropy": 5.514835929870605,
      "epoch": 0.9724238026124818,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000491266630527036,
      "loss": 5.2124,
      "mean_token_accuracy": 0.189374640583992,
      "num_tokens": 22251795.0,
      "step": 10720
    },
    {
      "entropy": 5.549368095397949,
      "epoch": 0.972877358490566,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004912577010057638,
      "loss": 5.257,
      "mean_token_accuracy": 0.18586982041597366,
      "num_tokens": 22262465.0,
      "step": 10725
    },
    {
      "entropy": 5.599781894683838,
      "epoch": 0.9733309143686503,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004912487670122273,
      "loss": 5.3566,
      "mean_token_accuracy": 0.18138834536075593,
      "num_tokens": 22272399.0,
      "step": 10730
    },
    {
      "entropy": 5.5480303287506105,
      "epoch": 0.9737844702467344,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004912398285466111,
      "loss": 5.2657,
      "mean_token_accuracy": 0.18869337886571885,
      "num_tokens": 22282545.0,
      "step": 10735
    },
    {
      "entropy": 5.521822595596314,
      "epoch": 0.9742380261248186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004912308856091002,
      "loss": 5.2233,
      "mean_token_accuracy": 0.19182628244161606,
      "num_tokens": 22293154.0,
      "step": 10740
    },
    {
      "entropy": 5.5208924293518065,
      "epoch": 0.9746915820029027,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004912219381998795,
      "loss": 5.143,
      "mean_token_accuracy": 0.19636851102113723,
      "num_tokens": 22302309.0,
      "step": 10745
    },
    {
      "entropy": 5.558265972137451,
      "epoch": 0.9751451378809869,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004912129863191339,
      "loss": 5.2559,
      "mean_token_accuracy": 0.178457310795784,
      "num_tokens": 22312316.0,
      "step": 10750
    },
    {
      "entropy": 5.517702722549439,
      "epoch": 0.9755986937590712,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004912040299670487,
      "loss": 5.2337,
      "mean_token_accuracy": 0.1895495042204857,
      "num_tokens": 22322387.0,
      "step": 10755
    },
    {
      "entropy": 5.497195672988892,
      "epoch": 0.9760522496371553,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004911950691438091,
      "loss": 5.22,
      "mean_token_accuracy": 0.19919940531253816,
      "num_tokens": 22332759.0,
      "step": 10760
    },
    {
      "entropy": 5.516394901275635,
      "epoch": 0.9765058055152395,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004911861038496004,
      "loss": 5.2637,
      "mean_token_accuracy": 0.18964977264404298,
      "num_tokens": 22344056.0,
      "step": 10765
    },
    {
      "entropy": 5.451268243789673,
      "epoch": 0.9769593613933236,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004911771340846079,
      "loss": 5.2089,
      "mean_token_accuracy": 0.1823955789208412,
      "num_tokens": 22353877.0,
      "step": 10770
    },
    {
      "entropy": 5.557831621170044,
      "epoch": 0.9774129172714079,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911681598490172,
      "loss": 5.2351,
      "mean_token_accuracy": 0.18687503039836884,
      "num_tokens": 22363549.0,
      "step": 10775
    },
    {
      "entropy": 5.593202066421509,
      "epoch": 0.977866473149492,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911591811430139,
      "loss": 5.2692,
      "mean_token_accuracy": 0.18995759636163712,
      "num_tokens": 22374040.0,
      "step": 10780
    },
    {
      "entropy": 5.54970760345459,
      "epoch": 0.9783200290275762,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911501979667836,
      "loss": 5.3196,
      "mean_token_accuracy": 0.1861735463142395,
      "num_tokens": 22384819.0,
      "step": 10785
    },
    {
      "entropy": 5.487377309799195,
      "epoch": 0.9787735849056604,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004911412103205122,
      "loss": 5.1873,
      "mean_token_accuracy": 0.1898432657122612,
      "num_tokens": 22395130.0,
      "step": 10790
    },
    {
      "entropy": 5.503686380386353,
      "epoch": 0.9792271407837445,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004911322182043855,
      "loss": 5.2081,
      "mean_token_accuracy": 0.18677622675895691,
      "num_tokens": 22405093.0,
      "step": 10795
    },
    {
      "entropy": 5.51582441329956,
      "epoch": 0.9796806966618288,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004911232216185893,
      "loss": 5.2269,
      "mean_token_accuracy": 0.18393291234970094,
      "num_tokens": 22415008.0,
      "step": 10800
    },
    {
      "entropy": 5.5575262069702145,
      "epoch": 0.9801342525399129,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004911142205633099,
      "loss": 5.2651,
      "mean_token_accuracy": 0.18396617025136947,
      "num_tokens": 22426486.0,
      "step": 10805
    },
    {
      "entropy": 5.4911479473114015,
      "epoch": 0.9805878084179971,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004911052150387334,
      "loss": 5.1708,
      "mean_token_accuracy": 0.19237590581178665,
      "num_tokens": 22436844.0,
      "step": 10810
    },
    {
      "entropy": 5.568431949615478,
      "epoch": 0.9810413642960812,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004910962050450458,
      "loss": 5.3003,
      "mean_token_accuracy": 0.1872747138142586,
      "num_tokens": 22446630.0,
      "step": 10815
    },
    {
      "entropy": 5.56194052696228,
      "epoch": 0.9814949201741655,
      "grad_norm": 1.0,
      "learning_rate": 0.0004910871905824338,
      "loss": 5.1379,
      "mean_token_accuracy": 0.1935226395726204,
      "num_tokens": 22456406.0,
      "step": 10820
    },
    {
      "entropy": 5.460543298721314,
      "epoch": 0.9819484760522497,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004910781716510837,
      "loss": 5.1251,
      "mean_token_accuracy": 0.19543748646974562,
      "num_tokens": 22466210.0,
      "step": 10825
    },
    {
      "entropy": 5.49596643447876,
      "epoch": 0.9824020319303338,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004910691482511817,
      "loss": 5.2273,
      "mean_token_accuracy": 0.1921169638633728,
      "num_tokens": 22476978.0,
      "step": 10830
    },
    {
      "entropy": 5.588016319274902,
      "epoch": 0.982855587808418,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004910601203829148,
      "loss": 5.1929,
      "mean_token_accuracy": 0.18680335581302643,
      "num_tokens": 22488613.0,
      "step": 10835
    },
    {
      "entropy": 5.482051849365234,
      "epoch": 0.9833091436865021,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004910510880464696,
      "loss": 5.0861,
      "mean_token_accuracy": 0.2025364950299263,
      "num_tokens": 22498312.0,
      "step": 10840
    },
    {
      "entropy": 5.485942792892456,
      "epoch": 0.9837626995645864,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004910420512420327,
      "loss": 5.2127,
      "mean_token_accuracy": 0.1877727434039116,
      "num_tokens": 22508131.0,
      "step": 10845
    },
    {
      "entropy": 5.513615369796753,
      "epoch": 0.9842162554426706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004910330099697913,
      "loss": 5.1943,
      "mean_token_accuracy": 0.1947124034166336,
      "num_tokens": 22517842.0,
      "step": 10850
    },
    {
      "entropy": 5.540058898925781,
      "epoch": 0.9846698113207547,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004910239642299321,
      "loss": 5.1716,
      "mean_token_accuracy": 0.19113553911447526,
      "num_tokens": 22528239.0,
      "step": 10855
    },
    {
      "entropy": 5.472961711883545,
      "epoch": 0.9851233671988389,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004910149140226422,
      "loss": 5.1559,
      "mean_token_accuracy": 0.19106702506542206,
      "num_tokens": 22539734.0,
      "step": 10860
    },
    {
      "entropy": 5.481329441070557,
      "epoch": 0.9855769230769231,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004910058593481089,
      "loss": 5.1549,
      "mean_token_accuracy": 0.19573762267827988,
      "num_tokens": 22549796.0,
      "step": 10865
    },
    {
      "entropy": 5.457813835144043,
      "epoch": 0.9860304789550073,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004909968002065195,
      "loss": 5.137,
      "mean_token_accuracy": 0.19690064191818238,
      "num_tokens": 22559934.0,
      "step": 10870
    },
    {
      "entropy": 5.597090387344361,
      "epoch": 0.9864840348330914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490987736598061,
      "loss": 5.26,
      "mean_token_accuracy": 0.1757528841495514,
      "num_tokens": 22569864.0,
      "step": 10875
    },
    {
      "entropy": 5.525681018829346,
      "epoch": 0.9869375907111756,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909786685229211,
      "loss": 5.1874,
      "mean_token_accuracy": 0.19665819704532622,
      "num_tokens": 22580687.0,
      "step": 10880
    },
    {
      "entropy": 5.555548810958863,
      "epoch": 0.9873911465892597,
      "grad_norm": 1.0,
      "learning_rate": 0.0004909695959812875,
      "loss": 5.2377,
      "mean_token_accuracy": 0.18621382713317872,
      "num_tokens": 22590507.0,
      "step": 10885
    },
    {
      "entropy": 5.540699434280396,
      "epoch": 0.987844702467344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004909605189733474,
      "loss": 5.2675,
      "mean_token_accuracy": 0.19158151596784592,
      "num_tokens": 22601901.0,
      "step": 10890
    },
    {
      "entropy": 5.478123521804809,
      "epoch": 0.9882982583454282,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004909514374992889,
      "loss": 5.2022,
      "mean_token_accuracy": 0.18934640884399415,
      "num_tokens": 22613003.0,
      "step": 10895
    },
    {
      "entropy": 5.444561386108399,
      "epoch": 0.9887518142235123,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909423515592996,
      "loss": 5.0531,
      "mean_token_accuracy": 0.1984160289168358,
      "num_tokens": 22623401.0,
      "step": 10900
    },
    {
      "entropy": 5.533397102355957,
      "epoch": 0.9892053701015965,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004909332611535675,
      "loss": 5.2067,
      "mean_token_accuracy": 0.1915249705314636,
      "num_tokens": 22633147.0,
      "step": 10905
    },
    {
      "entropy": 5.486864948272705,
      "epoch": 0.9896589259796807,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909241662822804,
      "loss": 5.0547,
      "mean_token_accuracy": 0.20758973360061644,
      "num_tokens": 22642990.0,
      "step": 10910
    },
    {
      "entropy": 5.489868307113648,
      "epoch": 0.9901124818577649,
      "grad_norm": 1.0,
      "learning_rate": 0.0004909150669456265,
      "loss": 5.2756,
      "mean_token_accuracy": 0.18449527472257615,
      "num_tokens": 22652734.0,
      "step": 10915
    },
    {
      "entropy": 5.500755023956299,
      "epoch": 0.9905660377358491,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000490905963143794,
      "loss": 5.2837,
      "mean_token_accuracy": 0.18280779719352722,
      "num_tokens": 22662968.0,
      "step": 10920
    },
    {
      "entropy": 5.448152780532837,
      "epoch": 0.9910195936139332,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004908968548769713,
      "loss": 5.1258,
      "mean_token_accuracy": 0.20330245196819305,
      "num_tokens": 22673157.0,
      "step": 10925
    },
    {
      "entropy": 5.517993688583374,
      "epoch": 0.9914731494920174,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004908877421453466,
      "loss": 5.2567,
      "mean_token_accuracy": 0.19120330810546876,
      "num_tokens": 22683105.0,
      "step": 10930
    },
    {
      "entropy": 5.48880033493042,
      "epoch": 0.9919267053701016,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004908786249491082,
      "loss": 5.1906,
      "mean_token_accuracy": 0.19872619658708573,
      "num_tokens": 22693793.0,
      "step": 10935
    },
    {
      "entropy": 5.597190093994141,
      "epoch": 0.9923802612481858,
      "grad_norm": 1.015625,
      "learning_rate": 0.000490869503288445,
      "loss": 5.2387,
      "mean_token_accuracy": 0.19000887721776963,
      "num_tokens": 22703884.0,
      "step": 10940
    },
    {
      "entropy": 5.532585954666137,
      "epoch": 0.99283381712627,
      "grad_norm": 1.0,
      "learning_rate": 0.0004908603771635454,
      "loss": 5.2745,
      "mean_token_accuracy": 0.18670693933963775,
      "num_tokens": 22713699.0,
      "step": 10945
    },
    {
      "entropy": 5.5036516189575195,
      "epoch": 0.9932873730043541,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004908512465745982,
      "loss": 5.1147,
      "mean_token_accuracy": 0.19385309815406798,
      "num_tokens": 22722871.0,
      "step": 10950
    },
    {
      "entropy": 5.504653596878052,
      "epoch": 0.9937409288824384,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004908421115217922,
      "loss": 5.1247,
      "mean_token_accuracy": 0.18782687783241273,
      "num_tokens": 22733314.0,
      "step": 10955
    },
    {
      "entropy": 5.431400537490845,
      "epoch": 0.9941944847605225,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004908329720053164,
      "loss": 5.1209,
      "mean_token_accuracy": 0.1992374211549759,
      "num_tokens": 22743255.0,
      "step": 10960
    },
    {
      "entropy": 5.420324754714966,
      "epoch": 0.9946480406386067,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004908238280253597,
      "loss": 5.1006,
      "mean_token_accuracy": 0.19695128947496415,
      "num_tokens": 22753857.0,
      "step": 10965
    },
    {
      "entropy": 5.496116971969604,
      "epoch": 0.9951015965166908,
      "grad_norm": 1.0,
      "learning_rate": 0.0004908146795821112,
      "loss": 5.1057,
      "mean_token_accuracy": 0.19125611186027527,
      "num_tokens": 22764326.0,
      "step": 10970
    },
    {
      "entropy": 5.502989768981934,
      "epoch": 0.995555152394775,
      "grad_norm": 1.0,
      "learning_rate": 0.0004908055266757602,
      "loss": 5.1898,
      "mean_token_accuracy": 0.19550111889839172,
      "num_tokens": 22774958.0,
      "step": 10975
    },
    {
      "entropy": 5.408831548690796,
      "epoch": 0.9960087082728593,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004907963693064958,
      "loss": 5.0892,
      "mean_token_accuracy": 0.1952947497367859,
      "num_tokens": 22785616.0,
      "step": 10980
    },
    {
      "entropy": 5.565986251831054,
      "epoch": 0.9964622641509434,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004907872074745075,
      "loss": 5.2869,
      "mean_token_accuracy": 0.18623248487710953,
      "num_tokens": 22795967.0,
      "step": 10985
    },
    {
      "entropy": 5.514637279510498,
      "epoch": 0.9969158200290276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004907780411799849,
      "loss": 5.2722,
      "mean_token_accuracy": 0.18776020854711534,
      "num_tokens": 22806335.0,
      "step": 10990
    },
    {
      "entropy": 5.554602861404419,
      "epoch": 0.9973693759071117,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004907688704231173,
      "loss": 5.2089,
      "mean_token_accuracy": 0.19053253084421157,
      "num_tokens": 22816390.0,
      "step": 10995
    },
    {
      "entropy": 5.597799396514892,
      "epoch": 0.997822931785196,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004907596952040945,
      "loss": 5.2777,
      "mean_token_accuracy": 0.18841657638549805,
      "num_tokens": 22826395.0,
      "step": 11000
    },
    {
      "entropy": 5.564309597015381,
      "epoch": 0.9982764876632801,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004907505155231063,
      "loss": 5.2449,
      "mean_token_accuracy": 0.18635112643241883,
      "num_tokens": 22836790.0,
      "step": 11005
    },
    {
      "entropy": 5.494991683959961,
      "epoch": 0.9987300435413643,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004907413313803424,
      "loss": 5.1386,
      "mean_token_accuracy": 0.19103351086378098,
      "num_tokens": 22846687.0,
      "step": 11010
    },
    {
      "entropy": 5.538417148590088,
      "epoch": 0.9991835994194485,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004907321427759928,
      "loss": 5.1779,
      "mean_token_accuracy": 0.19118976593017578,
      "num_tokens": 22857208.0,
      "step": 11015
    },
    {
      "entropy": 5.539492750167847,
      "epoch": 0.9996371552975326,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004907229497102475,
      "loss": 5.2696,
      "mean_token_accuracy": 0.1830685555934906,
      "num_tokens": 22868556.0,
      "step": 11020
    },
    {
      "entropy": 5.554223203659058,
      "epoch": 1.0000907111756168,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004907137521832966,
      "loss": 5.2441,
      "mean_token_accuracy": 0.18953093886375427,
      "num_tokens": 22880081.0,
      "step": 11025
    },
    {
      "entropy": 5.496646070480347,
      "epoch": 1.000544267053701,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004907045501953304,
      "loss": 5.0765,
      "mean_token_accuracy": 0.19675455540418624,
      "num_tokens": 22892141.0,
      "step": 11030
    },
    {
      "entropy": 5.515450572967529,
      "epoch": 1.0009978229317853,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000490695343746539,
      "loss": 5.0916,
      "mean_token_accuracy": 0.19881475418806077,
      "num_tokens": 22902688.0,
      "step": 11035
    },
    {
      "entropy": 5.431853437423706,
      "epoch": 1.0014513788098693,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004906861328371131,
      "loss": 5.0387,
      "mean_token_accuracy": 0.19932159036397934,
      "num_tokens": 22912453.0,
      "step": 11040
    },
    {
      "entropy": 5.550728511810303,
      "epoch": 1.0019049346879536,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004906769174672429,
      "loss": 5.1846,
      "mean_token_accuracy": 0.18711721748113633,
      "num_tokens": 22923167.0,
      "step": 11045
    },
    {
      "entropy": 5.52794919013977,
      "epoch": 1.0023584905660377,
      "grad_norm": 0.96875,
      "learning_rate": 0.000490667697637119,
      "loss": 5.1534,
      "mean_token_accuracy": 0.18911569565534592,
      "num_tokens": 22934753.0,
      "step": 11050
    },
    {
      "entropy": 5.456609916687012,
      "epoch": 1.002812046444122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004906584733469324,
      "loss": 5.0764,
      "mean_token_accuracy": 0.20015866309404373,
      "num_tokens": 22945277.0,
      "step": 11055
    },
    {
      "entropy": 5.559716701507568,
      "epoch": 1.0032656023222062,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004906492445968735,
      "loss": 5.1634,
      "mean_token_accuracy": 0.18062861263751984,
      "num_tokens": 22955797.0,
      "step": 11060
    },
    {
      "entropy": 5.544274425506591,
      "epoch": 1.0037191582002902,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004906400113871332,
      "loss": 5.132,
      "mean_token_accuracy": 0.19073731899261476,
      "num_tokens": 22965969.0,
      "step": 11065
    },
    {
      "entropy": 5.493846797943116,
      "epoch": 1.0041727140783745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004906307737179026,
      "loss": 5.1033,
      "mean_token_accuracy": 0.20635355561971663,
      "num_tokens": 22975323.0,
      "step": 11070
    },
    {
      "entropy": 5.447731399536133,
      "epoch": 1.0046262699564585,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004906215315893726,
      "loss": 5.108,
      "mean_token_accuracy": 0.19876318871974946,
      "num_tokens": 22985063.0,
      "step": 11075
    },
    {
      "entropy": 5.5594007015228275,
      "epoch": 1.0050798258345428,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004906122850017344,
      "loss": 5.2382,
      "mean_token_accuracy": 0.18069565892219544,
      "num_tokens": 22996001.0,
      "step": 11080
    },
    {
      "entropy": 5.500598669052124,
      "epoch": 1.005533381712627,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004906030339551792,
      "loss": 5.1468,
      "mean_token_accuracy": 0.19543736428022385,
      "num_tokens": 23006896.0,
      "step": 11085
    },
    {
      "entropy": 5.514100217819214,
      "epoch": 1.0059869375907111,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004905937784498983,
      "loss": 5.1205,
      "mean_token_accuracy": 0.1983324557542801,
      "num_tokens": 23017080.0,
      "step": 11090
    },
    {
      "entropy": 5.526914501190186,
      "epoch": 1.0064404934687954,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905845184860831,
      "loss": 5.0812,
      "mean_token_accuracy": 0.19216082692146302,
      "num_tokens": 23027168.0,
      "step": 11095
    },
    {
      "entropy": 5.4901073455810545,
      "epoch": 1.0068940493468794,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004905752540639252,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19402697086334228,
      "num_tokens": 23037364.0,
      "step": 11100
    },
    {
      "entropy": 5.444597339630127,
      "epoch": 1.0073476052249637,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004905659851836161,
      "loss": 5.0147,
      "mean_token_accuracy": 0.2009423241019249,
      "num_tokens": 23047609.0,
      "step": 11105
    },
    {
      "entropy": 5.493267250061035,
      "epoch": 1.007801161103048,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004905567118453475,
      "loss": 5.1704,
      "mean_token_accuracy": 0.18859416395425796,
      "num_tokens": 23057942.0,
      "step": 11110
    },
    {
      "entropy": 5.469370317459107,
      "epoch": 1.008254716981132,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004905474340493111,
      "loss": 5.0764,
      "mean_token_accuracy": 0.1908912554383278,
      "num_tokens": 23069119.0,
      "step": 11115
    },
    {
      "entropy": 5.45906629562378,
      "epoch": 1.0087082728592163,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004905381517956988,
      "loss": 5.0678,
      "mean_token_accuracy": 0.20038813650608062,
      "num_tokens": 23080060.0,
      "step": 11120
    },
    {
      "entropy": 5.434975862503052,
      "epoch": 1.0091618287373005,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004905288650847027,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19017168432474135,
      "num_tokens": 23090645.0,
      "step": 11125
    },
    {
      "entropy": 5.517891359329224,
      "epoch": 1.0096153846153846,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004905195739165146,
      "loss": 5.108,
      "mean_token_accuracy": 0.19634836316108703,
      "num_tokens": 23100449.0,
      "step": 11130
    },
    {
      "entropy": 5.488073348999023,
      "epoch": 1.0100689404934688,
      "grad_norm": 1.0,
      "learning_rate": 0.000490510278291327,
      "loss": 5.0931,
      "mean_token_accuracy": 0.1908187672495842,
      "num_tokens": 23110415.0,
      "step": 11135
    },
    {
      "entropy": 5.465580987930298,
      "epoch": 1.010522496371553,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004905009782093318,
      "loss": 5.08,
      "mean_token_accuracy": 0.19305008202791213,
      "num_tokens": 23120336.0,
      "step": 11140
    },
    {
      "entropy": 5.442973518371582,
      "epoch": 1.0109760522496372,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904916736707214,
      "loss": 5.157,
      "mean_token_accuracy": 0.18449659645557404,
      "num_tokens": 23130568.0,
      "step": 11145
    },
    {
      "entropy": 5.543105840682983,
      "epoch": 1.0114296081277214,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004904823646756885,
      "loss": 5.112,
      "mean_token_accuracy": 0.1943865343928337,
      "num_tokens": 23140080.0,
      "step": 11150
    },
    {
      "entropy": 5.483130693435669,
      "epoch": 1.0118831640058055,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004904730512244252,
      "loss": 5.0452,
      "mean_token_accuracy": 0.1947724238038063,
      "num_tokens": 23150360.0,
      "step": 11155
    },
    {
      "entropy": 5.424828720092774,
      "epoch": 1.0123367198838897,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004904637333171243,
      "loss": 5.0556,
      "mean_token_accuracy": 0.19245949983596802,
      "num_tokens": 23160620.0,
      "step": 11160
    },
    {
      "entropy": 5.550334787368774,
      "epoch": 1.0127902757619738,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004904544109539785,
      "loss": 5.1532,
      "mean_token_accuracy": 0.19757008105516433,
      "num_tokens": 23171920.0,
      "step": 11165
    },
    {
      "entropy": 5.491248846054077,
      "epoch": 1.013243831640058,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004904450841351805,
      "loss": 5.1023,
      "mean_token_accuracy": 0.1991717129945755,
      "num_tokens": 23182929.0,
      "step": 11170
    },
    {
      "entropy": 5.514200067520141,
      "epoch": 1.0136973875181423,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004904357528609236,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19128975570201873,
      "num_tokens": 23193020.0,
      "step": 11175
    },
    {
      "entropy": 5.448251628875733,
      "epoch": 1.0141509433962264,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004904264171314,
      "loss": 5.0941,
      "mean_token_accuracy": 0.20164622366428375,
      "num_tokens": 23204073.0,
      "step": 11180
    },
    {
      "entropy": 5.421363639831543,
      "epoch": 1.0146044992743106,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004904170769468035,
      "loss": 4.9913,
      "mean_token_accuracy": 0.19620340317487717,
      "num_tokens": 23214376.0,
      "step": 11185
    },
    {
      "entropy": 5.5082276344299315,
      "epoch": 1.0150580551523947,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004904077323073268,
      "loss": 5.1497,
      "mean_token_accuracy": 0.19836536943912506,
      "num_tokens": 23224304.0,
      "step": 11190
    },
    {
      "entropy": 5.492418527603149,
      "epoch": 1.015511611030479,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004903983832131633,
      "loss": 5.1656,
      "mean_token_accuracy": 0.1937434956431389,
      "num_tokens": 23235324.0,
      "step": 11195
    },
    {
      "entropy": 5.493016624450684,
      "epoch": 1.0159651669085632,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004903890296645063,
      "loss": 5.1331,
      "mean_token_accuracy": 0.19382944256067275,
      "num_tokens": 23245774.0,
      "step": 11200
    },
    {
      "entropy": 5.507924365997314,
      "epoch": 1.0164187227866472,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004903796716615495,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19998153001070024,
      "num_tokens": 23256274.0,
      "step": 11205
    },
    {
      "entropy": 5.483238792419433,
      "epoch": 1.0168722786647315,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004903703092044861,
      "loss": 5.0927,
      "mean_token_accuracy": 0.19459670037031174,
      "num_tokens": 23266304.0,
      "step": 11210
    },
    {
      "entropy": 5.416814374923706,
      "epoch": 1.0173258345428158,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004903609422935097,
      "loss": 5.0196,
      "mean_token_accuracy": 0.19843120574951173,
      "num_tokens": 23276552.0,
      "step": 11215
    },
    {
      "entropy": 5.462734937667847,
      "epoch": 1.0177793904208998,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004903515709288143,
      "loss": 5.1272,
      "mean_token_accuracy": 0.1896622136235237,
      "num_tokens": 23287799.0,
      "step": 11220
    },
    {
      "entropy": 5.425547027587891,
      "epoch": 1.018232946298984,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004903421951105933,
      "loss": 4.9804,
      "mean_token_accuracy": 0.2000567466020584,
      "num_tokens": 23297958.0,
      "step": 11225
    },
    {
      "entropy": 5.490804672241211,
      "epoch": 1.0186865021770681,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004903328148390411,
      "loss": 5.1606,
      "mean_token_accuracy": 0.18824684619903564,
      "num_tokens": 23307523.0,
      "step": 11230
    },
    {
      "entropy": 5.533596992492676,
      "epoch": 1.0191400580551524,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004903234301143513,
      "loss": 5.1305,
      "mean_token_accuracy": 0.1967354640364647,
      "num_tokens": 23317970.0,
      "step": 11235
    },
    {
      "entropy": 5.465874528884887,
      "epoch": 1.0195936139332367,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004903140409367181,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19505420625209807,
      "num_tokens": 23328612.0,
      "step": 11240
    },
    {
      "entropy": 5.415059518814087,
      "epoch": 1.0200471698113207,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004903046473063357,
      "loss": 5.0724,
      "mean_token_accuracy": 0.18936332017183305,
      "num_tokens": 23338104.0,
      "step": 11245
    },
    {
      "entropy": 5.445874977111816,
      "epoch": 1.020500725689405,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004902952492233983,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21072589308023454,
      "num_tokens": 23348105.0,
      "step": 11250
    },
    {
      "entropy": 5.442014884948731,
      "epoch": 1.020954281567489,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004902858466881002,
      "loss": 4.9613,
      "mean_token_accuracy": 0.19927150905132293,
      "num_tokens": 23357150.0,
      "step": 11255
    },
    {
      "entropy": 5.465661239624024,
      "epoch": 1.0214078374455733,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490276439700636,
      "loss": 5.1693,
      "mean_token_accuracy": 0.1910180225968361,
      "num_tokens": 23367223.0,
      "step": 11260
    },
    {
      "entropy": 5.492317008972168,
      "epoch": 1.0218613933236576,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004902670282612002,
      "loss": 5.0993,
      "mean_token_accuracy": 0.1948421135544777,
      "num_tokens": 23377664.0,
      "step": 11265
    },
    {
      "entropy": 5.424583578109742,
      "epoch": 1.0223149492017416,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004902576123699873,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20003152638673782,
      "num_tokens": 23388057.0,
      "step": 11270
    },
    {
      "entropy": 5.3437347412109375,
      "epoch": 1.0227685050798259,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902481920271922,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20076482594013215,
      "num_tokens": 23397808.0,
      "step": 11275
    },
    {
      "entropy": 5.3924784660339355,
      "epoch": 1.02322206095791,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004902387672330097,
      "loss": 5.0345,
      "mean_token_accuracy": 0.19002968221902847,
      "num_tokens": 23406979.0,
      "step": 11280
    },
    {
      "entropy": 5.443336629867554,
      "epoch": 1.0236756168359942,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004902293379876345,
      "loss": 5.0838,
      "mean_token_accuracy": 0.19750918447971344,
      "num_tokens": 23416545.0,
      "step": 11285
    },
    {
      "entropy": 5.479193782806396,
      "epoch": 1.0241291727140784,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004902199042912618,
      "loss": 5.1001,
      "mean_token_accuracy": 0.18947370052337648,
      "num_tokens": 23427464.0,
      "step": 11290
    },
    {
      "entropy": 5.497503757476807,
      "epoch": 1.0245827285921625,
      "grad_norm": 1.0,
      "learning_rate": 0.0004902104661440868,
      "loss": 5.1211,
      "mean_token_accuracy": 0.19749965518712997,
      "num_tokens": 23436690.0,
      "step": 11295
    },
    {
      "entropy": 5.464810943603515,
      "epoch": 1.0250362844702468,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902010235463042,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19768522083759307,
      "num_tokens": 23446701.0,
      "step": 11300
    },
    {
      "entropy": 5.534040021896362,
      "epoch": 1.025489840348331,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004901915764981098,
      "loss": 5.1473,
      "mean_token_accuracy": 0.19012537896633147,
      "num_tokens": 23456966.0,
      "step": 11305
    },
    {
      "entropy": 5.495480489730835,
      "epoch": 1.025943396226415,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004901821249996988,
      "loss": 5.1077,
      "mean_token_accuracy": 0.19729931354522706,
      "num_tokens": 23467300.0,
      "step": 11310
    },
    {
      "entropy": 5.5246494770050045,
      "epoch": 1.0263969521044993,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004901726690512665,
      "loss": 5.1462,
      "mean_token_accuracy": 0.1901402324438095,
      "num_tokens": 23478457.0,
      "step": 11315
    },
    {
      "entropy": 5.463982629776001,
      "epoch": 1.0268505079825834,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004901632086530087,
      "loss": 5.0403,
      "mean_token_accuracy": 0.19925176948308945,
      "num_tokens": 23488066.0,
      "step": 11320
    },
    {
      "entropy": 5.4931072235107425,
      "epoch": 1.0273040638606676,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004901537438051207,
      "loss": 5.1304,
      "mean_token_accuracy": 0.1888103410601616,
      "num_tokens": 23498856.0,
      "step": 11325
    },
    {
      "entropy": 5.479607105255127,
      "epoch": 1.027757619738752,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004901442745077986,
      "loss": 5.0765,
      "mean_token_accuracy": 0.196649968624115,
      "num_tokens": 23509845.0,
      "step": 11330
    },
    {
      "entropy": 5.557180452346802,
      "epoch": 1.028211175616836,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004901348007612381,
      "loss": 5.2062,
      "mean_token_accuracy": 0.19007708877325058,
      "num_tokens": 23520322.0,
      "step": 11335
    },
    {
      "entropy": 5.499701595306396,
      "epoch": 1.0286647314949202,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901253225656351,
      "loss": 5.1263,
      "mean_token_accuracy": 0.19733863621950148,
      "num_tokens": 23530030.0,
      "step": 11340
    },
    {
      "entropy": 5.559247732162476,
      "epoch": 1.0291182873730043,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004901158399211855,
      "loss": 5.1838,
      "mean_token_accuracy": 0.19454871863126755,
      "num_tokens": 23541063.0,
      "step": 11345
    },
    {
      "entropy": 5.469040298461914,
      "epoch": 1.0295718432510885,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004901063528280856,
      "loss": 5.0867,
      "mean_token_accuracy": 0.1939779043197632,
      "num_tokens": 23551693.0,
      "step": 11350
    },
    {
      "entropy": 5.4881370067596436,
      "epoch": 1.0300253991291728,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900968612865314,
      "loss": 5.1351,
      "mean_token_accuracy": 0.19694630354642867,
      "num_tokens": 23561758.0,
      "step": 11355
    },
    {
      "entropy": 5.429140520095825,
      "epoch": 1.0304789550072568,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900873652967195,
      "loss": 5.0771,
      "mean_token_accuracy": 0.18856350779533387,
      "num_tokens": 23571912.0,
      "step": 11360
    },
    {
      "entropy": 5.442412996292115,
      "epoch": 1.030932510885341,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004900778648588459,
      "loss": 5.0333,
      "mean_token_accuracy": 0.20146304965019227,
      "num_tokens": 23582600.0,
      "step": 11365
    },
    {
      "entropy": 5.472327375411988,
      "epoch": 1.0313860667634251,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004900683599731073,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20228659957647324,
      "num_tokens": 23592761.0,
      "step": 11370
    },
    {
      "entropy": 5.564744472503662,
      "epoch": 1.0318396226415094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900588506397002,
      "loss": 5.159,
      "mean_token_accuracy": 0.19236104786396027,
      "num_tokens": 23603119.0,
      "step": 11375
    },
    {
      "entropy": 5.404077386856079,
      "epoch": 1.0322931785195937,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004900493368588213,
      "loss": 5.068,
      "mean_token_accuracy": 0.20131881684064865,
      "num_tokens": 23613916.0,
      "step": 11380
    },
    {
      "entropy": 5.535603141784668,
      "epoch": 1.0327467343976777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004900398186306672,
      "loss": 5.2023,
      "mean_token_accuracy": 0.19502190947532655,
      "num_tokens": 23623956.0,
      "step": 11385
    },
    {
      "entropy": 5.452536010742188,
      "epoch": 1.033200290275762,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000490030295955435,
      "loss": 5.0969,
      "mean_token_accuracy": 0.19548225551843643,
      "num_tokens": 23634666.0,
      "step": 11390
    },
    {
      "entropy": 5.414992809295654,
      "epoch": 1.0336538461538463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004900207688333215,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19136851727962495,
      "num_tokens": 23645601.0,
      "step": 11395
    },
    {
      "entropy": 5.4911417961120605,
      "epoch": 1.0341074020319303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900112372645236,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19436682015657425,
      "num_tokens": 23656339.0,
      "step": 11400
    },
    {
      "entropy": 5.497488927841187,
      "epoch": 1.0345609579100146,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004900017012492386,
      "loss": 5.0899,
      "mean_token_accuracy": 0.1950056940317154,
      "num_tokens": 23667433.0,
      "step": 11405
    },
    {
      "entropy": 5.528972911834717,
      "epoch": 1.0350145137880986,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004899921607876634,
      "loss": 5.1392,
      "mean_token_accuracy": 0.1908893272280693,
      "num_tokens": 23678186.0,
      "step": 11410
    },
    {
      "entropy": 5.481750106811523,
      "epoch": 1.0354680696661829,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004899826158799958,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19224554300308228,
      "num_tokens": 23688317.0,
      "step": 11415
    },
    {
      "entropy": 5.4784728527069095,
      "epoch": 1.0359216255442671,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004899730665264328,
      "loss": 5.0224,
      "mean_token_accuracy": 0.19699247032403946,
      "num_tokens": 23697752.0,
      "step": 11420
    },
    {
      "entropy": 5.447753953933716,
      "epoch": 1.0363751814223512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899635127271721,
      "loss": 4.9843,
      "mean_token_accuracy": 0.20179656744003296,
      "num_tokens": 23707803.0,
      "step": 11425
    },
    {
      "entropy": 5.401472043991089,
      "epoch": 1.0368287373004355,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004899539544824111,
      "loss": 5.0116,
      "mean_token_accuracy": 0.2051843896508217,
      "num_tokens": 23717630.0,
      "step": 11430
    },
    {
      "entropy": 5.473198318481446,
      "epoch": 1.0372822931785195,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004899443917923476,
      "loss": 5.1287,
      "mean_token_accuracy": 0.19700049012899398,
      "num_tokens": 23728836.0,
      "step": 11435
    },
    {
      "entropy": 5.452097368240357,
      "epoch": 1.0377358490566038,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004899348246571792,
      "loss": 5.1174,
      "mean_token_accuracy": 0.1917497396469116,
      "num_tokens": 23740145.0,
      "step": 11440
    },
    {
      "entropy": 5.4935462951660154,
      "epoch": 1.038189404934688,
      "grad_norm": 0.96875,
      "learning_rate": 0.000489925253077104,
      "loss": 5.0886,
      "mean_token_accuracy": 0.19191719591617584,
      "num_tokens": 23751162.0,
      "step": 11445
    },
    {
      "entropy": 5.504214334487915,
      "epoch": 1.038642960812772,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004899156770523196,
      "loss": 5.0929,
      "mean_token_accuracy": 0.19370600432157517,
      "num_tokens": 23761634.0,
      "step": 11450
    },
    {
      "entropy": 5.47665491104126,
      "epoch": 1.0390965166908563,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004899060965830244,
      "loss": 5.0611,
      "mean_token_accuracy": 0.20089899897575378,
      "num_tokens": 23772483.0,
      "step": 11455
    },
    {
      "entropy": 5.452212142944336,
      "epoch": 1.0395500725689404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004898965116694162,
      "loss": 5.0453,
      "mean_token_accuracy": 0.1946641504764557,
      "num_tokens": 23782298.0,
      "step": 11460
    },
    {
      "entropy": 5.494813919067383,
      "epoch": 1.0400036284470247,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004898869223116935,
      "loss": 5.1309,
      "mean_token_accuracy": 0.18946383744478226,
      "num_tokens": 23792492.0,
      "step": 11465
    },
    {
      "entropy": 5.54174313545227,
      "epoch": 1.040457184325109,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004898773285100546,
      "loss": 5.256,
      "mean_token_accuracy": 0.18755862563848497,
      "num_tokens": 23802999.0,
      "step": 11470
    },
    {
      "entropy": 5.487423419952393,
      "epoch": 1.040910740203193,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004898677302646975,
      "loss": 5.0665,
      "mean_token_accuracy": 0.1984709843993187,
      "num_tokens": 23812541.0,
      "step": 11475
    },
    {
      "entropy": 5.466430902481079,
      "epoch": 1.0413642960812772,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004898581275758211,
      "loss": 5.1706,
      "mean_token_accuracy": 0.18461524546146393,
      "num_tokens": 23823244.0,
      "step": 11480
    },
    {
      "entropy": 5.467811203002929,
      "epoch": 1.0418178519593615,
      "grad_norm": 1.0,
      "learning_rate": 0.0004898485204436239,
      "loss": 5.0044,
      "mean_token_accuracy": 0.20237330198287964,
      "num_tokens": 23832346.0,
      "step": 11485
    },
    {
      "entropy": 5.515140390396118,
      "epoch": 1.0422714078374455,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004898389088683047,
      "loss": 5.1301,
      "mean_token_accuracy": 0.20025138109922408,
      "num_tokens": 23842554.0,
      "step": 11490
    },
    {
      "entropy": 5.392171144485474,
      "epoch": 1.0427249637155298,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004898292928500619,
      "loss": 4.9725,
      "mean_token_accuracy": 0.20212062299251557,
      "num_tokens": 23852938.0,
      "step": 11495
    },
    {
      "entropy": 5.421096229553223,
      "epoch": 1.0431785195936139,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004898196723890947,
      "loss": 4.9989,
      "mean_token_accuracy": 0.20295965373516084,
      "num_tokens": 23862367.0,
      "step": 11500
    },
    {
      "entropy": 5.464499378204346,
      "epoch": 1.0436320754716981,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004898100474856019,
      "loss": 5.1109,
      "mean_token_accuracy": 0.1982620269060135,
      "num_tokens": 23873249.0,
      "step": 11505
    },
    {
      "entropy": 5.475397109985352,
      "epoch": 1.0440856313497824,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004898004181397824,
      "loss": 5.1431,
      "mean_token_accuracy": 0.1877864271402359,
      "num_tokens": 23883830.0,
      "step": 11510
    },
    {
      "entropy": 5.5174485206604,
      "epoch": 1.0445391872278664,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004897907843518358,
      "loss": 5.0633,
      "mean_token_accuracy": 0.19433184564113617,
      "num_tokens": 23894359.0,
      "step": 11515
    },
    {
      "entropy": 5.46463451385498,
      "epoch": 1.0449927431059507,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004897811461219609,
      "loss": 5.0531,
      "mean_token_accuracy": 0.19569095522165297,
      "num_tokens": 23903779.0,
      "step": 11520
    },
    {
      "entropy": 5.448574495315552,
      "epoch": 1.0454462989840347,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004897715034503571,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19483239948749542,
      "num_tokens": 23914523.0,
      "step": 11525
    },
    {
      "entropy": 5.456052017211914,
      "epoch": 1.045899854862119,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000489761856337224,
      "loss": 5.0868,
      "mean_token_accuracy": 0.19752579927444458,
      "num_tokens": 23924996.0,
      "step": 11530
    },
    {
      "entropy": 5.469106912612915,
      "epoch": 1.0463534107402033,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004897522047827609,
      "loss": 5.0567,
      "mean_token_accuracy": 0.20350758582353592,
      "num_tokens": 23935843.0,
      "step": 11535
    },
    {
      "entropy": 5.433482885360718,
      "epoch": 1.0468069666182873,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004897425487871676,
      "loss": 5.0343,
      "mean_token_accuracy": 0.1945498988032341,
      "num_tokens": 23945044.0,
      "step": 11540
    },
    {
      "entropy": 5.449752473831177,
      "epoch": 1.0472605224963716,
      "grad_norm": 1.0,
      "learning_rate": 0.0004897328883506436,
      "loss": 5.0772,
      "mean_token_accuracy": 0.19367363899946213,
      "num_tokens": 23955933.0,
      "step": 11545
    },
    {
      "entropy": 5.481488513946533,
      "epoch": 1.0477140783744556,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004897232234733887,
      "loss": 5.1493,
      "mean_token_accuracy": 0.18738794922828675,
      "num_tokens": 23966320.0,
      "step": 11550
    },
    {
      "entropy": 5.456253623962402,
      "epoch": 1.04816763425254,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897135541556028,
      "loss": 5.0468,
      "mean_token_accuracy": 0.20325277298688887,
      "num_tokens": 23976233.0,
      "step": 11555
    },
    {
      "entropy": 5.49235692024231,
      "epoch": 1.0486211901306242,
      "grad_norm": 0.84375,
      "learning_rate": 0.000489703880397486,
      "loss": 5.11,
      "mean_token_accuracy": 0.19220408350229262,
      "num_tokens": 23987854.0,
      "step": 11560
    },
    {
      "entropy": 5.457041692733765,
      "epoch": 1.0490747460087082,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004896942021992382,
      "loss": 5.1154,
      "mean_token_accuracy": 0.19077137857675552,
      "num_tokens": 23997973.0,
      "step": 11565
    },
    {
      "entropy": 5.482615661621094,
      "epoch": 1.0495283018867925,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004896845195610595,
      "loss": 5.1169,
      "mean_token_accuracy": 0.1928719624876976,
      "num_tokens": 24007313.0,
      "step": 11570
    },
    {
      "entropy": 5.476922655105591,
      "epoch": 1.0499818577648767,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004896748324831504,
      "loss": 5.1244,
      "mean_token_accuracy": 0.19298154413700103,
      "num_tokens": 24018254.0,
      "step": 11575
    },
    {
      "entropy": 5.492932891845703,
      "epoch": 1.0504354136429608,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004896651409657109,
      "loss": 5.1368,
      "mean_token_accuracy": 0.19103600829839706,
      "num_tokens": 24029799.0,
      "step": 11580
    },
    {
      "entropy": 5.414491605758667,
      "epoch": 1.050888969521045,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004896554450089417,
      "loss": 5.0258,
      "mean_token_accuracy": 0.19909809976816178,
      "num_tokens": 24040972.0,
      "step": 11585
    },
    {
      "entropy": 5.435051107406617,
      "epoch": 1.051342525399129,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004896457446130431,
      "loss": 4.958,
      "mean_token_accuracy": 0.2022219255566597,
      "num_tokens": 24050781.0,
      "step": 11590
    },
    {
      "entropy": 5.418264770507813,
      "epoch": 1.0517960812772134,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004896360397782159,
      "loss": 5.1011,
      "mean_token_accuracy": 0.1896982803940773,
      "num_tokens": 24060184.0,
      "step": 11595
    },
    {
      "entropy": 5.513707160949707,
      "epoch": 1.0522496371552976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004896263305046607,
      "loss": 5.1531,
      "mean_token_accuracy": 0.1932757616043091,
      "num_tokens": 24070777.0,
      "step": 11600
    },
    {
      "entropy": 5.481814289093018,
      "epoch": 1.0527031930333817,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004896166167925782,
      "loss": 5.1822,
      "mean_token_accuracy": 0.19279300570487976,
      "num_tokens": 24080878.0,
      "step": 11605
    },
    {
      "entropy": 5.519545316696167,
      "epoch": 1.053156748911466,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004896068986421693,
      "loss": 5.1407,
      "mean_token_accuracy": 0.19935741722583772,
      "num_tokens": 24090038.0,
      "step": 11610
    },
    {
      "entropy": 5.561168384552002,
      "epoch": 1.05361030478955,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004895971760536351,
      "loss": 5.1541,
      "mean_token_accuracy": 0.19477766305208205,
      "num_tokens": 24100425.0,
      "step": 11615
    },
    {
      "entropy": 5.538584423065186,
      "epoch": 1.0540638606676342,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004895874490271768,
      "loss": 5.2186,
      "mean_token_accuracy": 0.18994971215724946,
      "num_tokens": 24112182.0,
      "step": 11620
    },
    {
      "entropy": 5.481795358657837,
      "epoch": 1.0545174165457185,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004895777175629952,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1948103278875351,
      "num_tokens": 24122182.0,
      "step": 11625
    },
    {
      "entropy": 5.5433516025543215,
      "epoch": 1.0549709724238026,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004895679816612917,
      "loss": 5.0685,
      "mean_token_accuracy": 0.2016040116548538,
      "num_tokens": 24131930.0,
      "step": 11630
    },
    {
      "entropy": 5.5081311702728275,
      "epoch": 1.0554245283018868,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004895582413222678,
      "loss": 5.0693,
      "mean_token_accuracy": 0.20441591888666152,
      "num_tokens": 24141575.0,
      "step": 11635
    },
    {
      "entropy": 5.475733232498169,
      "epoch": 1.0558780841799709,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004895484965461247,
      "loss": 5.2086,
      "mean_token_accuracy": 0.1897387534379959,
      "num_tokens": 24151720.0,
      "step": 11640
    },
    {
      "entropy": 5.4533342838287355,
      "epoch": 1.0563316400580551,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489538747333064,
      "loss": 5.089,
      "mean_token_accuracy": 0.19781393110752105,
      "num_tokens": 24161881.0,
      "step": 11645
    },
    {
      "entropy": 5.43628077507019,
      "epoch": 1.0567851959361394,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895289936832874,
      "loss": 5.0461,
      "mean_token_accuracy": 0.20072407871484757,
      "num_tokens": 24171474.0,
      "step": 11650
    },
    {
      "entropy": 5.535496854782105,
      "epoch": 1.0572387518142234,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004895192355969966,
      "loss": 5.1662,
      "mean_token_accuracy": 0.19123655706644058,
      "num_tokens": 24182186.0,
      "step": 11655
    },
    {
      "entropy": 5.506235218048095,
      "epoch": 1.0576923076923077,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004895094730743932,
      "loss": 5.0906,
      "mean_token_accuracy": 0.19827568531036377,
      "num_tokens": 24192318.0,
      "step": 11660
    },
    {
      "entropy": 5.52546124458313,
      "epoch": 1.058145863570392,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004894997061156793,
      "loss": 5.1616,
      "mean_token_accuracy": 0.19303964376449584,
      "num_tokens": 24202918.0,
      "step": 11665
    },
    {
      "entropy": 5.439881277084351,
      "epoch": 1.058599419448476,
      "grad_norm": 1.0,
      "learning_rate": 0.0004894899347210567,
      "loss": 5.0427,
      "mean_token_accuracy": 0.20049711018800737,
      "num_tokens": 24213218.0,
      "step": 11670
    },
    {
      "entropy": 5.467959308624268,
      "epoch": 1.0590529753265603,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004894801588907278,
      "loss": 5.0387,
      "mean_token_accuracy": 0.198946575820446,
      "num_tokens": 24223229.0,
      "step": 11675
    },
    {
      "entropy": 5.52664999961853,
      "epoch": 1.0595065312046443,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004894703786248944,
      "loss": 5.1111,
      "mean_token_accuracy": 0.19159167855978013,
      "num_tokens": 24233637.0,
      "step": 11680
    },
    {
      "entropy": 5.478523540496826,
      "epoch": 1.0599600870827286,
      "grad_norm": 1.0,
      "learning_rate": 0.000489460593923759,
      "loss": 5.1528,
      "mean_token_accuracy": 0.1973455563187599,
      "num_tokens": 24244251.0,
      "step": 11685
    },
    {
      "entropy": 5.44070839881897,
      "epoch": 1.0604136429608129,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004894508047875237,
      "loss": 5.0526,
      "mean_token_accuracy": 0.19362615346908568,
      "num_tokens": 24254380.0,
      "step": 11690
    },
    {
      "entropy": 5.355499649047852,
      "epoch": 1.060867198838897,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004894410112163913,
      "loss": 5.0026,
      "mean_token_accuracy": 0.2004845768213272,
      "num_tokens": 24264841.0,
      "step": 11695
    },
    {
      "entropy": 5.414450836181641,
      "epoch": 1.0613207547169812,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004894312132105641,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20084351897239686,
      "num_tokens": 24274383.0,
      "step": 11700
    },
    {
      "entropy": 5.5343756675720215,
      "epoch": 1.0617743105950652,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004894214107702447,
      "loss": 5.0948,
      "mean_token_accuracy": 0.19480592012405396,
      "num_tokens": 24284140.0,
      "step": 11705
    },
    {
      "entropy": 5.590111207962036,
      "epoch": 1.0622278664731495,
      "grad_norm": 1.109375,
      "learning_rate": 0.000489411603895636,
      "loss": 5.1771,
      "mean_token_accuracy": 0.19371833950281142,
      "num_tokens": 24295044.0,
      "step": 11710
    },
    {
      "entropy": 5.4418053150177,
      "epoch": 1.0626814223512338,
      "grad_norm": 1.0,
      "learning_rate": 0.0004894017925869406,
      "loss": 5.08,
      "mean_token_accuracy": 0.19503361880779266,
      "num_tokens": 24305080.0,
      "step": 11715
    },
    {
      "entropy": 5.4726255416870115,
      "epoch": 1.0631349782293178,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004893919768443615,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19226745814085006,
      "num_tokens": 24316306.0,
      "step": 11720
    },
    {
      "entropy": 5.4423802375793455,
      "epoch": 1.063588534107402,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004893821566681017,
      "loss": 5.1073,
      "mean_token_accuracy": 0.19782619476318358,
      "num_tokens": 24326740.0,
      "step": 11725
    },
    {
      "entropy": 5.503826522827149,
      "epoch": 1.0640420899854863,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004893723320583644,
      "loss": 5.1457,
      "mean_token_accuracy": 0.18667171746492386,
      "num_tokens": 24337621.0,
      "step": 11730
    },
    {
      "entropy": 5.399421215057373,
      "epoch": 1.0644956458635704,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004893625030153525,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20595765560865403,
      "num_tokens": 24347732.0,
      "step": 11735
    },
    {
      "entropy": 5.4055805683135985,
      "epoch": 1.0649492017416546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004893526695392694,
      "loss": 5.0222,
      "mean_token_accuracy": 0.20757206678390502,
      "num_tokens": 24359173.0,
      "step": 11740
    },
    {
      "entropy": 5.422066068649292,
      "epoch": 1.0654027576197387,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004893428316303185,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19757631570100784,
      "num_tokens": 24369465.0,
      "step": 11745
    },
    {
      "entropy": 5.412892198562622,
      "epoch": 1.065856313497823,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004893329892887031,
      "loss": 5.0619,
      "mean_token_accuracy": 0.1981126308441162,
      "num_tokens": 24378899.0,
      "step": 11750
    },
    {
      "entropy": 5.484836912155151,
      "epoch": 1.0663098693759072,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004893231425146271,
      "loss": 5.1799,
      "mean_token_accuracy": 0.19192711710929872,
      "num_tokens": 24388919.0,
      "step": 11755
    },
    {
      "entropy": 5.4312244892120365,
      "epoch": 1.0667634252539913,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004893132913082937,
      "loss": 5.0478,
      "mean_token_accuracy": 0.1983172506093979,
      "num_tokens": 24399632.0,
      "step": 11760
    },
    {
      "entropy": 5.488347053527832,
      "epoch": 1.0672169811320755,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004893034356699069,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19779288917779922,
      "num_tokens": 24410817.0,
      "step": 11765
    },
    {
      "entropy": 5.417764663696289,
      "epoch": 1.0676705370101596,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004892935755996705,
      "loss": 5.0904,
      "mean_token_accuracy": 0.1991182193160057,
      "num_tokens": 24421058.0,
      "step": 11770
    },
    {
      "entropy": 5.408098793029785,
      "epoch": 1.0681240928882438,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004892837110977883,
      "loss": 5.0112,
      "mean_token_accuracy": 0.2003073051571846,
      "num_tokens": 24431522.0,
      "step": 11775
    },
    {
      "entropy": 5.462185478210449,
      "epoch": 1.068577648766328,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004892738421644644,
      "loss": 5.0522,
      "mean_token_accuracy": 0.19488768875598908,
      "num_tokens": 24441746.0,
      "step": 11780
    },
    {
      "entropy": 5.397897434234619,
      "epoch": 1.0690312046444121,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004892639687999027,
      "loss": 4.9586,
      "mean_token_accuracy": 0.20120252817869186,
      "num_tokens": 24451916.0,
      "step": 11785
    },
    {
      "entropy": 5.439264392852783,
      "epoch": 1.0694847605224964,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004892540910043076,
      "loss": 5.0829,
      "mean_token_accuracy": 0.20249505788087846,
      "num_tokens": 24461935.0,
      "step": 11790
    },
    {
      "entropy": 5.457518911361694,
      "epoch": 1.0699383164005805,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004892442087778833,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20152089595794678,
      "num_tokens": 24472437.0,
      "step": 11795
    },
    {
      "entropy": 5.457108974456787,
      "epoch": 1.0703918722786647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004892343221208342,
      "loss": 5.0965,
      "mean_token_accuracy": 0.20119676738977432,
      "num_tokens": 24481726.0,
      "step": 11800
    },
    {
      "entropy": 5.491053771972656,
      "epoch": 1.070845428156749,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004892244310333646,
      "loss": 5.1704,
      "mean_token_accuracy": 0.19343004524707794,
      "num_tokens": 24492340.0,
      "step": 11805
    },
    {
      "entropy": 5.452194881439209,
      "epoch": 1.071298984034833,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004892145355156793,
      "loss": 5.0366,
      "mean_token_accuracy": 0.19774752855300903,
      "num_tokens": 24502138.0,
      "step": 11810
    },
    {
      "entropy": 5.4785271167755125,
      "epoch": 1.0717525399129173,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004892046355679829,
      "loss": 5.1006,
      "mean_token_accuracy": 0.19350825250148773,
      "num_tokens": 24512723.0,
      "step": 11815
    },
    {
      "entropy": 5.504654312133789,
      "epoch": 1.0722060957910013,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004891947311904799,
      "loss": 5.1181,
      "mean_token_accuracy": 0.19506742358207702,
      "num_tokens": 24523766.0,
      "step": 11820
    },
    {
      "entropy": 5.500041055679321,
      "epoch": 1.0726596516690856,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004891848223833753,
      "loss": 5.1052,
      "mean_token_accuracy": 0.19698472768068315,
      "num_tokens": 24534571.0,
      "step": 11825
    },
    {
      "entropy": 5.482295513153076,
      "epoch": 1.0731132075471699,
      "grad_norm": 1.03125,
      "learning_rate": 0.000489174909146874,
      "loss": 5.1054,
      "mean_token_accuracy": 0.19994303733110427,
      "num_tokens": 24545265.0,
      "step": 11830
    },
    {
      "entropy": 5.5266765594482425,
      "epoch": 1.073566763425254,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004891649914811811,
      "loss": 5.1877,
      "mean_token_accuracy": 0.18802001029253007,
      "num_tokens": 24555524.0,
      "step": 11835
    },
    {
      "entropy": 5.397583484649658,
      "epoch": 1.0740203193033382,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004891550693865014,
      "loss": 5.0404,
      "mean_token_accuracy": 0.21245964467525483,
      "num_tokens": 24565420.0,
      "step": 11840
    },
    {
      "entropy": 5.469696044921875,
      "epoch": 1.0744738751814222,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004891451428630405,
      "loss": 5.0954,
      "mean_token_accuracy": 0.1935744747519493,
      "num_tokens": 24576794.0,
      "step": 11845
    },
    {
      "entropy": 5.500405263900757,
      "epoch": 1.0749274310595065,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004891352119110033,
      "loss": 5.0969,
      "mean_token_accuracy": 0.19519586563110353,
      "num_tokens": 24586320.0,
      "step": 11850
    },
    {
      "entropy": 5.446904039382934,
      "epoch": 1.0753809869375908,
      "grad_norm": 1.0,
      "learning_rate": 0.0004891252765305954,
      "loss": 5.0769,
      "mean_token_accuracy": 0.2048339292407036,
      "num_tokens": 24596524.0,
      "step": 11855
    },
    {
      "entropy": 5.4598876953125,
      "epoch": 1.0758345428156748,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004891153367220222,
      "loss": 5.016,
      "mean_token_accuracy": 0.20519405454397202,
      "num_tokens": 24605877.0,
      "step": 11860
    },
    {
      "entropy": 5.514345026016235,
      "epoch": 1.076288098693759,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004891053924854894,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1887012705206871,
      "num_tokens": 24616719.0,
      "step": 11865
    },
    {
      "entropy": 5.420893049240112,
      "epoch": 1.0767416545718433,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004890954438212024,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20212965607643127,
      "num_tokens": 24626882.0,
      "step": 11870
    },
    {
      "entropy": 5.469028377532959,
      "epoch": 1.0771952104499274,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004890854907293671,
      "loss": 5.0921,
      "mean_token_accuracy": 0.19691340178251265,
      "num_tokens": 24637884.0,
      "step": 11875
    },
    {
      "entropy": 5.448418807983399,
      "epoch": 1.0776487663280117,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004890755332101893,
      "loss": 5.0798,
      "mean_token_accuracy": 0.1930866315960884,
      "num_tokens": 24647389.0,
      "step": 11880
    },
    {
      "entropy": 5.449471950531006,
      "epoch": 1.0781023222060957,
      "grad_norm": 0.875,
      "learning_rate": 0.0004890655712638749,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19339723885059357,
      "num_tokens": 24657595.0,
      "step": 11885
    },
    {
      "entropy": 5.456168222427368,
      "epoch": 1.07855587808418,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048905560489063,
      "loss": 5.0011,
      "mean_token_accuracy": 0.19946667104959487,
      "num_tokens": 24667683.0,
      "step": 11890
    },
    {
      "entropy": 5.491331672668457,
      "epoch": 1.0790094339622642,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004890456340906605,
      "loss": 5.157,
      "mean_token_accuracy": 0.1912636011838913,
      "num_tokens": 24678041.0,
      "step": 11895
    },
    {
      "entropy": 5.445166206359863,
      "epoch": 1.0794629898403483,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004890356588641729,
      "loss": 5.1521,
      "mean_token_accuracy": 0.20095144957304,
      "num_tokens": 24688569.0,
      "step": 11900
    },
    {
      "entropy": 5.475362777709961,
      "epoch": 1.0799165457184325,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004890256792113733,
      "loss": 5.0967,
      "mean_token_accuracy": 0.19915975630283356,
      "num_tokens": 24699913.0,
      "step": 11905
    },
    {
      "entropy": 5.547744035720825,
      "epoch": 1.0803701015965168,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004890156951324681,
      "loss": 5.1223,
      "mean_token_accuracy": 0.19804803878068925,
      "num_tokens": 24710682.0,
      "step": 11910
    },
    {
      "entropy": 5.558585834503174,
      "epoch": 1.0808236574746009,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004890057066276636,
      "loss": 5.1974,
      "mean_token_accuracy": 0.18895184248685837,
      "num_tokens": 24721087.0,
      "step": 11915
    },
    {
      "entropy": 5.424392366409302,
      "epoch": 1.0812772133526851,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004889957136971668,
      "loss": 5.015,
      "mean_token_accuracy": 0.20212464183568954,
      "num_tokens": 24731323.0,
      "step": 11920
    },
    {
      "entropy": 5.424842023849488,
      "epoch": 1.0817307692307692,
      "grad_norm": 1.0,
      "learning_rate": 0.000488985716341184,
      "loss": 5.1052,
      "mean_token_accuracy": 0.1901997834444046,
      "num_tokens": 24742559.0,
      "step": 11925
    },
    {
      "entropy": 5.492843246459961,
      "epoch": 1.0821843251088534,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004889757145599222,
      "loss": 5.0918,
      "mean_token_accuracy": 0.1986592561006546,
      "num_tokens": 24752307.0,
      "step": 11930
    },
    {
      "entropy": 5.383856630325317,
      "epoch": 1.0826378809869377,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000488965708353588,
      "loss": 5.0516,
      "mean_token_accuracy": 0.19449937641620635,
      "num_tokens": 24762792.0,
      "step": 11935
    },
    {
      "entropy": 5.493750333786011,
      "epoch": 1.0830914368650217,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004889556977223883,
      "loss": 5.1593,
      "mean_token_accuracy": 0.19200471192598342,
      "num_tokens": 24774377.0,
      "step": 11940
    },
    {
      "entropy": 5.503545761108398,
      "epoch": 1.083544992743106,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004889456826665304,
      "loss": 5.078,
      "mean_token_accuracy": 0.19847277849912642,
      "num_tokens": 24785919.0,
      "step": 11945
    },
    {
      "entropy": 5.51596851348877,
      "epoch": 1.08399854862119,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004889356631862211,
      "loss": 5.115,
      "mean_token_accuracy": 0.18979937434196473,
      "num_tokens": 24796594.0,
      "step": 11950
    },
    {
      "entropy": 5.460939598083496,
      "epoch": 1.0844521044992743,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000488925639281668,
      "loss": 5.1161,
      "mean_token_accuracy": 0.1949406310915947,
      "num_tokens": 24807322.0,
      "step": 11955
    },
    {
      "entropy": 5.417977905273437,
      "epoch": 1.0849056603773586,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004889156109530781,
      "loss": 4.9998,
      "mean_token_accuracy": 0.19890448153018953,
      "num_tokens": 24817593.0,
      "step": 11960
    },
    {
      "entropy": 5.4494860649108885,
      "epoch": 1.0853592162554426,
      "grad_norm": 1.0,
      "learning_rate": 0.0004889055782006588,
      "loss": 5.0115,
      "mean_token_accuracy": 0.20230089426040648,
      "num_tokens": 24827577.0,
      "step": 11965
    },
    {
      "entropy": 5.441661405563354,
      "epoch": 1.085812772133527,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004888955410246176,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19938754588365554,
      "num_tokens": 24838148.0,
      "step": 11970
    },
    {
      "entropy": 5.443475580215454,
      "epoch": 1.086266328011611,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004888854994251621,
      "loss": 5.0108,
      "mean_token_accuracy": 0.19843291789293288,
      "num_tokens": 24848451.0,
      "step": 11975
    },
    {
      "entropy": 5.492741966247559,
      "epoch": 1.0867198838896952,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004888754534025,
      "loss": 5.1404,
      "mean_token_accuracy": 0.18598994314670564,
      "num_tokens": 24859619.0,
      "step": 11980
    },
    {
      "entropy": 5.499451589584351,
      "epoch": 1.0871734397677795,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004888654029568391,
      "loss": 5.1711,
      "mean_token_accuracy": 0.18404650390148164,
      "num_tokens": 24869863.0,
      "step": 11985
    },
    {
      "entropy": 5.457519960403443,
      "epoch": 1.0876269956458635,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004888553480883871,
      "loss": 5.1094,
      "mean_token_accuracy": 0.1899152174592018,
      "num_tokens": 24880524.0,
      "step": 11990
    },
    {
      "entropy": 5.415886926651001,
      "epoch": 1.0880805515239478,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000488845288797352,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19850811511278152,
      "num_tokens": 24891514.0,
      "step": 11995
    },
    {
      "entropy": 5.4387938499450685,
      "epoch": 1.0885341074020318,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004888352250839418,
      "loss": 5.0251,
      "mean_token_accuracy": 0.19936315417289735,
      "num_tokens": 24900842.0,
      "step": 12000
    },
    {
      "epoch": 1.0885341074020318,
      "eval_entropy": 5.235634247102349,
      "eval_loss": 5.16903829574585,
      "eval_mean_token_accuracy": 0.20154844025714636,
      "eval_num_tokens": 24900842.0,
      "eval_runtime": 20.6538,
      "eval_samples_per_second": 1712.03,
      "eval_steps_per_second": 214.004,
      "step": 12000
    },
    {
      "entropy": 5.4245994091033936,
      "epoch": 1.088987663280116,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004888251569483647,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20064777582883836,
      "num_tokens": 24911845.0,
      "step": 12005
    },
    {
      "entropy": 5.382421827316284,
      "epoch": 1.0894412191582004,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004888150843908288,
      "loss": 5.0402,
      "mean_token_accuracy": 0.20876970440149306,
      "num_tokens": 24922211.0,
      "step": 12010
    },
    {
      "entropy": 5.518812799453736,
      "epoch": 1.0898947750362844,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004888050074115426,
      "loss": 5.1569,
      "mean_token_accuracy": 0.1951376900076866,
      "num_tokens": 24932129.0,
      "step": 12015
    },
    {
      "entropy": 5.550019693374634,
      "epoch": 1.0903483309143687,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004887949260107143,
      "loss": 5.1401,
      "mean_token_accuracy": 0.19230424612760544,
      "num_tokens": 24942869.0,
      "step": 12020
    },
    {
      "entropy": 5.4164728164672855,
      "epoch": 1.0908018867924527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004887848401885524,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19477627873420716,
      "num_tokens": 24952621.0,
      "step": 12025
    },
    {
      "entropy": 5.556595325469971,
      "epoch": 1.091255442670537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004887747499452656,
      "loss": 5.195,
      "mean_token_accuracy": 0.1833399146795273,
      "num_tokens": 24963587.0,
      "step": 12030
    },
    {
      "entropy": 5.515872573852539,
      "epoch": 1.0917089985486212,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004887646552810624,
      "loss": 5.0398,
      "mean_token_accuracy": 0.2013555198907852,
      "num_tokens": 24974122.0,
      "step": 12035
    },
    {
      "entropy": 5.483128356933594,
      "epoch": 1.0921625544267053,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004887545561961517,
      "loss": 5.1285,
      "mean_token_accuracy": 0.1937446415424347,
      "num_tokens": 24985145.0,
      "step": 12040
    },
    {
      "entropy": 5.429468202590942,
      "epoch": 1.0926161103047896,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004887444526907424,
      "loss": 5.0495,
      "mean_token_accuracy": 0.1973423257470131,
      "num_tokens": 24995511.0,
      "step": 12045
    },
    {
      "entropy": 5.477809238433838,
      "epoch": 1.0930696661828738,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004887343447650432,
      "loss": 5.0753,
      "mean_token_accuracy": 0.19301147162914276,
      "num_tokens": 25006193.0,
      "step": 12050
    },
    {
      "entropy": 5.505136442184448,
      "epoch": 1.0935232220609579,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004887242324192634,
      "loss": 5.1009,
      "mean_token_accuracy": 0.19657760858535767,
      "num_tokens": 25016444.0,
      "step": 12055
    },
    {
      "entropy": 5.528617286682129,
      "epoch": 1.0939767779390421,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004887141156536119,
      "loss": 5.0888,
      "mean_token_accuracy": 0.19392160177230836,
      "num_tokens": 25027898.0,
      "step": 12060
    },
    {
      "entropy": 5.447785425186157,
      "epoch": 1.0944303338171262,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004887039944682981,
      "loss": 4.9868,
      "mean_token_accuracy": 0.20039299726486207,
      "num_tokens": 25038201.0,
      "step": 12065
    },
    {
      "entropy": 5.549479961395264,
      "epoch": 1.0948838896952104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004886938688635313,
      "loss": 5.1931,
      "mean_token_accuracy": 0.1823815271258354,
      "num_tokens": 25048466.0,
      "step": 12070
    },
    {
      "entropy": 5.569545698165894,
      "epoch": 1.0953374455732947,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004886837388395207,
      "loss": 5.1347,
      "mean_token_accuracy": 0.19485244303941726,
      "num_tokens": 25059064.0,
      "step": 12075
    },
    {
      "entropy": 5.543216323852539,
      "epoch": 1.0957910014513788,
      "grad_norm": 1.078125,
      "learning_rate": 0.000488673604396476,
      "loss": 5.2136,
      "mean_token_accuracy": 0.18050719946622848,
      "num_tokens": 25068834.0,
      "step": 12080
    },
    {
      "entropy": 5.506651210784912,
      "epoch": 1.096244557329463,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004886634655346067,
      "loss": 5.155,
      "mean_token_accuracy": 0.18910235464572905,
      "num_tokens": 25080058.0,
      "step": 12085
    },
    {
      "entropy": 5.440000104904175,
      "epoch": 1.0966981132075473,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004886533222541225,
      "loss": 5.0182,
      "mean_token_accuracy": 0.20164837390184404,
      "num_tokens": 25089751.0,
      "step": 12090
    },
    {
      "entropy": 5.478103876113892,
      "epoch": 1.0971516690856313,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886431745552331,
      "loss": 5.1127,
      "mean_token_accuracy": 0.19630647152662278,
      "num_tokens": 25099055.0,
      "step": 12095
    },
    {
      "entropy": 5.54117488861084,
      "epoch": 1.0976052249637156,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004886330224381485,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18562830686569215,
      "num_tokens": 25109793.0,
      "step": 12100
    },
    {
      "entropy": 5.521268510818482,
      "epoch": 1.0980587808417996,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004886228659030785,
      "loss": 5.1163,
      "mean_token_accuracy": 0.19521568566560746,
      "num_tokens": 25119814.0,
      "step": 12105
    },
    {
      "entropy": 5.5408790588378904,
      "epoch": 1.098512336719884,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004886127049502331,
      "loss": 5.1591,
      "mean_token_accuracy": 0.19464686661958694,
      "num_tokens": 25130747.0,
      "step": 12110
    },
    {
      "entropy": 5.454093742370605,
      "epoch": 1.0989658925979682,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004886025395798228,
      "loss": 5.0522,
      "mean_token_accuracy": 0.1918802246451378,
      "num_tokens": 25142235.0,
      "step": 12115
    },
    {
      "entropy": 5.366468334197998,
      "epoch": 1.0994194484760522,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004885923697920573,
      "loss": 4.9928,
      "mean_token_accuracy": 0.205183207988739,
      "num_tokens": 25152047.0,
      "step": 12120
    },
    {
      "entropy": 5.431280422210693,
      "epoch": 1.0998730043541365,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004885821955871472,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19808025658130646,
      "num_tokens": 25162678.0,
      "step": 12125
    },
    {
      "entropy": 5.49901065826416,
      "epoch": 1.1003265602322205,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004885720169653028,
      "loss": 5.0507,
      "mean_token_accuracy": 0.19658662378787994,
      "num_tokens": 25172973.0,
      "step": 12130
    },
    {
      "entropy": 5.507623529434204,
      "epoch": 1.1007801161103048,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004885618339267348,
      "loss": 5.1716,
      "mean_token_accuracy": 0.1899319440126419,
      "num_tokens": 25183458.0,
      "step": 12135
    },
    {
      "entropy": 5.445809078216553,
      "epoch": 1.101233671988389,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004885516464716537,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19627972692251205,
      "num_tokens": 25193827.0,
      "step": 12140
    },
    {
      "entropy": 5.5591919898986815,
      "epoch": 1.101687227866473,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048854145460027,
      "loss": 5.1016,
      "mean_token_accuracy": 0.19065470844507218,
      "num_tokens": 25204448.0,
      "step": 12145
    },
    {
      "entropy": 5.519454479217529,
      "epoch": 1.1021407837445574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004885312583127946,
      "loss": 5.0433,
      "mean_token_accuracy": 0.20112146586179733,
      "num_tokens": 25215041.0,
      "step": 12150
    },
    {
      "entropy": 5.444091129302978,
      "epoch": 1.1025943396226414,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004885210576094385,
      "loss": 5.0821,
      "mean_token_accuracy": 0.1961773082613945,
      "num_tokens": 25225937.0,
      "step": 12155
    },
    {
      "entropy": 5.418184804916382,
      "epoch": 1.1030478955007257,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004885108524904124,
      "loss": 4.995,
      "mean_token_accuracy": 0.1989628866314888,
      "num_tokens": 25235270.0,
      "step": 12160
    },
    {
      "entropy": 5.467058181762695,
      "epoch": 1.10350145137881,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004885006429559275,
      "loss": 5.101,
      "mean_token_accuracy": 0.18867916166782378,
      "num_tokens": 25247080.0,
      "step": 12165
    },
    {
      "entropy": 5.534938716888428,
      "epoch": 1.103955007256894,
      "grad_norm": 1.125,
      "learning_rate": 0.0004884904290061949,
      "loss": 5.0948,
      "mean_token_accuracy": 0.18896181732416154,
      "num_tokens": 25256718.0,
      "step": 12170
    },
    {
      "entropy": 5.4963476181030275,
      "epoch": 1.1044085631349783,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004884802106414258,
      "loss": 5.2132,
      "mean_token_accuracy": 0.1859460324048996,
      "num_tokens": 25266632.0,
      "step": 12175
    },
    {
      "entropy": 5.516941118240356,
      "epoch": 1.1048621190130623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884699878618316,
      "loss": 5.0693,
      "mean_token_accuracy": 0.19886712729930878,
      "num_tokens": 25275598.0,
      "step": 12180
    },
    {
      "entropy": 5.540178489685059,
      "epoch": 1.1053156748911466,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004884597606676236,
      "loss": 5.1522,
      "mean_token_accuracy": 0.1940113976597786,
      "num_tokens": 25286598.0,
      "step": 12185
    },
    {
      "entropy": 5.459210348129273,
      "epoch": 1.1057692307692308,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004884495290590134,
      "loss": 5.0831,
      "mean_token_accuracy": 0.19530683755874634,
      "num_tokens": 25296730.0,
      "step": 12190
    },
    {
      "entropy": 5.447724723815918,
      "epoch": 1.1062227866473149,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004884392930362124,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20545725524425507,
      "num_tokens": 25306326.0,
      "step": 12195
    },
    {
      "entropy": 5.494918966293335,
      "epoch": 1.1066763425253991,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004884290525994326,
      "loss": 5.0998,
      "mean_token_accuracy": 0.19826816022396088,
      "num_tokens": 25316342.0,
      "step": 12200
    },
    {
      "entropy": 5.579006290435791,
      "epoch": 1.1071298984034832,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884188077488855,
      "loss": 5.1396,
      "mean_token_accuracy": 0.1880671426653862,
      "num_tokens": 25326129.0,
      "step": 12205
    },
    {
      "entropy": 5.491881370544434,
      "epoch": 1.1075834542815675,
      "grad_norm": 1.0,
      "learning_rate": 0.0004884085584847831,
      "loss": 5.1088,
      "mean_token_accuracy": 0.18871529400348663,
      "num_tokens": 25337156.0,
      "step": 12210
    },
    {
      "entropy": 5.553006219863891,
      "epoch": 1.1080370101596517,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004883983048073375,
      "loss": 5.2199,
      "mean_token_accuracy": 0.18386245816946029,
      "num_tokens": 25347632.0,
      "step": 12215
    },
    {
      "entropy": 5.458075666427613,
      "epoch": 1.1084905660377358,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004883880467167603,
      "loss": 4.9776,
      "mean_token_accuracy": 0.20122691094875336,
      "num_tokens": 25359186.0,
      "step": 12220
    },
    {
      "entropy": 5.496602344512939,
      "epoch": 1.10894412191582,
      "grad_norm": 1.15625,
      "learning_rate": 0.000488377784213264,
      "loss": 5.0975,
      "mean_token_accuracy": 0.1946451857686043,
      "num_tokens": 25368993.0,
      "step": 12225
    },
    {
      "entropy": 5.43999285697937,
      "epoch": 1.1093976777939043,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004883675172970608,
      "loss": 5.0739,
      "mean_token_accuracy": 0.19383164942264558,
      "num_tokens": 25379880.0,
      "step": 12230
    },
    {
      "entropy": 5.533678007125855,
      "epoch": 1.1098512336719883,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000488357245968363,
      "loss": 5.2312,
      "mean_token_accuracy": 0.18909236490726472,
      "num_tokens": 25389942.0,
      "step": 12235
    },
    {
      "entropy": 5.496716022491455,
      "epoch": 1.1103047895500726,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004883469702273829,
      "loss": 5.045,
      "mean_token_accuracy": 0.19642258435487747,
      "num_tokens": 25400836.0,
      "step": 12240
    },
    {
      "entropy": 5.479660177230835,
      "epoch": 1.1107583454281567,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004883366900743331,
      "loss": 5.0513,
      "mean_token_accuracy": 0.20178468227386476,
      "num_tokens": 25410502.0,
      "step": 12245
    },
    {
      "entropy": 5.4430962085723875,
      "epoch": 1.111211901306241,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004883264055094262,
      "loss": 5.0206,
      "mean_token_accuracy": 0.20242179334163665,
      "num_tokens": 25420301.0,
      "step": 12250
    },
    {
      "entropy": 5.464849042892456,
      "epoch": 1.1116654571843252,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004883161165328749,
      "loss": 5.1091,
      "mean_token_accuracy": 0.19221176207065582,
      "num_tokens": 25431107.0,
      "step": 12255
    },
    {
      "entropy": 5.491602897644043,
      "epoch": 1.1121190130624092,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000488305823144892,
      "loss": 5.0869,
      "mean_token_accuracy": 0.1981050953269005,
      "num_tokens": 25441256.0,
      "step": 12260
    },
    {
      "entropy": 5.526266479492188,
      "epoch": 1.1125725689404935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004882955253456902,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20367671996355058,
      "num_tokens": 25451627.0,
      "step": 12265
    },
    {
      "entropy": 5.511344575881958,
      "epoch": 1.1130261248185778,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004882852231354826,
      "loss": 5.1622,
      "mean_token_accuracy": 0.18984195441007615,
      "num_tokens": 25461273.0,
      "step": 12270
    },
    {
      "entropy": 5.3871073722839355,
      "epoch": 1.1134796806966618,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004882749165144822,
      "loss": 5.0437,
      "mean_token_accuracy": 0.2021900862455368,
      "num_tokens": 25471949.0,
      "step": 12275
    },
    {
      "entropy": 5.513221025466919,
      "epoch": 1.113933236574746,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004882646054829023,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19517802745103835,
      "num_tokens": 25482175.0,
      "step": 12280
    },
    {
      "entropy": 5.539937973022461,
      "epoch": 1.1143867924528301,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048825429004095597,
      "loss": 5.1398,
      "mean_token_accuracy": 0.1953285738825798,
      "num_tokens": 25491786.0,
      "step": 12285
    },
    {
      "entropy": 5.4934309959411625,
      "epoch": 1.1148403483309144,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004882439701888566,
      "loss": 5.0614,
      "mean_token_accuracy": 0.2028183802962303,
      "num_tokens": 25502103.0,
      "step": 12290
    },
    {
      "entropy": 5.489936685562133,
      "epoch": 1.1152939042089987,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048823364592681754,
      "loss": 5.0345,
      "mean_token_accuracy": 0.20078462064266206,
      "num_tokens": 25512530.0,
      "step": 12295
    },
    {
      "entropy": 5.453662967681884,
      "epoch": 1.1157474600870827,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048822331725505234,
      "loss": 5.0326,
      "mean_token_accuracy": 0.1963891550898552,
      "num_tokens": 25522905.0,
      "step": 12300
    },
    {
      "entropy": 5.455841970443726,
      "epoch": 1.116201015965167,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048821298417377456,
      "loss": 5.0072,
      "mean_token_accuracy": 0.2051209181547165,
      "num_tokens": 25532963.0,
      "step": 12305
    },
    {
      "entropy": 5.400241136550903,
      "epoch": 1.116654571843251,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000488202646683198,
      "loss": 5.0859,
      "mean_token_accuracy": 0.19909514039754866,
      "num_tokens": 25542583.0,
      "step": 12310
    },
    {
      "entropy": 5.4752288341522215,
      "epoch": 1.1171081277213353,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004881923047835364,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19813199937343598,
      "num_tokens": 25553367.0,
      "step": 12315
    },
    {
      "entropy": 5.614972066879273,
      "epoch": 1.1175616835994195,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004881819584750036,
      "loss": 5.158,
      "mean_token_accuracy": 0.18798512667417527,
      "num_tokens": 25564823.0,
      "step": 12320
    },
    {
      "entropy": 5.5026294708251955,
      "epoch": 1.1180152394775036,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048817160775781356,
      "loss": 5.1412,
      "mean_token_accuracy": 0.19281987398862838,
      "num_tokens": 25574946.0,
      "step": 12325
    },
    {
      "entropy": 5.5025474548339846,
      "epoch": 1.1184687953555879,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048816125263218023,
      "loss": 5.1247,
      "mean_token_accuracy": 0.19272203445434571,
      "num_tokens": 25584779.0,
      "step": 12330
    },
    {
      "entropy": 5.55981478691101,
      "epoch": 1.118922351233672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048815089309831803,
      "loss": 5.1748,
      "mean_token_accuracy": 0.1895157814025879,
      "num_tokens": 25594725.0,
      "step": 12335
    },
    {
      "entropy": 5.490996026992798,
      "epoch": 1.1193759071117562,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000488140529156441,
      "loss": 5.1544,
      "mean_token_accuracy": 0.19034090340137483,
      "num_tokens": 25605499.0,
      "step": 12340
    },
    {
      "entropy": 5.4678388118743895,
      "epoch": 1.1198294629898404,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004881301608067635,
      "loss": 5.099,
      "mean_token_accuracy": 0.19867289513349534,
      "num_tokens": 25616443.0,
      "step": 12345
    },
    {
      "entropy": 5.508192157745361,
      "epoch": 1.1202830188679245,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048811978804949997,
      "loss": 5.0662,
      "mean_token_accuracy": 0.19435406923294068,
      "num_tokens": 25627762.0,
      "step": 12350
    },
    {
      "entropy": 5.572100734710693,
      "epoch": 1.1207365747460087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000488109410884865,
      "loss": 5.1585,
      "mean_token_accuracy": 0.19211570024490357,
      "num_tokens": 25637904.0,
      "step": 12355
    },
    {
      "entropy": 5.481449365615845,
      "epoch": 1.1211901306240928,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048809902931307294,
      "loss": 5.0221,
      "mean_token_accuracy": 0.20028771758079528,
      "num_tokens": 25649058.0,
      "step": 12360
    },
    {
      "entropy": 5.452664804458618,
      "epoch": 1.121643686502177,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004880886433343388,
      "loss": 5.0991,
      "mean_token_accuracy": 0.19796349555253984,
      "num_tokens": 25659161.0,
      "step": 12365
    },
    {
      "entropy": 5.456683731079101,
      "epoch": 1.1220972423802613,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004880782529488771,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20418920665979384,
      "num_tokens": 25669336.0,
      "step": 12370
    },
    {
      "entropy": 5.531540966033935,
      "epoch": 1.1225507982583454,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004880678581569029,
      "loss": 5.1348,
      "mean_token_accuracy": 0.19526318609714508,
      "num_tokens": 25680443.0,
      "step": 12375
    },
    {
      "entropy": 5.457948493957519,
      "epoch": 1.1230043541364296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048805745895863105,
      "loss": 5.0458,
      "mean_token_accuracy": 0.1991736814379692,
      "num_tokens": 25690202.0,
      "step": 12380
    },
    {
      "entropy": 5.457512521743775,
      "epoch": 1.1234579100145137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004880470553542767,
      "loss": 5.1085,
      "mean_token_accuracy": 0.1963775172829628,
      "num_tokens": 25699255.0,
      "step": 12385
    },
    {
      "entropy": 5.506438302993774,
      "epoch": 1.123911465892598,
      "grad_norm": 0.875,
      "learning_rate": 0.00048803664734405495,
      "loss": 5.0875,
      "mean_token_accuracy": 0.18788405805826186,
      "num_tokens": 25710714.0,
      "step": 12390
    },
    {
      "entropy": 5.501832008361816,
      "epoch": 1.1243650217706822,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000488026234928181,
      "loss": 5.0692,
      "mean_token_accuracy": 0.1955667182803154,
      "num_tokens": 25721722.0,
      "step": 12395
    },
    {
      "entropy": 5.461233329772949,
      "epoch": 1.1248185776487662,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004880158181068703,
      "loss": 5.0539,
      "mean_token_accuracy": 0.20558907687664033,
      "num_tokens": 25732096.0,
      "step": 12400
    },
    {
      "entropy": 5.483524560928345,
      "epoch": 1.1252721335268505,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048800539688033807,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19797305017709732,
      "num_tokens": 25741499.0,
      "step": 12405
    },
    {
      "entropy": 5.534729719161987,
      "epoch": 1.1257256894049348,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004879949712488001,
      "loss": 5.1901,
      "mean_token_accuracy": 0.1946818009018898,
      "num_tokens": 25751913.0,
      "step": 12410
    },
    {
      "entropy": 5.53972749710083,
      "epoch": 1.1261792452830188,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048798454121247177,
      "loss": 5.131,
      "mean_token_accuracy": 0.19206802695989608,
      "num_tokens": 25762602.0,
      "step": 12415
    },
    {
      "entropy": 5.440932893753052,
      "epoch": 1.126632801161103,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004879741067715688,
      "loss": 5.0934,
      "mean_token_accuracy": 0.19573008865118027,
      "num_tokens": 25773872.0,
      "step": 12420
    },
    {
      "entropy": 5.512517929077148,
      "epoch": 1.1270863570391871,
      "grad_norm": 1.0,
      "learning_rate": 0.00048796366792630715,
      "loss": 5.0935,
      "mean_token_accuracy": 0.19780833125114441,
      "num_tokens": 25784723.0,
      "step": 12425
    },
    {
      "entropy": 5.485009574890137,
      "epoch": 1.1275399129172714,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004879532246769025,
      "loss": 5.1044,
      "mean_token_accuracy": 0.19494618028402327,
      "num_tokens": 25794897.0,
      "step": 12430
    },
    {
      "entropy": 5.4504047393798825,
      "epoch": 1.1279934687953557,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004879427770235709,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20593176931142806,
      "num_tokens": 25804445.0,
      "step": 12435
    },
    {
      "entropy": 5.5492393493652346,
      "epoch": 1.1284470246734397,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048793232496652846,
      "loss": 5.1673,
      "mean_token_accuracy": 0.1864362522959709,
      "num_tokens": 25815495.0,
      "step": 12440
    },
    {
      "entropy": 5.485481834411621,
      "epoch": 1.128900580551524,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048792186850599124,
      "loss": 5.0865,
      "mean_token_accuracy": 0.20277364999055864,
      "num_tokens": 25826120.0,
      "step": 12445
    },
    {
      "entropy": 5.441395616531372,
      "epoch": 1.1293541364296082,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004879114076421755,
      "loss": 5.0743,
      "mean_token_accuracy": 0.2006630629301071,
      "num_tokens": 25835926.0,
      "step": 12450
    },
    {
      "entropy": 5.51393871307373,
      "epoch": 1.1298076923076923,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048790094237529763,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19224480241537095,
      "num_tokens": 25846095.0,
      "step": 12455
    },
    {
      "entropy": 5.568841123580933,
      "epoch": 1.1302612481857766,
      "grad_norm": 1.015625,
      "learning_rate": 0.000487890472705574,
      "loss": 5.1286,
      "mean_token_accuracy": 0.19136714339256286,
      "num_tokens": 25858541.0,
      "step": 12460
    },
    {
      "entropy": 5.4886997699737545,
      "epoch": 1.1307148040638606,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048787999863322113,
      "loss": 5.0166,
      "mean_token_accuracy": 0.2051388591527939,
      "num_tokens": 25867664.0,
      "step": 12465
    },
    {
      "entropy": 5.474790382385254,
      "epoch": 1.1311683599419449,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004878695201584557,
      "loss": 5.1188,
      "mean_token_accuracy": 0.19833892285823823,
      "num_tokens": 25877300.0,
      "step": 12470
    },
    {
      "entropy": 5.444518423080444,
      "epoch": 1.1316219158200291,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004878590372814943,
      "loss": 5.0776,
      "mean_token_accuracy": 0.20307615250349045,
      "num_tokens": 25888009.0,
      "step": 12475
    },
    {
      "entropy": 5.532946348190308,
      "epoch": 1.1320754716981132,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048784855000255374,
      "loss": 5.1264,
      "mean_token_accuracy": 0.18937529623508453,
      "num_tokens": 25897285.0,
      "step": 12480
    },
    {
      "entropy": 5.478674602508545,
      "epoch": 1.1325290275761974,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048783805832185097,
      "loss": 5.0157,
      "mean_token_accuracy": 0.201152403652668,
      "num_tokens": 25907355.0,
      "step": 12485
    },
    {
      "entropy": 5.494596529006958,
      "epoch": 1.1329825834542815,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004878275622396029,
      "loss": 5.0919,
      "mean_token_accuracy": 0.20091527253389357,
      "num_tokens": 25918566.0,
      "step": 12490
    },
    {
      "entropy": 5.486082220077515,
      "epoch": 1.1334361393323658,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048781706175602665,
      "loss": 5.1331,
      "mean_token_accuracy": 0.19996363818645477,
      "num_tokens": 25929145.0,
      "step": 12495
    },
    {
      "entropy": 5.460976219177246,
      "epoch": 1.13388969521045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048780655687133927,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20151668787002563,
      "num_tokens": 25938163.0,
      "step": 12500
    },
    {
      "entropy": 5.457347440719604,
      "epoch": 1.134343251088534,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048779604758575806,
      "loss": 5.046,
      "mean_token_accuracy": 0.19633741229772567,
      "num_tokens": 25948962.0,
      "step": 12505
    },
    {
      "entropy": 5.419883298873901,
      "epoch": 1.1347968069666183,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004877855338995004,
      "loss": 5.033,
      "mean_token_accuracy": 0.19193218201398848,
      "num_tokens": 25959653.0,
      "step": 12510
    },
    {
      "entropy": 5.5417643070220945,
      "epoch": 1.1352503628447024,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048777501581278367,
      "loss": 5.0924,
      "mean_token_accuracy": 0.19666285663843155,
      "num_tokens": 25970268.0,
      "step": 12515
    },
    {
      "entropy": 5.472268581390381,
      "epoch": 1.1357039187227866,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004877644933258254,
      "loss": 5.0664,
      "mean_token_accuracy": 0.1935422420501709,
      "num_tokens": 25980321.0,
      "step": 12520
    },
    {
      "entropy": 5.421860313415527,
      "epoch": 1.136157474600871,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004877539664388432,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19183527082204818,
      "num_tokens": 25992466.0,
      "step": 12525
    },
    {
      "entropy": 5.475657939910889,
      "epoch": 1.136611030478955,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048774343515205464,
      "loss": 5.0452,
      "mean_token_accuracy": 0.2010459080338478,
      "num_tokens": 26003553.0,
      "step": 12530
    },
    {
      "entropy": 5.488823318481446,
      "epoch": 1.1370645863570392,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004877328994656778,
      "loss": 5.0703,
      "mean_token_accuracy": 0.19986444860696792,
      "num_tokens": 26013299.0,
      "step": 12535
    },
    {
      "entropy": 5.454457712173462,
      "epoch": 1.1375181422351233,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004877223593799303,
      "loss": 5.0957,
      "mean_token_accuracy": 0.19835387021303177,
      "num_tokens": 26022913.0,
      "step": 12540
    },
    {
      "entropy": 5.421596193313599,
      "epoch": 1.1379716981132075,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004877118148950302,
      "loss": 5.0688,
      "mean_token_accuracy": 0.19932219684123992,
      "num_tokens": 26033191.0,
      "step": 12545
    },
    {
      "entropy": 5.6301860332489015,
      "epoch": 1.1384252539912918,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048770126601119555,
      "loss": 5.2148,
      "mean_token_accuracy": 0.18730915635824202,
      "num_tokens": 26043517.0,
      "step": 12550
    },
    {
      "entropy": 5.577491188049317,
      "epoch": 1.1388788098693758,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048769071272864454,
      "loss": 5.1624,
      "mean_token_accuracy": 0.19309705942869188,
      "num_tokens": 26054173.0,
      "step": 12555
    },
    {
      "entropy": 5.546094846725464,
      "epoch": 1.13933236574746,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048768015504759536,
      "loss": 5.1488,
      "mean_token_accuracy": 0.1877343937754631,
      "num_tokens": 26064929.0,
      "step": 12560
    },
    {
      "entropy": 5.445685005187988,
      "epoch": 1.1397859216255442,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004876695929682665,
      "loss": 5.0438,
      "mean_token_accuracy": 0.200204998254776,
      "num_tokens": 26074872.0,
      "step": 12565
    },
    {
      "entropy": 5.451004409790039,
      "epoch": 1.1402394775036284,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048765902649087605,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20139967799186706,
      "num_tokens": 26086328.0,
      "step": 12570
    },
    {
      "entropy": 5.524568271636963,
      "epoch": 1.1406930333817127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048764845561564285,
      "loss": 5.0762,
      "mean_token_accuracy": 0.19644412994384766,
      "num_tokens": 26096477.0,
      "step": 12575
    },
    {
      "entropy": 5.468381452560425,
      "epoch": 1.1411465892597967,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048763788034278545,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20553031712770461,
      "num_tokens": 26106183.0,
      "step": 12580
    },
    {
      "entropy": 5.454225015640259,
      "epoch": 1.141600145137881,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004876273006725225,
      "loss": 5.084,
      "mean_token_accuracy": 0.1978095680475235,
      "num_tokens": 26117932.0,
      "step": 12585
    },
    {
      "entropy": 5.446175527572632,
      "epoch": 1.1420537010159653,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048761671660507266,
      "loss": 5.0421,
      "mean_token_accuracy": 0.20245159715414046,
      "num_tokens": 26128616.0,
      "step": 12590
    },
    {
      "entropy": 5.502018308639526,
      "epoch": 1.1425072568940493,
      "grad_norm": 0.953125,
      "learning_rate": 0.000487606128140655,
      "loss": 5.0585,
      "mean_token_accuracy": 0.19867226481437683,
      "num_tokens": 26138758.0,
      "step": 12595
    },
    {
      "entropy": 5.4364453792572025,
      "epoch": 1.1429608127721336,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004875955352794885,
      "loss": 5.065,
      "mean_token_accuracy": 0.19398841261863708,
      "num_tokens": 26149199.0,
      "step": 12600
    },
    {
      "entropy": 5.430867147445679,
      "epoch": 1.1434143686502176,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004875849380217921,
      "loss": 5.1569,
      "mean_token_accuracy": 0.19733918458223343,
      "num_tokens": 26160276.0,
      "step": 12605
    },
    {
      "entropy": 5.500235271453858,
      "epoch": 1.1438679245283019,
      "grad_norm": 1.046875,
      "learning_rate": 0.000487574336367785,
      "loss": 5.0283,
      "mean_token_accuracy": 0.19995264261960982,
      "num_tokens": 26169564.0,
      "step": 12610
    },
    {
      "entropy": 5.42881498336792,
      "epoch": 1.1443214804063861,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004875637303176865,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20410393923521042,
      "num_tokens": 26179135.0,
      "step": 12615
    },
    {
      "entropy": 5.4993397235870365,
      "epoch": 1.1447750362844702,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048755311987171585,
      "loss": 5.1338,
      "mean_token_accuracy": 0.19475421607494353,
      "num_tokens": 26190116.0,
      "step": 12620
    },
    {
      "entropy": 5.511917066574097,
      "epoch": 1.1452285921625545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048754250503009244,
      "loss": 5.1992,
      "mean_token_accuracy": 0.18957407921552658,
      "num_tokens": 26200847.0,
      "step": 12625
    },
    {
      "entropy": 5.599766254425049,
      "epoch": 1.1456821480406387,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048753188579303596,
      "loss": 5.155,
      "mean_token_accuracy": 0.19580694288015366,
      "num_tokens": 26211151.0,
      "step": 12630
    },
    {
      "entropy": 5.524540090560913,
      "epoch": 1.1461357039187228,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004875212621607659,
      "loss": 5.0664,
      "mean_token_accuracy": 0.2049766629934311,
      "num_tokens": 26221273.0,
      "step": 12635
    },
    {
      "entropy": 5.4686830043792725,
      "epoch": 1.146589259796807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048751063413350203,
      "loss": 5.1737,
      "mean_token_accuracy": 0.19148514568805694,
      "num_tokens": 26231770.0,
      "step": 12640
    },
    {
      "entropy": 5.545756721496582,
      "epoch": 1.147042815674891,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000487500001711464,
      "loss": 5.1424,
      "mean_token_accuracy": 0.18901480883359909,
      "num_tokens": 26241386.0,
      "step": 12645
    },
    {
      "entropy": 5.642010736465454,
      "epoch": 1.1474963715529753,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004874893648948719,
      "loss": 5.2114,
      "mean_token_accuracy": 0.19006966054439545,
      "num_tokens": 26251991.0,
      "step": 12650
    },
    {
      "entropy": 5.627796792984009,
      "epoch": 1.1479499274310596,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048747872368394546,
      "loss": 5.2532,
      "mean_token_accuracy": 0.19527844488620758,
      "num_tokens": 26262331.0,
      "step": 12655
    },
    {
      "entropy": 5.561524677276611,
      "epoch": 1.1484034833091437,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048746807807890496,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19464536160230636,
      "num_tokens": 26272939.0,
      "step": 12660
    },
    {
      "entropy": 5.510550403594971,
      "epoch": 1.148857039187228,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048745742807997046,
      "loss": 5.0807,
      "mean_token_accuracy": 0.2004735991358757,
      "num_tokens": 26282849.0,
      "step": 12665
    },
    {
      "entropy": 5.467938804626465,
      "epoch": 1.149310595065312,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048744677368736216,
      "loss": 5.0885,
      "mean_token_accuracy": 0.2036068171262741,
      "num_tokens": 26292449.0,
      "step": 12670
    },
    {
      "entropy": 5.4095618724823,
      "epoch": 1.1497641509433962,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048743611490130044,
      "loss": 5.034,
      "mean_token_accuracy": 0.19792603999376296,
      "num_tokens": 26302506.0,
      "step": 12675
    },
    {
      "entropy": 5.50772762298584,
      "epoch": 1.1502177068214805,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048742545172200576,
      "loss": 5.1061,
      "mean_token_accuracy": 0.2004696920514107,
      "num_tokens": 26312346.0,
      "step": 12680
    },
    {
      "entropy": 5.542489147186279,
      "epoch": 1.1506712626995645,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004874147841496986,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19783084094524384,
      "num_tokens": 26323173.0,
      "step": 12685
    },
    {
      "entropy": 5.547864484786987,
      "epoch": 1.1511248185776488,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004874041121845996,
      "loss": 5.1629,
      "mean_token_accuracy": 0.187813962996006,
      "num_tokens": 26334371.0,
      "step": 12690
    },
    {
      "entropy": 5.475184345245362,
      "epoch": 1.1515783744557329,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004873934358269294,
      "loss": 5.0875,
      "mean_token_accuracy": 0.1992534339427948,
      "num_tokens": 26344506.0,
      "step": 12695
    },
    {
      "entropy": 5.564321422576905,
      "epoch": 1.1520319303338171,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004873827550769088,
      "loss": 5.1226,
      "mean_token_accuracy": 0.1877172037959099,
      "num_tokens": 26354534.0,
      "step": 12700
    },
    {
      "entropy": 5.496835136413575,
      "epoch": 1.1524854862119014,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004873720699347587,
      "loss": 5.0639,
      "mean_token_accuracy": 0.19809603989124297,
      "num_tokens": 26365682.0,
      "step": 12705
    },
    {
      "entropy": 5.570609045028687,
      "epoch": 1.1529390420899854,
      "grad_norm": 1.0,
      "learning_rate": 0.0004873613804007001,
      "loss": 5.1936,
      "mean_token_accuracy": 0.19731561243534088,
      "num_tokens": 26376702.0,
      "step": 12710
    },
    {
      "entropy": 5.4602783679962155,
      "epoch": 1.1533925979680697,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004873506864749541,
      "loss": 5.0067,
      "mean_token_accuracy": 0.19946054369211197,
      "num_tokens": 26387154.0,
      "step": 12715
    },
    {
      "entropy": 5.552713298797608,
      "epoch": 1.1538461538461537,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004873399881577417,
      "loss": 5.1788,
      "mean_token_accuracy": 0.19176933765411378,
      "num_tokens": 26397883.0,
      "step": 12720
    },
    {
      "entropy": 5.427668762207031,
      "epoch": 1.154299709724238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004873292854492842,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20469096153974534,
      "num_tokens": 26407613.0,
      "step": 12725
    },
    {
      "entropy": 5.396756219863891,
      "epoch": 1.1547532656023223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004873185783498031,
      "loss": 5.0023,
      "mean_token_accuracy": 0.20199891477823256,
      "num_tokens": 26417221.0,
      "step": 12730
    },
    {
      "entropy": 5.4925392150878904,
      "epoch": 1.1552068214804063,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004873078668595197,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19460710734128953,
      "num_tokens": 26427976.0,
      "step": 12735
    },
    {
      "entropy": 5.484690761566162,
      "epoch": 1.1556603773584906,
      "grad_norm": 1.0,
      "learning_rate": 0.00048729715097865534,
      "loss": 5.1287,
      "mean_token_accuracy": 0.19827549159526825,
      "num_tokens": 26439213.0,
      "step": 12740
    },
    {
      "entropy": 5.478594827651977,
      "epoch": 1.1561139332365746,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004872864307074319,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19928483068943023,
      "num_tokens": 26449763.0,
      "step": 12745
    },
    {
      "entropy": 5.546472692489624,
      "epoch": 1.156567489114659,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048727570604607096,
      "loss": 5.1252,
      "mean_token_accuracy": 0.19816723614931106,
      "num_tokens": 26460133.0,
      "step": 12750
    },
    {
      "entropy": 5.5383056640625,
      "epoch": 1.1570210449927432,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048726497699479437,
      "loss": 5.0922,
      "mean_token_accuracy": 0.19691014140844346,
      "num_tokens": 26470164.0,
      "step": 12755
    },
    {
      "entropy": 5.475792503356933,
      "epoch": 1.1574746008708272,
      "grad_norm": 0.921875,
      "learning_rate": 0.000487254243553824,
      "loss": 5.1044,
      "mean_token_accuracy": 0.19779773503541948,
      "num_tokens": 26480314.0,
      "step": 12760
    },
    {
      "entropy": 5.404000902175904,
      "epoch": 1.1579281567489115,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004872435057233817,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20014721006155015,
      "num_tokens": 26491184.0,
      "step": 12765
    },
    {
      "entropy": 5.449006414413452,
      "epoch": 1.1583817126269957,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048723276350368957,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20397381633520126,
      "num_tokens": 26501432.0,
      "step": 12770
    },
    {
      "entropy": 5.49757866859436,
      "epoch": 1.1588352685050798,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048722201689496984,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19721294641494752,
      "num_tokens": 26510379.0,
      "step": 12775
    },
    {
      "entropy": 5.46393346786499,
      "epoch": 1.159288824383164,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004872112658974447,
      "loss": 5.0777,
      "mean_token_accuracy": 0.20006239414215088,
      "num_tokens": 26519454.0,
      "step": 12780
    },
    {
      "entropy": 5.439549779891967,
      "epoch": 1.159742380261248,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048720051051133657,
      "loss": 4.9887,
      "mean_token_accuracy": 0.19962143003940583,
      "num_tokens": 26531325.0,
      "step": 12785
    },
    {
      "entropy": 5.460226392745971,
      "epoch": 1.1601959361393324,
      "grad_norm": 1.0,
      "learning_rate": 0.0004871897507368678,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19462902396917342,
      "num_tokens": 26541510.0,
      "step": 12790
    },
    {
      "entropy": 5.522227096557617,
      "epoch": 1.1606494920174166,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048717898657426093,
      "loss": 5.2614,
      "mean_token_accuracy": 0.18738264143466948,
      "num_tokens": 26553400.0,
      "step": 12795
    },
    {
      "entropy": 5.558783531188965,
      "epoch": 1.1611030478955007,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048716821802373856,
      "loss": 5.0776,
      "mean_token_accuracy": 0.1960080683231354,
      "num_tokens": 26564340.0,
      "step": 12800
    },
    {
      "entropy": 5.512545728683472,
      "epoch": 1.161556603773585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048715744508552334,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19665155112743377,
      "num_tokens": 26573615.0,
      "step": 12805
    },
    {
      "entropy": 5.526737642288208,
      "epoch": 1.1620101596516692,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004871466677598381,
      "loss": 5.198,
      "mean_token_accuracy": 0.19085295498371124,
      "num_tokens": 26584149.0,
      "step": 12810
    },
    {
      "entropy": 5.475385999679565,
      "epoch": 1.1624637155297532,
      "grad_norm": 1.0,
      "learning_rate": 0.0004871358860469057,
      "loss": 5.1147,
      "mean_token_accuracy": 0.18644985556602478,
      "num_tokens": 26595064.0,
      "step": 12815
    },
    {
      "entropy": 5.487485313415528,
      "epoch": 1.1629172714078375,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048712509994694903,
      "loss": 5.1004,
      "mean_token_accuracy": 0.1979370340704918,
      "num_tokens": 26605636.0,
      "step": 12820
    },
    {
      "entropy": 5.573808574676514,
      "epoch": 1.1633708272859216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004871143094601914,
      "loss": 5.196,
      "mean_token_accuracy": 0.19262153655290604,
      "num_tokens": 26616065.0,
      "step": 12825
    },
    {
      "entropy": 5.570623970031738,
      "epoch": 1.1638243831640058,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004871035145868557,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19735285639762878,
      "num_tokens": 26627246.0,
      "step": 12830
    },
    {
      "entropy": 5.500070810317993,
      "epoch": 1.16427793904209,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048709271532716524,
      "loss": 5.1282,
      "mean_token_accuracy": 0.19226828664541246,
      "num_tokens": 26637630.0,
      "step": 12835
    },
    {
      "entropy": 5.54427752494812,
      "epoch": 1.1647314949201741,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004870819116813434,
      "loss": 5.159,
      "mean_token_accuracy": 0.19256802797317504,
      "num_tokens": 26648683.0,
      "step": 12840
    },
    {
      "entropy": 5.546771287918091,
      "epoch": 1.1651850507982584,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048707110364961363,
      "loss": 5.1174,
      "mean_token_accuracy": 0.195330211520195,
      "num_tokens": 26659573.0,
      "step": 12845
    },
    {
      "entropy": 5.445801401138306,
      "epoch": 1.1656386066763424,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048706029123219937,
      "loss": 5.0211,
      "mean_token_accuracy": 0.2026770994067192,
      "num_tokens": 26669384.0,
      "step": 12850
    },
    {
      "entropy": 5.5239941596984865,
      "epoch": 1.1660921625544267,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048704947442932426,
      "loss": 5.195,
      "mean_token_accuracy": 0.1876979649066925,
      "num_tokens": 26679721.0,
      "step": 12855
    },
    {
      "entropy": 5.582709121704101,
      "epoch": 1.166545718432511,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048703865324121196,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20531719774007798,
      "num_tokens": 26689923.0,
      "step": 12860
    },
    {
      "entropy": 5.421995735168457,
      "epoch": 1.166999274310595,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048702782766808637,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19754623025655746,
      "num_tokens": 26699953.0,
      "step": 12865
    },
    {
      "entropy": 5.512957429885864,
      "epoch": 1.1674528301886793,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004870169977101711,
      "loss": 5.2723,
      "mean_token_accuracy": 0.18300295621156693,
      "num_tokens": 26712099.0,
      "step": 12870
    },
    {
      "entropy": 5.5708049774169925,
      "epoch": 1.1679063860667633,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048700616336769037,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19754792302846907,
      "num_tokens": 26722334.0,
      "step": 12875
    },
    {
      "entropy": 5.538000631332397,
      "epoch": 1.1683599419448476,
      "grad_norm": 1.0,
      "learning_rate": 0.00048699532464086817,
      "loss": 5.0668,
      "mean_token_accuracy": 0.19858947843313218,
      "num_tokens": 26732470.0,
      "step": 12880
    },
    {
      "entropy": 5.49172830581665,
      "epoch": 1.1688134978229319,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048698448152992865,
      "loss": 5.0952,
      "mean_token_accuracy": 0.19598378092050553,
      "num_tokens": 26742015.0,
      "step": 12885
    },
    {
      "entropy": 5.415465831756592,
      "epoch": 1.169267053701016,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048697363403509605,
      "loss": 5.1135,
      "mean_token_accuracy": 0.19333448261022568,
      "num_tokens": 26753231.0,
      "step": 12890
    },
    {
      "entropy": 5.515175199508667,
      "epoch": 1.1697206095791002,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004869627821565946,
      "loss": 5.0734,
      "mean_token_accuracy": 0.1952577516436577,
      "num_tokens": 26763895.0,
      "step": 12895
    },
    {
      "entropy": 5.568731832504272,
      "epoch": 1.1701741654571842,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004869519258946489,
      "loss": 5.1751,
      "mean_token_accuracy": 0.19621602445840836,
      "num_tokens": 26774208.0,
      "step": 12900
    },
    {
      "entropy": 5.484986734390259,
      "epoch": 1.1706277213352685,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048694106524948324,
      "loss": 5.1065,
      "mean_token_accuracy": 0.19364592730998992,
      "num_tokens": 26785554.0,
      "step": 12905
    },
    {
      "entropy": 5.521382761001587,
      "epoch": 1.1710812772133528,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048693020022132243,
      "loss": 5.0204,
      "mean_token_accuracy": 0.1997543454170227,
      "num_tokens": 26795488.0,
      "step": 12910
    },
    {
      "entropy": 5.552805805206299,
      "epoch": 1.1715348330914368,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000486919330810391,
      "loss": 5.1837,
      "mean_token_accuracy": 0.18431039452552794,
      "num_tokens": 26806801.0,
      "step": 12915
    },
    {
      "entropy": 5.510914707183838,
      "epoch": 1.171988388969521,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004869084570169139,
      "loss": 5.1009,
      "mean_token_accuracy": 0.19960006475448608,
      "num_tokens": 26816869.0,
      "step": 12920
    },
    {
      "entropy": 5.481137228012085,
      "epoch": 1.172441944847605,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048689757884111585,
      "loss": 5.0411,
      "mean_token_accuracy": 0.1998990297317505,
      "num_tokens": 26827003.0,
      "step": 12925
    },
    {
      "entropy": 5.433103561401367,
      "epoch": 1.1728955007256894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048688669628322185,
      "loss": 5.0472,
      "mean_token_accuracy": 0.19865425080060958,
      "num_tokens": 26837065.0,
      "step": 12930
    },
    {
      "entropy": 5.410325002670288,
      "epoch": 1.1733490566037736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000486875809343457,
      "loss": 4.9967,
      "mean_token_accuracy": 0.19845315665006638,
      "num_tokens": 26847487.0,
      "step": 12935
    },
    {
      "entropy": 5.475924348831176,
      "epoch": 1.1738026124818577,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048686491802204645,
      "loss": 5.0499,
      "mean_token_accuracy": 0.2035646066069603,
      "num_tokens": 26858407.0,
      "step": 12940
    },
    {
      "entropy": 5.54250168800354,
      "epoch": 1.174256168359942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004868540223192154,
      "loss": 5.1214,
      "mean_token_accuracy": 0.19381057024002074,
      "num_tokens": 26868469.0,
      "step": 12945
    },
    {
      "entropy": 5.530332469940186,
      "epoch": 1.1747097242380262,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004868431222351892,
      "loss": 5.1229,
      "mean_token_accuracy": 0.19985790103673934,
      "num_tokens": 26879580.0,
      "step": 12950
    },
    {
      "entropy": 5.5076759338378904,
      "epoch": 1.1751632801161103,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048683221777019313,
      "loss": 5.0943,
      "mean_token_accuracy": 0.2020096644759178,
      "num_tokens": 26889513.0,
      "step": 12955
    },
    {
      "entropy": 5.486361122131347,
      "epoch": 1.1756168359941945,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048682130892445293,
      "loss": 5.1776,
      "mean_token_accuracy": 0.19359630793333055,
      "num_tokens": 26900792.0,
      "step": 12960
    },
    {
      "entropy": 5.453381633758545,
      "epoch": 1.1760703918722786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000486810395698194,
      "loss": 5.0583,
      "mean_token_accuracy": 0.2014825612306595,
      "num_tokens": 26911298.0,
      "step": 12965
    },
    {
      "entropy": 5.533105802536011,
      "epoch": 1.1765239477503628,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048679947809164223,
      "loss": 5.1139,
      "mean_token_accuracy": 0.19168972223997116,
      "num_tokens": 26921545.0,
      "step": 12970
    },
    {
      "entropy": 5.528624534606934,
      "epoch": 1.176977503628447,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004867885561050232,
      "loss": 5.08,
      "mean_token_accuracy": 0.20872305780649186,
      "num_tokens": 26932140.0,
      "step": 12975
    },
    {
      "entropy": 5.496430492401123,
      "epoch": 1.1774310595065312,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004867776297385629,
      "loss": 5.1293,
      "mean_token_accuracy": 0.18912578672170638,
      "num_tokens": 26942634.0,
      "step": 12980
    },
    {
      "entropy": 5.506092357635498,
      "epoch": 1.1778846153846154,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048676669899248723,
      "loss": 5.0802,
      "mean_token_accuracy": 0.20136663019657136,
      "num_tokens": 26952758.0,
      "step": 12985
    },
    {
      "entropy": 5.442593050003052,
      "epoch": 1.1783381712626997,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004867557638670222,
      "loss": 5.0338,
      "mean_token_accuracy": 0.20214931666851044,
      "num_tokens": 26963439.0,
      "step": 12990
    },
    {
      "entropy": 5.461094951629638,
      "epoch": 1.1787917271407837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004867448243623941,
      "loss": 5.099,
      "mean_token_accuracy": 0.19264838695526124,
      "num_tokens": 26974331.0,
      "step": 12995
    },
    {
      "entropy": 5.447110414505005,
      "epoch": 1.179245283018868,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000486733880478829,
      "loss": 4.967,
      "mean_token_accuracy": 0.2060053676366806,
      "num_tokens": 26983416.0,
      "step": 13000
    },
    {
      "entropy": 5.51430835723877,
      "epoch": 1.179698838896952,
      "grad_norm": 1.0,
      "learning_rate": 0.00048672293221655333,
      "loss": 5.1047,
      "mean_token_accuracy": 0.19942551702260972,
      "num_tokens": 26994176.0,
      "step": 13005
    },
    {
      "entropy": 5.465344715118408,
      "epoch": 1.1801523947750363,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004867119795757935,
      "loss": 5.085,
      "mean_token_accuracy": 0.1964393898844719,
      "num_tokens": 27003295.0,
      "step": 13010
    },
    {
      "entropy": 5.476764297485351,
      "epoch": 1.1806059506531206,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000486701022556776,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19700005501508713,
      "num_tokens": 27014018.0,
      "step": 13015
    },
    {
      "entropy": 5.484636688232422,
      "epoch": 1.1810595065312046,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004866900611597273,
      "loss": 5.071,
      "mean_token_accuracy": 0.19935835003852845,
      "num_tokens": 27024706.0,
      "step": 13020
    },
    {
      "entropy": 5.532466125488281,
      "epoch": 1.1815130624092889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048667909538487427,
      "loss": 5.0869,
      "mean_token_accuracy": 0.20182140171527863,
      "num_tokens": 27035801.0,
      "step": 13025
    },
    {
      "entropy": 5.4090821743011475,
      "epoch": 1.181966618287373,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004866681252324436,
      "loss": 4.9705,
      "mean_token_accuracy": 0.20396371632814408,
      "num_tokens": 27046596.0,
      "step": 13030
    },
    {
      "entropy": 5.454617643356324,
      "epoch": 1.1824201741654572,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048665715070266214,
      "loss": 5.0772,
      "mean_token_accuracy": 0.1967050701379776,
      "num_tokens": 27057057.0,
      "step": 13035
    },
    {
      "entropy": 5.402617454528809,
      "epoch": 1.1828737300435415,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048664617179575685,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20518202632665633,
      "num_tokens": 27067947.0,
      "step": 13040
    },
    {
      "entropy": 5.530779886245727,
      "epoch": 1.1833272859216255,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004866351885119548,
      "loss": 5.0709,
      "mean_token_accuracy": 0.19731810688972473,
      "num_tokens": 27077503.0,
      "step": 13045
    },
    {
      "entropy": 5.535081672668457,
      "epoch": 1.1837808417997098,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004866242008514832,
      "loss": 5.1897,
      "mean_token_accuracy": 0.1931100070476532,
      "num_tokens": 27088298.0,
      "step": 13050
    },
    {
      "entropy": 5.475606679916382,
      "epoch": 1.1842343976777938,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048661320881456904,
      "loss": 5.0551,
      "mean_token_accuracy": 0.19839247316122055,
      "num_tokens": 27097841.0,
      "step": 13055
    },
    {
      "entropy": 5.406886243820191,
      "epoch": 1.184687953555878,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004866022124014399,
      "loss": 4.994,
      "mean_token_accuracy": 0.20290929526090623,
      "num_tokens": 27108263.0,
      "step": 13060
    },
    {
      "entropy": 5.4522699356079105,
      "epoch": 1.1851415094339623,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048659121161232313,
      "loss": 5.0589,
      "mean_token_accuracy": 0.20132856369018554,
      "num_tokens": 27119339.0,
      "step": 13065
    },
    {
      "entropy": 5.508952474594116,
      "epoch": 1.1855950653120464,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004865802064474461,
      "loss": 5.1084,
      "mean_token_accuracy": 0.19557663500308992,
      "num_tokens": 27130393.0,
      "step": 13070
    },
    {
      "entropy": 5.534052133560181,
      "epoch": 1.1860486211901307,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048656919690703653,
      "loss": 5.1063,
      "mean_token_accuracy": 0.19433179795742034,
      "num_tokens": 27141156.0,
      "step": 13075
    },
    {
      "entropy": 5.542227840423584,
      "epoch": 1.1865021770682147,
      "grad_norm": 1.015625,
      "learning_rate": 0.000486558182991322,
      "loss": 5.1431,
      "mean_token_accuracy": 0.19479037076234818,
      "num_tokens": 27151630.0,
      "step": 13080
    },
    {
      "entropy": 5.474994468688965,
      "epoch": 1.186955732946299,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004865471647005304,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20152200609445572,
      "num_tokens": 27160734.0,
      "step": 13085
    },
    {
      "entropy": 5.467565727233887,
      "epoch": 1.1874092888243832,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048653614203488947,
      "loss": 5.0981,
      "mean_token_accuracy": 0.19816022962331772,
      "num_tokens": 27171584.0,
      "step": 13090
    },
    {
      "entropy": 5.464590835571289,
      "epoch": 1.1878628447024673,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004865251149946273,
      "loss": 5.005,
      "mean_token_accuracy": 0.20658984631299973,
      "num_tokens": 27180817.0,
      "step": 13095
    },
    {
      "entropy": 5.542662477493286,
      "epoch": 1.1883164005805515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048651408357997173,
      "loss": 5.0887,
      "mean_token_accuracy": 0.1919817328453064,
      "num_tokens": 27190363.0,
      "step": 13100
    },
    {
      "entropy": 5.467042398452759,
      "epoch": 1.1887699564586356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048650304779115114,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20326141864061356,
      "num_tokens": 27202181.0,
      "step": 13105
    },
    {
      "entropy": 5.43560643196106,
      "epoch": 1.1892235123367199,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004864920076283935,
      "loss": 5.0895,
      "mean_token_accuracy": 0.19626662284135818,
      "num_tokens": 27212514.0,
      "step": 13110
    },
    {
      "entropy": 5.550775957107544,
      "epoch": 1.1896770682148041,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004864809630919273,
      "loss": 5.0584,
      "mean_token_accuracy": 0.20287249684333802,
      "num_tokens": 27223352.0,
      "step": 13115
    },
    {
      "entropy": 5.518157911300659,
      "epoch": 1.1901306240928882,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004864699141819809,
      "loss": 5.0064,
      "mean_token_accuracy": 0.2069821238517761,
      "num_tokens": 27232643.0,
      "step": 13120
    },
    {
      "entropy": 5.41427264213562,
      "epoch": 1.1905841799709724,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048645886089878276,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20254046469926834,
      "num_tokens": 27243793.0,
      "step": 13125
    },
    {
      "entropy": 5.4802384853363035,
      "epoch": 1.1910377358490567,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048644780324256153,
      "loss": 5.0796,
      "mean_token_accuracy": 0.1999229371547699,
      "num_tokens": 27254554.0,
      "step": 13130
    },
    {
      "entropy": 5.487999629974365,
      "epoch": 1.1914912917271407,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004864367412135458,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19469751119613649,
      "num_tokens": 27265842.0,
      "step": 13135
    },
    {
      "entropy": 5.496365976333618,
      "epoch": 1.191944847605225,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048642567481196446,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19370853155851364,
      "num_tokens": 27277617.0,
      "step": 13140
    },
    {
      "entropy": 5.44011173248291,
      "epoch": 1.192398403483309,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004864146040380462,
      "loss": 5.045,
      "mean_token_accuracy": 0.19978711754083633,
      "num_tokens": 27287450.0,
      "step": 13145
    },
    {
      "entropy": 5.47112717628479,
      "epoch": 1.1928519593613933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004864035288920202,
      "loss": 5.0671,
      "mean_token_accuracy": 0.20327106565237046,
      "num_tokens": 27297010.0,
      "step": 13150
    },
    {
      "entropy": 5.463698434829712,
      "epoch": 1.1933055152394776,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048639244937411516,
      "loss": 5.1043,
      "mean_token_accuracy": 0.19503520727157592,
      "num_tokens": 27306874.0,
      "step": 13155
    },
    {
      "entropy": 5.5497582912445065,
      "epoch": 1.1937590711175616,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048638136548456045,
      "loss": 5.1019,
      "mean_token_accuracy": 0.19800865054130554,
      "num_tokens": 27318762.0,
      "step": 13160
    },
    {
      "entropy": 5.442694759368896,
      "epoch": 1.194212626995646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048637027722358536,
      "loss": 4.981,
      "mean_token_accuracy": 0.21250585168600084,
      "num_tokens": 27327231.0,
      "step": 13165
    },
    {
      "entropy": 5.385290479660034,
      "epoch": 1.1946661828737302,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048635918459141893,
      "loss": 5.0192,
      "mean_token_accuracy": 0.19913043081760406,
      "num_tokens": 27338517.0,
      "step": 13170
    },
    {
      "entropy": 5.528426170349121,
      "epoch": 1.1951197387518142,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048634808758829073,
      "loss": 5.0883,
      "mean_token_accuracy": 0.19603673368692398,
      "num_tokens": 27348861.0,
      "step": 13175
    },
    {
      "entropy": 5.511196565628052,
      "epoch": 1.1955732946298985,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048633698621443024,
      "loss": 5.0483,
      "mean_token_accuracy": 0.20928379148244858,
      "num_tokens": 27358610.0,
      "step": 13180
    },
    {
      "entropy": 5.475197076797485,
      "epoch": 1.1960268505079825,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000486325880470067,
      "loss": 5.1288,
      "mean_token_accuracy": 0.1985629081726074,
      "num_tokens": 27369707.0,
      "step": 13185
    },
    {
      "entropy": 5.4696228981018065,
      "epoch": 1.1964804063860668,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004863147703554307,
      "loss": 5.0816,
      "mean_token_accuracy": 0.19705890119075775,
      "num_tokens": 27381030.0,
      "step": 13190
    },
    {
      "entropy": 5.519057321548462,
      "epoch": 1.196933962264151,
      "grad_norm": 1.0,
      "learning_rate": 0.00048630365587075115,
      "loss": 5.0961,
      "mean_token_accuracy": 0.19812251776456832,
      "num_tokens": 27390866.0,
      "step": 13195
    },
    {
      "entropy": 5.493912649154663,
      "epoch": 1.197387518142235,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048629253701625816,
      "loss": 5.0357,
      "mean_token_accuracy": 0.20332785397768022,
      "num_tokens": 27400724.0,
      "step": 13200
    },
    {
      "entropy": 5.473924732208252,
      "epoch": 1.1978410740203194,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004862814137921816,
      "loss": 5.0576,
      "mean_token_accuracy": 0.19959327727556228,
      "num_tokens": 27411212.0,
      "step": 13205
    },
    {
      "entropy": 5.467393922805786,
      "epoch": 1.1982946298984034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048627028619875157,
      "loss": 5.0834,
      "mean_token_accuracy": 0.1994468241930008,
      "num_tokens": 27421279.0,
      "step": 13210
    },
    {
      "entropy": 5.483545017242432,
      "epoch": 1.1987481857764877,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004862591542361982,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19779787510633468,
      "num_tokens": 27432520.0,
      "step": 13215
    },
    {
      "entropy": 5.4180910110473635,
      "epoch": 1.199201741654572,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004862480179047516,
      "loss": 4.989,
      "mean_token_accuracy": 0.20875904560089112,
      "num_tokens": 27442246.0,
      "step": 13220
    },
    {
      "entropy": 5.482895421981811,
      "epoch": 1.199655297532656,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004862368772046423,
      "loss": 5.0607,
      "mean_token_accuracy": 0.19732182919979097,
      "num_tokens": 27452134.0,
      "step": 13225
    },
    {
      "entropy": 5.496697950363159,
      "epoch": 1.2001088534107403,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048622573213610057,
      "loss": 5.1461,
      "mean_token_accuracy": 0.188792884349823,
      "num_tokens": 27462253.0,
      "step": 13230
    },
    {
      "entropy": 5.511988878250122,
      "epoch": 1.2005624092888243,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048621458269935676,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20037435442209245,
      "num_tokens": 27473131.0,
      "step": 13235
    },
    {
      "entropy": 5.489395809173584,
      "epoch": 1.2010159651669086,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048620342889464153,
      "loss": 5.0756,
      "mean_token_accuracy": 0.2012531042098999,
      "num_tokens": 27483683.0,
      "step": 13240
    },
    {
      "entropy": 5.384080266952514,
      "epoch": 1.2014695210449928,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048619227072218565,
      "loss": 5.052,
      "mean_token_accuracy": 0.1951747328042984,
      "num_tokens": 27493679.0,
      "step": 13245
    },
    {
      "entropy": 5.4413477897644045,
      "epoch": 1.2019230769230769,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004861811081822198,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20437403321266173,
      "num_tokens": 27503897.0,
      "step": 13250
    },
    {
      "entropy": 5.532030296325684,
      "epoch": 1.2023766328011611,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004861699412749748,
      "loss": 5.1421,
      "mean_token_accuracy": 0.192557655274868,
      "num_tokens": 27514560.0,
      "step": 13255
    },
    {
      "entropy": 5.4998753547668455,
      "epoch": 1.2028301886792452,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004861587700006816,
      "loss": 5.0427,
      "mean_token_accuracy": 0.1978370800614357,
      "num_tokens": 27525594.0,
      "step": 13260
    },
    {
      "entropy": 5.426636695861816,
      "epoch": 1.2032837445573294,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048614759435957114,
      "loss": 4.9909,
      "mean_token_accuracy": 0.2014489009976387,
      "num_tokens": 27535676.0,
      "step": 13265
    },
    {
      "entropy": 5.413556241989136,
      "epoch": 1.2037373004354137,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004861364143518747,
      "loss": 5.0281,
      "mean_token_accuracy": 0.19157828241586686,
      "num_tokens": 27545520.0,
      "step": 13270
    },
    {
      "entropy": 5.573331117630005,
      "epoch": 1.2041908563134978,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004861252299778234,
      "loss": 5.0973,
      "mean_token_accuracy": 0.18991344869136811,
      "num_tokens": 27556866.0,
      "step": 13275
    },
    {
      "entropy": 5.47489185333252,
      "epoch": 1.204644412191582,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048611404123764855,
      "loss": 5.0296,
      "mean_token_accuracy": 0.19900879561901091,
      "num_tokens": 27566951.0,
      "step": 13280
    },
    {
      "entropy": 5.470123195648194,
      "epoch": 1.205097968069666,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048610284813158154,
      "loss": 5.1365,
      "mean_token_accuracy": 0.1957337498664856,
      "num_tokens": 27576796.0,
      "step": 13285
    },
    {
      "entropy": 5.402993488311767,
      "epoch": 1.2055515239477503,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004860916506598538,
      "loss": 5.0034,
      "mean_token_accuracy": 0.1991758719086647,
      "num_tokens": 27586902.0,
      "step": 13290
    },
    {
      "entropy": 5.585123109817505,
      "epoch": 1.2060050798258346,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048608044882269695,
      "loss": 5.1957,
      "mean_token_accuracy": 0.18857692927122116,
      "num_tokens": 27599232.0,
      "step": 13295
    },
    {
      "entropy": 5.482767248153687,
      "epoch": 1.2064586357039186,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004860692426203427,
      "loss": 4.9931,
      "mean_token_accuracy": 0.1983819544315338,
      "num_tokens": 27609022.0,
      "step": 13300
    },
    {
      "entropy": 5.515181207656861,
      "epoch": 1.206912191582003,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004860580320530226,
      "loss": 5.1371,
      "mean_token_accuracy": 0.19917869865894317,
      "num_tokens": 27619829.0,
      "step": 13305
    },
    {
      "entropy": 5.515312433242798,
      "epoch": 1.2073657474600872,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004860468171209687,
      "loss": 5.1298,
      "mean_token_accuracy": 0.20136113315820695,
      "num_tokens": 27629337.0,
      "step": 13310
    },
    {
      "entropy": 5.546958637237549,
      "epoch": 1.2078193033381712,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004860355978244128,
      "loss": 5.1584,
      "mean_token_accuracy": 0.19766863137483598,
      "num_tokens": 27639377.0,
      "step": 13315
    },
    {
      "entropy": 5.500257062911987,
      "epoch": 1.2082728592162555,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000486024374163587,
      "loss": 5.0404,
      "mean_token_accuracy": 0.19715555906295776,
      "num_tokens": 27649989.0,
      "step": 13320
    },
    {
      "entropy": 5.49214277267456,
      "epoch": 1.2087264150943395,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048601314613872334,
      "loss": 5.1191,
      "mean_token_accuracy": 0.19113418459892273,
      "num_tokens": 27659967.0,
      "step": 13325
    },
    {
      "entropy": 5.509144020080567,
      "epoch": 1.2091799709724238,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000486001913750054,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19962707608938218,
      "num_tokens": 27670099.0,
      "step": 13330
    },
    {
      "entropy": 5.463675832748413,
      "epoch": 1.209633526850508,
      "grad_norm": 1.0,
      "learning_rate": 0.0004859906769978113,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1901517853140831,
      "num_tokens": 27681497.0,
      "step": 13335
    },
    {
      "entropy": 5.572259140014649,
      "epoch": 1.2100870827285921,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048597943588222776,
      "loss": 5.2052,
      "mean_token_accuracy": 0.19154019355773927,
      "num_tokens": 27691828.0,
      "step": 13340
    },
    {
      "entropy": 5.538029193878174,
      "epoch": 1.2105406386066764,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048596819040353564,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20628093034029008,
      "num_tokens": 27701426.0,
      "step": 13345
    },
    {
      "entropy": 5.493690490722656,
      "epoch": 1.2109941944847606,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859569405619676,
      "loss": 5.0153,
      "mean_token_accuracy": 0.20149533450603485,
      "num_tokens": 27711112.0,
      "step": 13350
    },
    {
      "entropy": 5.52337875366211,
      "epoch": 1.2114477503628447,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004859456863577562,
      "loss": 5.1357,
      "mean_token_accuracy": 0.19508373439311982,
      "num_tokens": 27723160.0,
      "step": 13355
    },
    {
      "entropy": 5.522914218902588,
      "epoch": 1.211901306240929,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048593442779113437,
      "loss": 5.1034,
      "mean_token_accuracy": 0.19623059332370757,
      "num_tokens": 27733143.0,
      "step": 13360
    },
    {
      "entropy": 5.5026891231536865,
      "epoch": 1.212354862119013,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004859231648623348,
      "loss": 5.0884,
      "mean_token_accuracy": 0.19636708945035936,
      "num_tokens": 27743956.0,
      "step": 13365
    },
    {
      "entropy": 5.522414779663086,
      "epoch": 1.2128084179970973,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048591189757159035,
      "loss": 5.0851,
      "mean_token_accuracy": 0.2020998165011406,
      "num_tokens": 27754192.0,
      "step": 13370
    },
    {
      "entropy": 5.487301158905029,
      "epoch": 1.2132619738751815,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004859006259191342,
      "loss": 5.1205,
      "mean_token_accuracy": 0.18899906873703004,
      "num_tokens": 27763939.0,
      "step": 13375
    },
    {
      "entropy": 5.485107564926148,
      "epoch": 1.2137155297532656,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004858893499051993,
      "loss": 5.0894,
      "mean_token_accuracy": 0.20167715549468995,
      "num_tokens": 27774743.0,
      "step": 13380
    },
    {
      "entropy": 5.508217763900757,
      "epoch": 1.2141690856313498,
      "grad_norm": 1.125,
      "learning_rate": 0.0004858780695300189,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19787210524082183,
      "num_tokens": 27785344.0,
      "step": 13385
    },
    {
      "entropy": 5.520311784744263,
      "epoch": 1.2146226415094339,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004858667847938263,
      "loss": 5.1449,
      "mean_token_accuracy": 0.19476209133863448,
      "num_tokens": 27797101.0,
      "step": 13390
    },
    {
      "entropy": 5.452220058441162,
      "epoch": 1.2150761973875182,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048585549569685486,
      "loss": 5.0338,
      "mean_token_accuracy": 0.2015383094549179,
      "num_tokens": 27807581.0,
      "step": 13395
    },
    {
      "entropy": 5.462075805664062,
      "epoch": 1.2155297532656024,
      "grad_norm": 1.0,
      "learning_rate": 0.00048584420223933805,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19964679926633835,
      "num_tokens": 27818170.0,
      "step": 13400
    },
    {
      "entropy": 5.432729625701905,
      "epoch": 1.2159833091436865,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004858329044215094,
      "loss": 5.029,
      "mean_token_accuracy": 0.20274798572063446,
      "num_tokens": 27828153.0,
      "step": 13405
    },
    {
      "entropy": 5.487482118606567,
      "epoch": 1.2164368650217707,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048582160224360253,
      "loss": 4.9698,
      "mean_token_accuracy": 0.206896835565567,
      "num_tokens": 27837681.0,
      "step": 13410
    },
    {
      "entropy": 5.494462537765503,
      "epoch": 1.2168904208998548,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004858102957058513,
      "loss": 5.108,
      "mean_token_accuracy": 0.19647631645202637,
      "num_tokens": 27847447.0,
      "step": 13415
    },
    {
      "entropy": 5.444395112991333,
      "epoch": 1.217343976777939,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048579898480848933,
      "loss": 4.9716,
      "mean_token_accuracy": 0.2026344954967499,
      "num_tokens": 27858586.0,
      "step": 13420
    },
    {
      "entropy": 5.495455551147461,
      "epoch": 1.2177975326560233,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004857876695517507,
      "loss": 5.0958,
      "mean_token_accuracy": 0.19661417752504348,
      "num_tokens": 27868832.0,
      "step": 13425
    },
    {
      "entropy": 5.492994785308838,
      "epoch": 1.2182510885341074,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048577634993586923,
      "loss": 5.0952,
      "mean_token_accuracy": 0.19540033489465714,
      "num_tokens": 27879339.0,
      "step": 13430
    },
    {
      "entropy": 5.515705108642578,
      "epoch": 1.2187046444121916,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048576502596107924,
      "loss": 5.0881,
      "mean_token_accuracy": 0.19840986430644988,
      "num_tokens": 27889964.0,
      "step": 13435
    },
    {
      "entropy": 5.5269997119903564,
      "epoch": 1.2191582002902757,
      "grad_norm": 1.0,
      "learning_rate": 0.0004857536976276148,
      "loss": 5.1418,
      "mean_token_accuracy": 0.19223980158567427,
      "num_tokens": 27901004.0,
      "step": 13440
    },
    {
      "entropy": 5.484214353561401,
      "epoch": 1.21961175616836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004857423649357102,
      "loss": 5.0613,
      "mean_token_accuracy": 0.1969600275158882,
      "num_tokens": 27910788.0,
      "step": 13445
    },
    {
      "entropy": 5.544168424606323,
      "epoch": 1.2200653120464442,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048573102788559974,
      "loss": 5.1035,
      "mean_token_accuracy": 0.2005355179309845,
      "num_tokens": 27921360.0,
      "step": 13450
    },
    {
      "entropy": 5.482368850708008,
      "epoch": 1.2205188679245282,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004857196864775179,
      "loss": 5.0467,
      "mean_token_accuracy": 0.20181792825460435,
      "num_tokens": 27931889.0,
      "step": 13455
    },
    {
      "entropy": 5.456076622009277,
      "epoch": 1.2209724238026125,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048570834071169926,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20518947392702103,
      "num_tokens": 27942049.0,
      "step": 13460
    },
    {
      "entropy": 5.538109827041626,
      "epoch": 1.2214259796806965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048569699058837856,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1968417227268219,
      "num_tokens": 27952416.0,
      "step": 13465
    },
    {
      "entropy": 5.467518997192383,
      "epoch": 1.2218795355587808,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004856856361077903,
      "loss": 5.0307,
      "mean_token_accuracy": 0.20182372778654098,
      "num_tokens": 27962009.0,
      "step": 13470
    },
    {
      "entropy": 5.550679159164429,
      "epoch": 1.222333091436865,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004856742772701694,
      "loss": 5.0959,
      "mean_token_accuracy": 0.19606231898069382,
      "num_tokens": 27972175.0,
      "step": 13475
    },
    {
      "entropy": 5.5194611072540285,
      "epoch": 1.2227866473149491,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048566291407575073,
      "loss": 5.1141,
      "mean_token_accuracy": 0.2015317127108574,
      "num_tokens": 27982899.0,
      "step": 13480
    },
    {
      "entropy": 5.513904333114624,
      "epoch": 1.2232402031930334,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048565154652476937,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20250820070505143,
      "num_tokens": 27993231.0,
      "step": 13485
    },
    {
      "entropy": 5.432183265686035,
      "epoch": 1.2236937590711174,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004856401746174603,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20159731060266495,
      "num_tokens": 28003623.0,
      "step": 13490
    },
    {
      "entropy": 5.488606214523315,
      "epoch": 1.2241473149492017,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048562879835405874,
      "loss": 5.0291,
      "mean_token_accuracy": 0.20178948640823363,
      "num_tokens": 28013878.0,
      "step": 13495
    },
    {
      "entropy": 5.447732400894165,
      "epoch": 1.224600870827286,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004856174177348,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19857164174318315,
      "num_tokens": 28024224.0,
      "step": 13500
    },
    {
      "entropy": 5.470084142684937,
      "epoch": 1.22505442670537,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004856060327599194,
      "loss": 5.0161,
      "mean_token_accuracy": 0.1966568797826767,
      "num_tokens": 28034819.0,
      "step": 13505
    },
    {
      "entropy": 5.463940000534057,
      "epoch": 1.2255079825834543,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048559464342965233,
      "loss": 5.0775,
      "mean_token_accuracy": 0.19706240743398667,
      "num_tokens": 28044240.0,
      "step": 13510
    },
    {
      "entropy": 5.565493249893189,
      "epoch": 1.2259615384615385,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004855832497442344,
      "loss": 5.2113,
      "mean_token_accuracy": 0.18766228705644608,
      "num_tokens": 28055046.0,
      "step": 13515
    },
    {
      "entropy": 5.462039518356323,
      "epoch": 1.2264150943396226,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004855718517039012,
      "loss": 5.0052,
      "mean_token_accuracy": 0.205356565117836,
      "num_tokens": 28065272.0,
      "step": 13520
    },
    {
      "entropy": 5.517233610153198,
      "epoch": 1.2268686502177069,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048556044930888844,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2079099714756012,
      "num_tokens": 28076255.0,
      "step": 13525
    },
    {
      "entropy": 5.515420389175415,
      "epoch": 1.2273222060957911,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004855490425594319,
      "loss": 5.1358,
      "mean_token_accuracy": 0.19787602573633195,
      "num_tokens": 28086258.0,
      "step": 13530
    },
    {
      "entropy": 5.494659328460694,
      "epoch": 1.2277757619738752,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004855376314557675,
      "loss": 5.0512,
      "mean_token_accuracy": 0.20258419066667557,
      "num_tokens": 28096497.0,
      "step": 13535
    },
    {
      "entropy": 5.510882329940796,
      "epoch": 1.2282293178519594,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004855262159981313,
      "loss": 5.0602,
      "mean_token_accuracy": 0.2041010618209839,
      "num_tokens": 28106282.0,
      "step": 13540
    },
    {
      "entropy": 5.450311946868896,
      "epoch": 1.2286828737300435,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048551479618675925,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19611413329839705,
      "num_tokens": 28115771.0,
      "step": 13545
    },
    {
      "entropy": 5.526232290267944,
      "epoch": 1.2291364296081277,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048550337202188756,
      "loss": 5.0407,
      "mean_token_accuracy": 0.2037350058555603,
      "num_tokens": 28125477.0,
      "step": 13550
    },
    {
      "entropy": 5.55634880065918,
      "epoch": 1.229589985486212,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004854919435037525,
      "loss": 5.0788,
      "mean_token_accuracy": 0.19526431262493132,
      "num_tokens": 28135800.0,
      "step": 13555
    },
    {
      "entropy": 5.429639768600464,
      "epoch": 1.230043541364296,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004854805106325904,
      "loss": 4.9507,
      "mean_token_accuracy": 0.2002708464860916,
      "num_tokens": 28146779.0,
      "step": 13560
    },
    {
      "entropy": 5.4886777877807615,
      "epoch": 1.2304970972423803,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004854690734086378,
      "loss": 5.1104,
      "mean_token_accuracy": 0.19459391683340072,
      "num_tokens": 28156819.0,
      "step": 13565
    },
    {
      "entropy": 5.470524597167969,
      "epoch": 1.2309506531204644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004854576318321311,
      "loss": 5.0871,
      "mean_token_accuracy": 0.1952335223555565,
      "num_tokens": 28168553.0,
      "step": 13570
    },
    {
      "entropy": 5.530834293365478,
      "epoch": 1.2314042089985486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048544618590330685,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19884106814861296,
      "num_tokens": 28178064.0,
      "step": 13575
    },
    {
      "entropy": 5.565642642974853,
      "epoch": 1.231857764876633,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048543473562240194,
      "loss": 5.1965,
      "mean_token_accuracy": 0.19776492416858674,
      "num_tokens": 28188580.0,
      "step": 13580
    },
    {
      "entropy": 5.524669981002807,
      "epoch": 1.232311320754717,
      "grad_norm": 0.984375,
      "learning_rate": 0.000485423280989653,
      "loss": 5.1629,
      "mean_token_accuracy": 0.19327968508005142,
      "num_tokens": 28199955.0,
      "step": 13585
    },
    {
      "entropy": 5.5645153522491455,
      "epoch": 1.2327648766328012,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000485411822005297,
      "loss": 5.1159,
      "mean_token_accuracy": 0.1941158130764961,
      "num_tokens": 28210506.0,
      "step": 13590
    },
    {
      "entropy": 5.493167734146118,
      "epoch": 1.2332184325108853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048540035866957097,
      "loss": 5.0342,
      "mean_token_accuracy": 0.2134607493877411,
      "num_tokens": 28220486.0,
      "step": 13595
    },
    {
      "entropy": 5.521177005767822,
      "epoch": 1.2336719883889695,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004853888909827118,
      "loss": 5.1623,
      "mean_token_accuracy": 0.18835390061140062,
      "num_tokens": 28232668.0,
      "step": 13600
    },
    {
      "entropy": 5.507457447052002,
      "epoch": 1.2341255442670538,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004853774189449568,
      "loss": 5.1072,
      "mean_token_accuracy": 0.1947883576154709,
      "num_tokens": 28243733.0,
      "step": 13605
    },
    {
      "entropy": 5.572214269638062,
      "epoch": 1.2345791001451378,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048536594255654315,
      "loss": 5.1413,
      "mean_token_accuracy": 0.19594713300466537,
      "num_tokens": 28254953.0,
      "step": 13610
    },
    {
      "entropy": 5.530012893676758,
      "epoch": 1.235032656023222,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004853544618177083,
      "loss": 5.0832,
      "mean_token_accuracy": 0.20127384960651398,
      "num_tokens": 28264959.0,
      "step": 13615
    },
    {
      "entropy": 5.54983115196228,
      "epoch": 1.2354862119013061,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048534297672868945,
      "loss": 5.0693,
      "mean_token_accuracy": 0.19841146767139434,
      "num_tokens": 28275509.0,
      "step": 13620
    },
    {
      "entropy": 5.500857210159301,
      "epoch": 1.2359397677793904,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048533148728972434,
      "loss": 5.0277,
      "mean_token_accuracy": 0.2020431414246559,
      "num_tokens": 28285983.0,
      "step": 13625
    },
    {
      "entropy": 5.510316038131714,
      "epoch": 1.2363933236574747,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048531999350105037,
      "loss": 5.1071,
      "mean_token_accuracy": 0.1961878165602684,
      "num_tokens": 28296178.0,
      "step": 13630
    },
    {
      "entropy": 5.491313409805298,
      "epoch": 1.2368468795355587,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048530849536290537,
      "loss": 5.0739,
      "mean_token_accuracy": 0.19448001086711883,
      "num_tokens": 28305741.0,
      "step": 13635
    },
    {
      "entropy": 5.42606258392334,
      "epoch": 1.237300435413643,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004852969928755271,
      "loss": 5.0352,
      "mean_token_accuracy": 0.2011515825986862,
      "num_tokens": 28315607.0,
      "step": 13640
    },
    {
      "entropy": 5.420935916900635,
      "epoch": 1.237753991291727,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048528548603915344,
      "loss": 5.0552,
      "mean_token_accuracy": 0.20115619599819184,
      "num_tokens": 28327349.0,
      "step": 13645
    },
    {
      "entropy": 5.52898759841919,
      "epoch": 1.2382075471698113,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048527397485402236,
      "loss": 5.0485,
      "mean_token_accuracy": 0.19680120646953583,
      "num_tokens": 28339949.0,
      "step": 13650
    },
    {
      "entropy": 5.542382383346558,
      "epoch": 1.2386611030478956,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048526245932037186,
      "loss": 5.0745,
      "mean_token_accuracy": 0.20272745341062545,
      "num_tokens": 28351340.0,
      "step": 13655
    },
    {
      "entropy": 5.385624742507934,
      "epoch": 1.2391146589259796,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048525093943844013,
      "loss": 4.9409,
      "mean_token_accuracy": 0.20911214351654053,
      "num_tokens": 28360929.0,
      "step": 13660
    },
    {
      "entropy": 5.426797580718994,
      "epoch": 1.2395682148040639,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004852394152084654,
      "loss": 5.0391,
      "mean_token_accuracy": 0.2021891802549362,
      "num_tokens": 28370477.0,
      "step": 13665
    },
    {
      "entropy": 5.446661186218262,
      "epoch": 1.240021770682148,
      "grad_norm": 0.984375,
      "learning_rate": 0.000485227886630686,
      "loss": 4.966,
      "mean_token_accuracy": 0.21486925035715104,
      "num_tokens": 28380876.0,
      "step": 13670
    },
    {
      "entropy": 5.394935607910156,
      "epoch": 1.2404753265602322,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004852163537053403,
      "loss": 4.9434,
      "mean_token_accuracy": 0.20210244953632356,
      "num_tokens": 28390685.0,
      "step": 13675
    },
    {
      "entropy": 5.438671875,
      "epoch": 1.2409288824383164,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004852048164326669,
      "loss": 5.0539,
      "mean_token_accuracy": 0.19393928349018097,
      "num_tokens": 28400236.0,
      "step": 13680
    },
    {
      "entropy": 5.503456926345825,
      "epoch": 1.2413824383164005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004851932748129043,
      "loss": 5.0938,
      "mean_token_accuracy": 0.20332229733467103,
      "num_tokens": 28410353.0,
      "step": 13685
    },
    {
      "entropy": 5.507759523391724,
      "epoch": 1.2418359941944848,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004851817288462912,
      "loss": 5.0384,
      "mean_token_accuracy": 0.1971975803375244,
      "num_tokens": 28420241.0,
      "step": 13690
    },
    {
      "entropy": 5.471080160140991,
      "epoch": 1.242289550072569,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004851701785330663,
      "loss": 5.0943,
      "mean_token_accuracy": 0.19966191947460174,
      "num_tokens": 28430610.0,
      "step": 13695
    },
    {
      "entropy": 5.473397445678711,
      "epoch": 1.242743105950653,
      "grad_norm": 1.0,
      "learning_rate": 0.0004851586238734687,
      "loss": 5.0184,
      "mean_token_accuracy": 0.19985776245594025,
      "num_tokens": 28440026.0,
      "step": 13700
    },
    {
      "entropy": 5.48423581123352,
      "epoch": 1.2431966618287373,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004851470648677371,
      "loss": 5.0913,
      "mean_token_accuracy": 0.1909073993563652,
      "num_tokens": 28450707.0,
      "step": 13705
    },
    {
      "entropy": 5.484880018234253,
      "epoch": 1.2436502177068216,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004851355015161107,
      "loss": 5.04,
      "mean_token_accuracy": 0.19988873600959778,
      "num_tokens": 28462378.0,
      "step": 13710
    },
    {
      "entropy": 5.562987518310547,
      "epoch": 1.2441037735849056,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004851239338188286,
      "loss": 5.1173,
      "mean_token_accuracy": 0.19390659928321838,
      "num_tokens": 28473084.0,
      "step": 13715
    },
    {
      "entropy": 5.484537076950073,
      "epoch": 1.24455732946299,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048511236177612994,
      "loss": 5.041,
      "mean_token_accuracy": 0.1961607202887535,
      "num_tokens": 28482767.0,
      "step": 13720
    },
    {
      "entropy": 5.44678144454956,
      "epoch": 1.245010885341074,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048510078538825413,
      "loss": 5.0312,
      "mean_token_accuracy": 0.20292534232139586,
      "num_tokens": 28493274.0,
      "step": 13725
    },
    {
      "entropy": 5.520133829116821,
      "epoch": 1.2454644412191582,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048508920465544056,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20110096782445908,
      "num_tokens": 28503823.0,
      "step": 13730
    },
    {
      "entropy": 5.439535331726074,
      "epoch": 1.2459179970972425,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004850776195779286,
      "loss": 5.1174,
      "mean_token_accuracy": 0.20129793733358384,
      "num_tokens": 28514209.0,
      "step": 13735
    },
    {
      "entropy": 5.532947778701782,
      "epoch": 1.2463715529753265,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000485066030155958,
      "loss": 5.0946,
      "mean_token_accuracy": 0.19237526655197143,
      "num_tokens": 28524665.0,
      "step": 13740
    },
    {
      "entropy": 5.485327291488647,
      "epoch": 1.2468251088534108,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004850544363897684,
      "loss": 4.9957,
      "mean_token_accuracy": 0.20430061221122742,
      "num_tokens": 28533639.0,
      "step": 13745
    },
    {
      "entropy": 5.405498456954956,
      "epoch": 1.2472786647314948,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048504283827959947,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20359645783901215,
      "num_tokens": 28544469.0,
      "step": 13750
    },
    {
      "entropy": 5.532805395126343,
      "epoch": 1.2477322206095791,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004850312358256912,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20535819679498674,
      "num_tokens": 28554637.0,
      "step": 13755
    },
    {
      "entropy": 5.463185548782349,
      "epoch": 1.2481857764876634,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004850196290282834,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20495663583278656,
      "num_tokens": 28564752.0,
      "step": 13760
    },
    {
      "entropy": 5.503740978240967,
      "epoch": 1.2486393323657474,
      "grad_norm": 1.0,
      "learning_rate": 0.00048500801788761613,
      "loss": 5.1586,
      "mean_token_accuracy": 0.19918418526649476,
      "num_tokens": 28575007.0,
      "step": 13765
    },
    {
      "entropy": 5.5523582935333256,
      "epoch": 1.2490928882438317,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004849964024039296,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19649922996759414,
      "num_tokens": 28584907.0,
      "step": 13770
    },
    {
      "entropy": 5.572278308868408,
      "epoch": 1.2495464441219157,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048498478257746387,
      "loss": 5.1286,
      "mean_token_accuracy": 0.20043385326862334,
      "num_tokens": 28594244.0,
      "step": 13775
    },
    {
      "entropy": 5.506260251998901,
      "epoch": 1.25,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048497315840845934,
      "loss": 5.0921,
      "mean_token_accuracy": 0.1991775080561638,
      "num_tokens": 28603816.0,
      "step": 13780
    },
    {
      "entropy": 5.560829496383667,
      "epoch": 1.2504535558780843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004849615298971565,
      "loss": 5.1309,
      "mean_token_accuracy": 0.19353427439928056,
      "num_tokens": 28613067.0,
      "step": 13785
    },
    {
      "entropy": 5.500925207138062,
      "epoch": 1.2509071117561683,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048494989704379556,
      "loss": 5.0649,
      "mean_token_accuracy": 0.20042569041252137,
      "num_tokens": 28623752.0,
      "step": 13790
    },
    {
      "entropy": 5.450966548919678,
      "epoch": 1.2513606676342526,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004849382598486173,
      "loss": 5.0284,
      "mean_token_accuracy": 0.2054324969649315,
      "num_tokens": 28633190.0,
      "step": 13795
    },
    {
      "entropy": 5.589546823501587,
      "epoch": 1.2518142235123366,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048492661831186243,
      "loss": 5.1193,
      "mean_token_accuracy": 0.19111987203359604,
      "num_tokens": 28644567.0,
      "step": 13800
    },
    {
      "entropy": 5.544061851501465,
      "epoch": 1.2522677793904209,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004849149724337715,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19287900030612945,
      "num_tokens": 28654785.0,
      "step": 13805
    },
    {
      "entropy": 5.419747304916382,
      "epoch": 1.2527213352685052,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048490332221458543,
      "loss": 4.9955,
      "mean_token_accuracy": 0.19568415731191635,
      "num_tokens": 28665286.0,
      "step": 13810
    },
    {
      "entropy": 5.51972770690918,
      "epoch": 1.2531748911465892,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004848916676545452,
      "loss": 5.1157,
      "mean_token_accuracy": 0.19823004603385924,
      "num_tokens": 28675830.0,
      "step": 13815
    },
    {
      "entropy": 5.568089532852173,
      "epoch": 1.2536284470246735,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048488000875389174,
      "loss": 5.1713,
      "mean_token_accuracy": 0.18988842964172364,
      "num_tokens": 28686640.0,
      "step": 13820
    },
    {
      "entropy": 5.505718946456909,
      "epoch": 1.2540820029027575,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004848683455128663,
      "loss": 5.0695,
      "mean_token_accuracy": 0.20179916322231292,
      "num_tokens": 28697677.0,
      "step": 13825
    },
    {
      "entropy": 5.494057321548462,
      "epoch": 1.2545355587808418,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048485667793170994,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19772837162017823,
      "num_tokens": 28708831.0,
      "step": 13830
    },
    {
      "entropy": 5.5372607707977295,
      "epoch": 1.254989114658926,
      "grad_norm": 1.0,
      "learning_rate": 0.000484845006010664,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19325800836086274,
      "num_tokens": 28718959.0,
      "step": 13835
    },
    {
      "entropy": 5.494212627410889,
      "epoch": 1.25544267053701,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004848333297499699,
      "loss": 5.0586,
      "mean_token_accuracy": 0.20306683331727982,
      "num_tokens": 28728719.0,
      "step": 13840
    },
    {
      "entropy": 5.565085124969483,
      "epoch": 1.2558962264150944,
      "grad_norm": 1.0,
      "learning_rate": 0.00048482164914986903,
      "loss": 5.2087,
      "mean_token_accuracy": 0.18758925199508666,
      "num_tokens": 28740032.0,
      "step": 13845
    },
    {
      "entropy": 5.449249124526977,
      "epoch": 1.2563497822931784,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048480996421060307,
      "loss": 5.0358,
      "mean_token_accuracy": 0.2051217496395111,
      "num_tokens": 28749374.0,
      "step": 13850
    },
    {
      "entropy": 5.466462850570679,
      "epoch": 1.2568033381712627,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048479827493241347,
      "loss": 5.0566,
      "mean_token_accuracy": 0.19801553338766098,
      "num_tokens": 28760038.0,
      "step": 13855
    },
    {
      "entropy": 5.49345588684082,
      "epoch": 1.257256894049347,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048478658131554216,
      "loss": 5.0803,
      "mean_token_accuracy": 0.19837768524885177,
      "num_tokens": 28770247.0,
      "step": 13860
    },
    {
      "entropy": 5.488525819778443,
      "epoch": 1.2577104499274312,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004847748833602308,
      "loss": 5.0425,
      "mean_token_accuracy": 0.20438348203897477,
      "num_tokens": 28780251.0,
      "step": 13865
    },
    {
      "entropy": 5.443554973602295,
      "epoch": 1.2581640058055152,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004847631810667215,
      "loss": 5.0331,
      "mean_token_accuracy": 0.2038698151707649,
      "num_tokens": 28790829.0,
      "step": 13870
    },
    {
      "entropy": 5.4590709686279295,
      "epoch": 1.2586175616835993,
      "grad_norm": 1.0,
      "learning_rate": 0.0004847514744352561,
      "loss": 4.9865,
      "mean_token_accuracy": 0.2023104190826416,
      "num_tokens": 28801036.0,
      "step": 13875
    },
    {
      "entropy": 5.403080654144287,
      "epoch": 1.2590711175616836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004847397634660768,
      "loss": 4.9986,
      "mean_token_accuracy": 0.20570943951606752,
      "num_tokens": 28811013.0,
      "step": 13880
    },
    {
      "entropy": 5.440387964248657,
      "epoch": 1.2595246734397678,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048472804815942575,
      "loss": 5.0242,
      "mean_token_accuracy": 0.20063267648220062,
      "num_tokens": 28821367.0,
      "step": 13885
    },
    {
      "entropy": 5.50364317893982,
      "epoch": 1.259978229317852,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048471632851554517,
      "loss": 5.0643,
      "mean_token_accuracy": 0.20409667789936065,
      "num_tokens": 28831077.0,
      "step": 13890
    },
    {
      "entropy": 5.5035868167877195,
      "epoch": 1.2604317851959361,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004847046045346775,
      "loss": 5.0865,
      "mean_token_accuracy": 0.1920575052499771,
      "num_tokens": 28841490.0,
      "step": 13895
    },
    {
      "entropy": 5.508995771408081,
      "epoch": 1.2608853410740204,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048469287621706516,
      "loss": 5.0859,
      "mean_token_accuracy": 0.19784659445285796,
      "num_tokens": 28852510.0,
      "step": 13900
    },
    {
      "entropy": 5.496377563476562,
      "epoch": 1.2613388969521044,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004846811435629507,
      "loss": 5.0724,
      "mean_token_accuracy": 0.20141980946063995,
      "num_tokens": 28863388.0,
      "step": 13905
    },
    {
      "entropy": 5.401807069778442,
      "epoch": 1.2617924528301887,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004846694065725768,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20184410065412522,
      "num_tokens": 28873798.0,
      "step": 13910
    },
    {
      "entropy": 5.483473587036133,
      "epoch": 1.262246008708273,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004846576652461861,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19962013363838196,
      "num_tokens": 28883482.0,
      "step": 13915
    },
    {
      "entropy": 5.4887090682983395,
      "epoch": 1.262699564586357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048464591958402147,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20471181720495224,
      "num_tokens": 28893573.0,
      "step": 13920
    },
    {
      "entropy": 5.560998773574829,
      "epoch": 1.2631531204644413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048463416958632573,
      "loss": 5.1454,
      "mean_token_accuracy": 0.1967838451266289,
      "num_tokens": 28904064.0,
      "step": 13925
    },
    {
      "entropy": 5.505468559265137,
      "epoch": 1.2636066763425253,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048462241525334206,
      "loss": 5.1,
      "mean_token_accuracy": 0.19794451743364333,
      "num_tokens": 28915329.0,
      "step": 13930
    },
    {
      "entropy": 5.49501748085022,
      "epoch": 1.2640602322206096,
      "grad_norm": 1.0,
      "learning_rate": 0.0004846106565853133,
      "loss": 5.1859,
      "mean_token_accuracy": 0.19487343728542328,
      "num_tokens": 28925228.0,
      "step": 13935
    },
    {
      "entropy": 5.477839565277099,
      "epoch": 1.2645137880986939,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048459889358248284,
      "loss": 5.023,
      "mean_token_accuracy": 0.19917779564857482,
      "num_tokens": 28935891.0,
      "step": 13940
    },
    {
      "entropy": 5.4873875141143795,
      "epoch": 1.264967343976778,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048458712624509383,
      "loss": 4.985,
      "mean_token_accuracy": 0.20131113827228547,
      "num_tokens": 28946372.0,
      "step": 13945
    },
    {
      "entropy": 5.5058043003082275,
      "epoch": 1.2654208998548622,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048457535457338964,
      "loss": 5.0739,
      "mean_token_accuracy": 0.1947830557823181,
      "num_tokens": 28957874.0,
      "step": 13950
    },
    {
      "entropy": 5.5295580387115475,
      "epoch": 1.2658744557329462,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004845635785676137,
      "loss": 5.0528,
      "mean_token_accuracy": 0.19305006712675093,
      "num_tokens": 28968023.0,
      "step": 13955
    },
    {
      "entropy": 5.480109596252442,
      "epoch": 1.2663280116110305,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048455179822800957,
      "loss": 4.9497,
      "mean_token_accuracy": 0.2045432895421982,
      "num_tokens": 28977744.0,
      "step": 13960
    },
    {
      "entropy": 5.56711974143982,
      "epoch": 1.2667815674891147,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004845400135548208,
      "loss": 5.2127,
      "mean_token_accuracy": 0.191379314661026,
      "num_tokens": 28988875.0,
      "step": 13965
    },
    {
      "entropy": 5.415848731994629,
      "epoch": 1.2672351233671988,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004845282245482912,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20306566059589387,
      "num_tokens": 28998562.0,
      "step": 13970
    },
    {
      "entropy": 5.4908302307128904,
      "epoch": 1.267688679245283,
      "grad_norm": 1.0,
      "learning_rate": 0.0004845164312086646,
      "loss": 5.0358,
      "mean_token_accuracy": 0.20640750229358673,
      "num_tokens": 29008783.0,
      "step": 13975
    },
    {
      "entropy": 5.436105823516845,
      "epoch": 1.268142235123367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004845046335361847,
      "loss": 4.9642,
      "mean_token_accuracy": 0.19867230355739593,
      "num_tokens": 29019303.0,
      "step": 13980
    },
    {
      "entropy": 5.5175525665283205,
      "epoch": 1.2685957910014514,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048449283153109566,
      "loss": 5.1912,
      "mean_token_accuracy": 0.18820945173501968,
      "num_tokens": 29029843.0,
      "step": 13985
    },
    {
      "entropy": 5.526941013336182,
      "epoch": 1.2690493468795356,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004844810251936413,
      "loss": 5.0977,
      "mean_token_accuracy": 0.19493067264556885,
      "num_tokens": 29040603.0,
      "step": 13990
    },
    {
      "entropy": 5.484497308731079,
      "epoch": 1.2695029027576197,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004844692145240661,
      "loss": 5.0479,
      "mean_token_accuracy": 0.20133136510848998,
      "num_tokens": 29050599.0,
      "step": 13995
    },
    {
      "entropy": 5.470039415359497,
      "epoch": 1.269956458635704,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004844573995226142,
      "loss": 5.0875,
      "mean_token_accuracy": 0.195917047560215,
      "num_tokens": 29061292.0,
      "step": 14000
    },
    {
      "entropy": 5.417318868637085,
      "epoch": 1.270410014513788,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048444558018952984,
      "loss": 5.0016,
      "mean_token_accuracy": 0.2037899911403656,
      "num_tokens": 29071779.0,
      "step": 14005
    },
    {
      "entropy": 5.4463835716247555,
      "epoch": 1.2708635703918723,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048443375652505754,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20027604550123215,
      "num_tokens": 29082073.0,
      "step": 14010
    },
    {
      "entropy": 5.47108268737793,
      "epoch": 1.2713171262699565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004844219285294418,
      "loss": 4.9638,
      "mean_token_accuracy": 0.20421914309263228,
      "num_tokens": 29092032.0,
      "step": 14015
    },
    {
      "entropy": 5.4538318634033205,
      "epoch": 1.2717706821480406,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048441009620292714,
      "loss": 5.0123,
      "mean_token_accuracy": 0.2062854215502739,
      "num_tokens": 29102936.0,
      "step": 14020
    },
    {
      "entropy": 5.460636281967163,
      "epoch": 1.2722242380261248,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004843982595457584,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20491472631692886,
      "num_tokens": 29113385.0,
      "step": 14025
    },
    {
      "entropy": 5.457720708847046,
      "epoch": 1.2726777939042089,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048438641855818026,
      "loss": 5.0877,
      "mean_token_accuracy": 0.19535764306783676,
      "num_tokens": 29125075.0,
      "step": 14030
    },
    {
      "entropy": 5.503909730911255,
      "epoch": 1.2731313497822931,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004843745732404376,
      "loss": 5.0182,
      "mean_token_accuracy": 0.1980168715119362,
      "num_tokens": 29135098.0,
      "step": 14035
    },
    {
      "entropy": 5.545841121673584,
      "epoch": 1.2735849056603774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004843627235927755,
      "loss": 5.1158,
      "mean_token_accuracy": 0.19769282042980194,
      "num_tokens": 29145577.0,
      "step": 14040
    },
    {
      "entropy": 5.463726425170899,
      "epoch": 1.2740384615384617,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004843508696154388,
      "loss": 5.0457,
      "mean_token_accuracy": 0.20328806787729264,
      "num_tokens": 29155905.0,
      "step": 14045
    },
    {
      "entropy": 5.406875324249268,
      "epoch": 1.2744920174165457,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048433901130867283,
      "loss": 4.9801,
      "mean_token_accuracy": 0.20879150778055192,
      "num_tokens": 29164926.0,
      "step": 14050
    },
    {
      "entropy": 5.431298923492432,
      "epoch": 1.2749455732946298,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004843271486727228,
      "loss": 5.0057,
      "mean_token_accuracy": 0.2010043293237686,
      "num_tokens": 29175050.0,
      "step": 14055
    },
    {
      "entropy": 5.49884524345398,
      "epoch": 1.275399129172714,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048431528170783395,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19557402282953262,
      "num_tokens": 29186255.0,
      "step": 14060
    },
    {
      "entropy": 5.4883829116821286,
      "epoch": 1.2758526850507983,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004843034104142517,
      "loss": 5.0718,
      "mean_token_accuracy": 0.2044265389442444,
      "num_tokens": 29195639.0,
      "step": 14065
    },
    {
      "entropy": 5.477971744537354,
      "epoch": 1.2763062409288826,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004842915347922216,
      "loss": 5.0892,
      "mean_token_accuracy": 0.1940547063946724,
      "num_tokens": 29206482.0,
      "step": 14070
    },
    {
      "entropy": 5.554836750030518,
      "epoch": 1.2767597968069666,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004842796548419893,
      "loss": 5.0606,
      "mean_token_accuracy": 0.19817731380462647,
      "num_tokens": 29217433.0,
      "step": 14075
    },
    {
      "entropy": 5.548083972930908,
      "epoch": 1.2772133526850509,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048426777056380035,
      "loss": 5.1262,
      "mean_token_accuracy": 0.18795846849679948,
      "num_tokens": 29226193.0,
      "step": 14080
    },
    {
      "entropy": 5.44550051689148,
      "epoch": 1.277666908563135,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004842558819579006,
      "loss": 5.0557,
      "mean_token_accuracy": 0.20124146789312364,
      "num_tokens": 29237061.0,
      "step": 14085
    },
    {
      "entropy": 5.469652271270752,
      "epoch": 1.2781204644412192,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004842439890245358,
      "loss": 5.0297,
      "mean_token_accuracy": 0.20593013614416122,
      "num_tokens": 29247689.0,
      "step": 14090
    },
    {
      "entropy": 5.469670963287354,
      "epoch": 1.2785740203193035,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048423209176395204,
      "loss": 5.018,
      "mean_token_accuracy": 0.2025768503546715,
      "num_tokens": 29258235.0,
      "step": 14095
    },
    {
      "entropy": 5.454458236694336,
      "epoch": 1.2790275761973875,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004842201901763954,
      "loss": 4.9898,
      "mean_token_accuracy": 0.20647548735141755,
      "num_tokens": 29268884.0,
      "step": 14100
    },
    {
      "entropy": 5.475712776184082,
      "epoch": 1.2794811320754718,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004842082842621118,
      "loss": 5.1422,
      "mean_token_accuracy": 0.19233165234327315,
      "num_tokens": 29280484.0,
      "step": 14105
    },
    {
      "entropy": 5.504487562179565,
      "epoch": 1.2799346879535558,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004841963740213476,
      "loss": 5.0951,
      "mean_token_accuracy": 0.199015374481678,
      "num_tokens": 29291199.0,
      "step": 14110
    },
    {
      "entropy": 5.559184598922729,
      "epoch": 1.28038824383164,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004841844594543491,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19844071418046952,
      "num_tokens": 29302007.0,
      "step": 14115
    },
    {
      "entropy": 5.513727045059204,
      "epoch": 1.2808417997097243,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004841725405613626,
      "loss": 5.0442,
      "mean_token_accuracy": 0.20648347586393356,
      "num_tokens": 29312127.0,
      "step": 14120
    },
    {
      "entropy": 5.5265700340271,
      "epoch": 1.2812953555878084,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004841606173426347,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19536625295877458,
      "num_tokens": 29323691.0,
      "step": 14125
    },
    {
      "entropy": 5.46632194519043,
      "epoch": 1.2817489114658926,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048414868979841197,
      "loss": 5.0687,
      "mean_token_accuracy": 0.1954666033387184,
      "num_tokens": 29335203.0,
      "step": 14130
    },
    {
      "entropy": 5.524719381332398,
      "epoch": 1.2822024673439767,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048413675792894105,
      "loss": 5.0948,
      "mean_token_accuracy": 0.19794932007789612,
      "num_tokens": 29346422.0,
      "step": 14135
    },
    {
      "entropy": 5.46589846611023,
      "epoch": 1.282656023222061,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004841248217344686,
      "loss": 5.0215,
      "mean_token_accuracy": 0.20230863988399506,
      "num_tokens": 29356084.0,
      "step": 14140
    },
    {
      "entropy": 5.420564365386963,
      "epoch": 1.2831095791001452,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004841128812152415,
      "loss": 5.0331,
      "mean_token_accuracy": 0.20358027517795563,
      "num_tokens": 29366093.0,
      "step": 14145
    },
    {
      "entropy": 5.505201244354248,
      "epoch": 1.2835631349782293,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004841009363715068,
      "loss": 5.0912,
      "mean_token_accuracy": 0.19935815930366516,
      "num_tokens": 29376621.0,
      "step": 14150
    },
    {
      "entropy": 5.556482458114624,
      "epoch": 1.2840166908563135,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004840889872035115,
      "loss": 5.1318,
      "mean_token_accuracy": 0.19459037482738495,
      "num_tokens": 29387119.0,
      "step": 14155
    },
    {
      "entropy": 5.484884357452392,
      "epoch": 1.2844702467343976,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004840770337115026,
      "loss": 5.0552,
      "mean_token_accuracy": 0.20135198682546615,
      "num_tokens": 29396981.0,
      "step": 14160
    },
    {
      "entropy": 5.525591802597046,
      "epoch": 1.2849238026124818,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048406507589572747,
      "loss": 5.0417,
      "mean_token_accuracy": 0.20318254977464675,
      "num_tokens": 29406737.0,
      "step": 14165
    },
    {
      "entropy": 5.500586366653442,
      "epoch": 1.2853773584905661,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004840531137564331,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19997634142637252,
      "num_tokens": 29417300.0,
      "step": 14170
    },
    {
      "entropy": 5.476732492446899,
      "epoch": 1.2858309143686502,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004840411472938672,
      "loss": 4.9697,
      "mean_token_accuracy": 0.2080481082201004,
      "num_tokens": 29426749.0,
      "step": 14175
    },
    {
      "entropy": 5.405505704879761,
      "epoch": 1.2862844702467344,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004840291765082771,
      "loss": 5.0034,
      "mean_token_accuracy": 0.19875230491161347,
      "num_tokens": 29437264.0,
      "step": 14180
    },
    {
      "entropy": 5.474243116378784,
      "epoch": 1.2867380261248185,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048401720139991036,
      "loss": 4.9954,
      "mean_token_accuracy": 0.19903332740068436,
      "num_tokens": 29448192.0,
      "step": 14185
    },
    {
      "entropy": 5.5092840671539305,
      "epoch": 1.2871915820029027,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004840052219690146,
      "loss": 4.9105,
      "mean_token_accuracy": 0.20989975184202195,
      "num_tokens": 29457923.0,
      "step": 14190
    },
    {
      "entropy": 5.3856312274932865,
      "epoch": 1.287645137880987,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004839932382158376,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20904698222875595,
      "num_tokens": 29466914.0,
      "step": 14195
    },
    {
      "entropy": 5.449931573867798,
      "epoch": 1.288098693759071,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004839812501406272,
      "loss": 5.0846,
      "mean_token_accuracy": 0.1940669134259224,
      "num_tokens": 29477648.0,
      "step": 14200
    },
    {
      "entropy": 5.52697377204895,
      "epoch": 1.2885522496371553,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004839692577436313,
      "loss": 5.0756,
      "mean_token_accuracy": 0.2011621966958046,
      "num_tokens": 29487947.0,
      "step": 14205
    },
    {
      "entropy": 5.589422082901001,
      "epoch": 1.2890058055152394,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004839572610250978,
      "loss": 5.1548,
      "mean_token_accuracy": 0.1920746996998787,
      "num_tokens": 29497667.0,
      "step": 14210
    },
    {
      "entropy": 5.475475311279297,
      "epoch": 1.2894593613933236,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048394525998527496,
      "loss": 5.0363,
      "mean_token_accuracy": 0.1968516707420349,
      "num_tokens": 29507065.0,
      "step": 14215
    },
    {
      "entropy": 5.54481086730957,
      "epoch": 1.2899129172714079,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004839332546244109,
      "loss": 5.1283,
      "mean_token_accuracy": 0.1999591186642647,
      "num_tokens": 29517752.0,
      "step": 14220
    },
    {
      "entropy": 5.490834951400757,
      "epoch": 1.2903664731494922,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048392124494275385,
      "loss": 5.0457,
      "mean_token_accuracy": 0.19647222459316255,
      "num_tokens": 29527894.0,
      "step": 14225
    },
    {
      "entropy": 5.454130935668945,
      "epoch": 1.2908200290275762,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004839092309405523,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19558317065238953,
      "num_tokens": 29538136.0,
      "step": 14230
    },
    {
      "entropy": 5.496419763565063,
      "epoch": 1.2912735849056602,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048389721261805457,
      "loss": 5.0621,
      "mean_token_accuracy": 0.20487109571695328,
      "num_tokens": 29549193.0,
      "step": 14235
    },
    {
      "entropy": 5.461962795257568,
      "epoch": 1.2917271407837445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048388518997550926,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2094683215022087,
      "num_tokens": 29558889.0,
      "step": 14240
    },
    {
      "entropy": 5.4891866683959964,
      "epoch": 1.2921806966618288,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048387316301316493,
      "loss": 5.0708,
      "mean_token_accuracy": 0.1955946922302246,
      "num_tokens": 29569691.0,
      "step": 14245
    },
    {
      "entropy": 5.508250522613525,
      "epoch": 1.292634252539913,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004838611317312705,
      "loss": 5.1483,
      "mean_token_accuracy": 0.19559630900621414,
      "num_tokens": 29580459.0,
      "step": 14250
    },
    {
      "entropy": 5.51726884841919,
      "epoch": 1.293087808417997,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004838490961300745,
      "loss": 5.0637,
      "mean_token_accuracy": 0.19881267249584197,
      "num_tokens": 29590652.0,
      "step": 14255
    },
    {
      "entropy": 5.512678384780884,
      "epoch": 1.2935413642960814,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048383705620982606,
      "loss": 5.089,
      "mean_token_accuracy": 0.2042189583182335,
      "num_tokens": 29600353.0,
      "step": 14260
    },
    {
      "entropy": 5.496693420410156,
      "epoch": 1.2939949201741654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048382501197077405,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20726697742938996,
      "num_tokens": 29610561.0,
      "step": 14265
    },
    {
      "entropy": 5.438076114654541,
      "epoch": 1.2944484760522497,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004838129634131675,
      "loss": 5.0185,
      "mean_token_accuracy": 0.2033169150352478,
      "num_tokens": 29621741.0,
      "step": 14270
    },
    {
      "entropy": 5.44814715385437,
      "epoch": 1.294902031930334,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004838009105372558,
      "loss": 4.967,
      "mean_token_accuracy": 0.201059290766716,
      "num_tokens": 29633097.0,
      "step": 14275
    },
    {
      "entropy": 5.48220272064209,
      "epoch": 1.295355587808418,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048378885334328793,
      "loss": 5.088,
      "mean_token_accuracy": 0.19085302352905273,
      "num_tokens": 29644290.0,
      "step": 14280
    },
    {
      "entropy": 5.585373497009277,
      "epoch": 1.2958091436865022,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004837767918315135,
      "loss": 5.1258,
      "mean_token_accuracy": 0.19204142391681672,
      "num_tokens": 29654977.0,
      "step": 14285
    },
    {
      "entropy": 5.5744030475616455,
      "epoch": 1.2962626995645863,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048376472600218174,
      "loss": 5.1235,
      "mean_token_accuracy": 0.19505043774843217,
      "num_tokens": 29665020.0,
      "step": 14290
    },
    {
      "entropy": 5.475365161895752,
      "epoch": 1.2967162554426706,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004837526558555422,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20274775326251984,
      "num_tokens": 29675098.0,
      "step": 14295
    },
    {
      "entropy": 5.428688049316406,
      "epoch": 1.2971698113207548,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004837405813918446,
      "loss": 4.9932,
      "mean_token_accuracy": 0.2038100093603134,
      "num_tokens": 29685450.0,
      "step": 14300
    },
    {
      "entropy": 5.453429365158081,
      "epoch": 1.2976233671988389,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048372850261133856,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2004016637802124,
      "num_tokens": 29694818.0,
      "step": 14305
    },
    {
      "entropy": 5.5270758152008055,
      "epoch": 1.2980769230769231,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048371641951427393,
      "loss": 5.1108,
      "mean_token_accuracy": 0.19570721834897994,
      "num_tokens": 29705573.0,
      "step": 14310
    },
    {
      "entropy": 5.487916421890259,
      "epoch": 1.2985304789550072,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048370433210090055,
      "loss": 5.0108,
      "mean_token_accuracy": 0.19875363856554032,
      "num_tokens": 29716366.0,
      "step": 14315
    },
    {
      "entropy": 5.469477128982544,
      "epoch": 1.2989840348330914,
      "grad_norm": 1.0,
      "learning_rate": 0.0004836922403714683,
      "loss": 5.042,
      "mean_token_accuracy": 0.19777244478464126,
      "num_tokens": 29727181.0,
      "step": 14320
    },
    {
      "entropy": 5.452441024780273,
      "epoch": 1.2994375907111757,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048368014432622746,
      "loss": 5.0523,
      "mean_token_accuracy": 0.20698187947273256,
      "num_tokens": 29736960.0,
      "step": 14325
    },
    {
      "entropy": 5.445215511322021,
      "epoch": 1.2998911465892597,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000483668043965428,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21604397892951965,
      "num_tokens": 29746354.0,
      "step": 14330
    },
    {
      "entropy": 5.471987867355347,
      "epoch": 1.300344702467344,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004836559392893203,
      "loss": 5.0542,
      "mean_token_accuracy": 0.1957583099603653,
      "num_tokens": 29756627.0,
      "step": 14335
    },
    {
      "entropy": 5.464911985397339,
      "epoch": 1.300798258345428,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004836438302981545,
      "loss": 5.0398,
      "mean_token_accuracy": 0.20956512540578842,
      "num_tokens": 29766684.0,
      "step": 14340
    },
    {
      "entropy": 5.470847320556641,
      "epoch": 1.3012518142235123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004836317169921811,
      "loss": 5.0622,
      "mean_token_accuracy": 0.19975050240755082,
      "num_tokens": 29776556.0,
      "step": 14345
    },
    {
      "entropy": 5.553652000427246,
      "epoch": 1.3017053701015966,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048361959937165075,
      "loss": 5.1958,
      "mean_token_accuracy": 0.19425188899040222,
      "num_tokens": 29786896.0,
      "step": 14350
    },
    {
      "entropy": 5.496381521224976,
      "epoch": 1.3021589259796806,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004836074774368138,
      "loss": 4.9972,
      "mean_token_accuracy": 0.2041577160358429,
      "num_tokens": 29797332.0,
      "step": 14355
    },
    {
      "entropy": 5.503427219390869,
      "epoch": 1.302612481857765,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048359535118792116,
      "loss": 5.0903,
      "mean_token_accuracy": 0.20044789463281631,
      "num_tokens": 29806428.0,
      "step": 14360
    },
    {
      "entropy": 5.4628736019134525,
      "epoch": 1.303066037735849,
      "grad_norm": 1.0,
      "learning_rate": 0.0004835832206252235,
      "loss": 5.0628,
      "mean_token_accuracy": 0.19847192913293837,
      "num_tokens": 29816794.0,
      "step": 14365
    },
    {
      "entropy": 5.562831878662109,
      "epoch": 1.3035195936139332,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004835710857489716,
      "loss": 5.0886,
      "mean_token_accuracy": 0.20265541970729828,
      "num_tokens": 29827077.0,
      "step": 14370
    },
    {
      "entropy": 5.462686586380005,
      "epoch": 1.3039731494920175,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004835589465594165,
      "loss": 5.0597,
      "mean_token_accuracy": 0.1975301370024681,
      "num_tokens": 29837868.0,
      "step": 14375
    },
    {
      "entropy": 5.480144453048706,
      "epoch": 1.3044267053701015,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048354680305680926,
      "loss": 5.0848,
      "mean_token_accuracy": 0.2043821007013321,
      "num_tokens": 29846929.0,
      "step": 14380
    },
    {
      "entropy": 5.444469165802002,
      "epoch": 1.3048802612481858,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048353465524140105,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20302888005971909,
      "num_tokens": 29857905.0,
      "step": 14385
    },
    {
      "entropy": 5.474423599243164,
      "epoch": 1.3053338171262698,
      "grad_norm": 1.078125,
      "learning_rate": 0.000483522503113443,
      "loss": 5.0751,
      "mean_token_accuracy": 0.1963631957769394,
      "num_tokens": 29868230.0,
      "step": 14390
    },
    {
      "entropy": 5.482108020782471,
      "epoch": 1.305787373004354,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048351034667318634,
      "loss": 5.0315,
      "mean_token_accuracy": 0.20345700830221175,
      "num_tokens": 29879224.0,
      "step": 14395
    },
    {
      "entropy": 5.5577699661254885,
      "epoch": 1.3062409288824384,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004834981859208827,
      "loss": 5.1748,
      "mean_token_accuracy": 0.19315745383501054,
      "num_tokens": 29890416.0,
      "step": 14400
    },
    {
      "entropy": 5.52885799407959,
      "epoch": 1.3066944847605226,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048348602085678346,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19536293894052506,
      "num_tokens": 29901064.0,
      "step": 14405
    },
    {
      "entropy": 5.577747058868408,
      "epoch": 1.3071480406386067,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048347385148114004,
      "loss": 5.1489,
      "mean_token_accuracy": 0.19691503047943115,
      "num_tokens": 29911084.0,
      "step": 14410
    },
    {
      "entropy": 5.467166519165039,
      "epoch": 1.3076015965166907,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048346167779420436,
      "loss": 4.9916,
      "mean_token_accuracy": 0.21268735975027084,
      "num_tokens": 29921104.0,
      "step": 14415
    },
    {
      "entropy": 5.474800539016724,
      "epoch": 1.308055152394775,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048344949979622806,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20343236178159713,
      "num_tokens": 29932014.0,
      "step": 14420
    },
    {
      "entropy": 5.46609206199646,
      "epoch": 1.3085087082728593,
      "grad_norm": 1.125,
      "learning_rate": 0.00048343731748746294,
      "loss": 5.0518,
      "mean_token_accuracy": 0.2032724514603615,
      "num_tokens": 29941707.0,
      "step": 14425
    },
    {
      "entropy": 5.475191116333008,
      "epoch": 1.3089622641509435,
      "grad_norm": 1.09375,
      "learning_rate": 0.000483425130868161,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19366600662469863,
      "num_tokens": 29952082.0,
      "step": 14430
    },
    {
      "entropy": 5.422572040557862,
      "epoch": 1.3094158200290276,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004834129399385742,
      "loss": 4.8636,
      "mean_token_accuracy": 0.22088147699832916,
      "num_tokens": 29961744.0,
      "step": 14435
    },
    {
      "entropy": 5.479946565628052,
      "epoch": 1.3098693759071118,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004834007446989548,
      "loss": 5.0253,
      "mean_token_accuracy": 0.2101757764816284,
      "num_tokens": 29972611.0,
      "step": 14440
    },
    {
      "entropy": 5.459532022476196,
      "epoch": 1.3103229317851959,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048338854514955474,
      "loss": 4.9624,
      "mean_token_accuracy": 0.208077135682106,
      "num_tokens": 29982660.0,
      "step": 14445
    },
    {
      "entropy": 5.445099782943726,
      "epoch": 1.3107764876632801,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048337634129062655,
      "loss": 5.0747,
      "mean_token_accuracy": 0.1888832926750183,
      "num_tokens": 29993349.0,
      "step": 14450
    },
    {
      "entropy": 5.562231159210205,
      "epoch": 1.3112300435413644,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048336413312242255,
      "loss": 5.1659,
      "mean_token_accuracy": 0.19416691362857819,
      "num_tokens": 30004474.0,
      "step": 14455
    },
    {
      "entropy": 5.558346843719482,
      "epoch": 1.3116835994194485,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048335192064519514,
      "loss": 5.1042,
      "mean_token_accuracy": 0.19719318449497222,
      "num_tokens": 30015189.0,
      "step": 14460
    },
    {
      "entropy": 5.511405658721924,
      "epoch": 1.3121371552975327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048333970385919695,
      "loss": 4.9939,
      "mean_token_accuracy": 0.19969361424446105,
      "num_tokens": 30025491.0,
      "step": 14465
    },
    {
      "entropy": 5.478708028793335,
      "epoch": 1.3125907111756168,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048332748276468053,
      "loss": 5.0612,
      "mean_token_accuracy": 0.20458794683218,
      "num_tokens": 30034683.0,
      "step": 14470
    },
    {
      "entropy": 5.471246767044067,
      "epoch": 1.313044267053701,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004833152573618987,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20007718205451966,
      "num_tokens": 30044135.0,
      "step": 14475
    },
    {
      "entropy": 5.51534218788147,
      "epoch": 1.3134978229317853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048330302765110436,
      "loss": 5.1094,
      "mean_token_accuracy": 0.2011711835861206,
      "num_tokens": 30054520.0,
      "step": 14480
    },
    {
      "entropy": 5.505444955825806,
      "epoch": 1.3139513788098693,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048329079363255024,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19864712804555892,
      "num_tokens": 30065439.0,
      "step": 14485
    },
    {
      "entropy": 5.502534627914429,
      "epoch": 1.3144049346879536,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048327855530648934,
      "loss": 5.0746,
      "mean_token_accuracy": 0.1971224755048752,
      "num_tokens": 30075362.0,
      "step": 14490
    },
    {
      "entropy": 5.471915912628174,
      "epoch": 1.3148584905660377,
      "grad_norm": 1.015625,
      "learning_rate": 0.000483266312673175,
      "loss": 5.0361,
      "mean_token_accuracy": 0.19733360409736633,
      "num_tokens": 30084913.0,
      "step": 14495
    },
    {
      "entropy": 5.471127033233643,
      "epoch": 1.315312046444122,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004832540657328602,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19881338328123094,
      "num_tokens": 30095704.0,
      "step": 14500
    },
    {
      "entropy": 5.488482666015625,
      "epoch": 1.3157656023222062,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004832418144857982,
      "loss": 5.0117,
      "mean_token_accuracy": 0.20118715465068818,
      "num_tokens": 30105981.0,
      "step": 14505
    },
    {
      "entropy": 5.4902448654174805,
      "epoch": 1.3162191582002902,
      "grad_norm": 1.0,
      "learning_rate": 0.00048322955893224245,
      "loss": 5.0795,
      "mean_token_accuracy": 0.19059209823608397,
      "num_tokens": 30115772.0,
      "step": 14510
    },
    {
      "entropy": 5.492611694335937,
      "epoch": 1.3166727140783745,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048321729907244626,
      "loss": 5.016,
      "mean_token_accuracy": 0.2023007556796074,
      "num_tokens": 30127050.0,
      "step": 14515
    },
    {
      "entropy": 5.49431209564209,
      "epoch": 1.3171262699564585,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004832050349066634,
      "loss": 5.0692,
      "mean_token_accuracy": 0.20058691054582595,
      "num_tokens": 30138221.0,
      "step": 14520
    },
    {
      "entropy": 5.533912324905396,
      "epoch": 1.3175798258345428,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004831927664351473,
      "loss": 5.1336,
      "mean_token_accuracy": 0.195409394800663,
      "num_tokens": 30150181.0,
      "step": 14525
    },
    {
      "entropy": 5.490771627426147,
      "epoch": 1.318033381712627,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004831804936581518,
      "loss": 5.0948,
      "mean_token_accuracy": 0.20345331132411956,
      "num_tokens": 30160521.0,
      "step": 14530
    },
    {
      "entropy": 5.524169111251831,
      "epoch": 1.3184869375907111,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004831682165759305,
      "loss": 5.0772,
      "mean_token_accuracy": 0.19717661589384078,
      "num_tokens": 30170611.0,
      "step": 14535
    },
    {
      "entropy": 5.526472377777099,
      "epoch": 1.3189404934687954,
      "grad_norm": 1.0,
      "learning_rate": 0.00048315593518873757,
      "loss": 5.0375,
      "mean_token_accuracy": 0.209712053835392,
      "num_tokens": 30179914.0,
      "step": 14540
    },
    {
      "entropy": 5.466052293777466,
      "epoch": 1.3193940493468794,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004831436494968268,
      "loss": 5.0337,
      "mean_token_accuracy": 0.20637008249759675,
      "num_tokens": 30191426.0,
      "step": 14545
    },
    {
      "entropy": 5.4549181938171385,
      "epoch": 1.3198476052249637,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048313135950045234,
      "loss": 5.0122,
      "mean_token_accuracy": 0.19953661859035493,
      "num_tokens": 30202813.0,
      "step": 14550
    },
    {
      "entropy": 5.475877141952514,
      "epoch": 1.320301161103048,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048311906519986824,
      "loss": 4.9444,
      "mean_token_accuracy": 0.20676357299089432,
      "num_tokens": 30213044.0,
      "step": 14555
    },
    {
      "entropy": 5.5254028797149655,
      "epoch": 1.320754716981132,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000483106766595329,
      "loss": 5.0314,
      "mean_token_accuracy": 0.1970205068588257,
      "num_tokens": 30223419.0,
      "step": 14560
    },
    {
      "entropy": 5.553525733947754,
      "epoch": 1.3212082728592163,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004830944636870887,
      "loss": 5.0777,
      "mean_token_accuracy": 0.19376981556415557,
      "num_tokens": 30233203.0,
      "step": 14565
    },
    {
      "entropy": 5.487842988967896,
      "epoch": 1.3216618287373003,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048308215647540183,
      "loss": 5.0705,
      "mean_token_accuracy": 0.19618242979049683,
      "num_tokens": 30244604.0,
      "step": 14570
    },
    {
      "entropy": 5.5587037086486815,
      "epoch": 1.3221153846153846,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048306984496052297,
      "loss": 5.1187,
      "mean_token_accuracy": 0.20165631622076036,
      "num_tokens": 30255355.0,
      "step": 14575
    },
    {
      "entropy": 5.577435302734375,
      "epoch": 1.3225689404934688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048305752914270665,
      "loss": 5.1656,
      "mean_token_accuracy": 0.19230901449918747,
      "num_tokens": 30265665.0,
      "step": 14580
    },
    {
      "entropy": 5.564355754852295,
      "epoch": 1.3230224963715531,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048304520902220767,
      "loss": 5.129,
      "mean_token_accuracy": 0.20115920305252075,
      "num_tokens": 30274347.0,
      "step": 14585
    },
    {
      "entropy": 5.51794400215149,
      "epoch": 1.3234760522496372,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004830328845992808,
      "loss": 4.9679,
      "mean_token_accuracy": 0.20817530900239944,
      "num_tokens": 30284243.0,
      "step": 14590
    },
    {
      "entropy": 5.437684631347656,
      "epoch": 1.3239296081277212,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048302055587418074,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20018340796232223,
      "num_tokens": 30294327.0,
      "step": 14595
    },
    {
      "entropy": 5.445873594284057,
      "epoch": 1.3243831640058055,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004830082228471626,
      "loss": 5.0842,
      "mean_token_accuracy": 0.19650617390871047,
      "num_tokens": 30304188.0,
      "step": 14600
    },
    {
      "entropy": 5.5108343124389645,
      "epoch": 1.3248367198838897,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048299588551848146,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2086288645863533,
      "num_tokens": 30314917.0,
      "step": 14605
    },
    {
      "entropy": 5.482002687454224,
      "epoch": 1.325290275761974,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048298354388839237,
      "loss": 5.0159,
      "mean_token_accuracy": 0.2028503105044365,
      "num_tokens": 30325221.0,
      "step": 14610
    },
    {
      "entropy": 5.512210655212402,
      "epoch": 1.325743831640058,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004829711979571506,
      "loss": 5.1134,
      "mean_token_accuracy": 0.19776519536972045,
      "num_tokens": 30336196.0,
      "step": 14615
    },
    {
      "entropy": 5.444050025939942,
      "epoch": 1.3261973875181423,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004829588477250115,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21473861038684844,
      "num_tokens": 30345546.0,
      "step": 14620
    },
    {
      "entropy": 5.52281084060669,
      "epoch": 1.3266509433962264,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004829464931922304,
      "loss": 5.0554,
      "mean_token_accuracy": 0.1991855889558792,
      "num_tokens": 30355672.0,
      "step": 14625
    },
    {
      "entropy": 5.505038070678711,
      "epoch": 1.3271044992743106,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048293413435906285,
      "loss": 5.0484,
      "mean_token_accuracy": 0.20044143497943878,
      "num_tokens": 30366233.0,
      "step": 14630
    },
    {
      "entropy": 5.467237281799316,
      "epoch": 1.327558055152395,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004829217712257644,
      "loss": 5.0513,
      "mean_token_accuracy": 0.20227838754653932,
      "num_tokens": 30377366.0,
      "step": 14635
    },
    {
      "entropy": 5.473698282241822,
      "epoch": 1.328011611030479,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048290940379259073,
      "loss": 5.101,
      "mean_token_accuracy": 0.19721851050853728,
      "num_tokens": 30388022.0,
      "step": 14640
    },
    {
      "entropy": 5.483129501342773,
      "epoch": 1.3284651669085632,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048289703205979756,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20514749586582184,
      "num_tokens": 30399235.0,
      "step": 14645
    },
    {
      "entropy": 5.588254690170288,
      "epoch": 1.3289187227866472,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048288465602764084,
      "loss": 5.122,
      "mean_token_accuracy": 0.20071659833192826,
      "num_tokens": 30409112.0,
      "step": 14650
    },
    {
      "entropy": 5.452871370315552,
      "epoch": 1.3293722786647315,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048287227569637645,
      "loss": 5.0219,
      "mean_token_accuracy": 0.2019849956035614,
      "num_tokens": 30420202.0,
      "step": 14655
    },
    {
      "entropy": 5.532520389556884,
      "epoch": 1.3298258345428158,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048285989106626036,
      "loss": 5.0777,
      "mean_token_accuracy": 0.20122217237949372,
      "num_tokens": 30431210.0,
      "step": 14660
    },
    {
      "entropy": 5.497643756866455,
      "epoch": 1.3302793904208998,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004828475021375489,
      "loss": 5.0042,
      "mean_token_accuracy": 0.20784106552600862,
      "num_tokens": 30441291.0,
      "step": 14665
    },
    {
      "entropy": 5.457648134231567,
      "epoch": 1.330732946298984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004828351089104981,
      "loss": 4.964,
      "mean_token_accuracy": 0.20614739060401915,
      "num_tokens": 30450285.0,
      "step": 14670
    },
    {
      "entropy": 5.443894147872925,
      "epoch": 1.3311865021770681,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048282271138536424,
      "loss": 5.0606,
      "mean_token_accuracy": 0.1993221625685692,
      "num_tokens": 30460311.0,
      "step": 14675
    },
    {
      "entropy": 5.560179233551025,
      "epoch": 1.3316400580551524,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004828103095624037,
      "loss": 5.072,
      "mean_token_accuracy": 0.20389872342348098,
      "num_tokens": 30470188.0,
      "step": 14680
    },
    {
      "entropy": 5.470388412475586,
      "epoch": 1.3320936139332367,
      "grad_norm": 0.921875,
      "learning_rate": 0.000482797903441873,
      "loss": 5.024,
      "mean_token_accuracy": 0.1966773748397827,
      "num_tokens": 30481406.0,
      "step": 14685
    },
    {
      "entropy": 5.407244253158569,
      "epoch": 1.3325471698113207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004827854930240288,
      "loss": 5.0032,
      "mean_token_accuracy": 0.20737527012825013,
      "num_tokens": 30492314.0,
      "step": 14690
    },
    {
      "entropy": 5.552218914031982,
      "epoch": 1.333000725689405,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004827730783091276,
      "loss": 5.1477,
      "mean_token_accuracy": 0.19321692138910293,
      "num_tokens": 30503273.0,
      "step": 14695
    },
    {
      "entropy": 5.546647071838379,
      "epoch": 1.333454281567489,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048276065929742616,
      "loss": 5.0477,
      "mean_token_accuracy": 0.1968608483672142,
      "num_tokens": 30513582.0,
      "step": 14700
    },
    {
      "entropy": 5.440297269821167,
      "epoch": 1.3339078374455733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004827482359891814,
      "loss": 4.97,
      "mean_token_accuracy": 0.21266762018203736,
      "num_tokens": 30523261.0,
      "step": 14705
    },
    {
      "entropy": 5.500866556167603,
      "epoch": 1.3343613933236576,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048273580838465016,
      "loss": 5.0851,
      "mean_token_accuracy": 0.19987517595291138,
      "num_tokens": 30533774.0,
      "step": 14710
    },
    {
      "entropy": 5.502251482009887,
      "epoch": 1.3348149492017416,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004827233764840894,
      "loss": 5.1351,
      "mean_token_accuracy": 0.19586722254753114,
      "num_tokens": 30544580.0,
      "step": 14715
    },
    {
      "entropy": 5.525274324417114,
      "epoch": 1.3352685050798259,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004827109402877564,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20630441159009932,
      "num_tokens": 30555105.0,
      "step": 14720
    },
    {
      "entropy": 5.514051389694214,
      "epoch": 1.33572206095791,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004826984997959082,
      "loss": 5.1256,
      "mean_token_accuracy": 0.18905919790267944,
      "num_tokens": 30564912.0,
      "step": 14725
    },
    {
      "entropy": 5.463503980636597,
      "epoch": 1.3361756168359942,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000482686055008802,
      "loss": 5.0573,
      "mean_token_accuracy": 0.1945282906293869,
      "num_tokens": 30575862.0,
      "step": 14730
    },
    {
      "entropy": 5.500078582763672,
      "epoch": 1.3366291727140784,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048267360592669535,
      "loss": 5.1625,
      "mean_token_accuracy": 0.19513950496912003,
      "num_tokens": 30586951.0,
      "step": 14735
    },
    {
      "entropy": 5.529357385635376,
      "epoch": 1.3370827285921625,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004826611525498456,
      "loss": 5.0766,
      "mean_token_accuracy": 0.1965434119105339,
      "num_tokens": 30597604.0,
      "step": 14740
    },
    {
      "entropy": 5.536353826522827,
      "epoch": 1.3375362844702468,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004826486948785103,
      "loss": 5.1232,
      "mean_token_accuracy": 0.19308859407901763,
      "num_tokens": 30608157.0,
      "step": 14745
    },
    {
      "entropy": 5.521725034713745,
      "epoch": 1.3379898403483308,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048263623291294706,
      "loss": 5.0643,
      "mean_token_accuracy": 0.20122882276773452,
      "num_tokens": 30618527.0,
      "step": 14750
    },
    {
      "entropy": 5.541813707351684,
      "epoch": 1.338443396226415,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048262376665341355,
      "loss": 5.1482,
      "mean_token_accuracy": 0.1948135241866112,
      "num_tokens": 30629253.0,
      "step": 14755
    },
    {
      "entropy": 5.508347606658935,
      "epoch": 1.3388969521044993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004826112961001677,
      "loss": 5.1093,
      "mean_token_accuracy": 0.18560464829206466,
      "num_tokens": 30639015.0,
      "step": 14760
    },
    {
      "entropy": 5.557197237014771,
      "epoch": 1.3393505079825834,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048259882125346735,
      "loss": 5.1851,
      "mean_token_accuracy": 0.1878888040781021,
      "num_tokens": 30650035.0,
      "step": 14765
    },
    {
      "entropy": 5.557344198226929,
      "epoch": 1.3398040638606676,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048258634211357045,
      "loss": 5.0957,
      "mean_token_accuracy": 0.1993729680776596,
      "num_tokens": 30659813.0,
      "step": 14770
    },
    {
      "entropy": 5.504110193252563,
      "epoch": 1.3402576197387517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004825738586807351,
      "loss": 5.0459,
      "mean_token_accuracy": 0.1985906630754471,
      "num_tokens": 30670243.0,
      "step": 14775
    },
    {
      "entropy": 5.452295589447021,
      "epoch": 1.340711175616836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048256137095521945,
      "loss": 5.0215,
      "mean_token_accuracy": 0.1977318599820137,
      "num_tokens": 30680056.0,
      "step": 14780
    },
    {
      "entropy": 5.476433897018433,
      "epoch": 1.3411647314949202,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048254887893728175,
      "loss": 4.9717,
      "mean_token_accuracy": 0.2092755541205406,
      "num_tokens": 30690536.0,
      "step": 14785
    },
    {
      "entropy": 5.443423271179199,
      "epoch": 1.3416182873730045,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004825363826271804,
      "loss": 5.0582,
      "mean_token_accuracy": 0.1987844005227089,
      "num_tokens": 30700200.0,
      "step": 14790
    },
    {
      "entropy": 5.44862470626831,
      "epoch": 1.3420718432510885,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004825238820251737,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20584656149148942,
      "num_tokens": 30710576.0,
      "step": 14795
    },
    {
      "entropy": 5.482344055175782,
      "epoch": 1.3425253991291728,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048251137713152027,
      "loss": 5.0585,
      "mean_token_accuracy": 0.2019256144762039,
      "num_tokens": 30721495.0,
      "step": 14800
    },
    {
      "entropy": 5.519159507751465,
      "epoch": 1.3429789550072568,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004824988679464787,
      "loss": 5.0938,
      "mean_token_accuracy": 0.1964031144976616,
      "num_tokens": 30731700.0,
      "step": 14805
    },
    {
      "entropy": 5.527659940719604,
      "epoch": 1.343432510885341,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048248635447030767,
      "loss": 5.0728,
      "mean_token_accuracy": 0.19750734865665437,
      "num_tokens": 30743421.0,
      "step": 14810
    },
    {
      "entropy": 5.488374805450439,
      "epoch": 1.3438860667634254,
      "grad_norm": 1.0,
      "learning_rate": 0.0004824738367032659,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2078663155436516,
      "num_tokens": 30754238.0,
      "step": 14815
    },
    {
      "entropy": 5.480378723144531,
      "epoch": 1.3443396226415094,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004824613146456123,
      "loss": 5.0715,
      "mean_token_accuracy": 0.19379124492406846,
      "num_tokens": 30763938.0,
      "step": 14820
    },
    {
      "entropy": 5.573088788986206,
      "epoch": 1.3447931785195937,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004824487882976059,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19805540144443512,
      "num_tokens": 30773654.0,
      "step": 14825
    },
    {
      "entropy": 5.551896667480468,
      "epoch": 1.3452467343976777,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048243625765950564,
      "loss": 5.1175,
      "mean_token_accuracy": 0.19453760385513305,
      "num_tokens": 30784659.0,
      "step": 14830
    },
    {
      "entropy": 5.469422769546509,
      "epoch": 1.345700290275762,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004824237227315708,
      "loss": 4.992,
      "mean_token_accuracy": 0.19851766228675843,
      "num_tokens": 30795194.0,
      "step": 14835
    },
    {
      "entropy": 5.410362339019775,
      "epoch": 1.3461538461538463,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004824111835140604,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21054262965917586,
      "num_tokens": 30805623.0,
      "step": 14840
    },
    {
      "entropy": 5.441026401519776,
      "epoch": 1.3466074020319303,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000482398640007234,
      "loss": 4.9817,
      "mean_token_accuracy": 0.2019845500588417,
      "num_tokens": 30814899.0,
      "step": 14845
    },
    {
      "entropy": 5.4771339893341064,
      "epoch": 1.3470609579100146,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048238609221135076,
      "loss": 5.1093,
      "mean_token_accuracy": 0.19643137007951736,
      "num_tokens": 30825546.0,
      "step": 14850
    },
    {
      "entropy": 5.455335998535157,
      "epoch": 1.3475145137880986,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004823735401266704,
      "loss": 4.9939,
      "mean_token_accuracy": 0.19967146068811417,
      "num_tokens": 30836406.0,
      "step": 14855
    },
    {
      "entropy": 5.473223638534546,
      "epoch": 1.3479680696661829,
      "grad_norm": 1.0,
      "learning_rate": 0.00048236098375345227,
      "loss": 5.0861,
      "mean_token_accuracy": 0.19803501665592194,
      "num_tokens": 30847966.0,
      "step": 14860
    },
    {
      "entropy": 5.569869422912598,
      "epoch": 1.3484216255442671,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004823484230919563,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19745918065309526,
      "num_tokens": 30856831.0,
      "step": 14865
    },
    {
      "entropy": 5.444964838027954,
      "epoch": 1.3488751814223512,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048233585814244195,
      "loss": 4.8928,
      "mean_token_accuracy": 0.21360531598329544,
      "num_tokens": 30866076.0,
      "step": 14870
    },
    {
      "entropy": 5.4761899471282955,
      "epoch": 1.3493287373004355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004823232889051693,
      "loss": 5.0081,
      "mean_token_accuracy": 0.2073754221200943,
      "num_tokens": 30876845.0,
      "step": 14875
    },
    {
      "entropy": 5.450684928894043,
      "epoch": 1.3497822931785195,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048231071538039816,
      "loss": 5.0523,
      "mean_token_accuracy": 0.19669134616851808,
      "num_tokens": 30887448.0,
      "step": 14880
    },
    {
      "entropy": 5.546138429641724,
      "epoch": 1.3502358490566038,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048229813756838863,
      "loss": 5.0474,
      "mean_token_accuracy": 0.2002974957227707,
      "num_tokens": 30898414.0,
      "step": 14885
    },
    {
      "entropy": 5.486542892456055,
      "epoch": 1.350689404934688,
      "grad_norm": 1.0,
      "learning_rate": 0.0004822855554694007,
      "loss": 5.0582,
      "mean_token_accuracy": 0.20294709950685502,
      "num_tokens": 30908559.0,
      "step": 14890
    },
    {
      "entropy": 5.514283227920532,
      "epoch": 1.351142960812772,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048227296908369475,
      "loss": 5.0293,
      "mean_token_accuracy": 0.20701266676187516,
      "num_tokens": 30918655.0,
      "step": 14895
    },
    {
      "entropy": 5.453053760528564,
      "epoch": 1.3515965166908563,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048226037841153104,
      "loss": 5.0473,
      "mean_token_accuracy": 0.20324480831623076,
      "num_tokens": 30929188.0,
      "step": 14900
    },
    {
      "entropy": 5.520646667480468,
      "epoch": 1.3520500725689404,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004822477834531698,
      "loss": 5.1263,
      "mean_token_accuracy": 0.2014889895915985,
      "num_tokens": 30939902.0,
      "step": 14905
    },
    {
      "entropy": 5.496817684173584,
      "epoch": 1.3525036284470247,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004822351842088716,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20315656661987305,
      "num_tokens": 30951289.0,
      "step": 14910
    },
    {
      "entropy": 5.581944751739502,
      "epoch": 1.352957184325109,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004822225806788969,
      "loss": 5.1835,
      "mean_token_accuracy": 0.1878611609339714,
      "num_tokens": 30961233.0,
      "step": 14915
    },
    {
      "entropy": 5.486229324340821,
      "epoch": 1.353410740203193,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048220997286350654,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20341320782899858,
      "num_tokens": 30971922.0,
      "step": 14920
    },
    {
      "entropy": 5.460137271881104,
      "epoch": 1.3538642960812772,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048219736076296114,
      "loss": 5.0352,
      "mean_token_accuracy": 0.20320184230804444,
      "num_tokens": 30981358.0,
      "step": 14925
    },
    {
      "entropy": 5.512710285186768,
      "epoch": 1.3543178519593613,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004821847443775214,
      "loss": 5.1231,
      "mean_token_accuracy": 0.19762219935655595,
      "num_tokens": 30991984.0,
      "step": 14930
    },
    {
      "entropy": 5.4981756687164305,
      "epoch": 1.3547714078374455,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048217212370744853,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20261044204235076,
      "num_tokens": 31002790.0,
      "step": 14935
    },
    {
      "entropy": 5.4207916259765625,
      "epoch": 1.3552249637155298,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048215949875300327,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2109668642282486,
      "num_tokens": 31012827.0,
      "step": 14940
    },
    {
      "entropy": 5.430930042266846,
      "epoch": 1.3556785195936139,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004821468695144467,
      "loss": 5.0063,
      "mean_token_accuracy": 0.2006464809179306,
      "num_tokens": 31023075.0,
      "step": 14945
    },
    {
      "entropy": 5.520316028594971,
      "epoch": 1.3561320754716981,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048213423599204015,
      "loss": 5.0683,
      "mean_token_accuracy": 0.20287319719791413,
      "num_tokens": 31032454.0,
      "step": 14950
    },
    {
      "entropy": 5.57982611656189,
      "epoch": 1.3565856313497822,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048212159818604477,
      "loss": 5.1899,
      "mean_token_accuracy": 0.18757683485746385,
      "num_tokens": 31042498.0,
      "step": 14955
    },
    {
      "entropy": 5.469901990890503,
      "epoch": 1.3570391872278664,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000482108956096722,
      "loss": 5.0336,
      "mean_token_accuracy": 0.2002289518713951,
      "num_tokens": 31053188.0,
      "step": 14960
    },
    {
      "entropy": 5.497510814666748,
      "epoch": 1.3574927431059507,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048209630972433326,
      "loss": 5.012,
      "mean_token_accuracy": 0.2035995379090309,
      "num_tokens": 31062840.0,
      "step": 14965
    },
    {
      "entropy": 5.561136054992676,
      "epoch": 1.357946298984035,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048208365906914,
      "loss": 5.1582,
      "mean_token_accuracy": 0.1987423777580261,
      "num_tokens": 31072646.0,
      "step": 14970
    },
    {
      "entropy": 5.48033299446106,
      "epoch": 1.358399854862119,
      "grad_norm": 1.125,
      "learning_rate": 0.00048207100413140397,
      "loss": 4.9932,
      "mean_token_accuracy": 0.1994940683245659,
      "num_tokens": 31082700.0,
      "step": 14975
    },
    {
      "entropy": 5.541996240615845,
      "epoch": 1.3588534107402033,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004820583449113868,
      "loss": 5.0812,
      "mean_token_accuracy": 0.19881935864686967,
      "num_tokens": 31093743.0,
      "step": 14980
    },
    {
      "entropy": 5.503304290771484,
      "epoch": 1.3593069666182873,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048204568140935023,
      "loss": 5.0221,
      "mean_token_accuracy": 0.20480051189661025,
      "num_tokens": 31104293.0,
      "step": 14985
    },
    {
      "entropy": 5.443846797943115,
      "epoch": 1.3597605224963716,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004820330136255563,
      "loss": 5.0281,
      "mean_token_accuracy": 0.2066284880042076,
      "num_tokens": 31114244.0,
      "step": 14990
    },
    {
      "entropy": 5.46366810798645,
      "epoch": 1.3602140783744558,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048202034156026683,
      "loss": 5.0191,
      "mean_token_accuracy": 0.19951775372028352,
      "num_tokens": 31123596.0,
      "step": 14995
    },
    {
      "entropy": 5.5296409130096436,
      "epoch": 1.36066763425254,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048200766521374396,
      "loss": 5.0809,
      "mean_token_accuracy": 0.1923889234662056,
      "num_tokens": 31134258.0,
      "step": 15000
    },
    {
      "epoch": 1.36066763425254,
      "eval_entropy": 5.295666738009561,
      "eval_loss": 5.112661838531494,
      "eval_mean_token_accuracy": 0.2063279388017312,
      "eval_num_tokens": 31134258.0,
      "eval_runtime": 20.7071,
      "eval_samples_per_second": 1707.628,
      "eval_steps_per_second": 213.453,
      "step": 15000
    },
    {
      "entropy": 5.491491031646729,
      "epoch": 1.3611211901306242,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004819949845862498,
      "loss": 5.0584,
      "mean_token_accuracy": 0.2033733531832695,
      "num_tokens": 31144806.0,
      "step": 15005
    },
    {
      "entropy": 5.542679929733277,
      "epoch": 1.3615747460087082,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004819822996780466,
      "loss": 5.1565,
      "mean_token_accuracy": 0.1965965673327446,
      "num_tokens": 31155642.0,
      "step": 15010
    },
    {
      "entropy": 5.541298770904541,
      "epoch": 1.3620283018867925,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004819696104893967,
      "loss": 4.9988,
      "mean_token_accuracy": 0.20615871697664262,
      "num_tokens": 31165034.0,
      "step": 15015
    },
    {
      "entropy": 5.529401683807373,
      "epoch": 1.3624818577648767,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004819569170205625,
      "loss": 5.0616,
      "mean_token_accuracy": 0.1988907501101494,
      "num_tokens": 31176491.0,
      "step": 15020
    },
    {
      "entropy": 5.515815877914429,
      "epoch": 1.3629354136429608,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048194421927180654,
      "loss": 5.0767,
      "mean_token_accuracy": 0.1967637598514557,
      "num_tokens": 31186198.0,
      "step": 15025
    },
    {
      "entropy": 5.515884733200073,
      "epoch": 1.363388969521045,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004819315172433914,
      "loss": 5.0597,
      "mean_token_accuracy": 0.19971127957105636,
      "num_tokens": 31197094.0,
      "step": 15030
    },
    {
      "entropy": 5.552103185653687,
      "epoch": 1.363842525399129,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048191881093557977,
      "loss": 5.0781,
      "mean_token_accuracy": 0.19090546071529388,
      "num_tokens": 31207505.0,
      "step": 15035
    },
    {
      "entropy": 5.576592445373535,
      "epoch": 1.3642960812772134,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048190610034863434,
      "loss": 5.1897,
      "mean_token_accuracy": 0.18819540590047837,
      "num_tokens": 31218809.0,
      "step": 15040
    },
    {
      "entropy": 5.542851114273072,
      "epoch": 1.3647496371552976,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004818933854828181,
      "loss": 5.1072,
      "mean_token_accuracy": 0.19087555557489394,
      "num_tokens": 31230281.0,
      "step": 15045
    },
    {
      "entropy": 5.482969665527344,
      "epoch": 1.3652031930333817,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004818806663383939,
      "loss": 4.992,
      "mean_token_accuracy": 0.20569752156734467,
      "num_tokens": 31239573.0,
      "step": 15050
    },
    {
      "entropy": 5.471133422851563,
      "epoch": 1.365656748911466,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004818679429156248,
      "loss": 5.0185,
      "mean_token_accuracy": 0.2037968084216118,
      "num_tokens": 31249671.0,
      "step": 15055
    },
    {
      "entropy": 5.448171234130859,
      "epoch": 1.36611030478955,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048185521521477385,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19569021314382554,
      "num_tokens": 31259601.0,
      "step": 15060
    },
    {
      "entropy": 5.488245248794556,
      "epoch": 1.3665638606676342,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004818424832361044,
      "loss": 5.0233,
      "mean_token_accuracy": 0.2052937850356102,
      "num_tokens": 31269085.0,
      "step": 15065
    },
    {
      "entropy": 5.454237413406372,
      "epoch": 1.3670174165457185,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004818297469798797,
      "loss": 5.0559,
      "mean_token_accuracy": 0.19666333496570587,
      "num_tokens": 31280177.0,
      "step": 15070
    },
    {
      "entropy": 5.516110610961914,
      "epoch": 1.3674709724238026,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048181700644636314,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20633066445589066,
      "num_tokens": 31290545.0,
      "step": 15075
    },
    {
      "entropy": 5.434341859817505,
      "epoch": 1.3679245283018868,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048180426163581816,
      "loss": 4.9031,
      "mean_token_accuracy": 0.20979585647583007,
      "num_tokens": 31301143.0,
      "step": 15080
    },
    {
      "entropy": 5.4555473804473875,
      "epoch": 1.3683780841799709,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004817915125485083,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20275210589170456,
      "num_tokens": 31311071.0,
      "step": 15085
    },
    {
      "entropy": 5.551762104034424,
      "epoch": 1.3688316400580551,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048177875918469735,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19941299557685851,
      "num_tokens": 31322263.0,
      "step": 15090
    },
    {
      "entropy": 5.529943132400513,
      "epoch": 1.3692851959361394,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048176600154464897,
      "loss": 5.0144,
      "mean_token_accuracy": 0.1998817577958107,
      "num_tokens": 31332975.0,
      "step": 15095
    },
    {
      "entropy": 5.581736898422241,
      "epoch": 1.3697387518142234,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004817532396286269,
      "loss": 5.1303,
      "mean_token_accuracy": 0.1915099710226059,
      "num_tokens": 31345252.0,
      "step": 15100
    },
    {
      "entropy": 5.494141530990601,
      "epoch": 1.3701923076923077,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004817404734368952,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20686235874891282,
      "num_tokens": 31355479.0,
      "step": 15105
    },
    {
      "entropy": 5.5393835544586185,
      "epoch": 1.3706458635703918,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048172770296971785,
      "loss": 5.136,
      "mean_token_accuracy": 0.19656438529491424,
      "num_tokens": 31366020.0,
      "step": 15110
    },
    {
      "entropy": 5.513412237167358,
      "epoch": 1.371099419448476,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004817149282273589,
      "loss": 5.046,
      "mean_token_accuracy": 0.1999151214957237,
      "num_tokens": 31376055.0,
      "step": 15115
    },
    {
      "entropy": 5.548714685440063,
      "epoch": 1.3715529753265603,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004817021492100825,
      "loss": 5.178,
      "mean_token_accuracy": 0.18876050859689714,
      "num_tokens": 31386179.0,
      "step": 15120
    },
    {
      "entropy": 5.466830825805664,
      "epoch": 1.3720065312046443,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000481689365918153,
      "loss": 4.9776,
      "mean_token_accuracy": 0.20302161574363708,
      "num_tokens": 31395850.0,
      "step": 15125
    },
    {
      "entropy": 5.502423667907715,
      "epoch": 1.3724600870827286,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004816765783518348,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21120330691337585,
      "num_tokens": 31405053.0,
      "step": 15130
    },
    {
      "entropy": 5.439601993560791,
      "epoch": 1.3729136429608126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004816637865113923,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20858091861009598,
      "num_tokens": 31415153.0,
      "step": 15135
    },
    {
      "entropy": 5.446499586105347,
      "epoch": 1.373367198838897,
      "grad_norm": 1.0,
      "learning_rate": 0.00048165099039708994,
      "loss": 5.0592,
      "mean_token_accuracy": 0.19852253049612045,
      "num_tokens": 31425451.0,
      "step": 15140
    },
    {
      "entropy": 5.445763492584229,
      "epoch": 1.3738207547169812,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004816381900091925,
      "loss": 5.0296,
      "mean_token_accuracy": 0.20059703588485717,
      "num_tokens": 31436298.0,
      "step": 15145
    },
    {
      "entropy": 5.520010280609131,
      "epoch": 1.3742743105950654,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048162538534796465,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20523983836174012,
      "num_tokens": 31446932.0,
      "step": 15150
    },
    {
      "entropy": 5.551976919174194,
      "epoch": 1.3747278664731495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004816125764136711,
      "loss": 5.0848,
      "mean_token_accuracy": 0.19165421277284622,
      "num_tokens": 31457582.0,
      "step": 15155
    },
    {
      "entropy": 5.560914659500122,
      "epoch": 1.3751814223512338,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004815997632065769,
      "loss": 5.0433,
      "mean_token_accuracy": 0.201973794400692,
      "num_tokens": 31468168.0,
      "step": 15160
    },
    {
      "entropy": 5.4446817398071286,
      "epoch": 1.3756349782293178,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048158694572694694,
      "loss": 5.0329,
      "mean_token_accuracy": 0.207179257273674,
      "num_tokens": 31478524.0,
      "step": 15165
    },
    {
      "entropy": 5.456293964385987,
      "epoch": 1.376088534107402,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048157412397504623,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20530054569244385,
      "num_tokens": 31488608.0,
      "step": 15170
    },
    {
      "entropy": 5.493011617660523,
      "epoch": 1.3765420899854863,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004815612979511401,
      "loss": 5.0507,
      "mean_token_accuracy": 0.20859668999910355,
      "num_tokens": 31498984.0,
      "step": 15175
    },
    {
      "entropy": 5.5580322265625,
      "epoch": 1.3769956458635704,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048154846765549353,
      "loss": 5.0469,
      "mean_token_accuracy": 0.20434010922908782,
      "num_tokens": 31509110.0,
      "step": 15180
    },
    {
      "entropy": 5.479401540756226,
      "epoch": 1.3774492017416546,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048153563308837215,
      "loss": 5.0164,
      "mean_token_accuracy": 0.20211873799562455,
      "num_tokens": 31519266.0,
      "step": 15185
    },
    {
      "entropy": 5.491009759902954,
      "epoch": 1.3779027576197387,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048152279425004124,
      "loss": 5.0434,
      "mean_token_accuracy": 0.20235521346330643,
      "num_tokens": 31529043.0,
      "step": 15190
    },
    {
      "entropy": 5.520250082015991,
      "epoch": 1.378356313497823,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004815099511407663,
      "loss": 5.0376,
      "mean_token_accuracy": 0.20950424671173096,
      "num_tokens": 31538650.0,
      "step": 15195
    },
    {
      "entropy": 5.458363056182861,
      "epoch": 1.3788098693759072,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000481497103760813,
      "loss": 4.9837,
      "mean_token_accuracy": 0.20308187007904052,
      "num_tokens": 31549392.0,
      "step": 15200
    },
    {
      "entropy": 5.550771903991699,
      "epoch": 1.3792634252539913,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048148425211044684,
      "loss": 5.1663,
      "mean_token_accuracy": 0.18810785859823226,
      "num_tokens": 31559999.0,
      "step": 15205
    },
    {
      "entropy": 5.559471702575683,
      "epoch": 1.3797169811320755,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004814713961899338,
      "loss": 5.0556,
      "mean_token_accuracy": 0.1991157650947571,
      "num_tokens": 31570108.0,
      "step": 15210
    },
    {
      "entropy": 5.515550661087036,
      "epoch": 1.3801705370101596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048145853599953966,
      "loss": 5.0145,
      "mean_token_accuracy": 0.20343615710735322,
      "num_tokens": 31579698.0,
      "step": 15215
    },
    {
      "entropy": 5.457998752593994,
      "epoch": 1.3806240928882438,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048144567153953036,
      "loss": 5.0727,
      "mean_token_accuracy": 0.19539187401533126,
      "num_tokens": 31589292.0,
      "step": 15220
    },
    {
      "entropy": 5.559013748168946,
      "epoch": 1.381077648766328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048143280281017197,
      "loss": 5.1465,
      "mean_token_accuracy": 0.19716491550207138,
      "num_tokens": 31599848.0,
      "step": 15225
    },
    {
      "entropy": 5.541344213485718,
      "epoch": 1.3815312046444121,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004814199298117307,
      "loss": 5.0811,
      "mean_token_accuracy": 0.19916218519210815,
      "num_tokens": 31611109.0,
      "step": 15230
    },
    {
      "entropy": 5.4703916072845455,
      "epoch": 1.3819847605224964,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048140705254447256,
      "loss": 4.9954,
      "mean_token_accuracy": 0.203843155503273,
      "num_tokens": 31621140.0,
      "step": 15235
    },
    {
      "entropy": 5.583371543884278,
      "epoch": 1.3824383164005805,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048139417100866405,
      "loss": 5.0945,
      "mean_token_accuracy": 0.19653286635875702,
      "num_tokens": 31631268.0,
      "step": 15240
    },
    {
      "entropy": 5.551007652282715,
      "epoch": 1.3828918722786647,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048138128520457146,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20167967230081557,
      "num_tokens": 31640758.0,
      "step": 15245
    },
    {
      "entropy": 5.47036771774292,
      "epoch": 1.383345428156749,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004813683951324613,
      "loss": 5.0601,
      "mean_token_accuracy": 0.19761260747909545,
      "num_tokens": 31651060.0,
      "step": 15250
    },
    {
      "entropy": 5.516869258880615,
      "epoch": 1.383798984034833,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048135550079260014,
      "loss": 5.134,
      "mean_token_accuracy": 0.20071062743663787,
      "num_tokens": 31662282.0,
      "step": 15255
    },
    {
      "entropy": 5.504281425476075,
      "epoch": 1.3842525399129173,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004813426021852547,
      "loss": 5.0203,
      "mean_token_accuracy": 0.19429908096790313,
      "num_tokens": 31672364.0,
      "step": 15260
    },
    {
      "entropy": 5.517607593536377,
      "epoch": 1.3847060957910013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048132969931069156,
      "loss": 5.0738,
      "mean_token_accuracy": 0.1958681508898735,
      "num_tokens": 31683094.0,
      "step": 15265
    },
    {
      "entropy": 5.505269622802734,
      "epoch": 1.3851596516690856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004813167921691777,
      "loss": 5.0391,
      "mean_token_accuracy": 0.2022292897105217,
      "num_tokens": 31693087.0,
      "step": 15270
    },
    {
      "entropy": 5.5196010112762455,
      "epoch": 1.3856132075471699,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048130388076098003,
      "loss": 5.0301,
      "mean_token_accuracy": 0.19693872481584548,
      "num_tokens": 31703708.0,
      "step": 15275
    },
    {
      "entropy": 5.46891303062439,
      "epoch": 1.386066763425254,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048129096508636547,
      "loss": 4.9565,
      "mean_token_accuracy": 0.211345899105072,
      "num_tokens": 31713640.0,
      "step": 15280
    },
    {
      "entropy": 5.512295389175415,
      "epoch": 1.3865203193033382,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048127804514560126,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19729174673557281,
      "num_tokens": 31723526.0,
      "step": 15285
    },
    {
      "entropy": 5.444248962402344,
      "epoch": 1.3869738751814222,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048126512093895447,
      "loss": 4.9937,
      "mean_token_accuracy": 0.20083393156528473,
      "num_tokens": 31733265.0,
      "step": 15290
    },
    {
      "entropy": 5.519571924209595,
      "epoch": 1.3874274310595065,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004812521924666925,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20516951829195024,
      "num_tokens": 31743832.0,
      "step": 15295
    },
    {
      "entropy": 5.533500385284424,
      "epoch": 1.3878809869375908,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004812392597290825,
      "loss": 5.0691,
      "mean_token_accuracy": 0.2090713858604431,
      "num_tokens": 31754443.0,
      "step": 15300
    },
    {
      "entropy": 5.4089892387390135,
      "epoch": 1.3883345428156748,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004812263227263921,
      "loss": 5.019,
      "mean_token_accuracy": 0.20233907401561738,
      "num_tokens": 31765304.0,
      "step": 15305
    },
    {
      "entropy": 5.474977779388428,
      "epoch": 1.388788098693759,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004812133814588888,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19324804544448854,
      "num_tokens": 31775809.0,
      "step": 15310
    },
    {
      "entropy": 5.53752875328064,
      "epoch": 1.3892416545718431,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004812004359268402,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20598687529563903,
      "num_tokens": 31786271.0,
      "step": 15315
    },
    {
      "entropy": 5.527387714385986,
      "epoch": 1.3896952104499274,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004811874861305141,
      "loss": 5.0821,
      "mean_token_accuracy": 0.20123516917228698,
      "num_tokens": 31797358.0,
      "step": 15320
    },
    {
      "entropy": 5.475525903701782,
      "epoch": 1.3901487663280117,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004811745320701782,
      "loss": 5.0145,
      "mean_token_accuracy": 0.20617760568857194,
      "num_tokens": 31808303.0,
      "step": 15325
    },
    {
      "entropy": 5.433736848831177,
      "epoch": 1.390602322206096,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004811615737461004,
      "loss": 5.0313,
      "mean_token_accuracy": 0.20380660891532898,
      "num_tokens": 31818837.0,
      "step": 15330
    },
    {
      "entropy": 5.55443115234375,
      "epoch": 1.39105587808418,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048114861115854877,
      "loss": 5.0496,
      "mean_token_accuracy": 0.1971920147538185,
      "num_tokens": 31829729.0,
      "step": 15335
    },
    {
      "entropy": 5.499588584899902,
      "epoch": 1.3915094339622642,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048113564430779136,
      "loss": 4.9903,
      "mean_token_accuracy": 0.20167743265628815,
      "num_tokens": 31839996.0,
      "step": 15340
    },
    {
      "entropy": 5.401311779022217,
      "epoch": 1.3919629898403483,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004811226731940962,
      "loss": 4.9661,
      "mean_token_accuracy": 0.20950472056865693,
      "num_tokens": 31850298.0,
      "step": 15345
    },
    {
      "entropy": 5.48508653640747,
      "epoch": 1.3924165457184325,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048110969781773173,
      "loss": 5.0851,
      "mean_token_accuracy": 0.20816129744052886,
      "num_tokens": 31862170.0,
      "step": 15350
    },
    {
      "entropy": 5.5605326175689695,
      "epoch": 1.3928701015965168,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048109671817896614,
      "loss": 5.1177,
      "mean_token_accuracy": 0.20055409371852875,
      "num_tokens": 31872081.0,
      "step": 15355
    },
    {
      "entropy": 5.497061347961425,
      "epoch": 1.3933236574746009,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048108373427806797,
      "loss": 5.0376,
      "mean_token_accuracy": 0.2065151587128639,
      "num_tokens": 31882715.0,
      "step": 15360
    },
    {
      "entropy": 5.526398038864135,
      "epoch": 1.3937772133526851,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048107074611530555,
      "loss": 5.0203,
      "mean_token_accuracy": 0.20854706615209578,
      "num_tokens": 31891925.0,
      "step": 15365
    },
    {
      "entropy": 5.541769695281983,
      "epoch": 1.3942307692307692,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048105775369094756,
      "loss": 5.0347,
      "mean_token_accuracy": 0.20411615669727326,
      "num_tokens": 31902320.0,
      "step": 15370
    },
    {
      "entropy": 5.45694899559021,
      "epoch": 1.3946843251088534,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004810447570052628,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20837390571832656,
      "num_tokens": 31912726.0,
      "step": 15375
    },
    {
      "entropy": 5.457587814331054,
      "epoch": 1.3951378809869377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048103175605851997,
      "loss": 5.0727,
      "mean_token_accuracy": 0.19942245334386827,
      "num_tokens": 31924039.0,
      "step": 15380
    },
    {
      "entropy": 5.43327054977417,
      "epoch": 1.3955914368650217,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004810187508509879,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2109137788414955,
      "num_tokens": 31933280.0,
      "step": 15385
    },
    {
      "entropy": 5.482158946990967,
      "epoch": 1.396044992743106,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004810057413829355,
      "loss": 5.1531,
      "mean_token_accuracy": 0.19854603260755538,
      "num_tokens": 31943182.0,
      "step": 15390
    },
    {
      "entropy": 5.606261444091797,
      "epoch": 1.39649854862119,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048099272765463193,
      "loss": 5.0792,
      "mean_token_accuracy": 0.20472536087036133,
      "num_tokens": 31953256.0,
      "step": 15395
    },
    {
      "entropy": 5.608113241195679,
      "epoch": 1.3969521044992743,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004809797096663463,
      "loss": 5.1651,
      "mean_token_accuracy": 0.19737764447927475,
      "num_tokens": 31964231.0,
      "step": 15400
    },
    {
      "entropy": 5.514317989349365,
      "epoch": 1.3974056603773586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004809666874183477,
      "loss": 5.0173,
      "mean_token_accuracy": 0.20222418010234833,
      "num_tokens": 31974322.0,
      "step": 15405
    },
    {
      "entropy": 5.4731401920318605,
      "epoch": 1.3978592162554426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004809536609109056,
      "loss": 5.0691,
      "mean_token_accuracy": 0.2095978870987892,
      "num_tokens": 31984239.0,
      "step": 15410
    },
    {
      "entropy": 5.476027631759644,
      "epoch": 1.398312772133527,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048094063014428925,
      "loss": 5.0785,
      "mean_token_accuracy": 0.19442455768585204,
      "num_tokens": 31994366.0,
      "step": 15415
    },
    {
      "entropy": 5.566170167922974,
      "epoch": 1.398766328011611,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004809275951187682,
      "loss": 5.16,
      "mean_token_accuracy": 0.19522467702627183,
      "num_tokens": 32004921.0,
      "step": 15420
    },
    {
      "entropy": 5.466505813598633,
      "epoch": 1.3992198838896952,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048091455583461204,
      "loss": 5.0773,
      "mean_token_accuracy": 0.19962003231048583,
      "num_tokens": 32016385.0,
      "step": 15425
    },
    {
      "entropy": 5.520288276672363,
      "epoch": 1.3996734397677795,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048090151229209036,
      "loss": 5.0585,
      "mean_token_accuracy": 0.20249414145946504,
      "num_tokens": 32027398.0,
      "step": 15430
    },
    {
      "entropy": 5.5420714855194095,
      "epoch": 1.4001269956458635,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004808884644914729,
      "loss": 5.0636,
      "mean_token_accuracy": 0.20085171461105347,
      "num_tokens": 32038326.0,
      "step": 15435
    },
    {
      "entropy": 5.516499710083008,
      "epoch": 1.4005805515239478,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004808754124330296,
      "loss": 5.1277,
      "mean_token_accuracy": 0.19790878891944885,
      "num_tokens": 32048744.0,
      "step": 15440
    },
    {
      "entropy": 5.500717973709106,
      "epoch": 1.4010341074020318,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004808623561170302,
      "loss": 4.9933,
      "mean_token_accuracy": 0.20921683311462402,
      "num_tokens": 32059133.0,
      "step": 15445
    },
    {
      "entropy": 5.502972602844238,
      "epoch": 1.401487663280116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004808492955437449,
      "loss": 5.0357,
      "mean_token_accuracy": 0.20092233568429946,
      "num_tokens": 32069277.0,
      "step": 15450
    },
    {
      "entropy": 5.505456733703613,
      "epoch": 1.4019412191582004,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048083623071344374,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20528948158025742,
      "num_tokens": 32079802.0,
      "step": 15455
    },
    {
      "entropy": 5.508525562286377,
      "epoch": 1.4023947750362844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048082316162639685,
      "loss": 5.1042,
      "mean_token_accuracy": 0.1987621620297432,
      "num_tokens": 32090962.0,
      "step": 15460
    },
    {
      "entropy": 5.478303909301758,
      "epoch": 1.4028483309143687,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048081008828287447,
      "loss": 5.077,
      "mean_token_accuracy": 0.20106507390737532,
      "num_tokens": 32101058.0,
      "step": 15465
    },
    {
      "entropy": 5.55407304763794,
      "epoch": 1.4033018867924527,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048079701068314706,
      "loss": 5.0598,
      "mean_token_accuracy": 0.20511925965547562,
      "num_tokens": 32111763.0,
      "step": 15470
    },
    {
      "entropy": 5.53937177658081,
      "epoch": 1.403755442670537,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048078392882748507,
      "loss": 5.1272,
      "mean_token_accuracy": 0.2001139998435974,
      "num_tokens": 32123556.0,
      "step": 15475
    },
    {
      "entropy": 5.495906591415405,
      "epoch": 1.4042089985486212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048077084271615895,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20298214703798295,
      "num_tokens": 32134012.0,
      "step": 15480
    },
    {
      "entropy": 5.424602937698364,
      "epoch": 1.4046625544267053,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004807577523494394,
      "loss": 5.0046,
      "mean_token_accuracy": 0.2091123193502426,
      "num_tokens": 32144540.0,
      "step": 15485
    },
    {
      "entropy": 5.516141748428344,
      "epoch": 1.4051161103047896,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048074465772759707,
      "loss": 5.0977,
      "mean_token_accuracy": 0.20051607191562654,
      "num_tokens": 32154655.0,
      "step": 15490
    },
    {
      "entropy": 5.478973913192749,
      "epoch": 1.4055696661828736,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048073155885090276,
      "loss": 4.9875,
      "mean_token_accuracy": 0.20102933645248414,
      "num_tokens": 32165191.0,
      "step": 15495
    },
    {
      "entropy": 5.515355730056763,
      "epoch": 1.4060232220609579,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048071845571962733,
      "loss": 5.0933,
      "mean_token_accuracy": 0.1930096685886383,
      "num_tokens": 32175961.0,
      "step": 15500
    },
    {
      "entropy": 5.549034500122071,
      "epoch": 1.4064767779390421,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004807053483340419,
      "loss": 5.0587,
      "mean_token_accuracy": 0.2007862851023674,
      "num_tokens": 32187073.0,
      "step": 15505
    },
    {
      "entropy": 5.495439291000366,
      "epoch": 1.4069303338171264,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004806922366944175,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19969787448644638,
      "num_tokens": 32197426.0,
      "step": 15510
    },
    {
      "entropy": 5.536189460754395,
      "epoch": 1.4073838896952104,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048067912080102513,
      "loss": 5.032,
      "mean_token_accuracy": 0.2082691878080368,
      "num_tokens": 32206847.0,
      "step": 15515
    },
    {
      "entropy": 5.478679609298706,
      "epoch": 1.4078374455732947,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004806660006541362,
      "loss": 5.0616,
      "mean_token_accuracy": 0.1986638382077217,
      "num_tokens": 32217647.0,
      "step": 15520
    },
    {
      "entropy": 5.412083864212036,
      "epoch": 1.4082910014513788,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004806528762540219,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21444422006607056,
      "num_tokens": 32227133.0,
      "step": 15525
    },
    {
      "entropy": 5.521481990814209,
      "epoch": 1.408744557329463,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048063974760095374,
      "loss": 5.1197,
      "mean_token_accuracy": 0.19907931387424468,
      "num_tokens": 32237378.0,
      "step": 15530
    },
    {
      "entropy": 5.450902986526489,
      "epoch": 1.4091981132075473,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048062661469520327,
      "loss": 4.9959,
      "mean_token_accuracy": 0.2066905677318573,
      "num_tokens": 32247213.0,
      "step": 15535
    },
    {
      "entropy": 5.555349111557007,
      "epoch": 1.4096516690856313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048061347753704193,
      "loss": 5.0782,
      "mean_token_accuracy": 0.2022297203540802,
      "num_tokens": 32257100.0,
      "step": 15540
    },
    {
      "entropy": 5.5293927669525145,
      "epoch": 1.4101052249637156,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004806003361267415,
      "loss": 5.0423,
      "mean_token_accuracy": 0.1960933342576027,
      "num_tokens": 32267456.0,
      "step": 15545
    },
    {
      "entropy": 5.456585788726807,
      "epoch": 1.4105587808417996,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004805871904645737,
      "loss": 4.9779,
      "mean_token_accuracy": 0.2102513313293457,
      "num_tokens": 32278220.0,
      "step": 15550
    },
    {
      "entropy": 5.444500207901001,
      "epoch": 1.411012336719884,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004805740405508103,
      "loss": 5.0455,
      "mean_token_accuracy": 0.2062092110514641,
      "num_tokens": 32289036.0,
      "step": 15555
    },
    {
      "entropy": 5.468543577194214,
      "epoch": 1.4114658925979682,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004805608863857236,
      "loss": 4.9818,
      "mean_token_accuracy": 0.19995850175619126,
      "num_tokens": 32298909.0,
      "step": 15560
    },
    {
      "entropy": 5.513676548004151,
      "epoch": 1.4119194484760522,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.0129,
      "mean_token_accuracy": 0.20487444698810578,
      "num_tokens": 32309851.0,
      "step": 15565
    },
    {
      "entropy": 5.464004564285278,
      "epoch": 1.4123730043541365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004805345653026675,
      "loss": 4.99,
      "mean_token_accuracy": 0.21017162203788758,
      "num_tokens": 32319336.0,
      "step": 15570
    },
    {
      "entropy": 5.510065793991089,
      "epoch": 1.4128265602322205,
      "grad_norm": 1.0,
      "learning_rate": 0.00048052139838524263,
      "loss": 5.0422,
      "mean_token_accuracy": 0.20483591854572297,
      "num_tokens": 32329898.0,
      "step": 15575
    },
    {
      "entropy": 5.492889308929444,
      "epoch": 1.4132801161103048,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048050822721758274,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20117055475711823,
      "num_tokens": 32340100.0,
      "step": 15580
    },
    {
      "entropy": 5.4790565967559814,
      "epoch": 1.413733671988389,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004804950517999604,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19834206104278565,
      "num_tokens": 32349663.0,
      "step": 15585
    },
    {
      "entropy": 5.524758529663086,
      "epoch": 1.414187227866473,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004804818721326481,
      "loss": 5.0909,
      "mean_token_accuracy": 0.20059896111488343,
      "num_tokens": 32360461.0,
      "step": 15590
    },
    {
      "entropy": 5.545775079727173,
      "epoch": 1.4146407837445574,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004804686882159183,
      "loss": 5.0886,
      "mean_token_accuracy": 0.20205838233232498,
      "num_tokens": 32370910.0,
      "step": 15595
    },
    {
      "entropy": 5.578698396682739,
      "epoch": 1.4150943396226414,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048045550005004355,
      "loss": 5.0914,
      "mean_token_accuracy": 0.20864028632640838,
      "num_tokens": 32381669.0,
      "step": 15600
    },
    {
      "entropy": 5.462166213989258,
      "epoch": 1.4155478955007257,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004804423076352968,
      "loss": 5.0099,
      "mean_token_accuracy": 0.19899781793355942,
      "num_tokens": 32392024.0,
      "step": 15605
    },
    {
      "entropy": 5.416386508941651,
      "epoch": 1.41600145137881,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048042911097195075,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20628166496753692,
      "num_tokens": 32401872.0,
      "step": 15610
    },
    {
      "entropy": 5.426913595199585,
      "epoch": 1.416455007256894,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004804159100602784,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2084158256649971,
      "num_tokens": 32411749.0,
      "step": 15615
    },
    {
      "entropy": 5.584346342086792,
      "epoch": 1.4169085631349783,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048040270490055264,
      "loss": 5.1609,
      "mean_token_accuracy": 0.19357605278491974,
      "num_tokens": 32421358.0,
      "step": 15620
    },
    {
      "entropy": 5.543413543701172,
      "epoch": 1.4173621190130623,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004803894954930465,
      "loss": 5.055,
      "mean_token_accuracy": 0.19952864944934845,
      "num_tokens": 32432322.0,
      "step": 15625
    },
    {
      "entropy": 5.500397825241089,
      "epoch": 1.4178156748911466,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004803762818380334,
      "loss": 5.0745,
      "mean_token_accuracy": 0.19714537113904954,
      "num_tokens": 32440930.0,
      "step": 15630
    },
    {
      "entropy": 5.511843538284301,
      "epoch": 1.4182692307692308,
      "grad_norm": 1.0,
      "learning_rate": 0.00048036306393578644,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19900647252798082,
      "num_tokens": 32451124.0,
      "step": 15635
    },
    {
      "entropy": 5.549473667144776,
      "epoch": 1.4187227866473149,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004803498417865789,
      "loss": 5.0478,
      "mean_token_accuracy": 0.19772396683692933,
      "num_tokens": 32460636.0,
      "step": 15640
    },
    {
      "entropy": 5.521689081192017,
      "epoch": 1.4191763425253991,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004803366153906844,
      "loss": 5.0847,
      "mean_token_accuracy": 0.19507592916488647,
      "num_tokens": 32470501.0,
      "step": 15645
    },
    {
      "entropy": 5.53672423362732,
      "epoch": 1.4196298984034832,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004803233847483763,
      "loss": 5.1524,
      "mean_token_accuracy": 0.19808802306652068,
      "num_tokens": 32481559.0,
      "step": 15650
    },
    {
      "entropy": 5.6585461616516115,
      "epoch": 1.4200834542815675,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048031014985992825,
      "loss": 5.2276,
      "mean_token_accuracy": 0.19318990856409074,
      "num_tokens": 32493000.0,
      "step": 15655
    },
    {
      "entropy": 5.520326471328735,
      "epoch": 1.4205370101596517,
      "grad_norm": 1.03125,
      "learning_rate": 0.000480296910725614,
      "loss": 5.0037,
      "mean_token_accuracy": 0.21055828481912614,
      "num_tokens": 32502773.0,
      "step": 15660
    },
    {
      "entropy": 5.480014324188232,
      "epoch": 1.4209905660377358,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004802836673457074,
      "loss": 5.1155,
      "mean_token_accuracy": 0.19522669911384583,
      "num_tokens": 32513430.0,
      "step": 15665
    },
    {
      "entropy": 5.4377953052520756,
      "epoch": 1.42144412191582,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048027041972048225,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20099482089281082,
      "num_tokens": 32524409.0,
      "step": 15670
    },
    {
      "entropy": 5.553106594085693,
      "epoch": 1.421897677793904,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004802571678502124,
      "loss": 5.0725,
      "mean_token_accuracy": 0.19653349220752717,
      "num_tokens": 32534696.0,
      "step": 15675
    },
    {
      "entropy": 5.576407051086425,
      "epoch": 1.4223512336719883,
      "grad_norm": 0.9375,
      "learning_rate": 0.000480243911735172,
      "loss": 5.0582,
      "mean_token_accuracy": 0.20046114325523376,
      "num_tokens": 32544782.0,
      "step": 15680
    },
    {
      "entropy": 5.475587797164917,
      "epoch": 1.4228047895500726,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048023065137563534,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19665246605873107,
      "num_tokens": 32554818.0,
      "step": 15685
    },
    {
      "entropy": 5.473144388198852,
      "epoch": 1.4232583454281569,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004802173867718764,
      "loss": 5.0834,
      "mean_token_accuracy": 0.20439634025096892,
      "num_tokens": 32565007.0,
      "step": 15690
    },
    {
      "entropy": 5.523764944076538,
      "epoch": 1.423711901306241,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004802041179241697,
      "loss": 5.0382,
      "mean_token_accuracy": 0.19968950599431992,
      "num_tokens": 32575657.0,
      "step": 15695
    },
    {
      "entropy": 5.4944602966308596,
      "epoch": 1.4241654571843252,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004801908448327895,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19597385972738265,
      "num_tokens": 32586676.0,
      "step": 15700
    },
    {
      "entropy": 5.4961262226104735,
      "epoch": 1.4246190130624092,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004801775674980103,
      "loss": 5.021,
      "mean_token_accuracy": 0.20194448232650758,
      "num_tokens": 32596191.0,
      "step": 15705
    },
    {
      "entropy": 5.567626953125,
      "epoch": 1.4250725689404935,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048016428592010676,
      "loss": 5.0809,
      "mean_token_accuracy": 0.1960490435361862,
      "num_tokens": 32607178.0,
      "step": 15710
    },
    {
      "entropy": 5.550439023971558,
      "epoch": 1.4255261248185778,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048015100009935347,
      "loss": 5.0184,
      "mean_token_accuracy": 0.2020075500011444,
      "num_tokens": 32617301.0,
      "step": 15715
    },
    {
      "entropy": 5.430491018295288,
      "epoch": 1.4259796806966618,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004801377100360253,
      "loss": 5.0074,
      "mean_token_accuracy": 0.2087884947657585,
      "num_tokens": 32627395.0,
      "step": 15720
    },
    {
      "entropy": 5.501881122589111,
      "epoch": 1.426433236574746,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000480124415730397,
      "loss": 5.0287,
      "mean_token_accuracy": 0.206740340590477,
      "num_tokens": 32638460.0,
      "step": 15725
    },
    {
      "entropy": 5.589461088180542,
      "epoch": 1.4268867924528301,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004801111171827434,
      "loss": 5.0899,
      "mean_token_accuracy": 0.20110534280538558,
      "num_tokens": 32649319.0,
      "step": 15730
    },
    {
      "entropy": 5.498560190200806,
      "epoch": 1.4273403483309144,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004800978143933398,
      "loss": 5.0293,
      "mean_token_accuracy": 0.20413393080234526,
      "num_tokens": 32659976.0,
      "step": 15735
    },
    {
      "entropy": 5.433808660507202,
      "epoch": 1.4277939042089987,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048008450736246094,
      "loss": 4.9683,
      "mean_token_accuracy": 0.20098088234663009,
      "num_tokens": 32670772.0,
      "step": 15740
    },
    {
      "entropy": 5.5375607967376705,
      "epoch": 1.4282474600870827,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048007119609038237,
      "loss": 5.0968,
      "mean_token_accuracy": 0.19511828124523162,
      "num_tokens": 32682368.0,
      "step": 15745
    },
    {
      "entropy": 5.439204216003418,
      "epoch": 1.428701015965167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048005788057737907,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20723052322864532,
      "num_tokens": 32692649.0,
      "step": 15750
    },
    {
      "entropy": 5.533323192596436,
      "epoch": 1.429154571843251,
      "grad_norm": 1.0,
      "learning_rate": 0.0004800445608237266,
      "loss": 5.1454,
      "mean_token_accuracy": 0.20151389092206956,
      "num_tokens": 32704075.0,
      "step": 15755
    },
    {
      "entropy": 5.5563231945037845,
      "epoch": 1.4296081277213353,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048003123682970045,
      "loss": 5.1597,
      "mean_token_accuracy": 0.19443312734365464,
      "num_tokens": 32714670.0,
      "step": 15760
    },
    {
      "entropy": 5.512262058258057,
      "epoch": 1.4300616835994195,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048001790859557596,
      "loss": 4.9954,
      "mean_token_accuracy": 0.2086060270667076,
      "num_tokens": 32724601.0,
      "step": 15765
    },
    {
      "entropy": 5.458773851394653,
      "epoch": 1.4305152394775036,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004800045761216289,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20292164236307145,
      "num_tokens": 32735015.0,
      "step": 15770
    },
    {
      "entropy": 5.500267553329468,
      "epoch": 1.4309687953555879,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000479991239408135,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20426508635282517,
      "num_tokens": 32745899.0,
      "step": 15775
    },
    {
      "entropy": 5.536904525756836,
      "epoch": 1.431422351233672,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047997789845537003,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20640725493431092,
      "num_tokens": 32756052.0,
      "step": 15780
    },
    {
      "entropy": 5.4844502925872805,
      "epoch": 1.4318759071117562,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004799645532636098,
      "loss": 5.0253,
      "mean_token_accuracy": 0.2041822388768196,
      "num_tokens": 32767118.0,
      "step": 15785
    },
    {
      "entropy": 5.410877799987793,
      "epoch": 1.4323294629898404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047995120383313044,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21355968564748765,
      "num_tokens": 32777069.0,
      "step": 15790
    },
    {
      "entropy": 5.477860069274902,
      "epoch": 1.4327830188679245,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047993785016420796,
      "loss": 5.1263,
      "mean_token_accuracy": 0.1909132644534111,
      "num_tokens": 32786967.0,
      "step": 15795
    },
    {
      "entropy": 5.523899364471435,
      "epoch": 1.4332365747460087,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047992449225711854,
      "loss": 5.002,
      "mean_token_accuracy": 0.20324475318193436,
      "num_tokens": 32797563.0,
      "step": 15800
    },
    {
      "entropy": 5.5151763439178465,
      "epoch": 1.4336901306240928,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047991113011213837,
      "loss": 4.9764,
      "mean_token_accuracy": 0.20701471120119094,
      "num_tokens": 32808730.0,
      "step": 15805
    },
    {
      "entropy": 5.444163513183594,
      "epoch": 1.434143686502177,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004798977637295438,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20792941600084305,
      "num_tokens": 32818050.0,
      "step": 15810
    },
    {
      "entropy": 5.459983968734742,
      "epoch": 1.4345972423802613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004798843931096113,
      "loss": 5.0233,
      "mean_token_accuracy": 0.20102835297584534,
      "num_tokens": 32828993.0,
      "step": 15815
    },
    {
      "entropy": 5.4861907958984375,
      "epoch": 1.4350507982583454,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004798710182526173,
      "loss": 5.0203,
      "mean_token_accuracy": 0.20848754793405533,
      "num_tokens": 32839331.0,
      "step": 15820
    },
    {
      "entropy": 5.5293779373168945,
      "epoch": 1.4355043541364296,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047985763915883847,
      "loss": 5.0278,
      "mean_token_accuracy": 0.19959765970706939,
      "num_tokens": 32849946.0,
      "step": 15825
    },
    {
      "entropy": 5.438718461990357,
      "epoch": 1.4359579100145137,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004798442558285514,
      "loss": 4.965,
      "mean_token_accuracy": 0.2041826367378235,
      "num_tokens": 32859960.0,
      "step": 15830
    },
    {
      "entropy": 5.485565805435181,
      "epoch": 1.436411465892598,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004798308682620329,
      "loss": 4.9937,
      "mean_token_accuracy": 0.19982496798038482,
      "num_tokens": 32870317.0,
      "step": 15835
    },
    {
      "entropy": 5.587348508834839,
      "epoch": 1.4368650217706822,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047981747645955996,
      "loss": 5.1683,
      "mean_token_accuracy": 0.19669103622436523,
      "num_tokens": 32880906.0,
      "step": 15840
    },
    {
      "entropy": 5.593022584915161,
      "epoch": 1.4373185776487662,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004798040804214094,
      "loss": 5.1978,
      "mean_token_accuracy": 0.1926955610513687,
      "num_tokens": 32891038.0,
      "step": 15845
    },
    {
      "entropy": 5.585289764404297,
      "epoch": 1.4377721335268505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004797906801478582,
      "loss": 5.136,
      "mean_token_accuracy": 0.20103828907012938,
      "num_tokens": 32901013.0,
      "step": 15850
    },
    {
      "entropy": 5.58568844795227,
      "epoch": 1.4382256894049346,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004797772756391836,
      "loss": 5.0582,
      "mean_token_accuracy": 0.1981079399585724,
      "num_tokens": 32911592.0,
      "step": 15855
    },
    {
      "entropy": 5.4849381923675535,
      "epoch": 1.4386792452830188,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004797638668956627,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20855915248394014,
      "num_tokens": 32922087.0,
      "step": 15860
    },
    {
      "entropy": 5.482039165496826,
      "epoch": 1.439132801161103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004797504539175729,
      "loss": 5.0489,
      "mean_token_accuracy": 0.20400669425725937,
      "num_tokens": 32932039.0,
      "step": 15865
    },
    {
      "entropy": 5.499445915222168,
      "epoch": 1.4395863570391874,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004797370367051915,
      "loss": 5.0049,
      "mean_token_accuracy": 0.2056662067770958,
      "num_tokens": 32941896.0,
      "step": 15870
    },
    {
      "entropy": 5.50581727027893,
      "epoch": 1.4400399129172714,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000479723615258796,
      "loss": 5.0322,
      "mean_token_accuracy": 0.19427276104688646,
      "num_tokens": 32951675.0,
      "step": 15875
    },
    {
      "entropy": 5.531674242019653,
      "epoch": 1.4404934687953557,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047971018957866396,
      "loss": 5.1049,
      "mean_token_accuracy": 0.20822160840034484,
      "num_tokens": 32961378.0,
      "step": 15880
    },
    {
      "entropy": 5.497912979125976,
      "epoch": 1.4409470246734397,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047969675966507314,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20277913063764572,
      "num_tokens": 32971770.0,
      "step": 15885
    },
    {
      "entropy": 5.465924024581909,
      "epoch": 1.441400580551524,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000479683325518301,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2001161202788353,
      "num_tokens": 32982934.0,
      "step": 15890
    },
    {
      "entropy": 5.492756032943726,
      "epoch": 1.4418541364296082,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004796698871386257,
      "loss": 5.0294,
      "mean_token_accuracy": 0.1993359848856926,
      "num_tokens": 32993252.0,
      "step": 15895
    },
    {
      "entropy": 5.536276340484619,
      "epoch": 1.4423076923076923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004796564445263249,
      "loss": 5.0486,
      "mean_token_accuracy": 0.2013053610920906,
      "num_tokens": 33003610.0,
      "step": 15900
    },
    {
      "entropy": 5.499846506118774,
      "epoch": 1.4427612481857766,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047964299768167665,
      "loss": 5.0061,
      "mean_token_accuracy": 0.2022084638476372,
      "num_tokens": 33014037.0,
      "step": 15905
    },
    {
      "entropy": 5.495324516296387,
      "epoch": 1.4432148040638606,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004796295466049591,
      "loss": 5.031,
      "mean_token_accuracy": 0.1982549697160721,
      "num_tokens": 33023299.0,
      "step": 15910
    },
    {
      "entropy": 5.528369045257568,
      "epoch": 1.4436683599419449,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047961609129645036,
      "loss": 5.085,
      "mean_token_accuracy": 0.19658140540122987,
      "num_tokens": 33034194.0,
      "step": 15915
    },
    {
      "entropy": 5.492931365966797,
      "epoch": 1.4441219158200291,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047960263175642867,
      "loss": 4.9631,
      "mean_token_accuracy": 0.21053073406219483,
      "num_tokens": 33044715.0,
      "step": 15920
    },
    {
      "entropy": 5.429548358917236,
      "epoch": 1.4445754716981132,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004795891679851724,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21614315807819368,
      "num_tokens": 33055214.0,
      "step": 15925
    },
    {
      "entropy": 5.474163293838501,
      "epoch": 1.4450290275761974,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004795756999829601,
      "loss": 5.0022,
      "mean_token_accuracy": 0.20264820456504823,
      "num_tokens": 33065026.0,
      "step": 15930
    },
    {
      "entropy": 5.55479736328125,
      "epoch": 1.4454825834542815,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047956222775007016,
      "loss": 5.0705,
      "mean_token_accuracy": 0.1960279420018196,
      "num_tokens": 33075138.0,
      "step": 15935
    },
    {
      "entropy": 5.525855541229248,
      "epoch": 1.4459361393323658,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004795487512867812,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20468559861183167,
      "num_tokens": 33086625.0,
      "step": 15940
    },
    {
      "entropy": 5.53939790725708,
      "epoch": 1.44638969521045,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047953527059337194,
      "loss": 5.0861,
      "mean_token_accuracy": 0.19433015435934067,
      "num_tokens": 33097076.0,
      "step": 15945
    },
    {
      "entropy": 5.514823246002197,
      "epoch": 1.446843251088534,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004795217856701212,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20547079890966416,
      "num_tokens": 33107168.0,
      "step": 15950
    },
    {
      "entropy": 5.537642621994019,
      "epoch": 1.4472968069666183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047950829651730774,
      "loss": 5.0654,
      "mean_token_accuracy": 0.20469075739383696,
      "num_tokens": 33117862.0,
      "step": 15955
    },
    {
      "entropy": 5.523198652267456,
      "epoch": 1.4477503628447024,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004794948031352107,
      "loss": 4.9808,
      "mean_token_accuracy": 0.20375883877277373,
      "num_tokens": 33127144.0,
      "step": 15960
    },
    {
      "entropy": 5.587871456146241,
      "epoch": 1.4482039187227866,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047948130552410897,
      "loss": 5.1368,
      "mean_token_accuracy": 0.19235823452472686,
      "num_tokens": 33139180.0,
      "step": 15965
    },
    {
      "entropy": 5.546544790267944,
      "epoch": 1.448657474600871,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047946780368428175,
      "loss": 5.0157,
      "mean_token_accuracy": 0.1992461785674095,
      "num_tokens": 33149467.0,
      "step": 15970
    },
    {
      "entropy": 5.458764553070068,
      "epoch": 1.449111030478955,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004794542976160081,
      "loss": 4.8907,
      "mean_token_accuracy": 0.21425037384033202,
      "num_tokens": 33158752.0,
      "step": 15975
    },
    {
      "entropy": 5.4563695907592775,
      "epoch": 1.4495645863570392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047944078731956765,
      "loss": 4.978,
      "mean_token_accuracy": 0.2071639657020569,
      "num_tokens": 33168242.0,
      "step": 15980
    },
    {
      "entropy": 5.450711584091186,
      "epoch": 1.4500181422351233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047942727279523957,
      "loss": 4.995,
      "mean_token_accuracy": 0.20726195424795152,
      "num_tokens": 33179134.0,
      "step": 15985
    },
    {
      "entropy": 5.507195711135864,
      "epoch": 1.4504716981132075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004794137540433034,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19952171742916108,
      "num_tokens": 33188388.0,
      "step": 15990
    },
    {
      "entropy": 5.557099103927612,
      "epoch": 1.4509252539912918,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004794002310640386,
      "loss": 5.0144,
      "mean_token_accuracy": 0.2005619153380394,
      "num_tokens": 33198750.0,
      "step": 15995
    },
    {
      "entropy": 5.621443796157837,
      "epoch": 1.4513788098693758,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047938670385772503,
      "loss": 5.1347,
      "mean_token_accuracy": 0.19036090821027757,
      "num_tokens": 33209532.0,
      "step": 16000
    },
    {
      "entropy": 5.533468580245971,
      "epoch": 1.45183236574746,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047937317242464244,
      "loss": 4.9851,
      "mean_token_accuracy": 0.20237762480974197,
      "num_tokens": 33219675.0,
      "step": 16005
    },
    {
      "entropy": 5.518578052520752,
      "epoch": 1.4522859216255442,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004793596367650704,
      "loss": 5.0946,
      "mean_token_accuracy": 0.19858584105968474,
      "num_tokens": 33230919.0,
      "step": 16010
    },
    {
      "entropy": 5.495726633071899,
      "epoch": 1.4527394775036284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004793460968792891,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20685409456491471,
      "num_tokens": 33241146.0,
      "step": 16015
    },
    {
      "entropy": 5.535656595230103,
      "epoch": 1.4531930333817127,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047933255276757836,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19511209279298783,
      "num_tokens": 33251434.0,
      "step": 16020
    },
    {
      "entropy": 5.569311809539795,
      "epoch": 1.4536465892597967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004793190044302184,
      "loss": 5.1259,
      "mean_token_accuracy": 0.19259205162525178,
      "num_tokens": 33261868.0,
      "step": 16025
    },
    {
      "entropy": 5.501216983795166,
      "epoch": 1.454100145137881,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047930545186748936,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2036798745393753,
      "num_tokens": 33271798.0,
      "step": 16030
    },
    {
      "entropy": 5.495710563659668,
      "epoch": 1.454553701015965,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047929189507967153,
      "loss": 5.0757,
      "mean_token_accuracy": 0.20195723026990892,
      "num_tokens": 33282614.0,
      "step": 16035
    },
    {
      "entropy": 5.459695625305176,
      "epoch": 1.4550072568940493,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047927833406704516,
      "loss": 4.9892,
      "mean_token_accuracy": 0.19896757304668428,
      "num_tokens": 33292906.0,
      "step": 16040
    },
    {
      "entropy": 5.509294652938843,
      "epoch": 1.4554608127721336,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004792647688298909,
      "loss": 4.9944,
      "mean_token_accuracy": 0.20415859073400497,
      "num_tokens": 33302720.0,
      "step": 16045
    },
    {
      "entropy": 5.459511947631836,
      "epoch": 1.4559143686502178,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004792511993684891,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2102614790201187,
      "num_tokens": 33312959.0,
      "step": 16050
    },
    {
      "entropy": 5.46536283493042,
      "epoch": 1.4563679245283019,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047923762568312053,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20821652561426163,
      "num_tokens": 33322737.0,
      "step": 16055
    },
    {
      "entropy": 5.4693663120269775,
      "epoch": 1.4568214804063861,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004792240477740657,
      "loss": 4.9625,
      "mean_token_accuracy": 0.21078547537326814,
      "num_tokens": 33333185.0,
      "step": 16060
    },
    {
      "entropy": 5.519652271270752,
      "epoch": 1.4572750362844702,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004792104656416056,
      "loss": 5.0883,
      "mean_token_accuracy": 0.1959143102169037,
      "num_tokens": 33343414.0,
      "step": 16065
    },
    {
      "entropy": 5.496297025680542,
      "epoch": 1.4577285921625545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047919687928602104,
      "loss": 5.0021,
      "mean_token_accuracy": 0.2069736033678055,
      "num_tokens": 33352527.0,
      "step": 16070
    },
    {
      "entropy": 5.553343343734741,
      "epoch": 1.4581821480406387,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047918328870759293,
      "loss": 5.0628,
      "mean_token_accuracy": 0.19901107400655746,
      "num_tokens": 33363526.0,
      "step": 16075
    },
    {
      "entropy": 5.490126514434815,
      "epoch": 1.4586357039187228,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004791696939066024,
      "loss": 5.0925,
      "mean_token_accuracy": 0.20202584862709044,
      "num_tokens": 33374302.0,
      "step": 16080
    },
    {
      "entropy": 5.5225855827331545,
      "epoch": 1.459089259796807,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004791560948833306,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19944061189889908,
      "num_tokens": 33384786.0,
      "step": 16085
    },
    {
      "entropy": 5.548208236694336,
      "epoch": 1.459542815674891,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004791424916380586,
      "loss": 5.0331,
      "mean_token_accuracy": 0.2090217351913452,
      "num_tokens": 33394464.0,
      "step": 16090
    },
    {
      "entropy": 5.509014225006103,
      "epoch": 1.4599963715529753,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000479128884171068,
      "loss": 5.0608,
      "mean_token_accuracy": 0.20015767216682434,
      "num_tokens": 33405476.0,
      "step": 16095
    },
    {
      "entropy": 5.522106266021728,
      "epoch": 1.4604499274310596,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047911527248264003,
      "loss": 5.0873,
      "mean_token_accuracy": 0.20099602192640303,
      "num_tokens": 33415758.0,
      "step": 16100
    },
    {
      "entropy": 5.497673892974854,
      "epoch": 1.4609034833091437,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047910165657305613,
      "loss": 5.06,
      "mean_token_accuracy": 0.19588928818702697,
      "num_tokens": 33424845.0,
      "step": 16105
    },
    {
      "entropy": 5.547488403320313,
      "epoch": 1.461357039187228,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047908803644259804,
      "loss": 5.0802,
      "mean_token_accuracy": 0.20185463279485702,
      "num_tokens": 33435928.0,
      "step": 16110
    },
    {
      "entropy": 5.537172937393189,
      "epoch": 1.461810595065312,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047907441209154726,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19959668070077896,
      "num_tokens": 33447307.0,
      "step": 16115
    },
    {
      "entropy": 5.52874789237976,
      "epoch": 1.4622641509433962,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00047906078352018575,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19590460509061813,
      "num_tokens": 33458116.0,
      "step": 16120
    },
    {
      "entropy": 5.383693361282349,
      "epoch": 1.4627177068214805,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004790471507287952,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2084435611963272,
      "num_tokens": 33469065.0,
      "step": 16125
    },
    {
      "entropy": 5.535000514984131,
      "epoch": 1.4631712626995645,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047903351371765757,
      "loss": 5.0389,
      "mean_token_accuracy": 0.20379114598035813,
      "num_tokens": 33479118.0,
      "step": 16130
    },
    {
      "entropy": 5.554023694992066,
      "epoch": 1.4636248185776488,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047901987248705484,
      "loss": 5.064,
      "mean_token_accuracy": 0.19928945004940032,
      "num_tokens": 33489642.0,
      "step": 16135
    },
    {
      "entropy": 5.477730655670166,
      "epoch": 1.4640783744557329,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004790062270372693,
      "loss": 5.0816,
      "mean_token_accuracy": 0.19506554156541825,
      "num_tokens": 33499771.0,
      "step": 16140
    },
    {
      "entropy": 5.470936298370361,
      "epoch": 1.4645319303338171,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004789925773685828,
      "loss": 5.027,
      "mean_token_accuracy": 0.19897197037935258,
      "num_tokens": 33509975.0,
      "step": 16145
    },
    {
      "entropy": 5.5395927906036375,
      "epoch": 1.4649854862119014,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004789789234812779,
      "loss": 5.0686,
      "mean_token_accuracy": 0.1954951375722885,
      "num_tokens": 33520017.0,
      "step": 16150
    },
    {
      "entropy": 5.571431016921997,
      "epoch": 1.4654390420899854,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047896526537563686,
      "loss": 5.1316,
      "mean_token_accuracy": 0.1930692121386528,
      "num_tokens": 33531977.0,
      "step": 16155
    },
    {
      "entropy": 5.512016248703003,
      "epoch": 1.4658925979680697,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047895160305194223,
      "loss": 5.0951,
      "mean_token_accuracy": 0.19491859078407286,
      "num_tokens": 33543660.0,
      "step": 16160
    },
    {
      "entropy": 5.518211936950683,
      "epoch": 1.4663461538461537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004789379365104764,
      "loss": 5.0271,
      "mean_token_accuracy": 0.2014843299984932,
      "num_tokens": 33553814.0,
      "step": 16165
    },
    {
      "entropy": 5.52564640045166,
      "epoch": 1.466799709724238,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004789242657515221,
      "loss": 5.0457,
      "mean_token_accuracy": 0.2034329056739807,
      "num_tokens": 33564250.0,
      "step": 16170
    },
    {
      "entropy": 5.5023092269897464,
      "epoch": 1.4672532656023223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004789105907753621,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20267768502235411,
      "num_tokens": 33574755.0,
      "step": 16175
    },
    {
      "entropy": 5.480043220520019,
      "epoch": 1.4677068214804063,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000478896911582279,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2060309335589409,
      "num_tokens": 33585268.0,
      "step": 16180
    },
    {
      "entropy": 5.487579488754273,
      "epoch": 1.4681603773584906,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004788832281725558,
      "loss": 5.0891,
      "mean_token_accuracy": 0.2038735941052437,
      "num_tokens": 33595753.0,
      "step": 16185
    },
    {
      "entropy": 5.461152362823486,
      "epoch": 1.4686139332365746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004788695405464755,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20444866567850112,
      "num_tokens": 33605270.0,
      "step": 16190
    },
    {
      "entropy": 5.515757703781128,
      "epoch": 1.469067489114659,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047885584870432127,
      "loss": 5.0641,
      "mean_token_accuracy": 0.20079802423715593,
      "num_tokens": 33615096.0,
      "step": 16195
    },
    {
      "entropy": 5.5126025676727295,
      "epoch": 1.4695210449927432,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047884215264637606,
      "loss": 4.968,
      "mean_token_accuracy": 0.20840421617031096,
      "num_tokens": 33624770.0,
      "step": 16200
    },
    {
      "entropy": 5.545363903045654,
      "epoch": 1.4699746008708272,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004788284523729232,
      "loss": 5.1092,
      "mean_token_accuracy": 0.19470954835414886,
      "num_tokens": 33634545.0,
      "step": 16205
    },
    {
      "entropy": 5.569434070587159,
      "epoch": 1.4704281567489115,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047881474788424595,
      "loss": 5.0661,
      "mean_token_accuracy": 0.19960723966360092,
      "num_tokens": 33644950.0,
      "step": 16210
    },
    {
      "entropy": 5.549499034881592,
      "epoch": 1.4708817126269955,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047880103918062776,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20500340163707734,
      "num_tokens": 33654369.0,
      "step": 16215
    },
    {
      "entropy": 5.4006843090057375,
      "epoch": 1.4713352685050798,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047878732626235225,
      "loss": 5.0459,
      "mean_token_accuracy": 0.19897183328866958,
      "num_tokens": 33664687.0,
      "step": 16220
    },
    {
      "entropy": 5.475517129898071,
      "epoch": 1.471788824383164,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004787736091297028,
      "loss": 4.9914,
      "mean_token_accuracy": 0.2159730851650238,
      "num_tokens": 33675832.0,
      "step": 16225
    },
    {
      "entropy": 5.559790515899659,
      "epoch": 1.4722423802612483,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004787598877829633,
      "loss": 5.0893,
      "mean_token_accuracy": 0.20152770280838012,
      "num_tokens": 33686788.0,
      "step": 16230
    },
    {
      "entropy": 5.472647285461425,
      "epoch": 1.4726959361393324,
      "grad_norm": 1.125,
      "learning_rate": 0.0004787461622224173,
      "loss": 4.9776,
      "mean_token_accuracy": 0.20888569504022597,
      "num_tokens": 33696384.0,
      "step": 16235
    },
    {
      "entropy": 5.463289499282837,
      "epoch": 1.4731494920174166,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004787324324483488,
      "loss": 5.0216,
      "mean_token_accuracy": 0.2022559866309166,
      "num_tokens": 33707758.0,
      "step": 16240
    },
    {
      "entropy": 5.502991962432861,
      "epoch": 1.4736030478955007,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004787186984610416,
      "loss": 4.9811,
      "mean_token_accuracy": 0.2065095692873001,
      "num_tokens": 33717747.0,
      "step": 16245
    },
    {
      "entropy": 5.4916908740997314,
      "epoch": 1.474056603773585,
      "grad_norm": 1.0,
      "learning_rate": 0.00047870496026077984,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20281320214271545,
      "num_tokens": 33727960.0,
      "step": 16250
    },
    {
      "entropy": 5.448631668090821,
      "epoch": 1.4745101596516692,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004786912178478476,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20749327540397644,
      "num_tokens": 33738874.0,
      "step": 16255
    },
    {
      "entropy": 5.473556089401245,
      "epoch": 1.4749637155297532,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004786774712225291,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20806999057531356,
      "num_tokens": 33749071.0,
      "step": 16260
    },
    {
      "entropy": 5.497419309616089,
      "epoch": 1.4754172714078375,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004786637203851086,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20123066157102584,
      "num_tokens": 33759509.0,
      "step": 16265
    },
    {
      "entropy": 5.53020339012146,
      "epoch": 1.4758708272859216,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004786499653358705,
      "loss": 5.0372,
      "mean_token_accuracy": 0.20973011404275893,
      "num_tokens": 33769974.0,
      "step": 16270
    },
    {
      "entropy": 5.570294809341431,
      "epoch": 1.4763243831640058,
      "grad_norm": 1.0,
      "learning_rate": 0.0004786362060750991,
      "loss": 5.0746,
      "mean_token_accuracy": 0.20195408910512924,
      "num_tokens": 33780403.0,
      "step": 16275
    },
    {
      "entropy": 5.508374261856079,
      "epoch": 1.47677793904209,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047862244260307913,
      "loss": 5.0793,
      "mean_token_accuracy": 0.20363739430904387,
      "num_tokens": 33791364.0,
      "step": 16280
    },
    {
      "entropy": 5.478105258941651,
      "epoch": 1.4772314949201741,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004786086749200951,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20485219210386277,
      "num_tokens": 33801964.0,
      "step": 16285
    },
    {
      "entropy": 5.54928183555603,
      "epoch": 1.4776850507982584,
      "grad_norm": 1.0,
      "learning_rate": 0.0004785949030264319,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20498123168945312,
      "num_tokens": 33813132.0,
      "step": 16290
    },
    {
      "entropy": 5.491539859771729,
      "epoch": 1.4781386066763424,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004785811269223741,
      "loss": 5.0086,
      "mean_token_accuracy": 0.2109767660498619,
      "num_tokens": 33823069.0,
      "step": 16295
    },
    {
      "entropy": 5.547567510604859,
      "epoch": 1.4785921625544267,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004785673466082068,
      "loss": 5.0136,
      "mean_token_accuracy": 0.2017671912908554,
      "num_tokens": 33832626.0,
      "step": 16300
    },
    {
      "entropy": 5.47188663482666,
      "epoch": 1.479045718432511,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047855356208421484,
      "loss": 4.9957,
      "mean_token_accuracy": 0.20665343701839448,
      "num_tokens": 33843384.0,
      "step": 16305
    },
    {
      "entropy": 5.470773935317993,
      "epoch": 1.479499274310595,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004785397733506834,
      "loss": 4.9992,
      "mean_token_accuracy": 0.20403469055891038,
      "num_tokens": 33854286.0,
      "step": 16310
    },
    {
      "entropy": 5.445315361022949,
      "epoch": 1.4799528301886793,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004785259804078976,
      "loss": 4.965,
      "mean_token_accuracy": 0.2035436064004898,
      "num_tokens": 33863379.0,
      "step": 16315
    },
    {
      "entropy": 5.542968368530273,
      "epoch": 1.4804063860667633,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004785121832561426,
      "loss": 5.2038,
      "mean_token_accuracy": 0.1929345428943634,
      "num_tokens": 33875641.0,
      "step": 16320
    },
    {
      "entropy": 5.525194787979126,
      "epoch": 1.4808599419448476,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047849838189570375,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20928843170404435,
      "num_tokens": 33885023.0,
      "step": 16325
    },
    {
      "entropy": 5.618498134613037,
      "epoch": 1.4813134978229319,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004784845763268666,
      "loss": 5.1197,
      "mean_token_accuracy": 0.19594130516052247,
      "num_tokens": 33895562.0,
      "step": 16330
    },
    {
      "entropy": 5.488453531265259,
      "epoch": 1.481767053701016,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004784707665499165,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20698830634355544,
      "num_tokens": 33907095.0,
      "step": 16335
    },
    {
      "entropy": 5.407530736923218,
      "epoch": 1.4822206095791002,
      "grad_norm": 1.0,
      "learning_rate": 0.0004784569525651391,
      "loss": 4.8663,
      "mean_token_accuracy": 0.21668605953454972,
      "num_tokens": 33917475.0,
      "step": 16340
    },
    {
      "entropy": 5.445207118988037,
      "epoch": 1.4826741654571842,
      "grad_norm": 1.0,
      "learning_rate": 0.00047844313437282005,
      "loss": 5.0278,
      "mean_token_accuracy": 0.20546100437641143,
      "num_tokens": 33927740.0,
      "step": 16345
    },
    {
      "entropy": 5.458725690841675,
      "epoch": 1.4831277213352685,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047842931197324525,
      "loss": 4.9693,
      "mean_token_accuracy": 0.20588107407093048,
      "num_tokens": 33937416.0,
      "step": 16350
    },
    {
      "entropy": 5.481399345397949,
      "epoch": 1.4835812772133528,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047841548536670033,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21526847928762435,
      "num_tokens": 33946968.0,
      "step": 16355
    },
    {
      "entropy": 5.577416038513183,
      "epoch": 1.4840348330914368,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047840165455347136,
      "loss": 5.0829,
      "mean_token_accuracy": 0.19545600265264512,
      "num_tokens": 33957877.0,
      "step": 16360
    },
    {
      "entropy": 5.556639623641968,
      "epoch": 1.484488388969521,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047838781953384436,
      "loss": 5.0587,
      "mean_token_accuracy": 0.2009688213467598,
      "num_tokens": 33967890.0,
      "step": 16365
    },
    {
      "entropy": 5.463122367858887,
      "epoch": 1.484941944847605,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004783739803081054,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20946183651685715,
      "num_tokens": 33977117.0,
      "step": 16370
    },
    {
      "entropy": 5.498557996749878,
      "epoch": 1.4853955007256894,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047836013687654075,
      "loss": 5.1039,
      "mean_token_accuracy": 0.19548790007829667,
      "num_tokens": 33986943.0,
      "step": 16375
    },
    {
      "entropy": 5.518389129638672,
      "epoch": 1.4858490566037736,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047834628923943654,
      "loss": 5.0073,
      "mean_token_accuracy": 0.19906968772411346,
      "num_tokens": 33996499.0,
      "step": 16380
    },
    {
      "entropy": 5.553841161727905,
      "epoch": 1.4863026124818577,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004783324373970793,
      "loss": 5.1192,
      "mean_token_accuracy": 0.20226586610078812,
      "num_tokens": 34006749.0,
      "step": 16385
    },
    {
      "entropy": 5.565635728836059,
      "epoch": 1.486756168359942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004783185813497555,
      "loss": 5.1152,
      "mean_token_accuracy": 0.19189804196357726,
      "num_tokens": 34017194.0,
      "step": 16390
    },
    {
      "entropy": 5.573387622833252,
      "epoch": 1.487209724238026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004783047210977516,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20614250153303146,
      "num_tokens": 34027386.0,
      "step": 16395
    },
    {
      "entropy": 5.6038854122161865,
      "epoch": 1.4876632801161103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047829085664135424,
      "loss": 5.091,
      "mean_token_accuracy": 0.19450503885746,
      "num_tokens": 34037540.0,
      "step": 16400
    },
    {
      "entropy": 5.462180089950562,
      "epoch": 1.4881168359941945,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047827698798085015,
      "loss": 5.0099,
      "mean_token_accuracy": 0.20876026153564453,
      "num_tokens": 34048008.0,
      "step": 16405
    },
    {
      "entropy": 5.512149477005005,
      "epoch": 1.4885703918722788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004782631151165262,
      "loss": 5.0296,
      "mean_token_accuracy": 0.2054557114839554,
      "num_tokens": 34058009.0,
      "step": 16410
    },
    {
      "entropy": 5.547857332229614,
      "epoch": 1.4890239477503628,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047824923804866914,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20190584361553193,
      "num_tokens": 34067625.0,
      "step": 16415
    },
    {
      "entropy": 5.48768720626831,
      "epoch": 1.4894775036284469,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047823535677756607,
      "loss": 5.0204,
      "mean_token_accuracy": 0.19690577685832977,
      "num_tokens": 34077349.0,
      "step": 16420
    },
    {
      "entropy": 5.555475282669067,
      "epoch": 1.4899310595065312,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004782214713035041,
      "loss": 5.139,
      "mean_token_accuracy": 0.19575213491916657,
      "num_tokens": 34088636.0,
      "step": 16425
    },
    {
      "entropy": 5.538937330245972,
      "epoch": 1.4903846153846154,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047820758162677026,
      "loss": 5.0424,
      "mean_token_accuracy": 0.20493769496679307,
      "num_tokens": 34098806.0,
      "step": 16430
    },
    {
      "entropy": 5.502752494812012,
      "epoch": 1.4908381712626997,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047819368774765187,
      "loss": 5.0268,
      "mean_token_accuracy": 0.2061465784907341,
      "num_tokens": 34109713.0,
      "step": 16435
    },
    {
      "entropy": 5.475600719451904,
      "epoch": 1.4912917271407837,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047817978966643626,
      "loss": 4.9947,
      "mean_token_accuracy": 0.205531345307827,
      "num_tokens": 34119585.0,
      "step": 16440
    },
    {
      "entropy": 5.568090486526489,
      "epoch": 1.491745283018868,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004781658873834108,
      "loss": 5.0499,
      "mean_token_accuracy": 0.1953028440475464,
      "num_tokens": 34131245.0,
      "step": 16445
    },
    {
      "entropy": 5.523230075836182,
      "epoch": 1.492198838896952,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047815198089886296,
      "loss": 5.0271,
      "mean_token_accuracy": 0.2036787435412407,
      "num_tokens": 34141229.0,
      "step": 16450
    },
    {
      "entropy": 5.5111089706420895,
      "epoch": 1.4926523947750363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004781380702130805,
      "loss": 5.0672,
      "mean_token_accuracy": 0.19387389421463014,
      "num_tokens": 34151781.0,
      "step": 16455
    },
    {
      "entropy": 5.496467018127442,
      "epoch": 1.4931059506531206,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047812415532635087,
      "loss": 5.0182,
      "mean_token_accuracy": 0.20649980157613754,
      "num_tokens": 34162305.0,
      "step": 16460
    },
    {
      "entropy": 5.5562403202056885,
      "epoch": 1.4935595065312046,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000478110236238962,
      "loss": 5.0943,
      "mean_token_accuracy": 0.19907167553901672,
      "num_tokens": 34173613.0,
      "step": 16465
    },
    {
      "entropy": 5.488120412826538,
      "epoch": 1.4940130624092889,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047809631295120164,
      "loss": 5.021,
      "mean_token_accuracy": 0.20047862976789474,
      "num_tokens": 34183540.0,
      "step": 16470
    },
    {
      "entropy": 5.479513454437256,
      "epoch": 1.494466618287373,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047808238546335786,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20338803231716157,
      "num_tokens": 34193292.0,
      "step": 16475
    },
    {
      "entropy": 5.552792882919311,
      "epoch": 1.4949201741654572,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047806845377571856,
      "loss": 5.0713,
      "mean_token_accuracy": 0.21045733243227005,
      "num_tokens": 34202697.0,
      "step": 16480
    },
    {
      "entropy": 5.5124249935150145,
      "epoch": 1.4953737300435415,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047805451788857187,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20582942068576812,
      "num_tokens": 34213396.0,
      "step": 16485
    },
    {
      "entropy": 5.520803928375244,
      "epoch": 1.4958272859216255,
      "grad_norm": 1.15625,
      "learning_rate": 0.000478040577802206,
      "loss": 4.9824,
      "mean_token_accuracy": 0.20087953060865402,
      "num_tokens": 34223020.0,
      "step": 16490
    },
    {
      "entropy": 5.445318698883057,
      "epoch": 1.4962808417997098,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004780266335169093,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20431505143642426,
      "num_tokens": 34233249.0,
      "step": 16495
    },
    {
      "entropy": 5.581749963760376,
      "epoch": 1.4967343976777938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047801268503297,
      "loss": 5.0777,
      "mean_token_accuracy": 0.19654079228639604,
      "num_tokens": 34244091.0,
      "step": 16500
    },
    {
      "entropy": 5.560769701004029,
      "epoch": 1.497187953555878,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047799873235067667,
      "loss": 5.072,
      "mean_token_accuracy": 0.204006327688694,
      "num_tokens": 34255060.0,
      "step": 16505
    },
    {
      "entropy": 5.464819622039795,
      "epoch": 1.4976415094339623,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004779847754703178,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2080100730061531,
      "num_tokens": 34264361.0,
      "step": 16510
    },
    {
      "entropy": 5.492960977554321,
      "epoch": 1.4980950653120464,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004779708143921821,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20396582782268524,
      "num_tokens": 34274480.0,
      "step": 16515
    },
    {
      "entropy": 5.490897560119629,
      "epoch": 1.4985486211901307,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047795684911655824,
      "loss": 5.0171,
      "mean_token_accuracy": 0.20501819401979446,
      "num_tokens": 34285951.0,
      "step": 16520
    },
    {
      "entropy": 5.553942012786865,
      "epoch": 1.4990021770682147,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477942879643735,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19646671563386917,
      "num_tokens": 34296384.0,
      "step": 16525
    },
    {
      "entropy": 5.501747560501099,
      "epoch": 1.499455732946299,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047792890597400125,
      "loss": 5.0116,
      "mean_token_accuracy": 0.20746974349021913,
      "num_tokens": 34306776.0,
      "step": 16530
    },
    {
      "entropy": 5.557483053207397,
      "epoch": 1.4999092888243832,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047791492810764604,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20399336218833924,
      "num_tokens": 34317047.0,
      "step": 16535
    },
    {
      "entropy": 5.4808516025543215,
      "epoch": 1.5003628447024675,
      "grad_norm": 1.0,
      "learning_rate": 0.0004779009460449584,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21083079278469086,
      "num_tokens": 34326392.0,
      "step": 16540
    },
    {
      "entropy": 5.536378717422485,
      "epoch": 1.5008164005805515,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047788695978622756,
      "loss": 5.0283,
      "mean_token_accuracy": 0.206308513879776,
      "num_tokens": 34336580.0,
      "step": 16545
    },
    {
      "entropy": 5.4815671920776365,
      "epoch": 1.5012699564586356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047787296933174265,
      "loss": 5.0454,
      "mean_token_accuracy": 0.2009593978524208,
      "num_tokens": 34346792.0,
      "step": 16550
    },
    {
      "entropy": 5.531548976898193,
      "epoch": 1.5017235123367199,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000477858974681793,
      "loss": 5.073,
      "mean_token_accuracy": 0.19875523447990417,
      "num_tokens": 34356675.0,
      "step": 16555
    },
    {
      "entropy": 5.547983121871948,
      "epoch": 1.5021770682148041,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004778449758366681,
      "loss": 5.0216,
      "mean_token_accuracy": 0.2152870386838913,
      "num_tokens": 34366357.0,
      "step": 16560
    },
    {
      "entropy": 5.482605981826782,
      "epoch": 1.5026306240928884,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004778309727966574,
      "loss": 5.0302,
      "mean_token_accuracy": 0.2065482184290886,
      "num_tokens": 34377229.0,
      "step": 16565
    },
    {
      "entropy": 5.479886150360107,
      "epoch": 1.5030841799709724,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047781696556205045,
      "loss": 4.9632,
      "mean_token_accuracy": 0.2029298186302185,
      "num_tokens": 34387873.0,
      "step": 16570
    },
    {
      "entropy": 5.515002155303955,
      "epoch": 1.5035377358490565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047780295413313696,
      "loss": 4.9855,
      "mean_token_accuracy": 0.20847856849431992,
      "num_tokens": 34398889.0,
      "step": 16575
    },
    {
      "entropy": 5.46763858795166,
      "epoch": 1.5039912917271407,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047778893851020676,
      "loss": 4.9349,
      "mean_token_accuracy": 0.207319737970829,
      "num_tokens": 34408508.0,
      "step": 16580
    },
    {
      "entropy": 5.492764616012574,
      "epoch": 1.504444847605225,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004777749186935496,
      "loss": 5.0973,
      "mean_token_accuracy": 0.19956476986408234,
      "num_tokens": 34418928.0,
      "step": 16585
    },
    {
      "entropy": 5.5559296131134035,
      "epoch": 1.5048984034833093,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004777608946834555,
      "loss": 5.0717,
      "mean_token_accuracy": 0.20337532162666322,
      "num_tokens": 34429789.0,
      "step": 16590
    },
    {
      "entropy": 5.575822639465332,
      "epoch": 1.5053519593613933,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004777468664802144,
      "loss": 5.1585,
      "mean_token_accuracy": 0.19455186128616334,
      "num_tokens": 34440795.0,
      "step": 16595
    },
    {
      "entropy": 5.568664598464966,
      "epoch": 1.5058055152394774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047773283408411636,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19236605167388915,
      "num_tokens": 34450772.0,
      "step": 16600
    },
    {
      "entropy": 5.51489109992981,
      "epoch": 1.5062590711175616,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047771879749545175,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20160395056009292,
      "num_tokens": 34461273.0,
      "step": 16605
    },
    {
      "entropy": 5.520502233505249,
      "epoch": 1.506712626995646,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047770475671451073,
      "loss": 5.0666,
      "mean_token_accuracy": 0.20057346075773239,
      "num_tokens": 34470942.0,
      "step": 16610
    },
    {
      "entropy": 5.560528707504273,
      "epoch": 1.5071661828737302,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004776907117415837,
      "loss": 4.994,
      "mean_token_accuracy": 0.2034139022231102,
      "num_tokens": 34481117.0,
      "step": 16615
    },
    {
      "entropy": 5.469490098953247,
      "epoch": 1.5076197387518142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047767666257696103,
      "loss": 4.9706,
      "mean_token_accuracy": 0.20528270900249482,
      "num_tokens": 34490953.0,
      "step": 16620
    },
    {
      "entropy": 5.440951585769653,
      "epoch": 1.5080732946298983,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047766260922093337,
      "loss": 4.925,
      "mean_token_accuracy": 0.21110389679670333,
      "num_tokens": 34502765.0,
      "step": 16625
    },
    {
      "entropy": 5.6015337944030765,
      "epoch": 1.5085268505079825,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004776485516737913,
      "loss": 5.1148,
      "mean_token_accuracy": 0.19609806537628174,
      "num_tokens": 34514473.0,
      "step": 16630
    },
    {
      "entropy": 5.567944288253784,
      "epoch": 1.5089804063860668,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004776344899358256,
      "loss": 5.1329,
      "mean_token_accuracy": 0.1925711914896965,
      "num_tokens": 34524218.0,
      "step": 16635
    },
    {
      "entropy": 5.536440992355347,
      "epoch": 1.509433962264151,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004776204240073269,
      "loss": 5.0156,
      "mean_token_accuracy": 0.20336877703666686,
      "num_tokens": 34535697.0,
      "step": 16640
    },
    {
      "entropy": 5.533255910873413,
      "epoch": 1.509887518142235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047760635388858624,
      "loss": 5.0585,
      "mean_token_accuracy": 0.2072887033224106,
      "num_tokens": 34545130.0,
      "step": 16645
    },
    {
      "entropy": 5.475963020324707,
      "epoch": 1.5103410740203191,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047759227957989465,
      "loss": 4.9807,
      "mean_token_accuracy": 0.20716975182294844,
      "num_tokens": 34554974.0,
      "step": 16650
    },
    {
      "entropy": 5.521869230270386,
      "epoch": 1.5107946298984034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000477578201081543,
      "loss": 5.024,
      "mean_token_accuracy": 0.20081094801425933,
      "num_tokens": 34565660.0,
      "step": 16655
    },
    {
      "entropy": 5.5908403396606445,
      "epoch": 1.5112481857764877,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004775641183938225,
      "loss": 5.0622,
      "mean_token_accuracy": 0.1991403043270111,
      "num_tokens": 34576447.0,
      "step": 16660
    },
    {
      "entropy": 5.5940087795257565,
      "epoch": 1.511701741654572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047755003151702443,
      "loss": 5.1877,
      "mean_token_accuracy": 0.19777849465608596,
      "num_tokens": 34588333.0,
      "step": 16665
    },
    {
      "entropy": 5.453550243377686,
      "epoch": 1.512155297532656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047753594045144015,
      "loss": 4.9437,
      "mean_token_accuracy": 0.20184141397476196,
      "num_tokens": 34598166.0,
      "step": 16670
    },
    {
      "entropy": 5.485975837707519,
      "epoch": 1.5126088534107403,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004775218451973609,
      "loss": 4.968,
      "mean_token_accuracy": 0.20365387797355652,
      "num_tokens": 34608615.0,
      "step": 16675
    },
    {
      "entropy": 5.497578477859497,
      "epoch": 1.5130624092888243,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004775077457550784,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20195758193731309,
      "num_tokens": 34618618.0,
      "step": 16680
    },
    {
      "entropy": 5.5292384147644045,
      "epoch": 1.5135159651669086,
      "grad_norm": 1.0,
      "learning_rate": 0.00047749364212488403,
      "loss": 5.0519,
      "mean_token_accuracy": 0.19552540332078933,
      "num_tokens": 34629412.0,
      "step": 16685
    },
    {
      "entropy": 5.576356315612793,
      "epoch": 1.5139695210449928,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004774795343070696,
      "loss": 5.1105,
      "mean_token_accuracy": 0.20159600377082826,
      "num_tokens": 34640487.0,
      "step": 16690
    },
    {
      "entropy": 5.537057590484619,
      "epoch": 1.5144230769230769,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004774654223019268,
      "loss": 5.0397,
      "mean_token_accuracy": 0.20246966928243637,
      "num_tokens": 34650126.0,
      "step": 16695
    },
    {
      "entropy": 5.455569219589234,
      "epoch": 1.5148766328011611,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004774513061097474,
      "loss": 4.9078,
      "mean_token_accuracy": 0.20852980017662048,
      "num_tokens": 34659721.0,
      "step": 16700
    },
    {
      "entropy": 5.494598388671875,
      "epoch": 1.5153301886792452,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004774371857308235,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20723765790462495,
      "num_tokens": 34670611.0,
      "step": 16705
    },
    {
      "entropy": 5.525382232666016,
      "epoch": 1.5157837445573294,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047742306116544693,
      "loss": 5.007,
      "mean_token_accuracy": 0.2060346469283104,
      "num_tokens": 34680173.0,
      "step": 16710
    },
    {
      "entropy": 5.528834962844849,
      "epoch": 1.5162373004354137,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047740893241390994,
      "loss": 5.0667,
      "mean_token_accuracy": 0.2004269316792488,
      "num_tokens": 34690176.0,
      "step": 16715
    },
    {
      "entropy": 5.544941663742065,
      "epoch": 1.516690856313498,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004773947994765046,
      "loss": 4.9883,
      "mean_token_accuracy": 0.205637526512146,
      "num_tokens": 34700481.0,
      "step": 16720
    },
    {
      "entropy": 5.481034755706787,
      "epoch": 1.517144412191582,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004773806623535232,
      "loss": 5.1565,
      "mean_token_accuracy": 0.19801995754241944,
      "num_tokens": 34710643.0,
      "step": 16725
    },
    {
      "entropy": 5.548236465454101,
      "epoch": 1.517597968069666,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047736652104525816,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20002883970737456,
      "num_tokens": 34720998.0,
      "step": 16730
    },
    {
      "entropy": 5.598523807525635,
      "epoch": 1.5180515239477503,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004773523755520019,
      "loss": 5.0815,
      "mean_token_accuracy": 0.20067711025476456,
      "num_tokens": 34731071.0,
      "step": 16735
    },
    {
      "entropy": 5.519377326965332,
      "epoch": 1.5185050798258346,
      "grad_norm": 1.046875,
      "learning_rate": 0.000477338225874047,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20087372213602067,
      "num_tokens": 34741346.0,
      "step": 16740
    },
    {
      "entropy": 5.559559297561646,
      "epoch": 1.5189586357039189,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047732407201168595,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20293824523687362,
      "num_tokens": 34751092.0,
      "step": 16745
    },
    {
      "entropy": 5.604568767547607,
      "epoch": 1.519412191582003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004773099139652116,
      "loss": 5.0992,
      "mean_token_accuracy": 0.20174917578697205,
      "num_tokens": 34761477.0,
      "step": 16750
    },
    {
      "entropy": 5.471523380279541,
      "epoch": 1.519865747460087,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004772957517349167,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2046545147895813,
      "num_tokens": 34772491.0,
      "step": 16755
    },
    {
      "entropy": 5.5182709217071535,
      "epoch": 1.5203193033381712,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004772815853210939,
      "loss": 5.0698,
      "mean_token_accuracy": 0.19876878559589387,
      "num_tokens": 34783192.0,
      "step": 16760
    },
    {
      "entropy": 5.534299516677857,
      "epoch": 1.5207728592162555,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047726741472403655,
      "loss": 5.0353,
      "mean_token_accuracy": 0.2047033876180649,
      "num_tokens": 34794164.0,
      "step": 16765
    },
    {
      "entropy": 5.557278108596802,
      "epoch": 1.5212264150943398,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047725323994403753,
      "loss": 5.0829,
      "mean_token_accuracy": 0.19490733742713928,
      "num_tokens": 34804390.0,
      "step": 16770
    },
    {
      "entropy": 5.639868545532226,
      "epoch": 1.5216799709724238,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047723906098138993,
      "loss": 5.1752,
      "mean_token_accuracy": 0.19346656948328017,
      "num_tokens": 34814751.0,
      "step": 16775
    },
    {
      "entropy": 5.527458047866821,
      "epoch": 1.5221335268505078,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047722487783638697,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20670805424451827,
      "num_tokens": 34824977.0,
      "step": 16780
    },
    {
      "entropy": 5.530150556564331,
      "epoch": 1.5225870827285921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000477210690509322,
      "loss": 5.0629,
      "mean_token_accuracy": 0.20371615886688232,
      "num_tokens": 34835540.0,
      "step": 16785
    },
    {
      "entropy": 5.575498342514038,
      "epoch": 1.5230406386066764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047719649900048843,
      "loss": 5.0356,
      "mean_token_accuracy": 0.20107826590538025,
      "num_tokens": 34845233.0,
      "step": 16790
    },
    {
      "entropy": 5.594610977172851,
      "epoch": 1.5234941944847606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047718230331017975,
      "loss": 5.1333,
      "mean_token_accuracy": 0.19620306491851808,
      "num_tokens": 34855756.0,
      "step": 16795
    },
    {
      "entropy": 5.552404689788818,
      "epoch": 1.5239477503628447,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047716810343868956,
      "loss": 5.0549,
      "mean_token_accuracy": 0.19141069501638414,
      "num_tokens": 34867869.0,
      "step": 16800
    },
    {
      "entropy": 5.617313003540039,
      "epoch": 1.5244013062409287,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047715389938631143,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19923090040683747,
      "num_tokens": 34878625.0,
      "step": 16805
    },
    {
      "entropy": 5.596384859085083,
      "epoch": 1.524854862119013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047713969115333916,
      "loss": 5.101,
      "mean_token_accuracy": 0.1899292528629303,
      "num_tokens": 34889408.0,
      "step": 16810
    },
    {
      "entropy": 5.51461124420166,
      "epoch": 1.5253084179970973,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004771254787400666,
      "loss": 5.1016,
      "mean_token_accuracy": 0.19528281986713408,
      "num_tokens": 34899602.0,
      "step": 16815
    },
    {
      "entropy": 5.465745878219605,
      "epoch": 1.5257619738751815,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004771112621467876,
      "loss": 4.9482,
      "mean_token_accuracy": 0.20518770664930344,
      "num_tokens": 34910215.0,
      "step": 16820
    },
    {
      "entropy": 5.495348262786865,
      "epoch": 1.5262155297532656,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004770970413737961,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2078547567129135,
      "num_tokens": 34919641.0,
      "step": 16825
    },
    {
      "entropy": 5.527891826629639,
      "epoch": 1.5266690856313496,
      "grad_norm": 1.0,
      "learning_rate": 0.0004770828164213865,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2131982073187828,
      "num_tokens": 34930243.0,
      "step": 16830
    },
    {
      "entropy": 5.534064245223999,
      "epoch": 1.5271226415094339,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004770685872898525,
      "loss": 5.1001,
      "mean_token_accuracy": 0.20086862593889238,
      "num_tokens": 34941057.0,
      "step": 16835
    },
    {
      "entropy": 5.490671062469483,
      "epoch": 1.5275761973875182,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047705435397948876,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21125524044036864,
      "num_tokens": 34951150.0,
      "step": 16840
    },
    {
      "entropy": 5.574343013763428,
      "epoch": 1.5280297532656024,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004770401164905895,
      "loss": 5.0603,
      "mean_token_accuracy": 0.203806009888649,
      "num_tokens": 34961648.0,
      "step": 16845
    },
    {
      "entropy": 5.55496039390564,
      "epoch": 1.5284833091436865,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047702587482344915,
      "loss": 5.0569,
      "mean_token_accuracy": 0.1965765506029129,
      "num_tokens": 34971573.0,
      "step": 16850
    },
    {
      "entropy": 5.556545734405518,
      "epoch": 1.5289368650217707,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004770116289783622,
      "loss": 5.0458,
      "mean_token_accuracy": 0.1982572853565216,
      "num_tokens": 34981984.0,
      "step": 16855
    },
    {
      "entropy": 5.547101306915283,
      "epoch": 1.5293904208998548,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004769973789556233,
      "loss": 5.0634,
      "mean_token_accuracy": 0.1957728385925293,
      "num_tokens": 34990861.0,
      "step": 16860
    },
    {
      "entropy": 5.530113744735718,
      "epoch": 1.529843976777939,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047698312475552714,
      "loss": 5.0611,
      "mean_token_accuracy": 0.20197166949510575,
      "num_tokens": 35001504.0,
      "step": 16865
    },
    {
      "entropy": 5.549115228652954,
      "epoch": 1.5302975326560233,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004769688663783684,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2118733510375023,
      "num_tokens": 35012425.0,
      "step": 16870
    },
    {
      "entropy": 5.511075735092163,
      "epoch": 1.5307510885341074,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004769546038244421,
      "loss": 5.0061,
      "mean_token_accuracy": 0.21017370223999024,
      "num_tokens": 35022854.0,
      "step": 16875
    },
    {
      "entropy": 5.544015026092529,
      "epoch": 1.5312046444121916,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047694033709404313,
      "loss": 5.0909,
      "mean_token_accuracy": 0.196853344142437,
      "num_tokens": 35032694.0,
      "step": 16880
    },
    {
      "entropy": 5.522608804702759,
      "epoch": 1.5316582002902757,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047692606618746655,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20808610320091248,
      "num_tokens": 35042217.0,
      "step": 16885
    },
    {
      "entropy": 5.5298590660095215,
      "epoch": 1.53211175616836,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047691179110500735,
      "loss": 5.0889,
      "mean_token_accuracy": 0.1910834416747093,
      "num_tokens": 35052831.0,
      "step": 16890
    },
    {
      "entropy": 5.586901235580444,
      "epoch": 1.5325653120464442,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047689751184696097,
      "loss": 5.1123,
      "mean_token_accuracy": 0.1989368662238121,
      "num_tokens": 35061955.0,
      "step": 16895
    },
    {
      "entropy": 5.543337059020996,
      "epoch": 1.5330188679245285,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004768832284136226,
      "loss": 5.0147,
      "mean_token_accuracy": 0.2031451642513275,
      "num_tokens": 35072526.0,
      "step": 16900
    },
    {
      "entropy": 5.501619005203247,
      "epoch": 1.5334724238026125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004768689408052876,
      "loss": 4.9844,
      "mean_token_accuracy": 0.20547995269298552,
      "num_tokens": 35082415.0,
      "step": 16905
    },
    {
      "entropy": 5.475245141983033,
      "epoch": 1.5339259796806965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047685464902225144,
      "loss": 4.982,
      "mean_token_accuracy": 0.20835471153259277,
      "num_tokens": 35092805.0,
      "step": 16910
    },
    {
      "entropy": 5.542777490615845,
      "epoch": 1.5343795355587808,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004768403530648097,
      "loss": 5.0369,
      "mean_token_accuracy": 0.20715077519416808,
      "num_tokens": 35102963.0,
      "step": 16915
    },
    {
      "entropy": 5.54917802810669,
      "epoch": 1.534833091436865,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047682605293325806,
      "loss": 5.0578,
      "mean_token_accuracy": 0.2008623957633972,
      "num_tokens": 35114140.0,
      "step": 16920
    },
    {
      "entropy": 5.588230037689209,
      "epoch": 1.5352866473149493,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047681174862789215,
      "loss": 5.1263,
      "mean_token_accuracy": 0.20138452351093292,
      "num_tokens": 35125314.0,
      "step": 16925
    },
    {
      "entropy": 5.523447322845459,
      "epoch": 1.5357402031930334,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047679744014900795,
      "loss": 5.0299,
      "mean_token_accuracy": 0.2043992131948471,
      "num_tokens": 35134934.0,
      "step": 16930
    },
    {
      "entropy": 5.487507247924805,
      "epoch": 1.5361937590711174,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004767831274969011,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20349218994379042,
      "num_tokens": 35145341.0,
      "step": 16935
    },
    {
      "entropy": 5.455434989929199,
      "epoch": 1.5366473149492017,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047676881067186793,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2022953748703003,
      "num_tokens": 35156046.0,
      "step": 16940
    },
    {
      "entropy": 5.558455896377564,
      "epoch": 1.537100870827286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047675448967420426,
      "loss": 5.1179,
      "mean_token_accuracy": 0.19479937106370926,
      "num_tokens": 35168232.0,
      "step": 16945
    },
    {
      "entropy": 5.607294321060181,
      "epoch": 1.5375544267053702,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004767401645042064,
      "loss": 5.0675,
      "mean_token_accuracy": 0.20226863920688629,
      "num_tokens": 35178585.0,
      "step": 16950
    },
    {
      "entropy": 5.532871150970459,
      "epoch": 1.5380079825834543,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047672583516217043,
      "loss": 4.9313,
      "mean_token_accuracy": 0.20752260088920593,
      "num_tokens": 35189417.0,
      "step": 16955
    },
    {
      "entropy": 5.495153856277466,
      "epoch": 1.5384615384615383,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004767115016483928,
      "loss": 4.9993,
      "mean_token_accuracy": 0.2012806624174118,
      "num_tokens": 35200167.0,
      "step": 16960
    },
    {
      "entropy": 5.490826416015625,
      "epoch": 1.5389150943396226,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004766971639631699,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20241640508174896,
      "num_tokens": 35210711.0,
      "step": 16965
    },
    {
      "entropy": 5.481350374221802,
      "epoch": 1.5393686502177069,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047668282210679825,
      "loss": 5.0353,
      "mean_token_accuracy": 0.1973751053214073,
      "num_tokens": 35220866.0,
      "step": 16970
    },
    {
      "entropy": 5.4429432392120365,
      "epoch": 1.5398222060957911,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004766684760795745,
      "loss": 4.992,
      "mean_token_accuracy": 0.20854220241308213,
      "num_tokens": 35231478.0,
      "step": 16975
    },
    {
      "entropy": 5.476144599914551,
      "epoch": 1.5402757619738752,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047665412588179524,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21443094462156295,
      "num_tokens": 35242233.0,
      "step": 16980
    },
    {
      "entropy": 5.492010641098022,
      "epoch": 1.5407293178519592,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004766397715137573,
      "loss": 4.9949,
      "mean_token_accuracy": 0.20129960477352143,
      "num_tokens": 35253097.0,
      "step": 16985
    },
    {
      "entropy": 5.592983293533325,
      "epoch": 1.5411828737300435,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047662541297575743,
      "loss": 5.1592,
      "mean_token_accuracy": 0.19676952362060546,
      "num_tokens": 35264258.0,
      "step": 16990
    },
    {
      "entropy": 5.667860507965088,
      "epoch": 1.5416364296081277,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047661105026809274,
      "loss": 5.1293,
      "mean_token_accuracy": 0.19510892480611802,
      "num_tokens": 35273501.0,
      "step": 16995
    },
    {
      "entropy": 5.5120124340057375,
      "epoch": 1.542089985486212,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004765966833910601,
      "loss": 5.0757,
      "mean_token_accuracy": 0.19924862831830978,
      "num_tokens": 35284125.0,
      "step": 17000
    },
    {
      "entropy": 5.618932294845581,
      "epoch": 1.542543541364296,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004765823123449567,
      "loss": 5.073,
      "mean_token_accuracy": 0.1988307297229767,
      "num_tokens": 35296039.0,
      "step": 17005
    },
    {
      "entropy": 5.568522262573242,
      "epoch": 1.54299709724238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047656793713007965,
      "loss": 5.0089,
      "mean_token_accuracy": 0.19973995089530944,
      "num_tokens": 35306178.0,
      "step": 17010
    },
    {
      "entropy": 5.4798565864562985,
      "epoch": 1.5434506531204644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047655355774672636,
      "loss": 5.0433,
      "mean_token_accuracy": 0.20508571118116378,
      "num_tokens": 35316445.0,
      "step": 17015
    },
    {
      "entropy": 5.537961816787719,
      "epoch": 1.5439042089985486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047653917419519413,
      "loss": 5.0381,
      "mean_token_accuracy": 0.1992551326751709,
      "num_tokens": 35327274.0,
      "step": 17020
    },
    {
      "entropy": 5.526545619964599,
      "epoch": 1.544357764876633,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004765247864757804,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20724283307790756,
      "num_tokens": 35339007.0,
      "step": 17025
    },
    {
      "entropy": 5.55948657989502,
      "epoch": 1.544811320754717,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004765103945887827,
      "loss": 5.1208,
      "mean_token_accuracy": 0.19256114810705185,
      "num_tokens": 35349303.0,
      "step": 17030
    },
    {
      "entropy": 5.5475461959838865,
      "epoch": 1.5452648766328012,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004764959985344987,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2082339644432068,
      "num_tokens": 35359556.0,
      "step": 17035
    },
    {
      "entropy": 5.494559288024902,
      "epoch": 1.5457184325108853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004764815983132261,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20900323241949081,
      "num_tokens": 35370043.0,
      "step": 17040
    },
    {
      "entropy": 5.523154592514038,
      "epoch": 1.5461719883889695,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004764671939252628,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21071674376726152,
      "num_tokens": 35380031.0,
      "step": 17045
    },
    {
      "entropy": 5.532931280136109,
      "epoch": 1.5466255442670538,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004764527853709064,
      "loss": 5.0761,
      "mean_token_accuracy": 0.1993743047118187,
      "num_tokens": 35390481.0,
      "step": 17050
    },
    {
      "entropy": 5.51744589805603,
      "epoch": 1.5470791001451378,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004764383726504552,
      "loss": 5.06,
      "mean_token_accuracy": 0.20070597678422927,
      "num_tokens": 35401756.0,
      "step": 17055
    },
    {
      "entropy": 5.545562696456909,
      "epoch": 1.547532656023222,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004764239557642071,
      "loss": 4.9559,
      "mean_token_accuracy": 0.21284027099609376,
      "num_tokens": 35411045.0,
      "step": 17060
    },
    {
      "entropy": 5.501655769348145,
      "epoch": 1.5479862119013061,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004764095347124603,
      "loss": 5.0474,
      "mean_token_accuracy": 0.19385991096496583,
      "num_tokens": 35421256.0,
      "step": 17065
    },
    {
      "entropy": 5.5252477645874025,
      "epoch": 1.5484397677793904,
      "grad_norm": 1.0,
      "learning_rate": 0.0004763951094955129,
      "loss": 5.0484,
      "mean_token_accuracy": 0.20608636289834975,
      "num_tokens": 35430798.0,
      "step": 17070
    },
    {
      "entropy": 5.514478445053101,
      "epoch": 1.5488933236574747,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004763806801136634,
      "loss": 4.9919,
      "mean_token_accuracy": 0.2099059194326401,
      "num_tokens": 35440459.0,
      "step": 17075
    },
    {
      "entropy": 5.513672113418579,
      "epoch": 1.549346879535559,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004763662465672101,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20351307094097137,
      "num_tokens": 35451586.0,
      "step": 17080
    },
    {
      "entropy": 5.521123647689819,
      "epoch": 1.549800435413643,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047635180885645146,
      "loss": 5.0499,
      "mean_token_accuracy": 0.2032095104455948,
      "num_tokens": 35462197.0,
      "step": 17085
    },
    {
      "entropy": 5.549882459640503,
      "epoch": 1.550253991291727,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004763373669816862,
      "loss": 4.9743,
      "mean_token_accuracy": 0.20916669368743895,
      "num_tokens": 35471737.0,
      "step": 17090
    },
    {
      "entropy": 5.562024736404419,
      "epoch": 1.5507075471698113,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004763229209432128,
      "loss": 5.0796,
      "mean_token_accuracy": 0.1963101416826248,
      "num_tokens": 35482016.0,
      "step": 17095
    },
    {
      "entropy": 5.572429990768432,
      "epoch": 1.5511611030478956,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004763084707413301,
      "loss": 5.1605,
      "mean_token_accuracy": 0.18899236917495726,
      "num_tokens": 35492622.0,
      "step": 17100
    },
    {
      "entropy": 5.624724960327148,
      "epoch": 1.5516146589259798,
      "grad_norm": 1.125,
      "learning_rate": 0.00047629401637633696,
      "loss": 5.1854,
      "mean_token_accuracy": 0.19349545240402222,
      "num_tokens": 35504468.0,
      "step": 17105
    },
    {
      "entropy": 5.528221845626831,
      "epoch": 1.5520682148040639,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047627955784853226,
      "loss": 5.0141,
      "mean_token_accuracy": 0.199432609975338,
      "num_tokens": 35514272.0,
      "step": 17110
    },
    {
      "entropy": 5.519846725463867,
      "epoch": 1.552521770682148,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000476265095158215,
      "loss": 5.0043,
      "mean_token_accuracy": 0.1996614083647728,
      "num_tokens": 35524110.0,
      "step": 17115
    },
    {
      "entropy": 5.548833322525025,
      "epoch": 1.5529753265602322,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004762506283056843,
      "loss": 4.9249,
      "mean_token_accuracy": 0.21215675473213197,
      "num_tokens": 35535504.0,
      "step": 17120
    },
    {
      "entropy": 5.624251079559326,
      "epoch": 1.5534288824383164,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004762361572912393,
      "loss": 5.1219,
      "mean_token_accuracy": 0.19390741139650344,
      "num_tokens": 35544885.0,
      "step": 17125
    },
    {
      "entropy": 5.543760204315186,
      "epoch": 1.5538824383164007,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047622168211517926,
      "loss": 5.1441,
      "mean_token_accuracy": 0.20457107722759246,
      "num_tokens": 35555485.0,
      "step": 17130
    },
    {
      "entropy": 5.468977546691894,
      "epoch": 1.5543359941944848,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047620720277780353,
      "loss": 4.9531,
      "mean_token_accuracy": 0.20908924788236619,
      "num_tokens": 35567197.0,
      "step": 17135
    },
    {
      "entropy": 5.547029113769531,
      "epoch": 1.5547895500725688,
      "grad_norm": 1.0,
      "learning_rate": 0.00047619271927941165,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20580670535564421,
      "num_tokens": 35577576.0,
      "step": 17140
    },
    {
      "entropy": 5.532487440109253,
      "epoch": 1.555243105950653,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000476178231620303,
      "loss": 5.0558,
      "mean_token_accuracy": 0.2027304396033287,
      "num_tokens": 35588012.0,
      "step": 17145
    },
    {
      "entropy": 5.498031139373779,
      "epoch": 1.5556966618287373,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004761637398007774,
      "loss": 5.0329,
      "mean_token_accuracy": 0.19690879434347153,
      "num_tokens": 35598076.0,
      "step": 17150
    },
    {
      "entropy": 5.444629192352295,
      "epoch": 1.5561502177068216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004761492438211342,
      "loss": 4.8629,
      "mean_token_accuracy": 0.2136305958032608,
      "num_tokens": 35607170.0,
      "step": 17155
    },
    {
      "entropy": 5.514594650268554,
      "epoch": 1.5566037735849056,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047613474368167355,
      "loss": 4.9972,
      "mean_token_accuracy": 0.1959199920296669,
      "num_tokens": 35617865.0,
      "step": 17160
    },
    {
      "entropy": 5.528052568435669,
      "epoch": 1.5570573294629897,
      "grad_norm": 1.0625,
      "learning_rate": 0.000476120239382695,
      "loss": 5.0517,
      "mean_token_accuracy": 0.20201874971389772,
      "num_tokens": 35629329.0,
      "step": 17165
    },
    {
      "entropy": 5.517793083190918,
      "epoch": 1.557510885341074,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047610573092449877,
      "loss": 5.0536,
      "mean_token_accuracy": 0.19934562891721724,
      "num_tokens": 35640044.0,
      "step": 17170
    },
    {
      "entropy": 5.552383804321289,
      "epoch": 1.5579644412191582,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004760912183073847,
      "loss": 5.0881,
      "mean_token_accuracy": 0.20094917714595795,
      "num_tokens": 35650435.0,
      "step": 17175
    },
    {
      "entropy": 5.532912158966065,
      "epoch": 1.5584179970972425,
      "grad_norm": 0.953125,
      "learning_rate": 0.000476076701531653,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20271799117326736,
      "num_tokens": 35659613.0,
      "step": 17180
    },
    {
      "entropy": 5.494220638275147,
      "epoch": 1.5588715529753265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004760621805976039,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20519281178712845,
      "num_tokens": 35670987.0,
      "step": 17185
    },
    {
      "entropy": 5.498604202270508,
      "epoch": 1.5593251088534106,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047604765550553766,
      "loss": 5.0172,
      "mean_token_accuracy": 0.20504542142152787,
      "num_tokens": 35682505.0,
      "step": 17190
    },
    {
      "entropy": 5.552568626403809,
      "epoch": 1.5597786647314948,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047603312625575466,
      "loss": 4.9551,
      "mean_token_accuracy": 0.20232332348823548,
      "num_tokens": 35693056.0,
      "step": 17195
    },
    {
      "entropy": 5.518586921691894,
      "epoch": 1.5602322206095791,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047601859284855544,
      "loss": 4.9631,
      "mean_token_accuracy": 0.21319104433059693,
      "num_tokens": 35703659.0,
      "step": 17200
    },
    {
      "entropy": 5.4766816139221195,
      "epoch": 1.5606857764876634,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004760040552842403,
      "loss": 5.025,
      "mean_token_accuracy": 0.2015480637550354,
      "num_tokens": 35714367.0,
      "step": 17205
    },
    {
      "entropy": 5.4776341915130615,
      "epoch": 1.5611393323657474,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047598951356311035,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2010535180568695,
      "num_tokens": 35724773.0,
      "step": 17210
    },
    {
      "entropy": 5.569267797470093,
      "epoch": 1.5615928882438317,
      "grad_norm": 1.0,
      "learning_rate": 0.0004759749676854658,
      "loss": 5.1312,
      "mean_token_accuracy": 0.19799581170082092,
      "num_tokens": 35735622.0,
      "step": 17215
    },
    {
      "entropy": 5.543905401229859,
      "epoch": 1.5620464441219157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004759604176516079,
      "loss": 4.9875,
      "mean_token_accuracy": 0.2084849089384079,
      "num_tokens": 35744967.0,
      "step": 17220
    },
    {
      "entropy": 5.555732345581054,
      "epoch": 1.5625,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004759458634618372,
      "loss": 5.0294,
      "mean_token_accuracy": 0.20423816591501237,
      "num_tokens": 35755647.0,
      "step": 17225
    },
    {
      "entropy": 5.529198217391968,
      "epoch": 1.5629535558780843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004759313051164549,
      "loss": 5.063,
      "mean_token_accuracy": 0.1968506619334221,
      "num_tokens": 35765595.0,
      "step": 17230
    },
    {
      "entropy": 5.545705795288086,
      "epoch": 1.5634071117561683,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000475916742615762,
      "loss": 5.0265,
      "mean_token_accuracy": 0.2100907638669014,
      "num_tokens": 35775804.0,
      "step": 17235
    },
    {
      "entropy": 5.504986000061035,
      "epoch": 1.5638606676342526,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004759021759600597,
      "loss": 4.9768,
      "mean_token_accuracy": 0.20709893405437468,
      "num_tokens": 35786227.0,
      "step": 17240
    },
    {
      "entropy": 5.567078065872193,
      "epoch": 1.5643142235123366,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004758876051496492,
      "loss": 5.1063,
      "mean_token_accuracy": 0.20364011228084564,
      "num_tokens": 35797511.0,
      "step": 17245
    },
    {
      "entropy": 5.560240745544434,
      "epoch": 1.5647677793904209,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047587303018483177,
      "loss": 5.051,
      "mean_token_accuracy": 0.19859770983457564,
      "num_tokens": 35807248.0,
      "step": 17250
    },
    {
      "entropy": 5.52660493850708,
      "epoch": 1.5652213352685052,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047585845106590897,
      "loss": 5.0472,
      "mean_token_accuracy": 0.19810502827167512,
      "num_tokens": 35818164.0,
      "step": 17255
    },
    {
      "entropy": 5.515972757339478,
      "epoch": 1.5656748911465894,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047584386779318214,
      "loss": 5.0112,
      "mean_token_accuracy": 0.2109167367219925,
      "num_tokens": 35828746.0,
      "step": 17260
    },
    {
      "entropy": 5.546813488006592,
      "epoch": 1.5661284470246735,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000475829280366953,
      "loss": 5.0758,
      "mean_token_accuracy": 0.1961701601743698,
      "num_tokens": 35839325.0,
      "step": 17265
    },
    {
      "entropy": 5.536845350265503,
      "epoch": 1.5665820029027575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047581468878752313,
      "loss": 5.001,
      "mean_token_accuracy": 0.20544881075620652,
      "num_tokens": 35849406.0,
      "step": 17270
    },
    {
      "entropy": 5.454688835144043,
      "epoch": 1.5670355587808418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047580009305519427,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2047882094979286,
      "num_tokens": 35858715.0,
      "step": 17275
    },
    {
      "entropy": 5.4782956600189205,
      "epoch": 1.567489114658926,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004757854931702684,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2048191696405411,
      "num_tokens": 35869063.0,
      "step": 17280
    },
    {
      "entropy": 5.48824028968811,
      "epoch": 1.5679426705370103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047577088913304733,
      "loss": 4.9003,
      "mean_token_accuracy": 0.20872622430324556,
      "num_tokens": 35878428.0,
      "step": 17285
    },
    {
      "entropy": 5.558001232147217,
      "epoch": 1.5683962264150944,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047575628094383317,
      "loss": 5.035,
      "mean_token_accuracy": 0.2058235377073288,
      "num_tokens": 35888801.0,
      "step": 17290
    },
    {
      "entropy": 5.473654413223267,
      "epoch": 1.5688497822931784,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004757416686029279,
      "loss": 5.0448,
      "mean_token_accuracy": 0.20353106558322906,
      "num_tokens": 35899381.0,
      "step": 17295
    },
    {
      "entropy": 5.556720638275147,
      "epoch": 1.5693033381712627,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004757270521106338,
      "loss": 5.0738,
      "mean_token_accuracy": 0.20061299651861192,
      "num_tokens": 35909546.0,
      "step": 17300
    },
    {
      "entropy": 5.520714902877808,
      "epoch": 1.569756894049347,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004757124314672531,
      "loss": 5.0466,
      "mean_token_accuracy": 0.2015165463089943,
      "num_tokens": 35919301.0,
      "step": 17305
    },
    {
      "entropy": 5.58653039932251,
      "epoch": 1.5702104499274312,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047569780667308816,
      "loss": 5.0878,
      "mean_token_accuracy": 0.20005113929510115,
      "num_tokens": 35928603.0,
      "step": 17310
    },
    {
      "entropy": 5.58700156211853,
      "epoch": 1.5706640058055152,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047568317772844147,
      "loss": 5.067,
      "mean_token_accuracy": 0.20702016204595566,
      "num_tokens": 35939005.0,
      "step": 17315
    },
    {
      "entropy": 5.575606870651245,
      "epoch": 1.5711175616835993,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047566854463361544,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20989421457052232,
      "num_tokens": 35948471.0,
      "step": 17320
    },
    {
      "entropy": 5.5107934951782225,
      "epoch": 1.5715711175616836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004756539073889128,
      "loss": 5.0578,
      "mean_token_accuracy": 0.20375427305698396,
      "num_tokens": 35958488.0,
      "step": 17325
    },
    {
      "entropy": 5.51311354637146,
      "epoch": 1.5720246734397678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047563926599463623,
      "loss": 5.1086,
      "mean_token_accuracy": 0.19656799137592315,
      "num_tokens": 35968291.0,
      "step": 17330
    },
    {
      "entropy": 5.564052724838257,
      "epoch": 1.572478229317852,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004756246204510886,
      "loss": 5.0249,
      "mean_token_accuracy": 0.2046962261199951,
      "num_tokens": 35978565.0,
      "step": 17335
    },
    {
      "entropy": 5.559260129928589,
      "epoch": 1.5729317851959361,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047560997075857253,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20554231852293015,
      "num_tokens": 35987483.0,
      "step": 17340
    },
    {
      "entropy": 5.522519826889038,
      "epoch": 1.5733853410740202,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004755953169173913,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19977228343486786,
      "num_tokens": 35998355.0,
      "step": 17345
    },
    {
      "entropy": 5.510653686523438,
      "epoch": 1.5738388969521044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004755806589278477,
      "loss": 5.0015,
      "mean_token_accuracy": 0.2097836673259735,
      "num_tokens": 36008365.0,
      "step": 17350
    },
    {
      "entropy": 5.576305627822876,
      "epoch": 1.5742924528301887,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047556599679024497,
      "loss": 5.0831,
      "mean_token_accuracy": 0.19741983562707902,
      "num_tokens": 36018880.0,
      "step": 17355
    },
    {
      "entropy": 5.55736198425293,
      "epoch": 1.574746008708273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047555133050488637,
      "loss": 5.0977,
      "mean_token_accuracy": 0.19512535780668258,
      "num_tokens": 36029335.0,
      "step": 17360
    },
    {
      "entropy": 5.5182085037231445,
      "epoch": 1.575199564586357,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047553666007207516,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20793434828519822,
      "num_tokens": 36039774.0,
      "step": 17365
    },
    {
      "entropy": 5.588819742202759,
      "epoch": 1.575653120464441,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047552198549211463,
      "loss": 5.1406,
      "mean_token_accuracy": 0.201657672226429,
      "num_tokens": 36049864.0,
      "step": 17370
    },
    {
      "entropy": 5.533351182937622,
      "epoch": 1.5761066763425253,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004755073067653084,
      "loss": 5.0559,
      "mean_token_accuracy": 0.19713248908519745,
      "num_tokens": 36059224.0,
      "step": 17375
    },
    {
      "entropy": 5.431200313568115,
      "epoch": 1.5765602322206096,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004754926238919599,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21095066368579865,
      "num_tokens": 36069243.0,
      "step": 17380
    },
    {
      "entropy": 5.51810622215271,
      "epoch": 1.5770137880986939,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004754779368723729,
      "loss": 5.0266,
      "mean_token_accuracy": 0.1996670052409172,
      "num_tokens": 36079833.0,
      "step": 17385
    },
    {
      "entropy": 5.519850492477417,
      "epoch": 1.577467343976778,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004754632457068511,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20477883368730546,
      "num_tokens": 36090673.0,
      "step": 17390
    },
    {
      "entropy": 5.55263671875,
      "epoch": 1.5779208998548622,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004754485503956982,
      "loss": 5.0531,
      "mean_token_accuracy": 0.20432986617088317,
      "num_tokens": 36101719.0,
      "step": 17395
    },
    {
      "entropy": 5.518518877029419,
      "epoch": 1.5783744557329462,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004754338509392183,
      "loss": 5.0047,
      "mean_token_accuracy": 0.20799205005168914,
      "num_tokens": 36112352.0,
      "step": 17400
    },
    {
      "entropy": 5.403399133682251,
      "epoch": 1.5788280116110305,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004754191473377153,
      "loss": 4.9002,
      "mean_token_accuracy": 0.2080320432782173,
      "num_tokens": 36121797.0,
      "step": 17405
    },
    {
      "entropy": 5.507850456237793,
      "epoch": 1.5792815674891147,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047540443959149316,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20719701945781707,
      "num_tokens": 36132094.0,
      "step": 17410
    },
    {
      "entropy": 5.516197729110718,
      "epoch": 1.5797351233671988,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004753897277008562,
      "loss": 5.0261,
      "mean_token_accuracy": 0.2037370592355728,
      "num_tokens": 36142860.0,
      "step": 17415
    },
    {
      "entropy": 5.527059650421142,
      "epoch": 1.580188679245283,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004753750116661086,
      "loss": 5.0152,
      "mean_token_accuracy": 0.19879119247198104,
      "num_tokens": 36154555.0,
      "step": 17420
    },
    {
      "entropy": 5.602051639556885,
      "epoch": 1.580642235123367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047536029148755477,
      "loss": 5.145,
      "mean_token_accuracy": 0.1905045822262764,
      "num_tokens": 36165479.0,
      "step": 17425
    },
    {
      "entropy": 5.570267295837402,
      "epoch": 1.5810957910014514,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047534556716549893,
      "loss": 5.0555,
      "mean_token_accuracy": 0.1985606536269188,
      "num_tokens": 36175873.0,
      "step": 17430
    },
    {
      "entropy": 5.489364194869995,
      "epoch": 1.5815493468795356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047533083870024575,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20992728173732758,
      "num_tokens": 36186792.0,
      "step": 17435
    },
    {
      "entropy": 5.63928256034851,
      "epoch": 1.58200290275762,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004753161060920998,
      "loss": 5.1431,
      "mean_token_accuracy": 0.1986440509557724,
      "num_tokens": 36196691.0,
      "step": 17440
    },
    {
      "entropy": 5.606499433517456,
      "epoch": 1.582456458635704,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004753013693413657,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20787551999092102,
      "num_tokens": 36207173.0,
      "step": 17445
    },
    {
      "entropy": 5.48338131904602,
      "epoch": 1.582910014513788,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047528662844834827,
      "loss": 5.029,
      "mean_token_accuracy": 0.21019935011863708,
      "num_tokens": 36217912.0,
      "step": 17450
    },
    {
      "entropy": 5.420902919769287,
      "epoch": 1.5833635703918723,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047527188341335233,
      "loss": 4.9371,
      "mean_token_accuracy": 0.21276613771915437,
      "num_tokens": 36227714.0,
      "step": 17455
    },
    {
      "entropy": 5.499198532104492,
      "epoch": 1.5838171262699565,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004752571342366828,
      "loss": 5.0087,
      "mean_token_accuracy": 0.2063855767250061,
      "num_tokens": 36239114.0,
      "step": 17460
    },
    {
      "entropy": 5.530270004272461,
      "epoch": 1.5842706821480408,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004752423809186446,
      "loss": 5.0342,
      "mean_token_accuracy": 0.19996798485517503,
      "num_tokens": 36249436.0,
      "step": 17465
    },
    {
      "entropy": 5.529721927642822,
      "epoch": 1.5847242380261248,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047522762345954305,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20384032130241395,
      "num_tokens": 36259541.0,
      "step": 17470
    },
    {
      "entropy": 5.571310377120971,
      "epoch": 1.5851777939042089,
      "grad_norm": 1.0,
      "learning_rate": 0.0004752128618596832,
      "loss": 5.046,
      "mean_token_accuracy": 0.19789719432592393,
      "num_tokens": 36270689.0,
      "step": 17475
    },
    {
      "entropy": 5.60941514968872,
      "epoch": 1.5856313497822931,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047519809611937036,
      "loss": 5.0863,
      "mean_token_accuracy": 0.2013432890176773,
      "num_tokens": 36281191.0,
      "step": 17480
    },
    {
      "entropy": 5.47082896232605,
      "epoch": 1.5860849056603774,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004751833262389098,
      "loss": 4.9994,
      "mean_token_accuracy": 0.20849235653877257,
      "num_tokens": 36291022.0,
      "step": 17485
    },
    {
      "entropy": 5.536634874343872,
      "epoch": 1.5865384615384617,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047516855221860715,
      "loss": 5.0558,
      "mean_token_accuracy": 0.2028082311153412,
      "num_tokens": 36301237.0,
      "step": 17490
    },
    {
      "entropy": 5.563267469406128,
      "epoch": 1.5869920174165457,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047515377405876774,
      "loss": 5.021,
      "mean_token_accuracy": 0.1983124077320099,
      "num_tokens": 36311792.0,
      "step": 17495
    },
    {
      "entropy": 5.497449779510498,
      "epoch": 1.5874455732946298,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004751389917596973,
      "loss": 5.003,
      "mean_token_accuracy": 0.2117833748459816,
      "num_tokens": 36321847.0,
      "step": 17500
    },
    {
      "entropy": 5.534449243545533,
      "epoch": 1.587899129172714,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047512420532170154,
      "loss": 5.1105,
      "mean_token_accuracy": 0.19348909854888915,
      "num_tokens": 36332735.0,
      "step": 17505
    },
    {
      "entropy": 5.604989242553711,
      "epoch": 1.5883526850507983,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004751094147450862,
      "loss": 5.1031,
      "mean_token_accuracy": 0.19945639073848725,
      "num_tokens": 36343149.0,
      "step": 17510
    },
    {
      "entropy": 5.57188925743103,
      "epoch": 1.5888062409288826,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004750946200301571,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20454846769571305,
      "num_tokens": 36353081.0,
      "step": 17515
    },
    {
      "entropy": 5.513240242004395,
      "epoch": 1.5892597968069666,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047507982117722035,
      "loss": 5.0678,
      "mean_token_accuracy": 0.1978520154953003,
      "num_tokens": 36363624.0,
      "step": 17520
    },
    {
      "entropy": 5.562574768066407,
      "epoch": 1.5897133526850507,
      "grad_norm": 1.03125,
      "learning_rate": 0.000475065018186582,
      "loss": 5.0985,
      "mean_token_accuracy": 0.19767258316278458,
      "num_tokens": 36374081.0,
      "step": 17525
    },
    {
      "entropy": 5.577431631088257,
      "epoch": 1.590166908563135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047505021105854794,
      "loss": 5.1493,
      "mean_token_accuracy": 0.19680466055870055,
      "num_tokens": 36384360.0,
      "step": 17530
    },
    {
      "entropy": 5.541051292419434,
      "epoch": 1.5906204644412192,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047503539979342456,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20463062524795533,
      "num_tokens": 36395924.0,
      "step": 17535
    },
    {
      "entropy": 5.468084049224854,
      "epoch": 1.5910740203193035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004750205843915182,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21147719472646714,
      "num_tokens": 36405246.0,
      "step": 17540
    },
    {
      "entropy": 5.489252662658691,
      "epoch": 1.5915275761973875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004750057648531351,
      "loss": 4.9854,
      "mean_token_accuracy": 0.20914095193147658,
      "num_tokens": 36414964.0,
      "step": 17545
    },
    {
      "entropy": 5.516128301620483,
      "epoch": 1.5919811320754715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004749909411785819,
      "loss": 5.0587,
      "mean_token_accuracy": 0.19703762233257294,
      "num_tokens": 36425011.0,
      "step": 17550
    },
    {
      "entropy": 5.495929622650147,
      "epoch": 1.5924346879535558,
      "grad_norm": 1.078125,
      "learning_rate": 0.000474976113368165,
      "loss": 4.9487,
      "mean_token_accuracy": 0.2121456131339073,
      "num_tokens": 36434734.0,
      "step": 17555
    },
    {
      "entropy": 5.524063920974731,
      "epoch": 1.59288824383164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004749612814221912,
      "loss": 5.0448,
      "mean_token_accuracy": 0.2045255184173584,
      "num_tokens": 36444748.0,
      "step": 17560
    },
    {
      "entropy": 5.616097640991211,
      "epoch": 1.5933417997097243,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047494644534096703,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19496432393789292,
      "num_tokens": 36455076.0,
      "step": 17565
    },
    {
      "entropy": 5.543533039093018,
      "epoch": 1.5937953555878084,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004749316051247995,
      "loss": 4.9717,
      "mean_token_accuracy": 0.2077886164188385,
      "num_tokens": 36465302.0,
      "step": 17570
    },
    {
      "entropy": 5.510424327850342,
      "epoch": 1.5942489114658926,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047491676077399543,
      "loss": 5.0872,
      "mean_token_accuracy": 0.19867370873689652,
      "num_tokens": 36476666.0,
      "step": 17575
    },
    {
      "entropy": 5.487771701812744,
      "epoch": 1.5947024673439767,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004749019122888618,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2082185000181198,
      "num_tokens": 36487379.0,
      "step": 17580
    },
    {
      "entropy": 5.527111959457398,
      "epoch": 1.595156023222061,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047488705966970565,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2124447599053383,
      "num_tokens": 36497897.0,
      "step": 17585
    },
    {
      "entropy": 5.54830584526062,
      "epoch": 1.5956095791001452,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004748722029168342,
      "loss": 5.0678,
      "mean_token_accuracy": 0.2106671467423439,
      "num_tokens": 36508528.0,
      "step": 17590
    },
    {
      "entropy": 5.5429380416870115,
      "epoch": 1.5960631349782293,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004748573420305546,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19697727560997008,
      "num_tokens": 36519196.0,
      "step": 17595
    },
    {
      "entropy": 5.5159355163574215,
      "epoch": 1.5965166908563135,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047484247701117427,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20312255024909973,
      "num_tokens": 36529871.0,
      "step": 17600
    },
    {
      "entropy": 5.626338863372803,
      "epoch": 1.5969702467343976,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004748276078590006,
      "loss": 5.1236,
      "mean_token_accuracy": 0.1941250815987587,
      "num_tokens": 36541188.0,
      "step": 17605
    },
    {
      "entropy": 5.58647198677063,
      "epoch": 1.5974238026124818,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004748127345743411,
      "loss": 5.0966,
      "mean_token_accuracy": 0.19909974187612534,
      "num_tokens": 36551968.0,
      "step": 17610
    },
    {
      "entropy": 5.530744457244873,
      "epoch": 1.5978773584905661,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004747978571575033,
      "loss": 5.0348,
      "mean_token_accuracy": 0.20223728865385054,
      "num_tokens": 36562870.0,
      "step": 17615
    },
    {
      "entropy": 5.479086828231812,
      "epoch": 1.5983309143686504,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004747829756087949,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20745939463377,
      "num_tokens": 36573295.0,
      "step": 17620
    },
    {
      "entropy": 5.479890394210815,
      "epoch": 1.5987844702467344,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047476808992852364,
      "loss": 4.9653,
      "mean_token_accuracy": 0.20873234122991563,
      "num_tokens": 36583830.0,
      "step": 17625
    },
    {
      "entropy": 5.517432165145874,
      "epoch": 1.5992380261248185,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004747532001169974,
      "loss": 5.0462,
      "mean_token_accuracy": 0.2130327671766281,
      "num_tokens": 36594052.0,
      "step": 17630
    },
    {
      "entropy": 5.520537948608398,
      "epoch": 1.5996915820029027,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047473830617452415,
      "loss": 5.0696,
      "mean_token_accuracy": 0.20406648516654968,
      "num_tokens": 36604378.0,
      "step": 17635
    },
    {
      "entropy": 5.596294307708741,
      "epoch": 1.600145137880987,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047472340810141163,
      "loss": 5.1623,
      "mean_token_accuracy": 0.19078289419412614,
      "num_tokens": 36615588.0,
      "step": 17640
    },
    {
      "entropy": 5.626055479049683,
      "epoch": 1.6005986937590713,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047470850589796833,
      "loss": 5.0402,
      "mean_token_accuracy": 0.1986627995967865,
      "num_tokens": 36626369.0,
      "step": 17645
    },
    {
      "entropy": 5.5786247730255125,
      "epoch": 1.6010522496371553,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047469359956450215,
      "loss": 5.0791,
      "mean_token_accuracy": 0.1997091665863991,
      "num_tokens": 36636565.0,
      "step": 17650
    },
    {
      "entropy": 5.611868000030517,
      "epoch": 1.6015058055152394,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047467868910132144,
      "loss": 5.1591,
      "mean_token_accuracy": 0.1966838225722313,
      "num_tokens": 36647721.0,
      "step": 17655
    },
    {
      "entropy": 5.495266342163086,
      "epoch": 1.6019593613933236,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004746637745087346,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21024642437696456,
      "num_tokens": 36657887.0,
      "step": 17660
    },
    {
      "entropy": 5.604389238357544,
      "epoch": 1.6024129172714079,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047464885578705,
      "loss": 5.0494,
      "mean_token_accuracy": 0.19974400252103805,
      "num_tokens": 36667918.0,
      "step": 17665
    },
    {
      "entropy": 5.472668647766113,
      "epoch": 1.6028664731494922,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047463393293657616,
      "loss": 4.9976,
      "mean_token_accuracy": 0.2072445273399353,
      "num_tokens": 36678491.0,
      "step": 17670
    },
    {
      "entropy": 5.5294928550720215,
      "epoch": 1.6033200290275762,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004746190059576217,
      "loss": 5.0594,
      "mean_token_accuracy": 0.2025817811489105,
      "num_tokens": 36688638.0,
      "step": 17675
    },
    {
      "entropy": 5.555827474594116,
      "epoch": 1.6037735849056602,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004746040748504954,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20663116574287416,
      "num_tokens": 36699344.0,
      "step": 17680
    },
    {
      "entropy": 5.555319690704346,
      "epoch": 1.6042271407837445,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047458913961550593,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20222714096307753,
      "num_tokens": 36709606.0,
      "step": 17685
    },
    {
      "entropy": 5.463090229034424,
      "epoch": 1.6046806966618288,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004745742002529622,
      "loss": 4.9625,
      "mean_token_accuracy": 0.20935799330472946,
      "num_tokens": 36718740.0,
      "step": 17690
    },
    {
      "entropy": 5.522600412368774,
      "epoch": 1.605134252539913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004745592567631731,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19677862673997878,
      "num_tokens": 36729960.0,
      "step": 17695
    },
    {
      "entropy": 5.4693926811218265,
      "epoch": 1.605587808417997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047454430914644773,
      "loss": 4.9631,
      "mean_token_accuracy": 0.21306394040584564,
      "num_tokens": 36740221.0,
      "step": 17700
    },
    {
      "entropy": 5.427061891555786,
      "epoch": 1.6060413642960811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004745293574030952,
      "loss": 4.9615,
      "mean_token_accuracy": 0.213801009953022,
      "num_tokens": 36750841.0,
      "step": 17705
    },
    {
      "entropy": 5.497584199905395,
      "epoch": 1.6064949201741654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047451440153342477,
      "loss": 5.0296,
      "mean_token_accuracy": 0.19926677495241166,
      "num_tokens": 36760944.0,
      "step": 17710
    },
    {
      "entropy": 5.571436548233033,
      "epoch": 1.6069484760522497,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004744994415377457,
      "loss": 5.0151,
      "mean_token_accuracy": 0.20537643432617186,
      "num_tokens": 36770922.0,
      "step": 17715
    },
    {
      "entropy": 5.579338598251343,
      "epoch": 1.607402031930334,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047448447741636724,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20064235776662825,
      "num_tokens": 36780702.0,
      "step": 17720
    },
    {
      "entropy": 5.577346754074097,
      "epoch": 1.607855587808418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047446950916959905,
      "loss": 5.082,
      "mean_token_accuracy": 0.19528306573629378,
      "num_tokens": 36791868.0,
      "step": 17725
    },
    {
      "entropy": 5.585596323013306,
      "epoch": 1.608309143686502,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004744545367977505,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2042024776339531,
      "num_tokens": 36802330.0,
      "step": 17730
    },
    {
      "entropy": 5.558025169372558,
      "epoch": 1.6087626995645863,
      "grad_norm": 1.125,
      "learning_rate": 0.00047443956030113134,
      "loss": 5.0891,
      "mean_token_accuracy": 0.1996786117553711,
      "num_tokens": 36812936.0,
      "step": 17735
    },
    {
      "entropy": 5.577855634689331,
      "epoch": 1.6092162554426706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047442457968005123,
      "loss": 5.0469,
      "mean_token_accuracy": 0.20214971750974656,
      "num_tokens": 36823123.0,
      "step": 17740
    },
    {
      "entropy": 5.4617472171783445,
      "epoch": 1.6096698113207548,
      "grad_norm": 1.125,
      "learning_rate": 0.00047440959493481996,
      "loss": 4.9368,
      "mean_token_accuracy": 0.2160551816225052,
      "num_tokens": 36833050.0,
      "step": 17745
    },
    {
      "entropy": 5.5303796291351315,
      "epoch": 1.6101233671988389,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047439460606574745,
      "loss": 5.0548,
      "mean_token_accuracy": 0.20608482956886293,
      "num_tokens": 36842520.0,
      "step": 17750
    },
    {
      "entropy": 5.546290588378906,
      "epoch": 1.6105769230769231,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004743796130731437,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2015488028526306,
      "num_tokens": 36853024.0,
      "step": 17755
    },
    {
      "entropy": 5.547762823104859,
      "epoch": 1.6110304789550072,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004743646159573188,
      "loss": 4.9728,
      "mean_token_accuracy": 0.21105591356754302,
      "num_tokens": 36862534.0,
      "step": 17760
    },
    {
      "entropy": 5.516180324554443,
      "epoch": 1.6114840348330914,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004743496147185828,
      "loss": 4.9601,
      "mean_token_accuracy": 0.21105688959360122,
      "num_tokens": 36872356.0,
      "step": 17765
    },
    {
      "entropy": 5.461830472946167,
      "epoch": 1.6119375907111757,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047433460935724595,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20244371742010117,
      "num_tokens": 36882643.0,
      "step": 17770
    },
    {
      "entropy": 5.5579465389251705,
      "epoch": 1.6123911465892597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004743195998736186,
      "loss": 5.0052,
      "mean_token_accuracy": 0.20161498188972474,
      "num_tokens": 36893349.0,
      "step": 17775
    },
    {
      "entropy": 5.6073261260986325,
      "epoch": 1.612844702467344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047430458626801107,
      "loss": 5.0289,
      "mean_token_accuracy": 0.20888873934745789,
      "num_tokens": 36903713.0,
      "step": 17780
    },
    {
      "entropy": 5.512869548797608,
      "epoch": 1.613298258345428,
      "grad_norm": 1.109375,
      "learning_rate": 0.000474289568540734,
      "loss": 4.9383,
      "mean_token_accuracy": 0.21190437525510789,
      "num_tokens": 36912999.0,
      "step": 17785
    },
    {
      "entropy": 5.560475158691406,
      "epoch": 1.6137518142235123,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047427454669209785,
      "loss": 5.0708,
      "mean_token_accuracy": 0.2030900776386261,
      "num_tokens": 36923598.0,
      "step": 17790
    },
    {
      "entropy": 5.521582317352295,
      "epoch": 1.6142053701015966,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047425952072241326,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20506981611251832,
      "num_tokens": 36933036.0,
      "step": 17795
    },
    {
      "entropy": 5.536417770385742,
      "epoch": 1.6146589259796809,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000474244490631991,
      "loss": 5.0816,
      "mean_token_accuracy": 0.19817693680524825,
      "num_tokens": 36943137.0,
      "step": 17800
    },
    {
      "entropy": 5.649659061431885,
      "epoch": 1.615112481857765,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004742294564211419,
      "loss": 5.2031,
      "mean_token_accuracy": 0.19474415481090546,
      "num_tokens": 36955619.0,
      "step": 17805
    },
    {
      "entropy": 5.543700122833252,
      "epoch": 1.615566037735849,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004742144180901769,
      "loss": 4.973,
      "mean_token_accuracy": 0.21407785415649414,
      "num_tokens": 36966377.0,
      "step": 17810
    },
    {
      "entropy": 5.564962434768677,
      "epoch": 1.6160195936139332,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047419937563940697,
      "loss": 5.0333,
      "mean_token_accuracy": 0.2056509166955948,
      "num_tokens": 36977100.0,
      "step": 17815
    },
    {
      "entropy": 5.557198762893677,
      "epoch": 1.6164731494920175,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047418432906914314,
      "loss": 5.1014,
      "mean_token_accuracy": 0.20024712532758712,
      "num_tokens": 36987120.0,
      "step": 17820
    },
    {
      "entropy": 5.593332052230835,
      "epoch": 1.6169267053701017,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004741692783796967,
      "loss": 5.0526,
      "mean_token_accuracy": 0.20203897207975388,
      "num_tokens": 36996885.0,
      "step": 17825
    },
    {
      "entropy": 5.544119739532471,
      "epoch": 1.6173802612481858,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047415422357137876,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21137814223766327,
      "num_tokens": 37006570.0,
      "step": 17830
    },
    {
      "entropy": 5.594113206863403,
      "epoch": 1.6178338171262698,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004741391646445008,
      "loss": 5.0831,
      "mean_token_accuracy": 0.20022266656160354,
      "num_tokens": 37017185.0,
      "step": 17835
    },
    {
      "entropy": 5.547800064086914,
      "epoch": 1.618287373004354,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004741241015993741,
      "loss": 5.0493,
      "mean_token_accuracy": 0.20494076907634734,
      "num_tokens": 37028386.0,
      "step": 17840
    },
    {
      "entropy": 5.493295669555664,
      "epoch": 1.6187409288824384,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047410903443631026,
      "loss": 4.9522,
      "mean_token_accuracy": 0.20483069121837616,
      "num_tokens": 37038326.0,
      "step": 17845
    },
    {
      "entropy": 5.548564386367798,
      "epoch": 1.6191944847605226,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004740939631556209,
      "loss": 5.0458,
      "mean_token_accuracy": 0.20101993530988693,
      "num_tokens": 37048570.0,
      "step": 17850
    },
    {
      "entropy": 5.505143356323242,
      "epoch": 1.6196480406386067,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004740788877576176,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2146276816725731,
      "num_tokens": 37058909.0,
      "step": 17855
    },
    {
      "entropy": 5.533849716186523,
      "epoch": 1.6201015965166907,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004740638082426121,
      "loss": 5.0125,
      "mean_token_accuracy": 0.21495533287525176,
      "num_tokens": 37068957.0,
      "step": 17860
    },
    {
      "entropy": 5.539355087280273,
      "epoch": 1.620555152394775,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047404872461091643,
      "loss": 4.9764,
      "mean_token_accuracy": 0.20215528160333635,
      "num_tokens": 37080187.0,
      "step": 17865
    },
    {
      "entropy": 5.5432701587677,
      "epoch": 1.6210087082728593,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047403363686284236,
      "loss": 5.0616,
      "mean_token_accuracy": 0.20310061573982238,
      "num_tokens": 37090314.0,
      "step": 17870
    },
    {
      "entropy": 5.5888636112213135,
      "epoch": 1.6214622641509435,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047401854499870197,
      "loss": 5.0771,
      "mean_token_accuracy": 0.20264121890068054,
      "num_tokens": 37099907.0,
      "step": 17875
    },
    {
      "entropy": 5.561094284057617,
      "epoch": 1.6219158200290276,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047400344901880735,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1975569471716881,
      "num_tokens": 37110495.0,
      "step": 17880
    },
    {
      "entropy": 5.553846216201782,
      "epoch": 1.6223693759071116,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004739883489234707,
      "loss": 5.065,
      "mean_token_accuracy": 0.19802882373332978,
      "num_tokens": 37121957.0,
      "step": 17885
    },
    {
      "entropy": 5.539548778533936,
      "epoch": 1.6228229317851959,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047397324471300424,
      "loss": 4.9772,
      "mean_token_accuracy": 0.210438971221447,
      "num_tokens": 37132743.0,
      "step": 17890
    },
    {
      "entropy": 5.562828397750854,
      "epoch": 1.6232764876632801,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047395813638772047,
      "loss": 5.0776,
      "mean_token_accuracy": 0.1967625856399536,
      "num_tokens": 37143947.0,
      "step": 17895
    },
    {
      "entropy": 5.6736279964447025,
      "epoch": 1.6237300435413644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047394302394793164,
      "loss": 5.172,
      "mean_token_accuracy": 0.1933992624282837,
      "num_tokens": 37154316.0,
      "step": 17900
    },
    {
      "entropy": 5.604281616210938,
      "epoch": 1.6241835994194485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004739279073939504,
      "loss": 5.073,
      "mean_token_accuracy": 0.20024010837078093,
      "num_tokens": 37164999.0,
      "step": 17905
    },
    {
      "entropy": 5.56506929397583,
      "epoch": 1.6246371552975325,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004739127867260894,
      "loss": 5.075,
      "mean_token_accuracy": 0.20138372778892516,
      "num_tokens": 37174392.0,
      "step": 17910
    },
    {
      "entropy": 5.48519606590271,
      "epoch": 1.6250907111756168,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004738976619446613,
      "loss": 5.056,
      "mean_token_accuracy": 0.20201866924762726,
      "num_tokens": 37185341.0,
      "step": 17915
    },
    {
      "entropy": 5.580339765548706,
      "epoch": 1.625544267053701,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047388253304997874,
      "loss": 5.0779,
      "mean_token_accuracy": 0.1985635995864868,
      "num_tokens": 37196725.0,
      "step": 17920
    },
    {
      "entropy": 5.555712938308716,
      "epoch": 1.6259978229317853,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004738674000423548,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20997316688299178,
      "num_tokens": 37206877.0,
      "step": 17925
    },
    {
      "entropy": 5.458779954910279,
      "epoch": 1.6264513788098693,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004738522629221023,
      "loss": 4.9571,
      "mean_token_accuracy": 0.20690305680036544,
      "num_tokens": 37217277.0,
      "step": 17930
    },
    {
      "entropy": 5.475347471237183,
      "epoch": 1.6269049346879536,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004738371216895344,
      "loss": 4.934,
      "mean_token_accuracy": 0.2105925515294075,
      "num_tokens": 37227716.0,
      "step": 17935
    },
    {
      "entropy": 5.506228351593018,
      "epoch": 1.6273584905660377,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004738219763449641,
      "loss": 5.047,
      "mean_token_accuracy": 0.20263891220092772,
      "num_tokens": 37238178.0,
      "step": 17940
    },
    {
      "entropy": 5.525771284103394,
      "epoch": 1.627812046444122,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004738068268887047,
      "loss": 5.0732,
      "mean_token_accuracy": 0.2064121648669243,
      "num_tokens": 37248408.0,
      "step": 17945
    },
    {
      "entropy": 5.564480209350586,
      "epoch": 1.6282656023222062,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004737916733210694,
      "loss": 5.1805,
      "mean_token_accuracy": 0.1948343485593796,
      "num_tokens": 37259176.0,
      "step": 17950
    },
    {
      "entropy": 5.55413498878479,
      "epoch": 1.6287191582002902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047377651564237165,
      "loss": 4.9403,
      "mean_token_accuracy": 0.21000248938798904,
      "num_tokens": 37269205.0,
      "step": 17955
    },
    {
      "entropy": 5.556760597229004,
      "epoch": 1.6291727140783745,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047376135385292494,
      "loss": 5.0615,
      "mean_token_accuracy": 0.20742124766111375,
      "num_tokens": 37278841.0,
      "step": 17960
    },
    {
      "entropy": 5.571680688858033,
      "epoch": 1.6296262699564585,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004737461879530428,
      "loss": 5.0934,
      "mean_token_accuracy": 0.20085595697164535,
      "num_tokens": 37288487.0,
      "step": 17965
    },
    {
      "entropy": 5.593708944320679,
      "epoch": 1.6300798258345428,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047373101794303884,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2009998604655266,
      "num_tokens": 37300086.0,
      "step": 17970
    },
    {
      "entropy": 5.503368282318116,
      "epoch": 1.630533381712627,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004737158438232267,
      "loss": 5.0912,
      "mean_token_accuracy": 0.19563519060611725,
      "num_tokens": 37310072.0,
      "step": 17975
    },
    {
      "entropy": 5.5453534603118895,
      "epoch": 1.6309869375907113,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047370066559392033,
      "loss": 5.0245,
      "mean_token_accuracy": 0.21010988503694533,
      "num_tokens": 37320278.0,
      "step": 17980
    },
    {
      "entropy": 5.560372066497803,
      "epoch": 1.6314404934687954,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047368548325543366,
      "loss": 4.9913,
      "mean_token_accuracy": 0.19833416491746902,
      "num_tokens": 37330741.0,
      "step": 17985
    },
    {
      "entropy": 5.562993144989013,
      "epoch": 1.6318940493468794,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004736702968080804,
      "loss": 5.076,
      "mean_token_accuracy": 0.19929848611354828,
      "num_tokens": 37341213.0,
      "step": 17990
    },
    {
      "entropy": 5.496298837661743,
      "epoch": 1.6323476052249637,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047365510625217483,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20807789713144303,
      "num_tokens": 37350528.0,
      "step": 17995
    },
    {
      "entropy": 5.528117418289185,
      "epoch": 1.632801161103048,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047363991158803117,
      "loss": 5.0309,
      "mean_token_accuracy": 0.20054963678121568,
      "num_tokens": 37359938.0,
      "step": 18000
    },
    {
      "epoch": 1.632801161103048,
      "eval_entropy": 5.307341233098129,
      "eval_loss": 5.06912899017334,
      "eval_mean_token_accuracy": 0.21036261755180843,
      "eval_num_tokens": 37359938.0,
      "eval_runtime": 20.6219,
      "eval_samples_per_second": 1714.681,
      "eval_steps_per_second": 214.335,
      "step": 18000
    },
    {
      "entropy": 5.560543966293335,
      "epoch": 1.6332547169811322,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047362471281596334,
      "loss": 5.0245,
      "mean_token_accuracy": 0.20620202720165254,
      "num_tokens": 37369744.0,
      "step": 18005
    },
    {
      "entropy": 5.596627044677734,
      "epoch": 1.6337082728592163,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047360950993628585,
      "loss": 5.1155,
      "mean_token_accuracy": 0.20341364592313765,
      "num_tokens": 37380578.0,
      "step": 18010
    },
    {
      "entropy": 5.550653743743896,
      "epoch": 1.6341618287373003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004735943029493132,
      "loss": 5.0317,
      "mean_token_accuracy": 0.2036425366997719,
      "num_tokens": 37390237.0,
      "step": 18015
    },
    {
      "entropy": 5.574190616607666,
      "epoch": 1.6346153846153846,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047357909185535965,
      "loss": 5.0872,
      "mean_token_accuracy": 0.1972410872578621,
      "num_tokens": 37400261.0,
      "step": 18020
    },
    {
      "entropy": 5.560031843185425,
      "epoch": 1.6350689404934688,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004735638766547399,
      "loss": 5.0246,
      "mean_token_accuracy": 0.2051742523908615,
      "num_tokens": 37410012.0,
      "step": 18025
    },
    {
      "entropy": 5.576835584640503,
      "epoch": 1.6355224963715531,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047354865734776857,
      "loss": 5.1316,
      "mean_token_accuracy": 0.192862069606781,
      "num_tokens": 37420530.0,
      "step": 18030
    },
    {
      "entropy": 5.5550755023956295,
      "epoch": 1.6359760522496372,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004735334339347605,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21534911990165712,
      "num_tokens": 37431146.0,
      "step": 18035
    },
    {
      "entropy": 5.548226690292358,
      "epoch": 1.6364296081277212,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047351820641603035,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20894050747156143,
      "num_tokens": 37441253.0,
      "step": 18040
    },
    {
      "entropy": 5.463728237152099,
      "epoch": 1.6368831640058055,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004735029747918931,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20038239955902098,
      "num_tokens": 37451326.0,
      "step": 18045
    },
    {
      "entropy": 5.466217184066773,
      "epoch": 1.6373367198838897,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047348773906266374,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2092629000544548,
      "num_tokens": 37462462.0,
      "step": 18050
    },
    {
      "entropy": 5.486484622955322,
      "epoch": 1.637790275761974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004734724992286574,
      "loss": 4.9832,
      "mean_token_accuracy": 0.20355282127857208,
      "num_tokens": 37473016.0,
      "step": 18055
    },
    {
      "entropy": 5.569339752197266,
      "epoch": 1.638243831640058,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004734572552901892,
      "loss": 5.0969,
      "mean_token_accuracy": 0.20118896216154097,
      "num_tokens": 37483737.0,
      "step": 18060
    },
    {
      "entropy": 5.526478099822998,
      "epoch": 1.638697387518142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004734420072475743,
      "loss": 4.9928,
      "mean_token_accuracy": 0.20455218851566315,
      "num_tokens": 37493070.0,
      "step": 18065
    },
    {
      "entropy": 5.51882038116455,
      "epoch": 1.6391509433962264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004734267551011282,
      "loss": 4.995,
      "mean_token_accuracy": 0.20797305405139924,
      "num_tokens": 37503696.0,
      "step": 18070
    },
    {
      "entropy": 5.495640802383423,
      "epoch": 1.6396044992743106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004734114988511662,
      "loss": 5.0086,
      "mean_token_accuracy": 0.20355913639068604,
      "num_tokens": 37513468.0,
      "step": 18075
    },
    {
      "entropy": 5.449446725845337,
      "epoch": 1.640058055152395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004733962384980039,
      "loss": 4.8819,
      "mean_token_accuracy": 0.2130780801177025,
      "num_tokens": 37523631.0,
      "step": 18080
    },
    {
      "entropy": 5.569212102890015,
      "epoch": 1.640511611030479,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004733809740419568,
      "loss": 5.0759,
      "mean_token_accuracy": 0.20668964982032775,
      "num_tokens": 37534602.0,
      "step": 18085
    },
    {
      "entropy": 5.457507658004761,
      "epoch": 1.640965166908563,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047336570548334053,
      "loss": 4.9914,
      "mean_token_accuracy": 0.2021588295698166,
      "num_tokens": 37544547.0,
      "step": 18090
    },
    {
      "entropy": 5.548616027832031,
      "epoch": 1.6414187227866472,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047335043282247105,
      "loss": 5.0301,
      "mean_token_accuracy": 0.20326164364814758,
      "num_tokens": 37554903.0,
      "step": 18095
    },
    {
      "entropy": 5.55193772315979,
      "epoch": 1.6418722786647315,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047333515605966396,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2056132212281227,
      "num_tokens": 37565586.0,
      "step": 18100
    },
    {
      "entropy": 5.536584043502808,
      "epoch": 1.6423258345428158,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047331987519523533,
      "loss": 5.0357,
      "mean_token_accuracy": 0.2078459829092026,
      "num_tokens": 37575996.0,
      "step": 18105
    },
    {
      "entropy": 5.482731580734253,
      "epoch": 1.6427793904208998,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004733045902295012,
      "loss": 4.941,
      "mean_token_accuracy": 0.2008867606520653,
      "num_tokens": 37586441.0,
      "step": 18110
    },
    {
      "entropy": 5.551346015930176,
      "epoch": 1.643232946298984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047328930116277755,
      "loss": 5.055,
      "mean_token_accuracy": 0.2052132710814476,
      "num_tokens": 37598226.0,
      "step": 18115
    },
    {
      "entropy": 5.542083644866944,
      "epoch": 1.6436865021770681,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004732740079953806,
      "loss": 5.0445,
      "mean_token_accuracy": 0.20656532049179077,
      "num_tokens": 37609102.0,
      "step": 18120
    },
    {
      "entropy": 5.545897531509399,
      "epoch": 1.6441400580551524,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004732587107276267,
      "loss": 5.0499,
      "mean_token_accuracy": 0.2018830582499504,
      "num_tokens": 37619349.0,
      "step": 18125
    },
    {
      "entropy": 5.512788105010986,
      "epoch": 1.6445936139332367,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047324340935983225,
      "loss": 5.0458,
      "mean_token_accuracy": 0.20966263711452485,
      "num_tokens": 37629844.0,
      "step": 18130
    },
    {
      "entropy": 5.482539653778076,
      "epoch": 1.6450471698113207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047322810389231333,
      "loss": 4.9394,
      "mean_token_accuracy": 0.20833370983600616,
      "num_tokens": 37639339.0,
      "step": 18135
    },
    {
      "entropy": 5.502611827850342,
      "epoch": 1.645500725689405,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004732127943253868,
      "loss": 4.978,
      "mean_token_accuracy": 0.21004469245672225,
      "num_tokens": 37648836.0,
      "step": 18140
    },
    {
      "entropy": 5.584582233428955,
      "epoch": 1.645954281567489,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004731974806593693,
      "loss": 4.9896,
      "mean_token_accuracy": 0.2051238164305687,
      "num_tokens": 37658424.0,
      "step": 18145
    },
    {
      "entropy": 5.549859142303466,
      "epoch": 1.6464078374455733,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047318216289457727,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21039813309907912,
      "num_tokens": 37669849.0,
      "step": 18150
    },
    {
      "entropy": 5.505876684188843,
      "epoch": 1.6468613933236576,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047316684103132766,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20256547927856444,
      "num_tokens": 37678987.0,
      "step": 18155
    },
    {
      "entropy": 5.555306529998779,
      "epoch": 1.6473149492017418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004731515150699372,
      "loss": 5.0209,
      "mean_token_accuracy": 0.20816829204559326,
      "num_tokens": 37689692.0,
      "step": 18160
    },
    {
      "entropy": 5.585298776626587,
      "epoch": 1.6477685050798259,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047313618501072303,
      "loss": 5.1711,
      "mean_token_accuracy": 0.19286765158176422,
      "num_tokens": 37700092.0,
      "step": 18165
    },
    {
      "entropy": 5.5218486309051515,
      "epoch": 1.64822206095791,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047312085085400194,
      "loss": 5.065,
      "mean_token_accuracy": 0.20425543636083604,
      "num_tokens": 37710305.0,
      "step": 18170
    },
    {
      "entropy": 5.658654689788818,
      "epoch": 1.6486756168359942,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047310551260009126,
      "loss": 5.1307,
      "mean_token_accuracy": 0.1929813876748085,
      "num_tokens": 37720451.0,
      "step": 18175
    },
    {
      "entropy": 5.577204895019531,
      "epoch": 1.6491291727140784,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000473090170249308,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20589849203824998,
      "num_tokens": 37731952.0,
      "step": 18180
    },
    {
      "entropy": 5.489793586730957,
      "epoch": 1.6495827285921627,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004730748238019696,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20154588669538498,
      "num_tokens": 37742201.0,
      "step": 18185
    },
    {
      "entropy": 5.474636507034302,
      "epoch": 1.6500362844702468,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047305947325839337,
      "loss": 4.9928,
      "mean_token_accuracy": 0.20665855407714845,
      "num_tokens": 37753317.0,
      "step": 18190
    },
    {
      "entropy": 5.58493447303772,
      "epoch": 1.6504898403483308,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004730441186188968,
      "loss": 5.0549,
      "mean_token_accuracy": 0.2018160879611969,
      "num_tokens": 37763452.0,
      "step": 18195
    },
    {
      "entropy": 5.551719331741333,
      "epoch": 1.650943396226415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047302875988379734,
      "loss": 4.9661,
      "mean_token_accuracy": 0.2069983720779419,
      "num_tokens": 37772804.0,
      "step": 18200
    },
    {
      "entropy": 5.548997497558593,
      "epoch": 1.6513969521044993,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047301339705341266,
      "loss": 5.0422,
      "mean_token_accuracy": 0.19575041085481643,
      "num_tokens": 37784441.0,
      "step": 18205
    },
    {
      "entropy": 5.529232168197632,
      "epoch": 1.6518505079825836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047299803012806056,
      "loss": 5.0436,
      "mean_token_accuracy": 0.2065920427441597,
      "num_tokens": 37795103.0,
      "step": 18210
    },
    {
      "entropy": 5.487066984176636,
      "epoch": 1.6523040638606676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004729826591080586,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20657159388065338,
      "num_tokens": 37806654.0,
      "step": 18215
    },
    {
      "entropy": 5.526598072052002,
      "epoch": 1.6527576197387517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047296728399372483,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20016657412052155,
      "num_tokens": 37816673.0,
      "step": 18220
    },
    {
      "entropy": 5.555908298492431,
      "epoch": 1.653211175616836,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047295190478537725,
      "loss": 5.0503,
      "mean_token_accuracy": 0.2029751569032669,
      "num_tokens": 37827389.0,
      "step": 18225
    },
    {
      "entropy": 5.658416748046875,
      "epoch": 1.6536647314949202,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047293652148333376,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20344164222478867,
      "num_tokens": 37838087.0,
      "step": 18230
    },
    {
      "entropy": 5.4327747344970705,
      "epoch": 1.6541182873730045,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047292113408791266,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20138545483350753,
      "num_tokens": 37848630.0,
      "step": 18235
    },
    {
      "entropy": 5.453988885879516,
      "epoch": 1.6545718432510885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047290574259943195,
      "loss": 4.9281,
      "mean_token_accuracy": 0.22004650682210922,
      "num_tokens": 37858689.0,
      "step": 18240
    },
    {
      "entropy": 5.6077797412872314,
      "epoch": 1.6550253991291726,
      "grad_norm": 1.0,
      "learning_rate": 0.00047289034701821013,
      "loss": 5.0897,
      "mean_token_accuracy": 0.20234557688236238,
      "num_tokens": 37869049.0,
      "step": 18245
    },
    {
      "entropy": 5.5167817115783695,
      "epoch": 1.6554789550072568,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047287494734456545,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2003646969795227,
      "num_tokens": 37879495.0,
      "step": 18250
    },
    {
      "entropy": 5.544087553024292,
      "epoch": 1.655932510885341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047285954357881646,
      "loss": 5.0285,
      "mean_token_accuracy": 0.2015921652317047,
      "num_tokens": 37889442.0,
      "step": 18255
    },
    {
      "entropy": 5.598650503158569,
      "epoch": 1.6563860667634254,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004728441357212817,
      "loss": 5.1712,
      "mean_token_accuracy": 0.1892197161912918,
      "num_tokens": 37900147.0,
      "step": 18260
    },
    {
      "entropy": 5.6057878017425535,
      "epoch": 1.6568396226415094,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004728287237722798,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20190392732620238,
      "num_tokens": 37910307.0,
      "step": 18265
    },
    {
      "entropy": 5.492194032669067,
      "epoch": 1.6572931785195935,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004728133077321295,
      "loss": 4.9683,
      "mean_token_accuracy": 0.20334598273038865,
      "num_tokens": 37920838.0,
      "step": 18270
    },
    {
      "entropy": 5.546195602416992,
      "epoch": 1.6577467343976777,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047279788760114954,
      "loss": 5.0489,
      "mean_token_accuracy": 0.19957167655229568,
      "num_tokens": 37930717.0,
      "step": 18275
    },
    {
      "entropy": 5.565794944763184,
      "epoch": 1.658200290275762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004727824633796589,
      "loss": 5.0443,
      "mean_token_accuracy": 0.2027788817882538,
      "num_tokens": 37941254.0,
      "step": 18280
    },
    {
      "entropy": 5.510905408859253,
      "epoch": 1.6586538461538463,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004727670350679765,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20850671380758284,
      "num_tokens": 37952132.0,
      "step": 18285
    },
    {
      "entropy": 5.513995122909546,
      "epoch": 1.6591074020319303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004727516026664214,
      "loss": 5.0358,
      "mean_token_accuracy": 0.20742790400981903,
      "num_tokens": 37962026.0,
      "step": 18290
    },
    {
      "entropy": 5.5256506443023685,
      "epoch": 1.6595609579100146,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004727361661753128,
      "loss": 4.9523,
      "mean_token_accuracy": 0.20238326042890548,
      "num_tokens": 37972911.0,
      "step": 18295
    },
    {
      "entropy": 5.5586512088775635,
      "epoch": 1.6600145137880986,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004727207255949699,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20900478065013886,
      "num_tokens": 37983577.0,
      "step": 18300
    },
    {
      "entropy": 5.5725830078125,
      "epoch": 1.6604680696661829,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047270528092571197,
      "loss": 5.0672,
      "mean_token_accuracy": 0.20097507536411285,
      "num_tokens": 37994294.0,
      "step": 18305
    },
    {
      "entropy": 5.538248825073242,
      "epoch": 1.6609216255442671,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004726898321678584,
      "loss": 5.0291,
      "mean_token_accuracy": 0.20685731917619704,
      "num_tokens": 38005144.0,
      "step": 18310
    },
    {
      "entropy": 5.4818648338317875,
      "epoch": 1.6613751814223512,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004726743793217288,
      "loss": 5.0551,
      "mean_token_accuracy": 0.20080667734146118,
      "num_tokens": 38015277.0,
      "step": 18315
    },
    {
      "entropy": 5.5463134288787845,
      "epoch": 1.6618287373004355,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047265892238764265,
      "loss": 4.9716,
      "mean_token_accuracy": 0.20633944720029831,
      "num_tokens": 38027530.0,
      "step": 18320
    },
    {
      "entropy": 5.617530441284179,
      "epoch": 1.6622822931785195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004726434613659196,
      "loss": 5.0891,
      "mean_token_accuracy": 0.20411063283681868,
      "num_tokens": 38036895.0,
      "step": 18325
    },
    {
      "entropy": 5.583403873443603,
      "epoch": 1.6627358490566038,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004726279962568794,
      "loss": 5.1012,
      "mean_token_accuracy": 0.2044348746538162,
      "num_tokens": 38047001.0,
      "step": 18330
    },
    {
      "entropy": 5.554082775115967,
      "epoch": 1.663189404934688,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004726125270608419,
      "loss": 4.975,
      "mean_token_accuracy": 0.2108441025018692,
      "num_tokens": 38056815.0,
      "step": 18335
    },
    {
      "entropy": 5.568605566024781,
      "epoch": 1.6636429608127723,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047259705377812703,
      "loss": 5.0664,
      "mean_token_accuracy": 0.20433802753686905,
      "num_tokens": 38067139.0,
      "step": 18340
    },
    {
      "entropy": 5.548576307296753,
      "epoch": 1.6640965166908563,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004725815764090546,
      "loss": 4.9839,
      "mean_token_accuracy": 0.20833301097154616,
      "num_tokens": 38076824.0,
      "step": 18345
    },
    {
      "entropy": 5.536114835739136,
      "epoch": 1.6645500725689404,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047256609495394494,
      "loss": 5.0278,
      "mean_token_accuracy": 0.20151522010564804,
      "num_tokens": 38087517.0,
      "step": 18350
    },
    {
      "entropy": 5.568885850906372,
      "epoch": 1.6650036284470247,
      "grad_norm": 1.03125,
      "learning_rate": 0.000472550609413118,
      "loss": 5.0186,
      "mean_token_accuracy": 0.20044526159763337,
      "num_tokens": 38098499.0,
      "step": 18355
    },
    {
      "entropy": 5.48836989402771,
      "epoch": 1.665457184325109,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004725351197868942,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2095501482486725,
      "num_tokens": 38108327.0,
      "step": 18360
    },
    {
      "entropy": 5.598924016952514,
      "epoch": 1.6659107402031932,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047251962607559373,
      "loss": 5.0356,
      "mean_token_accuracy": 0.20512547791004182,
      "num_tokens": 38118667.0,
      "step": 18365
    },
    {
      "entropy": 5.5005449771881105,
      "epoch": 1.6663642960812772,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047250412827953717,
      "loss": 4.9998,
      "mean_token_accuracy": 0.20354425311088561,
      "num_tokens": 38128571.0,
      "step": 18370
    },
    {
      "entropy": 5.588938045501709,
      "epoch": 1.6668178519593613,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004724886263990448,
      "loss": 5.1049,
      "mean_token_accuracy": 0.195250703394413,
      "num_tokens": 38138921.0,
      "step": 18375
    },
    {
      "entropy": 5.591016340255737,
      "epoch": 1.6672714078374455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004724731204344372,
      "loss": 5.0368,
      "mean_token_accuracy": 0.20534040927886962,
      "num_tokens": 38148858.0,
      "step": 18380
    },
    {
      "entropy": 5.462696266174317,
      "epoch": 1.6677249637155298,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047245761038603536,
      "loss": 4.9635,
      "mean_token_accuracy": 0.20529389679431914,
      "num_tokens": 38159107.0,
      "step": 18385
    },
    {
      "entropy": 5.618238306045532,
      "epoch": 1.668178519593614,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047244209625415975,
      "loss": 5.0921,
      "mean_token_accuracy": 0.20097823292016984,
      "num_tokens": 38169228.0,
      "step": 18390
    },
    {
      "entropy": 5.603679180145264,
      "epoch": 1.6686320754716981,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004724265780391312,
      "loss": 5.0288,
      "mean_token_accuracy": 0.2068103551864624,
      "num_tokens": 38179175.0,
      "step": 18395
    },
    {
      "entropy": 5.556495428085327,
      "epoch": 1.6690856313497822,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004724110557412708,
      "loss": 5.0423,
      "mean_token_accuracy": 0.20790047347545623,
      "num_tokens": 38188537.0,
      "step": 18400
    },
    {
      "entropy": 5.514718437194825,
      "epoch": 1.6695391872278664,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047239552936089943,
      "loss": 5.0496,
      "mean_token_accuracy": 0.2000635713338852,
      "num_tokens": 38198972.0,
      "step": 18405
    },
    {
      "entropy": 5.53039960861206,
      "epoch": 1.6699927431059507,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047237999889833816,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2063676580786705,
      "num_tokens": 38209472.0,
      "step": 18410
    },
    {
      "entropy": 5.461790990829468,
      "epoch": 1.670446298984035,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047236446435390826,
      "loss": 4.9372,
      "mean_token_accuracy": 0.2051655277609825,
      "num_tokens": 38219977.0,
      "step": 18415
    },
    {
      "entropy": 5.519774389266968,
      "epoch": 1.670899854862119,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047234892572793093,
      "loss": 5.0577,
      "mean_token_accuracy": 0.2033174842596054,
      "num_tokens": 38230492.0,
      "step": 18420
    },
    {
      "entropy": 5.529501247406006,
      "epoch": 1.671353410740203,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004723333830207275,
      "loss": 5.0185,
      "mean_token_accuracy": 0.1979853630065918,
      "num_tokens": 38240799.0,
      "step": 18425
    },
    {
      "entropy": 5.518443870544433,
      "epoch": 1.6718069666182873,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047231783623261954,
      "loss": 4.9815,
      "mean_token_accuracy": 0.1997287839651108,
      "num_tokens": 38251439.0,
      "step": 18430
    },
    {
      "entropy": 5.596939945220948,
      "epoch": 1.6722605224963716,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004723022853639283,
      "loss": 5.1158,
      "mean_token_accuracy": 0.19885122925043106,
      "num_tokens": 38261795.0,
      "step": 18435
    },
    {
      "entropy": 5.516361951828003,
      "epoch": 1.6727140783744558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004722867304149756,
      "loss": 4.9921,
      "mean_token_accuracy": 0.20650445967912673,
      "num_tokens": 38272172.0,
      "step": 18440
    },
    {
      "entropy": 5.552573108673096,
      "epoch": 1.67316763425254,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004722711713860831,
      "loss": 5.083,
      "mean_token_accuracy": 0.19582296311855316,
      "num_tokens": 38282829.0,
      "step": 18445
    },
    {
      "entropy": 5.580194520950317,
      "epoch": 1.673621190130624,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004722556082775724,
      "loss": 5.0806,
      "mean_token_accuracy": 0.20674073845148086,
      "num_tokens": 38293793.0,
      "step": 18450
    },
    {
      "entropy": 5.5810974597930905,
      "epoch": 1.6740747460087082,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004722400410897655,
      "loss": 5.083,
      "mean_token_accuracy": 0.20138343721628188,
      "num_tokens": 38302577.0,
      "step": 18455
    },
    {
      "entropy": 5.53628568649292,
      "epoch": 1.6745283018867925,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004722244698229842,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20656457394361497,
      "num_tokens": 38312571.0,
      "step": 18460
    },
    {
      "entropy": 5.527368402481079,
      "epoch": 1.6749818577648767,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047220889447755073,
      "loss": 4.966,
      "mean_token_accuracy": 0.2101530909538269,
      "num_tokens": 38323544.0,
      "step": 18465
    },
    {
      "entropy": 5.582181930541992,
      "epoch": 1.6754354136429608,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004721933150537869,
      "loss": 5.0528,
      "mean_token_accuracy": 0.20263455361127852,
      "num_tokens": 38334509.0,
      "step": 18470
    },
    {
      "entropy": 5.493896722793579,
      "epoch": 1.675888969521045,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047217773155201526,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20418312698602675,
      "num_tokens": 38344747.0,
      "step": 18475
    },
    {
      "entropy": 5.586251354217529,
      "epoch": 1.676342525399129,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047216214397255776,
      "loss": 5.0781,
      "mean_token_accuracy": 0.1967146709561348,
      "num_tokens": 38355671.0,
      "step": 18480
    },
    {
      "entropy": 5.565404796600342,
      "epoch": 1.6767960812772134,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047214655231573693,
      "loss": 5.0391,
      "mean_token_accuracy": 0.20337298065423964,
      "num_tokens": 38366368.0,
      "step": 18485
    },
    {
      "entropy": 5.563398122787476,
      "epoch": 1.6772496371552976,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004721309565818752,
      "loss": 5.1038,
      "mean_token_accuracy": 0.20038592368364333,
      "num_tokens": 38377322.0,
      "step": 18490
    },
    {
      "entropy": 5.608231687545777,
      "epoch": 1.6777031930333817,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047211535677129496,
      "loss": 5.1022,
      "mean_token_accuracy": 0.20255047976970672,
      "num_tokens": 38389204.0,
      "step": 18495
    },
    {
      "entropy": 5.548760461807251,
      "epoch": 1.678156748911466,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047209975288431893,
      "loss": 5.053,
      "mean_token_accuracy": 0.2019473984837532,
      "num_tokens": 38399117.0,
      "step": 18500
    },
    {
      "entropy": 5.537803888320923,
      "epoch": 1.67861030478955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047208414492126986,
      "loss": 5.0154,
      "mean_token_accuracy": 0.2017965391278267,
      "num_tokens": 38410341.0,
      "step": 18505
    },
    {
      "entropy": 5.595143365859985,
      "epoch": 1.6790638606676342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047206853288247043,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19759543389081954,
      "num_tokens": 38422742.0,
      "step": 18510
    },
    {
      "entropy": 5.56275897026062,
      "epoch": 1.6795174165457185,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047205291676824345,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19663460105657576,
      "num_tokens": 38433022.0,
      "step": 18515
    },
    {
      "entropy": 5.593740558624267,
      "epoch": 1.6799709724238028,
      "grad_norm": 1.0,
      "learning_rate": 0.000472037296578912,
      "loss": 5.0561,
      "mean_token_accuracy": 0.20002095848321916,
      "num_tokens": 38444325.0,
      "step": 18520
    },
    {
      "entropy": 5.537397623062134,
      "epoch": 1.6804245283018868,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000472021672314799,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20704559981822968,
      "num_tokens": 38454576.0,
      "step": 18525
    },
    {
      "entropy": 5.525679063796997,
      "epoch": 1.6808780841799709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047200604397622763,
      "loss": 4.9752,
      "mean_token_accuracy": 0.20736278742551803,
      "num_tokens": 38465090.0,
      "step": 18530
    },
    {
      "entropy": 5.576778554916382,
      "epoch": 1.6813316400580551,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004719904115635211,
      "loss": 5.2162,
      "mean_token_accuracy": 0.1895391821861267,
      "num_tokens": 38477253.0,
      "step": 18535
    },
    {
      "entropy": 5.618865823745727,
      "epoch": 1.6817851959361394,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004719747750770026,
      "loss": 5.1115,
      "mean_token_accuracy": 0.19903451353311538,
      "num_tokens": 38488040.0,
      "step": 18540
    },
    {
      "entropy": 5.549099969863891,
      "epoch": 1.6822387518142237,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004719591345169956,
      "loss": 5.0326,
      "mean_token_accuracy": 0.20792275220155715,
      "num_tokens": 38498016.0,
      "step": 18545
    },
    {
      "entropy": 5.586939668655395,
      "epoch": 1.6826923076923077,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047194348988382345,
      "loss": 5.1154,
      "mean_token_accuracy": 0.20007387101650237,
      "num_tokens": 38508996.0,
      "step": 18550
    },
    {
      "entropy": 5.544355249404907,
      "epoch": 1.6831458635703918,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004719278411778097,
      "loss": 5.0735,
      "mean_token_accuracy": 0.20666104257106782,
      "num_tokens": 38519608.0,
      "step": 18555
    },
    {
      "entropy": 5.548975801467895,
      "epoch": 1.683599419448476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047191218839927817,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19572516828775405,
      "num_tokens": 38530737.0,
      "step": 18560
    },
    {
      "entropy": 5.543293809890747,
      "epoch": 1.6840529753265603,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004718965315485523,
      "loss": 4.9857,
      "mean_token_accuracy": 0.204212287068367,
      "num_tokens": 38540925.0,
      "step": 18565
    },
    {
      "entropy": 5.55231466293335,
      "epoch": 1.6845065312046446,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004718808706259561,
      "loss": 5.0631,
      "mean_token_accuracy": 0.19747941792011262,
      "num_tokens": 38553176.0,
      "step": 18570
    },
    {
      "entropy": 5.599540758132934,
      "epoch": 1.6849600870827286,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047186520563181325,
      "loss": 5.1021,
      "mean_token_accuracy": 0.19440075755119324,
      "num_tokens": 38563394.0,
      "step": 18575
    },
    {
      "entropy": 5.613511562347412,
      "epoch": 1.6854136429608126,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004718495365664477,
      "loss": 5.0467,
      "mean_token_accuracy": 0.20269798040390014,
      "num_tokens": 38573243.0,
      "step": 18580
    },
    {
      "entropy": 5.544748973846436,
      "epoch": 1.685867198838897,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047183386343018364,
      "loss": 5.0687,
      "mean_token_accuracy": 0.19938255101442337,
      "num_tokens": 38584014.0,
      "step": 18585
    },
    {
      "entropy": 5.572828197479248,
      "epoch": 1.6863207547169812,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004718181862233451,
      "loss": 4.996,
      "mean_token_accuracy": 0.20601304918527602,
      "num_tokens": 38594467.0,
      "step": 18590
    },
    {
      "entropy": 5.595120668411255,
      "epoch": 1.6867743105950654,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004718025049462563,
      "loss": 5.0625,
      "mean_token_accuracy": 0.20386261492967606,
      "num_tokens": 38604407.0,
      "step": 18595
    },
    {
      "entropy": 5.581459188461304,
      "epoch": 1.6872278664731495,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004717868195992416,
      "loss": 5.0629,
      "mean_token_accuracy": 0.20675850361585618,
      "num_tokens": 38614508.0,
      "step": 18600
    },
    {
      "entropy": 5.6191541194915775,
      "epoch": 1.6876814223512335,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004717711301826253,
      "loss": 5.0668,
      "mean_token_accuracy": 0.20611653476953506,
      "num_tokens": 38626244.0,
      "step": 18605
    },
    {
      "entropy": 5.520521974563598,
      "epoch": 1.6881349782293178,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047175543669673186,
      "loss": 5.0094,
      "mean_token_accuracy": 0.20494984090328217,
      "num_tokens": 38636694.0,
      "step": 18610
    },
    {
      "entropy": 5.6410387516021725,
      "epoch": 1.688588534107402,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047173973914188587,
      "loss": 5.1134,
      "mean_token_accuracy": 0.20223284512758255,
      "num_tokens": 38647414.0,
      "step": 18615
    },
    {
      "entropy": 5.5597906589508055,
      "epoch": 1.6890420899854863,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047172403751841196,
      "loss": 5.0246,
      "mean_token_accuracy": 0.20177806913852692,
      "num_tokens": 38657425.0,
      "step": 18620
    },
    {
      "entropy": 5.58294153213501,
      "epoch": 1.6894956458635704,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047170833182663485,
      "loss": 5.0429,
      "mean_token_accuracy": 0.2058634117245674,
      "num_tokens": 38667751.0,
      "step": 18625
    },
    {
      "entropy": 5.461794662475586,
      "epoch": 1.6899492017416544,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004716926220668793,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20780032575130464,
      "num_tokens": 38676918.0,
      "step": 18630
    },
    {
      "entropy": 5.486989974975586,
      "epoch": 1.6904027576197387,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047167690823947015,
      "loss": 5.025,
      "mean_token_accuracy": 0.21017662882804872,
      "num_tokens": 38685887.0,
      "step": 18635
    },
    {
      "entropy": 5.512678337097168,
      "epoch": 1.690856313497823,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004716611903447325,
      "loss": 5.0237,
      "mean_token_accuracy": 0.20788081139326095,
      "num_tokens": 38696179.0,
      "step": 18640
    },
    {
      "entropy": 5.5653375625610355,
      "epoch": 1.6913098693759072,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004716454683829912,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20205404609441757,
      "num_tokens": 38705777.0,
      "step": 18645
    },
    {
      "entropy": 5.53246579170227,
      "epoch": 1.6917634252539913,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047162974235457163,
      "loss": 5.023,
      "mean_token_accuracy": 0.20526780039072037,
      "num_tokens": 38717709.0,
      "step": 18650
    },
    {
      "entropy": 5.516905975341797,
      "epoch": 1.6922169811320755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004716140122597988,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20970158874988556,
      "num_tokens": 38729517.0,
      "step": 18655
    },
    {
      "entropy": 5.529263734817505,
      "epoch": 1.6926705370101596,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004715982780989981,
      "loss": 5.032,
      "mean_token_accuracy": 0.2079072117805481,
      "num_tokens": 38740878.0,
      "step": 18660
    },
    {
      "entropy": 5.517541885375977,
      "epoch": 1.6931240928882438,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047158253987249487,
      "loss": 4.9665,
      "mean_token_accuracy": 0.2089331105351448,
      "num_tokens": 38751291.0,
      "step": 18665
    },
    {
      "entropy": 5.509375333786011,
      "epoch": 1.693577648766328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047156679758061465,
      "loss": 4.959,
      "mean_token_accuracy": 0.2138681709766388,
      "num_tokens": 38760940.0,
      "step": 18670
    },
    {
      "entropy": 5.549563932418823,
      "epoch": 1.6940312046444121,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047155105122368297,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21167542934417724,
      "num_tokens": 38770240.0,
      "step": 18675
    },
    {
      "entropy": 5.466352033615112,
      "epoch": 1.6944847605224964,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004715353008020255,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20637205243110657,
      "num_tokens": 38780452.0,
      "step": 18680
    },
    {
      "entropy": 5.537557363510132,
      "epoch": 1.6949383164005805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047151954631596794,
      "loss": 5.0745,
      "mean_token_accuracy": 0.199966262280941,
      "num_tokens": 38790763.0,
      "step": 18685
    },
    {
      "entropy": 5.533513927459717,
      "epoch": 1.6953918722786647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047150378776583603,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21286385506391525,
      "num_tokens": 38802164.0,
      "step": 18690
    },
    {
      "entropy": 5.537321853637695,
      "epoch": 1.695845428156749,
      "grad_norm": 1.0,
      "learning_rate": 0.0004714880251519557,
      "loss": 5.0086,
      "mean_token_accuracy": 0.20215222984552383,
      "num_tokens": 38812901.0,
      "step": 18695
    },
    {
      "entropy": 5.508861923217774,
      "epoch": 1.6962989840348333,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000471472258474653,
      "loss": 4.9811,
      "mean_token_accuracy": 0.2096114531159401,
      "num_tokens": 38823305.0,
      "step": 18700
    },
    {
      "entropy": 5.539385175704956,
      "epoch": 1.6967525399129173,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047145648773425384,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20409904271364213,
      "num_tokens": 38833602.0,
      "step": 18705
    },
    {
      "entropy": 5.576952886581421,
      "epoch": 1.6972060957910013,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047144071293108456,
      "loss": 5.071,
      "mean_token_accuracy": 0.2042400449514389,
      "num_tokens": 38843950.0,
      "step": 18710
    },
    {
      "entropy": 5.52810320854187,
      "epoch": 1.6976596516690856,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004714249340654712,
      "loss": 4.9925,
      "mean_token_accuracy": 0.2058522433042526,
      "num_tokens": 38853753.0,
      "step": 18715
    },
    {
      "entropy": 5.467414999008179,
      "epoch": 1.6981132075471699,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004714091511377402,
      "loss": 4.9304,
      "mean_token_accuracy": 0.20456690341234207,
      "num_tokens": 38863394.0,
      "step": 18720
    },
    {
      "entropy": 5.521736574172974,
      "epoch": 1.6985667634252541,
      "grad_norm": 1.125,
      "learning_rate": 0.00047139336414821784,
      "loss": 5.024,
      "mean_token_accuracy": 0.20001382976770402,
      "num_tokens": 38872972.0,
      "step": 18725
    },
    {
      "entropy": 5.5153477668762205,
      "epoch": 1.6990203193033382,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047137757309723073,
      "loss": 5.0092,
      "mean_token_accuracy": 0.212299245595932,
      "num_tokens": 38883251.0,
      "step": 18730
    },
    {
      "entropy": 5.516787052154541,
      "epoch": 1.6994738751814222,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004713617779851053,
      "loss": 5.0281,
      "mean_token_accuracy": 0.20373117178678513,
      "num_tokens": 38893274.0,
      "step": 18735
    },
    {
      "entropy": 5.61175503730774,
      "epoch": 1.6999274310595065,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004713459788121683,
      "loss": 5.1158,
      "mean_token_accuracy": 0.19592057019472123,
      "num_tokens": 38903870.0,
      "step": 18740
    },
    {
      "entropy": 5.572488260269165,
      "epoch": 1.7003809869375908,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004713301755787464,
      "loss": 5.0541,
      "mean_token_accuracy": 0.1994834527373314,
      "num_tokens": 38914921.0,
      "step": 18745
    },
    {
      "entropy": 5.570324230194092,
      "epoch": 1.700834542815675,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004713143682851665,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20570874065160752,
      "num_tokens": 38924630.0,
      "step": 18750
    },
    {
      "entropy": 5.477497339248657,
      "epoch": 1.701288098693759,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004712985569317553,
      "loss": 4.9844,
      "mean_token_accuracy": 0.20439738780260086,
      "num_tokens": 38934926.0,
      "step": 18755
    },
    {
      "entropy": 5.603098392486572,
      "epoch": 1.7017416545718431,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004712827415188401,
      "loss": 5.1481,
      "mean_token_accuracy": 0.19430297911167144,
      "num_tokens": 38944513.0,
      "step": 18760
    },
    {
      "entropy": 5.575858306884766,
      "epoch": 1.7021952104499274,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004712669220467476,
      "loss": 5.0033,
      "mean_token_accuracy": 0.20788175910711287,
      "num_tokens": 38955918.0,
      "step": 18765
    },
    {
      "entropy": 5.568398094177246,
      "epoch": 1.7026487663280117,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047125109851580533,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20533353090286255,
      "num_tokens": 38966221.0,
      "step": 18770
    },
    {
      "entropy": 5.535524511337281,
      "epoch": 1.703102322206096,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047123527092634025,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20195710361003877,
      "num_tokens": 38976463.0,
      "step": 18775
    },
    {
      "entropy": 5.479678440093994,
      "epoch": 1.70355587808418,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004712194392786797,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20722113698720931,
      "num_tokens": 38986375.0,
      "step": 18780
    },
    {
      "entropy": 5.625212955474853,
      "epoch": 1.704009433962264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047120360357315127,
      "loss": 5.064,
      "mean_token_accuracy": 0.20052340477705002,
      "num_tokens": 38996116.0,
      "step": 18785
    },
    {
      "entropy": 5.509733486175537,
      "epoch": 1.7044629898403483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047118776381008216,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20391646027565002,
      "num_tokens": 39005846.0,
      "step": 18790
    },
    {
      "entropy": 5.573023700714112,
      "epoch": 1.7049165457184325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047117191998980026,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19692962765693664,
      "num_tokens": 39016744.0,
      "step": 18795
    },
    {
      "entropy": 5.5535383224487305,
      "epoch": 1.7053701015965168,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047115607211263295,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19864329397678376,
      "num_tokens": 39026725.0,
      "step": 18800
    },
    {
      "entropy": 5.5790722370147705,
      "epoch": 1.7058236574746009,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047114022017890814,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20246905237436294,
      "num_tokens": 39036851.0,
      "step": 18805
    },
    {
      "entropy": 5.526825666427612,
      "epoch": 1.706277213352685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004711243641889536,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2043837919831276,
      "num_tokens": 39047179.0,
      "step": 18810
    },
    {
      "entropy": 5.49183235168457,
      "epoch": 1.7067307692307692,
      "grad_norm": 1.0,
      "learning_rate": 0.00047110850414309714,
      "loss": 4.8502,
      "mean_token_accuracy": 0.2154756158590317,
      "num_tokens": 39057758.0,
      "step": 18815
    },
    {
      "entropy": 5.5515491485595705,
      "epoch": 1.7071843251088534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047109264004166696,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20021470189094542,
      "num_tokens": 39068142.0,
      "step": 18820
    },
    {
      "entropy": 5.5086527347564695,
      "epoch": 1.7076378809869377,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004710767718849909,
      "loss": 4.9844,
      "mean_token_accuracy": 0.19743564873933792,
      "num_tokens": 39078157.0,
      "step": 18825
    },
    {
      "entropy": 5.514392566680908,
      "epoch": 1.7080914368650217,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004710608996733973,
      "loss": 5.0044,
      "mean_token_accuracy": 0.21087042093276978,
      "num_tokens": 39089090.0,
      "step": 18830
    },
    {
      "entropy": 5.553383779525757,
      "epoch": 1.708544992743106,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047104502340721433,
      "loss": 5.0734,
      "mean_token_accuracy": 0.19710729867219925,
      "num_tokens": 39099079.0,
      "step": 18835
    },
    {
      "entropy": 5.547780752182007,
      "epoch": 1.70899854862119,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047102914308677033,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20297769010066985,
      "num_tokens": 39109927.0,
      "step": 18840
    },
    {
      "entropy": 5.581135034561157,
      "epoch": 1.7094521044992743,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047101325871239363,
      "loss": 5.0103,
      "mean_token_accuracy": 0.2026897057890892,
      "num_tokens": 39119806.0,
      "step": 18845
    },
    {
      "entropy": 5.597395563125611,
      "epoch": 1.7099056603773586,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004709973702844128,
      "loss": 5.0869,
      "mean_token_accuracy": 0.20226970463991165,
      "num_tokens": 39130147.0,
      "step": 18850
    },
    {
      "entropy": 5.476788091659546,
      "epoch": 1.7103592162554426,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047098147780315645,
      "loss": 4.902,
      "mean_token_accuracy": 0.21620516031980513,
      "num_tokens": 39139937.0,
      "step": 18855
    },
    {
      "entropy": 5.548338937759399,
      "epoch": 1.710812772133527,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004709655812689532,
      "loss": 5.023,
      "mean_token_accuracy": 0.20639331191778182,
      "num_tokens": 39149977.0,
      "step": 18860
    },
    {
      "entropy": 5.53417010307312,
      "epoch": 1.711266328011611,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004709496806821318,
      "loss": 5.0338,
      "mean_token_accuracy": 0.20468510836362838,
      "num_tokens": 39161076.0,
      "step": 18865
    },
    {
      "entropy": 5.57164454460144,
      "epoch": 1.7117198838896952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047093377604302106,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19266926050186156,
      "num_tokens": 39170776.0,
      "step": 18870
    },
    {
      "entropy": 5.578165769577026,
      "epoch": 1.7121734397677795,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004709178673519499,
      "loss": 5.0447,
      "mean_token_accuracy": 0.19984612464904786,
      "num_tokens": 39181134.0,
      "step": 18875
    },
    {
      "entropy": 5.5719358921051025,
      "epoch": 1.7126269956458637,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004709019546092473,
      "loss": 5.1256,
      "mean_token_accuracy": 0.19469009190797806,
      "num_tokens": 39193331.0,
      "step": 18880
    },
    {
      "entropy": 5.624257659912109,
      "epoch": 1.7130805515239478,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047088603781524237,
      "loss": 5.0845,
      "mean_token_accuracy": 0.19632348269224167,
      "num_tokens": 39204372.0,
      "step": 18885
    },
    {
      "entropy": 5.576696538925171,
      "epoch": 1.7135341074020318,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004708701169702642,
      "loss": 4.9966,
      "mean_token_accuracy": 0.2091236263513565,
      "num_tokens": 39214725.0,
      "step": 18890
    },
    {
      "entropy": 5.4728844165802,
      "epoch": 1.713987663280116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004708541920746422,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20199199467897416,
      "num_tokens": 39225119.0,
      "step": 18895
    },
    {
      "entropy": 5.503029870986938,
      "epoch": 1.7144412191582004,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004708382631287055,
      "loss": 4.9667,
      "mean_token_accuracy": 0.20356596112251282,
      "num_tokens": 39235021.0,
      "step": 18900
    },
    {
      "entropy": 5.5845558643341064,
      "epoch": 1.7148947750362846,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004708223301327837,
      "loss": 5.1454,
      "mean_token_accuracy": 0.19812427461147308,
      "num_tokens": 39245830.0,
      "step": 18905
    },
    {
      "entropy": 5.646688604354859,
      "epoch": 1.7153483309143687,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047080639308720617,
      "loss": 5.1165,
      "mean_token_accuracy": 0.2034643605351448,
      "num_tokens": 39255264.0,
      "step": 18910
    },
    {
      "entropy": 5.553671073913574,
      "epoch": 1.7158018867924527,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047079045199230254,
      "loss": 4.9903,
      "mean_token_accuracy": 0.20753138065338134,
      "num_tokens": 39264966.0,
      "step": 18915
    },
    {
      "entropy": 5.5494787216186525,
      "epoch": 1.716255442670537,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047077450684840247,
      "loss": 5.1004,
      "mean_token_accuracy": 0.1983710289001465,
      "num_tokens": 39275082.0,
      "step": 18920
    },
    {
      "entropy": 5.622508811950683,
      "epoch": 1.7167089985486212,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004707585576558357,
      "loss": 5.1766,
      "mean_token_accuracy": 0.19712687581777572,
      "num_tokens": 39285426.0,
      "step": 18925
    },
    {
      "entropy": 5.640629720687866,
      "epoch": 1.7171625544267055,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004707426044149321,
      "loss": 5.1012,
      "mean_token_accuracy": 0.20073680877685546,
      "num_tokens": 39296633.0,
      "step": 18930
    },
    {
      "entropy": 5.516549062728882,
      "epoch": 1.7176161103047896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047072664712602153,
      "loss": 4.9233,
      "mean_token_accuracy": 0.21242664009332657,
      "num_tokens": 39306775.0,
      "step": 18935
    },
    {
      "entropy": 5.538043451309204,
      "epoch": 1.7180696661828736,
      "grad_norm": 1.0625,
      "learning_rate": 0.000470710685789434,
      "loss": 5.0457,
      "mean_token_accuracy": 0.1989818185567856,
      "num_tokens": 39317917.0,
      "step": 18940
    },
    {
      "entropy": 5.572309398651123,
      "epoch": 1.7185232220609579,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047069472040549965,
      "loss": 5.0506,
      "mean_token_accuracy": 0.20398209542036055,
      "num_tokens": 39328387.0,
      "step": 18945
    },
    {
      "entropy": 5.482699537277222,
      "epoch": 1.7189767779390421,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004706787509745486,
      "loss": 4.8989,
      "mean_token_accuracy": 0.21320468336343765,
      "num_tokens": 39337639.0,
      "step": 18950
    },
    {
      "entropy": 5.552534103393555,
      "epoch": 1.7194303338171264,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047066277749691115,
      "loss": 5.0336,
      "mean_token_accuracy": 0.20591485649347305,
      "num_tokens": 39347971.0,
      "step": 18955
    },
    {
      "entropy": 5.576763486862182,
      "epoch": 1.7198838896952104,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004706467999729175,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20291970521211625,
      "num_tokens": 39358038.0,
      "step": 18960
    },
    {
      "entropy": 5.5647767066955565,
      "epoch": 1.7203374455732945,
      "grad_norm": 1.125,
      "learning_rate": 0.0004706308184028983,
      "loss": 5.0621,
      "mean_token_accuracy": 0.20378267467021943,
      "num_tokens": 39367834.0,
      "step": 18965
    },
    {
      "entropy": 5.572733592987061,
      "epoch": 1.7207910014513788,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004706148327871839,
      "loss": 5.0422,
      "mean_token_accuracy": 0.2063360780477524,
      "num_tokens": 39378924.0,
      "step": 18970
    },
    {
      "entropy": 5.635785675048828,
      "epoch": 1.721244557329463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047059884312610487,
      "loss": 5.1586,
      "mean_token_accuracy": 0.19298899471759795,
      "num_tokens": 39390082.0,
      "step": 18975
    },
    {
      "entropy": 5.568527746200561,
      "epoch": 1.7216981132075473,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047058284941999185,
      "loss": 5.0717,
      "mean_token_accuracy": 0.20253492444753646,
      "num_tokens": 39400485.0,
      "step": 18980
    },
    {
      "entropy": 5.5739670753479,
      "epoch": 1.7221516690856313,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004705668516691758,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20526162534952164,
      "num_tokens": 39411601.0,
      "step": 18985
    },
    {
      "entropy": 5.5383148193359375,
      "epoch": 1.7226052249637154,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047055084987398735,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20863622277975083,
      "num_tokens": 39421625.0,
      "step": 18990
    },
    {
      "entropy": 5.5457957744598385,
      "epoch": 1.7230587808417996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004705348440347575,
      "loss": 5.0186,
      "mean_token_accuracy": 0.19834876507520677,
      "num_tokens": 39432033.0,
      "step": 18995
    },
    {
      "entropy": 5.519546318054199,
      "epoch": 1.723512336719884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047051883415181724,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20518908947706221,
      "num_tokens": 39442514.0,
      "step": 19000
    },
    {
      "entropy": 5.529253387451172,
      "epoch": 1.7239658925979682,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004705028202254977,
      "loss": 4.897,
      "mean_token_accuracy": 0.21023046970367432,
      "num_tokens": 39452497.0,
      "step": 19005
    },
    {
      "entropy": 5.531752395629883,
      "epoch": 1.7244194484760522,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047048680225612996,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20395348966121674,
      "num_tokens": 39462173.0,
      "step": 19010
    },
    {
      "entropy": 5.590306568145752,
      "epoch": 1.7248730043541363,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004704707802440454,
      "loss": 5.0729,
      "mean_token_accuracy": 0.20942844450473785,
      "num_tokens": 39473123.0,
      "step": 19015
    },
    {
      "entropy": 5.588532257080078,
      "epoch": 1.7253265602322205,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047045475418957524,
      "loss": 5.0522,
      "mean_token_accuracy": 0.19402004182338714,
      "num_tokens": 39484142.0,
      "step": 19020
    },
    {
      "entropy": 5.568538188934326,
      "epoch": 1.7257801161103048,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004704387240930509,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19874946773052216,
      "num_tokens": 39494926.0,
      "step": 19025
    },
    {
      "entropy": 5.536000108718872,
      "epoch": 1.726233671988389,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004704226899548041,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20445523411035538,
      "num_tokens": 39505153.0,
      "step": 19030
    },
    {
      "entropy": 5.509835290908813,
      "epoch": 1.726687227866473,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047040665177516616,
      "loss": 4.9852,
      "mean_token_accuracy": 0.20506981313228606,
      "num_tokens": 39515905.0,
      "step": 19035
    },
    {
      "entropy": 5.510123252868652,
      "epoch": 1.7271407837445574,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047039060955446886,
      "loss": 4.9964,
      "mean_token_accuracy": 0.20858711749315262,
      "num_tokens": 39526762.0,
      "step": 19040
    },
    {
      "entropy": 5.503658151626587,
      "epoch": 1.7275943396226414,
      "grad_norm": 1.078125,
      "learning_rate": 0.000470374563293044,
      "loss": 4.988,
      "mean_token_accuracy": 0.20179152339696885,
      "num_tokens": 39537340.0,
      "step": 19045
    },
    {
      "entropy": 5.5714703559875485,
      "epoch": 1.7280478955007257,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004703585129912234,
      "loss": 5.0015,
      "mean_token_accuracy": 0.20270833969116211,
      "num_tokens": 39547144.0,
      "step": 19050
    },
    {
      "entropy": 5.653240966796875,
      "epoch": 1.72850145137881,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047034245864933894,
      "loss": 5.1215,
      "mean_token_accuracy": 0.19415885508060454,
      "num_tokens": 39557707.0,
      "step": 19055
    },
    {
      "entropy": 5.619681930541992,
      "epoch": 1.728955007256894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047032640026772274,
      "loss": 5.1011,
      "mean_token_accuracy": 0.2001511499285698,
      "num_tokens": 39567629.0,
      "step": 19060
    },
    {
      "entropy": 5.5302807807922365,
      "epoch": 1.7294085631349783,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004703103378467067,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20744604766368865,
      "num_tokens": 39577918.0,
      "step": 19065
    },
    {
      "entropy": 5.571910429000854,
      "epoch": 1.7298621190130623,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047029427138662314,
      "loss": 5.0371,
      "mean_token_accuracy": 0.21134694665670395,
      "num_tokens": 39587838.0,
      "step": 19070
    },
    {
      "entropy": 5.525185012817383,
      "epoch": 1.7303156748911466,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004702782008878042,
      "loss": 4.9286,
      "mean_token_accuracy": 0.20921002179384232,
      "num_tokens": 39597353.0,
      "step": 19075
    },
    {
      "entropy": 5.522932624816894,
      "epoch": 1.7307692307692308,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004702621263505824,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2087055891752243,
      "num_tokens": 39607349.0,
      "step": 19080
    },
    {
      "entropy": 5.562011480331421,
      "epoch": 1.731222786647315,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004702460477752901,
      "loss": 5.0562,
      "mean_token_accuracy": 0.20134317576885224,
      "num_tokens": 39617810.0,
      "step": 19085
    },
    {
      "entropy": 5.5258690357208256,
      "epoch": 1.7316763425253991,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004702299651622597,
      "loss": 4.942,
      "mean_token_accuracy": 0.21373670399188996,
      "num_tokens": 39626805.0,
      "step": 19090
    },
    {
      "entropy": 5.464047527313232,
      "epoch": 1.7321298984034832,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004702138785118239,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21195108890533448,
      "num_tokens": 39637839.0,
      "step": 19095
    },
    {
      "entropy": 5.493241214752198,
      "epoch": 1.7325834542815675,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004701977878243153,
      "loss": 4.9499,
      "mean_token_accuracy": 0.20881755650043488,
      "num_tokens": 39648697.0,
      "step": 19100
    },
    {
      "entropy": 5.490741825103759,
      "epoch": 1.7330370101596517,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004701816931000668,
      "loss": 4.991,
      "mean_token_accuracy": 0.20672685503959656,
      "num_tokens": 39658154.0,
      "step": 19105
    },
    {
      "entropy": 5.542169046401978,
      "epoch": 1.733490566037736,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047016559433941106,
      "loss": 4.9941,
      "mean_token_accuracy": 0.2038709044456482,
      "num_tokens": 39667939.0,
      "step": 19110
    },
    {
      "entropy": 5.519157743453979,
      "epoch": 1.73394412191582,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004701494915426812,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20618734061717986,
      "num_tokens": 39678289.0,
      "step": 19115
    },
    {
      "entropy": 5.559554719924927,
      "epoch": 1.734397677793904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047013338471021,
      "loss": 5.0078,
      "mean_token_accuracy": 0.2046155884861946,
      "num_tokens": 39689901.0,
      "step": 19120
    },
    {
      "entropy": 5.5800446510314945,
      "epoch": 1.7348512336719883,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047011727384233074,
      "loss": 5.0434,
      "mean_token_accuracy": 0.20256204903125763,
      "num_tokens": 39700831.0,
      "step": 19125
    },
    {
      "entropy": 5.562351942062378,
      "epoch": 1.7353047895500726,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004701011589393765,
      "loss": 5.0484,
      "mean_token_accuracy": 0.20052081495523452,
      "num_tokens": 39711794.0,
      "step": 19130
    },
    {
      "entropy": 5.521185445785522,
      "epoch": 1.7357583454281569,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047008504000168054,
      "loss": 4.9898,
      "mean_token_accuracy": 0.2035038337111473,
      "num_tokens": 39721889.0,
      "step": 19135
    },
    {
      "entropy": 5.497358751296997,
      "epoch": 1.736211901306241,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004700689170295763,
      "loss": 5.074,
      "mean_token_accuracy": 0.20340992957353593,
      "num_tokens": 39732562.0,
      "step": 19140
    },
    {
      "entropy": 5.610089254379273,
      "epoch": 1.736665457184325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047005279002339717,
      "loss": 5.0332,
      "mean_token_accuracy": 0.20637455582618713,
      "num_tokens": 39742286.0,
      "step": 19145
    },
    {
      "entropy": 5.5739748001098635,
      "epoch": 1.7371190130624092,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047003665898347663,
      "loss": 5.0318,
      "mean_token_accuracy": 0.20306781083345413,
      "num_tokens": 39754336.0,
      "step": 19150
    },
    {
      "entropy": 5.496878910064697,
      "epoch": 1.7375725689404935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004700205239101483,
      "loss": 4.936,
      "mean_token_accuracy": 0.21169695258140564,
      "num_tokens": 39763545.0,
      "step": 19155
    },
    {
      "entropy": 5.546338224411011,
      "epoch": 1.7380261248185778,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004700043848037458,
      "loss": 5.1084,
      "mean_token_accuracy": 0.19563563019037247,
      "num_tokens": 39774536.0,
      "step": 19160
    },
    {
      "entropy": 5.5876946449279785,
      "epoch": 1.7384796806966618,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004699882416646029,
      "loss": 4.9862,
      "mean_token_accuracy": 0.20746959298849105,
      "num_tokens": 39785377.0,
      "step": 19165
    },
    {
      "entropy": 5.5691807746887205,
      "epoch": 1.7389332365747459,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004699720944930535,
      "loss": 5.106,
      "mean_token_accuracy": 0.19034550935029984,
      "num_tokens": 39796692.0,
      "step": 19170
    },
    {
      "entropy": 5.473299121856689,
      "epoch": 1.7393867924528301,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046995594328943147,
      "loss": 5.0048,
      "mean_token_accuracy": 0.2059233844280243,
      "num_tokens": 39807927.0,
      "step": 19175
    },
    {
      "entropy": 5.669158411026001,
      "epoch": 1.7398403483309144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004699397880540709,
      "loss": 5.1929,
      "mean_token_accuracy": 0.20216887295246125,
      "num_tokens": 39817617.0,
      "step": 19180
    },
    {
      "entropy": 5.580893421173096,
      "epoch": 1.7402939042089987,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004699236287873058,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21136983335018159,
      "num_tokens": 39827092.0,
      "step": 19185
    },
    {
      "entropy": 5.529440689086914,
      "epoch": 1.7407474600870827,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004699074654894705,
      "loss": 4.9738,
      "mean_token_accuracy": 0.20527196526527405,
      "num_tokens": 39837378.0,
      "step": 19190
    },
    {
      "entropy": 5.53071403503418,
      "epoch": 1.7412010159651667,
      "grad_norm": 1.03125,
      "learning_rate": 0.000469891298160899,
      "loss": 5.0009,
      "mean_token_accuracy": 0.19492065608501435,
      "num_tokens": 39848191.0,
      "step": 19195
    },
    {
      "entropy": 5.512627172470093,
      "epoch": 1.741654571843251,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046987512680192586,
      "loss": 4.9916,
      "mean_token_accuracy": 0.204746875166893,
      "num_tokens": 39857886.0,
      "step": 19200
    },
    {
      "entropy": 5.600387763977051,
      "epoch": 1.7421081277213353,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046985895141288546,
      "loss": 5.0785,
      "mean_token_accuracy": 0.1978824257850647,
      "num_tokens": 39868348.0,
      "step": 19205
    },
    {
      "entropy": 5.620454978942871,
      "epoch": 1.7425616835994195,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046984277199411224,
      "loss": 5.0972,
      "mean_token_accuracy": 0.20216984152793885,
      "num_tokens": 39879129.0,
      "step": 19210
    },
    {
      "entropy": 5.51452603340149,
      "epoch": 1.7430152394775036,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046982658854594083,
      "loss": 4.984,
      "mean_token_accuracy": 0.20440545082092285,
      "num_tokens": 39890239.0,
      "step": 19215
    },
    {
      "entropy": 5.4759642601013185,
      "epoch": 1.7434687953555879,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046981040106870595,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2083138108253479,
      "num_tokens": 39900907.0,
      "step": 19220
    },
    {
      "entropy": 5.566823959350586,
      "epoch": 1.743922351233672,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004697942095627423,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20522894561290742,
      "num_tokens": 39911564.0,
      "step": 19225
    },
    {
      "entropy": 5.534403371810913,
      "epoch": 1.7443759071117562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004697780140283848,
      "loss": 4.9614,
      "mean_token_accuracy": 0.2044839709997177,
      "num_tokens": 39922168.0,
      "step": 19230
    },
    {
      "entropy": 5.536777639389038,
      "epoch": 1.7448294629898404,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004697618144659683,
      "loss": 4.9986,
      "mean_token_accuracy": 0.20845214426517486,
      "num_tokens": 39932690.0,
      "step": 19235
    },
    {
      "entropy": 5.645267963409424,
      "epoch": 1.7452830188679245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004697456108758278,
      "loss": 5.149,
      "mean_token_accuracy": 0.20028664767742158,
      "num_tokens": 39942970.0,
      "step": 19240
    },
    {
      "entropy": 5.525376605987549,
      "epoch": 1.7457365747460087,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004697294032582986,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20895958989858626,
      "num_tokens": 39953255.0,
      "step": 19245
    },
    {
      "entropy": 5.507604694366455,
      "epoch": 1.7461901306240928,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046971319161371564,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21043440699577332,
      "num_tokens": 39963521.0,
      "step": 19250
    },
    {
      "entropy": 5.5434962749481205,
      "epoch": 1.746643686502177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046969697594241426,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20170967131853104,
      "num_tokens": 39973494.0,
      "step": 19255
    },
    {
      "entropy": 5.534239292144775,
      "epoch": 1.7470972423802613,
      "grad_norm": 1.0,
      "learning_rate": 0.0004696807562447298,
      "loss": 5.1091,
      "mean_token_accuracy": 0.19806908071041107,
      "num_tokens": 39984652.0,
      "step": 19260
    },
    {
      "entropy": 5.531661176681519,
      "epoch": 1.7475507982583456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046966453252099763,
      "loss": 5.0623,
      "mean_token_accuracy": 0.20599662363529206,
      "num_tokens": 39995490.0,
      "step": 19265
    },
    {
      "entropy": 5.502593040466309,
      "epoch": 1.7480043541364296,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046964830477155335,
      "loss": 4.8317,
      "mean_token_accuracy": 0.22880762964487075,
      "num_tokens": 40006447.0,
      "step": 19270
    },
    {
      "entropy": 5.607112073898316,
      "epoch": 1.7484579100145137,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046963207299673257,
      "loss": 5.0919,
      "mean_token_accuracy": 0.20307234823703765,
      "num_tokens": 40015733.0,
      "step": 19275
    },
    {
      "entropy": 5.598391199111939,
      "epoch": 1.748911465892598,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046961583719687085,
      "loss": 4.9486,
      "mean_token_accuracy": 0.2105601742863655,
      "num_tokens": 40025844.0,
      "step": 19280
    },
    {
      "entropy": 5.516723155975342,
      "epoch": 1.7493650217706822,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046959959737230414,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21180037707090377,
      "num_tokens": 40035942.0,
      "step": 19285
    },
    {
      "entropy": 5.472798824310303,
      "epoch": 1.7498185776487665,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046958335352336806,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21567040979862212,
      "num_tokens": 40045551.0,
      "step": 19290
    },
    {
      "entropy": 5.558046770095825,
      "epoch": 1.7502721335268505,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004695671056503987,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2100541278719902,
      "num_tokens": 40055803.0,
      "step": 19295
    },
    {
      "entropy": 5.535646677017212,
      "epoch": 1.7507256894049346,
      "grad_norm": 1.03125,
      "learning_rate": 0.000469550853753732,
      "loss": 5.0046,
      "mean_token_accuracy": 0.2040293887257576,
      "num_tokens": 40066004.0,
      "step": 19300
    },
    {
      "entropy": 5.547567701339721,
      "epoch": 1.7511792452830188,
      "grad_norm": 1.09375,
      "learning_rate": 0.000469534597833704,
      "loss": 5.0056,
      "mean_token_accuracy": 0.2093586027622223,
      "num_tokens": 40075522.0,
      "step": 19305
    },
    {
      "entropy": 5.5241851806640625,
      "epoch": 1.751632801161103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000469518337890651,
      "loss": 5.0075,
      "mean_token_accuracy": 0.2129354938864708,
      "num_tokens": 40084839.0,
      "step": 19310
    },
    {
      "entropy": 5.571526432037354,
      "epoch": 1.7520863570391874,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046950207392490917,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20345304161310196,
      "num_tokens": 40094392.0,
      "step": 19315
    },
    {
      "entropy": 5.500971698760987,
      "epoch": 1.7525399129172714,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046948580593681496,
      "loss": 4.9803,
      "mean_token_accuracy": 0.20845315605401993,
      "num_tokens": 40105990.0,
      "step": 19320
    },
    {
      "entropy": 5.486760139465332,
      "epoch": 1.7529934687953554,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004694695339267047,
      "loss": 4.9008,
      "mean_token_accuracy": 0.21510082483291626,
      "num_tokens": 40116143.0,
      "step": 19325
    },
    {
      "entropy": 5.655039691925049,
      "epoch": 1.7534470246734397,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046945325789491483,
      "loss": 5.089,
      "mean_token_accuracy": 0.20208169966936113,
      "num_tokens": 40126849.0,
      "step": 19330
    },
    {
      "entropy": 5.551968955993653,
      "epoch": 1.753900580551524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004694369778417822,
      "loss": 5.0452,
      "mean_token_accuracy": 0.20378676950931549,
      "num_tokens": 40137259.0,
      "step": 19335
    },
    {
      "entropy": 5.43944149017334,
      "epoch": 1.7543541364296082,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004694206937676431,
      "loss": 4.9077,
      "mean_token_accuracy": 0.20813882648944854,
      "num_tokens": 40148578.0,
      "step": 19340
    },
    {
      "entropy": 5.605654096603393,
      "epoch": 1.7548076923076923,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004694044056728347,
      "loss": 5.0896,
      "mean_token_accuracy": 0.1949380248785019,
      "num_tokens": 40159754.0,
      "step": 19345
    },
    {
      "entropy": 5.568848085403443,
      "epoch": 1.7552612481857763,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046938811355769363,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20421231389045716,
      "num_tokens": 40169388.0,
      "step": 19350
    },
    {
      "entropy": 5.539310693740845,
      "epoch": 1.7557148040638606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046937181742255673,
      "loss": 5.0168,
      "mean_token_accuracy": 0.2036700502038002,
      "num_tokens": 40179271.0,
      "step": 19355
    },
    {
      "entropy": 5.590096855163575,
      "epoch": 1.7561683599419449,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004693555172677612,
      "loss": 5.0602,
      "mean_token_accuracy": 0.20551477372646332,
      "num_tokens": 40189292.0,
      "step": 19360
    },
    {
      "entropy": 5.516025447845459,
      "epoch": 1.7566219158200291,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000469339213093644,
      "loss": 5.0128,
      "mean_token_accuracy": 0.20286233127117156,
      "num_tokens": 40199271.0,
      "step": 19365
    },
    {
      "entropy": 5.531123161315918,
      "epoch": 1.7570754716981132,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046932290490054235,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20643097013235093,
      "num_tokens": 40210386.0,
      "step": 19370
    },
    {
      "entropy": 5.578841209411621,
      "epoch": 1.7575290275761972,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004693065926887936,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20037015676498413,
      "num_tokens": 40220892.0,
      "step": 19375
    },
    {
      "entropy": 5.576885557174682,
      "epoch": 1.7579825834542815,
      "grad_norm": 1.0,
      "learning_rate": 0.00046929027645873484,
      "loss": 4.9987,
      "mean_token_accuracy": 0.20828428864479065,
      "num_tokens": 40231737.0,
      "step": 19380
    },
    {
      "entropy": 5.5091084957122805,
      "epoch": 1.7584361393323658,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004692739562107038,
      "loss": 5.042,
      "mean_token_accuracy": 0.1980927422642708,
      "num_tokens": 40242218.0,
      "step": 19385
    },
    {
      "entropy": 5.525993061065674,
      "epoch": 1.75888969521045,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046925763194503775,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20683387964963912,
      "num_tokens": 40253659.0,
      "step": 19390
    },
    {
      "entropy": 5.507544612884521,
      "epoch": 1.759343251088534,
      "grad_norm": 1.0,
      "learning_rate": 0.0004692413036620744,
      "loss": 4.8609,
      "mean_token_accuracy": 0.21369356215000151,
      "num_tokens": 40264328.0,
      "step": 19395
    },
    {
      "entropy": 5.502504682540893,
      "epoch": 1.7597968069666183,
      "grad_norm": 1.0,
      "learning_rate": 0.0004692249713621514,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2095779672265053,
      "num_tokens": 40275188.0,
      "step": 19400
    },
    {
      "entropy": 5.608981037139893,
      "epoch": 1.7602503628447024,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046920863504560647,
      "loss": 5.1346,
      "mean_token_accuracy": 0.1956270754337311,
      "num_tokens": 40287094.0,
      "step": 19405
    },
    {
      "entropy": 5.5839334487915036,
      "epoch": 1.7607039187227866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004691922947127775,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20392605662345886,
      "num_tokens": 40296551.0,
      "step": 19410
    },
    {
      "entropy": 5.523501968383789,
      "epoch": 1.761157474600871,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004691759503640024,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20223620533943176,
      "num_tokens": 40306920.0,
      "step": 19415
    },
    {
      "entropy": 5.511826515197754,
      "epoch": 1.761611030478955,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004691596019996191,
      "loss": 4.9041,
      "mean_token_accuracy": 0.21371311396360398,
      "num_tokens": 40316797.0,
      "step": 19420
    },
    {
      "entropy": 5.458982658386231,
      "epoch": 1.7620645863570392,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004691432496199658,
      "loss": 4.951,
      "mean_token_accuracy": 0.20571546405553817,
      "num_tokens": 40327002.0,
      "step": 19425
    },
    {
      "entropy": 5.651781749725342,
      "epoch": 1.7625181422351233,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046912689322538064,
      "loss": 5.1108,
      "mean_token_accuracy": 0.20137424170970916,
      "num_tokens": 40337247.0,
      "step": 19430
    },
    {
      "entropy": 5.558581399917602,
      "epoch": 1.7629716981132075,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046911053281620187,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20686520040035247,
      "num_tokens": 40346960.0,
      "step": 19435
    },
    {
      "entropy": 5.588541746139526,
      "epoch": 1.7634252539912918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046909416839276775,
      "loss": 5.0532,
      "mean_token_accuracy": 0.2004593625664711,
      "num_tokens": 40357169.0,
      "step": 19440
    },
    {
      "entropy": 5.501184940338135,
      "epoch": 1.763878809869376,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004690777999554168,
      "loss": 4.8988,
      "mean_token_accuracy": 0.2171902596950531,
      "num_tokens": 40367660.0,
      "step": 19445
    },
    {
      "entropy": 5.5190881252288815,
      "epoch": 1.76433236574746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004690614275044875,
      "loss": 4.914,
      "mean_token_accuracy": 0.20759038627147675,
      "num_tokens": 40378217.0,
      "step": 19450
    },
    {
      "entropy": 5.514788818359375,
      "epoch": 1.7647859216255442,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004690450510403184,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20546943843364715,
      "num_tokens": 40388249.0,
      "step": 19455
    },
    {
      "entropy": 5.5890326499938965,
      "epoch": 1.7652394775036284,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004690286705632483,
      "loss": 5.064,
      "mean_token_accuracy": 0.1979004979133606,
      "num_tokens": 40398239.0,
      "step": 19460
    },
    {
      "entropy": 5.568636322021485,
      "epoch": 1.7656930333817127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004690122860736157,
      "loss": 5.0626,
      "mean_token_accuracy": 0.200815749168396,
      "num_tokens": 40408473.0,
      "step": 19465
    },
    {
      "entropy": 5.640263843536377,
      "epoch": 1.766146589259797,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046899589757175964,
      "loss": 5.1298,
      "mean_token_accuracy": 0.19818018674850463,
      "num_tokens": 40418674.0,
      "step": 19470
    },
    {
      "entropy": 5.501596450805664,
      "epoch": 1.766600145137881,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000468979505058019,
      "loss": 4.97,
      "mean_token_accuracy": 0.20711494535207747,
      "num_tokens": 40429815.0,
      "step": 19475
    },
    {
      "entropy": 5.587862634658814,
      "epoch": 1.767053701015965,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046896310853273274,
      "loss": 5.0866,
      "mean_token_accuracy": 0.20570432543754577,
      "num_tokens": 40439809.0,
      "step": 19480
    },
    {
      "entropy": 5.6429685115814205,
      "epoch": 1.7675072568940493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046894670799624,
      "loss": 5.0691,
      "mean_token_accuracy": 0.20091277956962586,
      "num_tokens": 40449503.0,
      "step": 19485
    },
    {
      "entropy": 5.549521589279175,
      "epoch": 1.7679608127721336,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046893030344887995,
      "loss": 5.0881,
      "mean_token_accuracy": 0.1946729063987732,
      "num_tokens": 40459794.0,
      "step": 19490
    },
    {
      "entropy": 5.5187304496765135,
      "epoch": 1.7684143686502178,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046891389489099166,
      "loss": 4.948,
      "mean_token_accuracy": 0.2143867790699005,
      "num_tokens": 40470044.0,
      "step": 19495
    },
    {
      "entropy": 5.54579644203186,
      "epoch": 1.7688679245283019,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046889748232291476,
      "loss": 5.002,
      "mean_token_accuracy": 0.2024047553539276,
      "num_tokens": 40480709.0,
      "step": 19500
    },
    {
      "entropy": 5.475869750976562,
      "epoch": 1.769321480406386,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046888106574498843,
      "loss": 4.843,
      "mean_token_accuracy": 0.21536738574504852,
      "num_tokens": 40490906.0,
      "step": 19505
    },
    {
      "entropy": 5.504793357849121,
      "epoch": 1.7697750362844702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004688646451575522,
      "loss": 4.9576,
      "mean_token_accuracy": 0.2155078321695328,
      "num_tokens": 40501971.0,
      "step": 19510
    },
    {
      "entropy": 5.525714445114136,
      "epoch": 1.7702285921625545,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046884822056094587,
      "loss": 5.0446,
      "mean_token_accuracy": 0.1978493884205818,
      "num_tokens": 40512685.0,
      "step": 19515
    },
    {
      "entropy": 5.506925630569458,
      "epoch": 1.7706821480406387,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046883179195550883,
      "loss": 4.9837,
      "mean_token_accuracy": 0.20833256095647812,
      "num_tokens": 40523460.0,
      "step": 19520
    },
    {
      "entropy": 5.576283740997314,
      "epoch": 1.7711357039187228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046881535934158094,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20880083590745926,
      "num_tokens": 40533594.0,
      "step": 19525
    },
    {
      "entropy": 5.504837560653686,
      "epoch": 1.7715892597968068,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004687989227195021,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20841706395149232,
      "num_tokens": 40543504.0,
      "step": 19530
    },
    {
      "entropy": 5.526325607299805,
      "epoch": 1.772042815674891,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046878248208961204,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2034996911883354,
      "num_tokens": 40553711.0,
      "step": 19535
    },
    {
      "entropy": 5.5185425758361815,
      "epoch": 1.7724963715529753,
      "grad_norm": 1.09375,
      "learning_rate": 0.000468766037452251,
      "loss": 5.0296,
      "mean_token_accuracy": 0.20368110984563828,
      "num_tokens": 40563731.0,
      "step": 19540
    },
    {
      "entropy": 5.586743545532227,
      "epoch": 1.7729499274310596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046874958880775886,
      "loss": 5.0377,
      "mean_token_accuracy": 0.20407156348228456,
      "num_tokens": 40574235.0,
      "step": 19545
    },
    {
      "entropy": 5.57427830696106,
      "epoch": 1.7734034833091437,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004687331361564758,
      "loss": 4.969,
      "mean_token_accuracy": 0.20999808609485626,
      "num_tokens": 40584459.0,
      "step": 19550
    },
    {
      "entropy": 5.531437730789184,
      "epoch": 1.7738570391872277,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004687166794987422,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20473010540008546,
      "num_tokens": 40594268.0,
      "step": 19555
    },
    {
      "entropy": 5.495528221130371,
      "epoch": 1.774310595065312,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004687002188348983,
      "loss": 4.9667,
      "mean_token_accuracy": 0.2113693654537201,
      "num_tokens": 40604673.0,
      "step": 19560
    },
    {
      "entropy": 5.454802703857422,
      "epoch": 1.7747641509433962,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046868375416528443,
      "loss": 4.9625,
      "mean_token_accuracy": 0.2076304167509079,
      "num_tokens": 40614998.0,
      "step": 19565
    },
    {
      "entropy": 5.547639560699463,
      "epoch": 1.7752177068214805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046866728549024123,
      "loss": 5.0423,
      "mean_token_accuracy": 0.20194907188415528,
      "num_tokens": 40624729.0,
      "step": 19570
    },
    {
      "entropy": 5.563374614715576,
      "epoch": 1.7756712626995645,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004686508128101092,
      "loss": 5.0957,
      "mean_token_accuracy": 0.20164481103420256,
      "num_tokens": 40635673.0,
      "step": 19575
    },
    {
      "entropy": 5.579333925247193,
      "epoch": 1.7761248185776488,
      "grad_norm": 0.984375,
      "learning_rate": 0.000468634336125229,
      "loss": 5.0702,
      "mean_token_accuracy": 0.20136806219816208,
      "num_tokens": 40645990.0,
      "step": 19580
    },
    {
      "entropy": 5.5818102836608885,
      "epoch": 1.7765783744557329,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046861785543594143,
      "loss": 5.0102,
      "mean_token_accuracy": 0.21454309225082396,
      "num_tokens": 40656339.0,
      "step": 19585
    },
    {
      "entropy": 5.520548629760742,
      "epoch": 1.7770319303338171,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004686013707425873,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20953194946050643,
      "num_tokens": 40666371.0,
      "step": 19590
    },
    {
      "entropy": 5.559475946426391,
      "epoch": 1.7774854862119014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046858488204550735,
      "loss": 4.9837,
      "mean_token_accuracy": 0.20916937738657,
      "num_tokens": 40676423.0,
      "step": 19595
    },
    {
      "entropy": 5.511946535110473,
      "epoch": 1.7779390420899854,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004685683893450428,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21064889430999756,
      "num_tokens": 40686331.0,
      "step": 19600
    },
    {
      "entropy": 5.496103239059448,
      "epoch": 1.7783925979680697,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004685518926415346,
      "loss": 4.9904,
      "mean_token_accuracy": 0.21334135830402373,
      "num_tokens": 40697408.0,
      "step": 19605
    },
    {
      "entropy": 5.439297533035278,
      "epoch": 1.7788461538461537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000468535391935324,
      "loss": 4.8678,
      "mean_token_accuracy": 0.21968182027339936,
      "num_tokens": 40707078.0,
      "step": 19610
    },
    {
      "entropy": 5.552730417251587,
      "epoch": 1.779299709724238,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004685188872267521,
      "loss": 4.9968,
      "mean_token_accuracy": 0.2048732966184616,
      "num_tokens": 40718116.0,
      "step": 19615
    },
    {
      "entropy": 5.568325138092041,
      "epoch": 1.7797532656023223,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004685023785161603,
      "loss": 5.0291,
      "mean_token_accuracy": 0.20626704692840575,
      "num_tokens": 40727737.0,
      "step": 19620
    },
    {
      "entropy": 5.5643861293792725,
      "epoch": 1.7802068214804065,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004684858658038901,
      "loss": 4.9754,
      "mean_token_accuracy": 0.20463858544826508,
      "num_tokens": 40737455.0,
      "step": 19625
    },
    {
      "entropy": 5.59386043548584,
      "epoch": 1.7806603773584906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004684693490902828,
      "loss": 5.0122,
      "mean_token_accuracy": 0.2064111277461052,
      "num_tokens": 40746715.0,
      "step": 19630
    },
    {
      "entropy": 5.49885196685791,
      "epoch": 1.7811139332365746,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046845282837568,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21309821605682372,
      "num_tokens": 40756432.0,
      "step": 19635
    },
    {
      "entropy": 5.446016025543213,
      "epoch": 1.781567489114659,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046843630366042355,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20433081090450286,
      "num_tokens": 40766629.0,
      "step": 19640
    },
    {
      "entropy": 5.593451738357544,
      "epoch": 1.7820210449927432,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046841977494485506,
      "loss": 5.0983,
      "mean_token_accuracy": 0.19296467453241348,
      "num_tokens": 40776650.0,
      "step": 19645
    },
    {
      "entropy": 5.60643949508667,
      "epoch": 1.7824746008708274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004684032422293162,
      "loss": 5.0026,
      "mean_token_accuracy": 0.21133604347705842,
      "num_tokens": 40786722.0,
      "step": 19650
    },
    {
      "entropy": 5.441478490829468,
      "epoch": 1.7829281567489115,
      "grad_norm": 1.015625,
      "learning_rate": 0.000468386705514149,
      "loss": 4.8511,
      "mean_token_accuracy": 0.22297567427158355,
      "num_tokens": 40797058.0,
      "step": 19655
    },
    {
      "entropy": 5.481105852127075,
      "epoch": 1.7833817126269955,
      "grad_norm": 1.125,
      "learning_rate": 0.0004683701647996956,
      "loss": 5.041,
      "mean_token_accuracy": 0.20970151722431182,
      "num_tokens": 40808207.0,
      "step": 19660
    },
    {
      "entropy": 5.586927700042724,
      "epoch": 1.7838352685050798,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004683536200862978,
      "loss": 5.0862,
      "mean_token_accuracy": 0.20260453671216966,
      "num_tokens": 40819655.0,
      "step": 19665
    },
    {
      "entropy": 5.59092755317688,
      "epoch": 1.784288824383164,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004683370713742979,
      "loss": 5.0747,
      "mean_token_accuracy": 0.20382409691810607,
      "num_tokens": 40831015.0,
      "step": 19670
    },
    {
      "entropy": 5.512878561019898,
      "epoch": 1.7847423802612483,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046832051866403814,
      "loss": 4.955,
      "mean_token_accuracy": 0.20847117751836777,
      "num_tokens": 40841763.0,
      "step": 19675
    },
    {
      "entropy": 5.506123781204224,
      "epoch": 1.7851959361393324,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004683039619558607,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2064852684736252,
      "num_tokens": 40852479.0,
      "step": 19680
    },
    {
      "entropy": 5.579826307296753,
      "epoch": 1.7856494920174164,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046828740125010813,
      "loss": 5.016,
      "mean_token_accuracy": 0.20882578641176225,
      "num_tokens": 40862486.0,
      "step": 19685
    },
    {
      "entropy": 5.611700773239136,
      "epoch": 1.7861030478955007,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004682708365471228,
      "loss": 5.0757,
      "mean_token_accuracy": 0.20229277312755584,
      "num_tokens": 40872182.0,
      "step": 19690
    },
    {
      "entropy": 5.558951568603516,
      "epoch": 1.786556603773585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046825426784724734,
      "loss": 5.0676,
      "mean_token_accuracy": 0.20649306178092958,
      "num_tokens": 40883036.0,
      "step": 19695
    },
    {
      "entropy": 5.594229221343994,
      "epoch": 1.7870101596516692,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004682376951508244,
      "loss": 5.1016,
      "mean_token_accuracy": 0.1905641257762909,
      "num_tokens": 40893684.0,
      "step": 19700
    },
    {
      "entropy": 5.629052925109863,
      "epoch": 1.7874637155297532,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046822111845819665,
      "loss": 5.0717,
      "mean_token_accuracy": 0.20329589396715164,
      "num_tokens": 40904540.0,
      "step": 19705
    },
    {
      "entropy": 5.60902042388916,
      "epoch": 1.7879172714078373,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046820453776970697,
      "loss": 5.0568,
      "mean_token_accuracy": 0.20841806381940842,
      "num_tokens": 40915269.0,
      "step": 19710
    },
    {
      "entropy": 5.5748247623443605,
      "epoch": 1.7883708272859216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046818795308569815,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19952248632907868,
      "num_tokens": 40925987.0,
      "step": 19715
    },
    {
      "entropy": 5.497185230255127,
      "epoch": 1.7888243831640058,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004681713644065133,
      "loss": 4.9734,
      "mean_token_accuracy": 0.21137189120054245,
      "num_tokens": 40936130.0,
      "step": 19720
    },
    {
      "entropy": 5.468851041793823,
      "epoch": 1.78927793904209,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004681547717324954,
      "loss": 4.8934,
      "mean_token_accuracy": 0.21224325299263,
      "num_tokens": 40945476.0,
      "step": 19725
    },
    {
      "entropy": 5.450008964538574,
      "epoch": 1.7897314949201741,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046813817506398755,
      "loss": 4.8696,
      "mean_token_accuracy": 0.217603899538517,
      "num_tokens": 40956492.0,
      "step": 19730
    },
    {
      "entropy": 5.5896382331848145,
      "epoch": 1.7901850507982582,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046812157440133303,
      "loss": 5.0792,
      "mean_token_accuracy": 0.20275839120149614,
      "num_tokens": 40966990.0,
      "step": 19735
    },
    {
      "entropy": 5.575714921951294,
      "epoch": 1.7906386066763424,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004681049697448752,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21206290721893312,
      "num_tokens": 40977065.0,
      "step": 19740
    },
    {
      "entropy": 5.519251680374145,
      "epoch": 1.7910921625544267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046808836109495734,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20591092705726624,
      "num_tokens": 40987207.0,
      "step": 19745
    },
    {
      "entropy": 5.5362590789794925,
      "epoch": 1.791545718432511,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004680717484519229,
      "loss": 4.9674,
      "mean_token_accuracy": 0.20424300730228423,
      "num_tokens": 40997541.0,
      "step": 19750
    },
    {
      "entropy": 5.5382311820983885,
      "epoch": 1.791999274310595,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004680551318161156,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20712293982505797,
      "num_tokens": 41009044.0,
      "step": 19755
    },
    {
      "entropy": 5.559739017486573,
      "epoch": 1.7924528301886793,
      "grad_norm": 1.0,
      "learning_rate": 0.00046803851118787897,
      "loss": 5.008,
      "mean_token_accuracy": 0.20548251569271087,
      "num_tokens": 41018165.0,
      "step": 19760
    },
    {
      "entropy": 5.497653388977051,
      "epoch": 1.7929063860667633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004680218865675567,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20819184333086013,
      "num_tokens": 41028972.0,
      "step": 19765
    },
    {
      "entropy": 5.507817316055298,
      "epoch": 1.7933599419448476,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004680052579554926,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20903472155332564,
      "num_tokens": 41039552.0,
      "step": 19770
    },
    {
      "entropy": 5.620876598358154,
      "epoch": 1.7938134978229319,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004679886253520307,
      "loss": 5.0528,
      "mean_token_accuracy": 0.1981100857257843,
      "num_tokens": 41049808.0,
      "step": 19775
    },
    {
      "entropy": 5.539993143081665,
      "epoch": 1.794267053701016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046797198875751474,
      "loss": 4.9395,
      "mean_token_accuracy": 0.2042449876666069,
      "num_tokens": 41060574.0,
      "step": 19780
    },
    {
      "entropy": 5.614523315429688,
      "epoch": 1.7947206095791002,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004679553481722889,
      "loss": 5.1255,
      "mean_token_accuracy": 0.20428696572780608,
      "num_tokens": 41071239.0,
      "step": 19785
    },
    {
      "entropy": 5.615858888626098,
      "epoch": 1.7951741654571842,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046793870359669733,
      "loss": 5.1343,
      "mean_token_accuracy": 0.20591361224651336,
      "num_tokens": 41080819.0,
      "step": 19790
    },
    {
      "entropy": 5.587080287933349,
      "epoch": 1.7956277213352685,
      "grad_norm": 1.25,
      "learning_rate": 0.0004679220550310842,
      "loss": 4.933,
      "mean_token_accuracy": 0.2169566422700882,
      "num_tokens": 41090457.0,
      "step": 19795
    },
    {
      "entropy": 5.530445909500122,
      "epoch": 1.7960812772133528,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004679054024757937,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21040300279855728,
      "num_tokens": 41101210.0,
      "step": 19800
    },
    {
      "entropy": 5.537619400024414,
      "epoch": 1.796534833091437,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004678887459311704,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20775469839572908,
      "num_tokens": 41110566.0,
      "step": 19805
    },
    {
      "entropy": 5.47896146774292,
      "epoch": 1.796988388969521,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004678720853975587,
      "loss": 4.9497,
      "mean_token_accuracy": 0.2113064005970955,
      "num_tokens": 41120000.0,
      "step": 19810
    },
    {
      "entropy": 5.469798278808594,
      "epoch": 1.797441944847605,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046785542087530307,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2107544869184494,
      "num_tokens": 41129944.0,
      "step": 19815
    },
    {
      "entropy": 5.485931491851806,
      "epoch": 1.7978955007256894,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004678387523647483,
      "loss": 4.9968,
      "mean_token_accuracy": 0.2056867852807045,
      "num_tokens": 41138803.0,
      "step": 19820
    },
    {
      "entropy": 5.560794973373413,
      "epoch": 1.7983490566037736,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046782207986623885,
      "loss": 5.0807,
      "mean_token_accuracy": 0.19582369774580002,
      "num_tokens": 41150365.0,
      "step": 19825
    },
    {
      "entropy": 5.664840650558472,
      "epoch": 1.798802612481858,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046780540338011974,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20080285668373107,
      "num_tokens": 41160310.0,
      "step": 19830
    },
    {
      "entropy": 5.512536096572876,
      "epoch": 1.799256168359942,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004677887229067358,
      "loss": 4.9104,
      "mean_token_accuracy": 0.20896457135677338,
      "num_tokens": 41170506.0,
      "step": 19835
    },
    {
      "entropy": 5.555658769607544,
      "epoch": 1.799709724238026,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046777203844643184,
      "loss": 5.0348,
      "mean_token_accuracy": 0.1990685984492302,
      "num_tokens": 41181293.0,
      "step": 19840
    },
    {
      "entropy": 5.489247608184814,
      "epoch": 1.8001632801161103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046775534999955303,
      "loss": 4.9688,
      "mean_token_accuracy": 0.20351264625787735,
      "num_tokens": 41191637.0,
      "step": 19845
    },
    {
      "entropy": 5.479731512069702,
      "epoch": 1.8006168359941945,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004677386575664445,
      "loss": 4.9665,
      "mean_token_accuracy": 0.2177043989300728,
      "num_tokens": 41201833.0,
      "step": 19850
    },
    {
      "entropy": 5.57675142288208,
      "epoch": 1.8010703918722788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046772196114745133,
      "loss": 5.0138,
      "mean_token_accuracy": 0.2109220579266548,
      "num_tokens": 41211243.0,
      "step": 19855
    },
    {
      "entropy": 5.58609938621521,
      "epoch": 1.8015239477503628,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004677052607429189,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21167860627174379,
      "num_tokens": 41221111.0,
      "step": 19860
    },
    {
      "entropy": 5.501392126083374,
      "epoch": 1.8019775036284469,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004676885563531927,
      "loss": 5.0146,
      "mean_token_accuracy": 0.2025648310780525,
      "num_tokens": 41231578.0,
      "step": 19865
    },
    {
      "entropy": 5.509230232238769,
      "epoch": 1.8024310595065312,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046767184797861795,
      "loss": 5.0598,
      "mean_token_accuracy": 0.20938555002212525,
      "num_tokens": 41242130.0,
      "step": 19870
    },
    {
      "entropy": 5.59357099533081,
      "epoch": 1.8028846153846154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046765513561954026,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19962058812379838,
      "num_tokens": 41252885.0,
      "step": 19875
    },
    {
      "entropy": 5.595680999755859,
      "epoch": 1.8033381712626997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046763841927630534,
      "loss": 5.0556,
      "mean_token_accuracy": 0.20472382009029388,
      "num_tokens": 41263532.0,
      "step": 19880
    },
    {
      "entropy": 5.59276933670044,
      "epoch": 1.8037917271407837,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046762169894925876,
      "loss": 5.0079,
      "mean_token_accuracy": 0.20627274066209794,
      "num_tokens": 41273469.0,
      "step": 19885
    },
    {
      "entropy": 5.55303692817688,
      "epoch": 1.8042452830188678,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004676049746387464,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20153881907463073,
      "num_tokens": 41283663.0,
      "step": 19890
    },
    {
      "entropy": 5.510688352584839,
      "epoch": 1.804698838896952,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004675882463451141,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21141158491373063,
      "num_tokens": 41294657.0,
      "step": 19895
    },
    {
      "entropy": 5.568413734436035,
      "epoch": 1.8051523947750363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004675715140687078,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20919404029846192,
      "num_tokens": 41305124.0,
      "step": 19900
    },
    {
      "entropy": 5.4937211036682125,
      "epoch": 1.8056059506531206,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046755477780987344,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20851007848978043,
      "num_tokens": 41315839.0,
      "step": 19905
    },
    {
      "entropy": 5.574371242523194,
      "epoch": 1.8060595065312046,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046753803756895726,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20170085430145263,
      "num_tokens": 41326388.0,
      "step": 19910
    },
    {
      "entropy": 5.527889442443848,
      "epoch": 1.8065130624092887,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004675212933463054,
      "loss": 5.0737,
      "mean_token_accuracy": 0.1926412522792816,
      "num_tokens": 41337333.0,
      "step": 19915
    },
    {
      "entropy": 5.618911981582642,
      "epoch": 1.806966618287373,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004675045451422641,
      "loss": 5.0657,
      "mean_token_accuracy": 0.19672111868858339,
      "num_tokens": 41348353.0,
      "step": 19920
    },
    {
      "entropy": 5.586848640441895,
      "epoch": 1.8074201741654572,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046748779295717973,
      "loss": 5.0304,
      "mean_token_accuracy": 0.213688226044178,
      "num_tokens": 41358797.0,
      "step": 19925
    },
    {
      "entropy": 5.5675770282745365,
      "epoch": 1.8078737300435415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004674710367913989,
      "loss": 5.0237,
      "mean_token_accuracy": 0.2066128134727478,
      "num_tokens": 41368818.0,
      "step": 19930
    },
    {
      "entropy": 5.5882336616516115,
      "epoch": 1.8083272859216255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004674542766452678,
      "loss": 5.0116,
      "mean_token_accuracy": 0.2071378633379936,
      "num_tokens": 41378949.0,
      "step": 19935
    },
    {
      "entropy": 5.566381454467773,
      "epoch": 1.8087808417997098,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046743751251913327,
      "loss": 4.9858,
      "mean_token_accuracy": 0.20560411214828492,
      "num_tokens": 41388892.0,
      "step": 19940
    },
    {
      "entropy": 5.5644384860992435,
      "epoch": 1.8092343976777938,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046742074441334197,
      "loss": 5.0086,
      "mean_token_accuracy": 0.20125612020492553,
      "num_tokens": 41399197.0,
      "step": 19945
    },
    {
      "entropy": 5.52111587524414,
      "epoch": 1.809687953555878,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004674039723282406,
      "loss": 5.0377,
      "mean_token_accuracy": 0.1966809868812561,
      "num_tokens": 41409551.0,
      "step": 19950
    },
    {
      "entropy": 5.518880939483642,
      "epoch": 1.8101415094339623,
      "grad_norm": 1.125,
      "learning_rate": 0.00046738719626417607,
      "loss": 4.9512,
      "mean_token_accuracy": 0.20883867889642715,
      "num_tokens": 41418900.0,
      "step": 19955
    },
    {
      "entropy": 5.52384581565857,
      "epoch": 1.8105950653120464,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046737041622149536,
      "loss": 4.949,
      "mean_token_accuracy": 0.20624009966850282,
      "num_tokens": 41428204.0,
      "step": 19960
    },
    {
      "entropy": 5.538127660751343,
      "epoch": 1.8110486211901307,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004673536322005454,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20278254002332688,
      "num_tokens": 41437922.0,
      "step": 19965
    },
    {
      "entropy": 5.561640787124634,
      "epoch": 1.8115021770682147,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004673368442016732,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20119996517896652,
      "num_tokens": 41447411.0,
      "step": 19970
    },
    {
      "entropy": 5.607109880447387,
      "epoch": 1.811955732946299,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004673200522252261,
      "loss": 5.093,
      "mean_token_accuracy": 0.20437152981758117,
      "num_tokens": 41458756.0,
      "step": 19975
    },
    {
      "entropy": 5.579017496109008,
      "epoch": 1.8124092888243832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004673032562715514,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2080783352255821,
      "num_tokens": 41468610.0,
      "step": 19980
    },
    {
      "entropy": 5.555026006698609,
      "epoch": 1.8128628447024675,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046728645634099627,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19671774953603743,
      "num_tokens": 41479849.0,
      "step": 19985
    },
    {
      "entropy": 5.526775121688843,
      "epoch": 1.8133164005805515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046726965243390825,
      "loss": 5.0188,
      "mean_token_accuracy": 0.2027814969420433,
      "num_tokens": 41490945.0,
      "step": 19990
    },
    {
      "entropy": 5.486761856079101,
      "epoch": 1.8137699564586356,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004672528445506348,
      "loss": 4.9515,
      "mean_token_accuracy": 0.20101569592952728,
      "num_tokens": 41501479.0,
      "step": 19995
    },
    {
      "entropy": 5.525984382629394,
      "epoch": 1.8142235123367199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004672360326915236,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21505804359912872,
      "num_tokens": 41512447.0,
      "step": 20000
    },
    {
      "entropy": 5.509391450881958,
      "epoch": 1.8146770682148041,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004672192168569222,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21713926196098327,
      "num_tokens": 41521905.0,
      "step": 20005
    },
    {
      "entropy": 5.505262660980224,
      "epoch": 1.8151306240928884,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046720239704717845,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20470809191465378,
      "num_tokens": 41533143.0,
      "step": 20010
    },
    {
      "entropy": 5.530465078353882,
      "epoch": 1.8155841799709724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046718557326264,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20491113215684892,
      "num_tokens": 41542234.0,
      "step": 20015
    },
    {
      "entropy": 5.579422378540039,
      "epoch": 1.8160377358490565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004671687455036551,
      "loss": 5.0773,
      "mean_token_accuracy": 0.2017393171787262,
      "num_tokens": 41552632.0,
      "step": 20020
    },
    {
      "entropy": 5.54284815788269,
      "epoch": 1.8164912917271407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004671519137705715,
      "loss": 5.016,
      "mean_token_accuracy": 0.21005022525787354,
      "num_tokens": 41563723.0,
      "step": 20025
    },
    {
      "entropy": 5.551046705245971,
      "epoch": 1.816944847605225,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046713507806373736,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21051034778356553,
      "num_tokens": 41573680.0,
      "step": 20030
    },
    {
      "entropy": 5.553695774078369,
      "epoch": 1.8173984034833093,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046711823838350086,
      "loss": 5.0444,
      "mean_token_accuracy": 0.20358869880437852,
      "num_tokens": 41584722.0,
      "step": 20035
    },
    {
      "entropy": 5.532588720321655,
      "epoch": 1.8178519593613933,
      "grad_norm": 1.0,
      "learning_rate": 0.00046710139473021025,
      "loss": 5.0871,
      "mean_token_accuracy": 0.19790272265672684,
      "num_tokens": 41595785.0,
      "step": 20040
    },
    {
      "entropy": 5.569847583770752,
      "epoch": 1.8183055152394774,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046708454710421376,
      "loss": 5.107,
      "mean_token_accuracy": 0.1956186145544052,
      "num_tokens": 41606023.0,
      "step": 20045
    },
    {
      "entropy": 5.566348218917847,
      "epoch": 1.8187590711175616,
      "grad_norm": 1.125,
      "learning_rate": 0.00046706769550586,
      "loss": 5.0154,
      "mean_token_accuracy": 0.20516480803489684,
      "num_tokens": 41616010.0,
      "step": 20050
    },
    {
      "entropy": 5.557003307342529,
      "epoch": 1.819212626995646,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046705083993549723,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20764904767274855,
      "num_tokens": 41625105.0,
      "step": 20055
    },
    {
      "entropy": 5.5250067710876465,
      "epoch": 1.8196661828737302,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004670339803934743,
      "loss": 5.0193,
      "mean_token_accuracy": 0.2096752032637596,
      "num_tokens": 41634989.0,
      "step": 20060
    },
    {
      "entropy": 5.5185634136199955,
      "epoch": 1.8201197387518142,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004670171168801396,
      "loss": 5.0033,
      "mean_token_accuracy": 0.2066364035010338,
      "num_tokens": 41645736.0,
      "step": 20065
    },
    {
      "entropy": 5.605870532989502,
      "epoch": 1.8205732946298983,
      "grad_norm": 1.0625,
      "learning_rate": 0.000467000249395842,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20920195430517197,
      "num_tokens": 41655505.0,
      "step": 20070
    },
    {
      "entropy": 5.528751707077026,
      "epoch": 1.8210268505079825,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004669833779409303,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20744871944189072,
      "num_tokens": 41665732.0,
      "step": 20075
    },
    {
      "entropy": 5.504740619659424,
      "epoch": 1.8214804063860668,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004669665025157535,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2010861560702324,
      "num_tokens": 41676050.0,
      "step": 20080
    },
    {
      "entropy": 5.614622354507446,
      "epoch": 1.821933962264151,
      "grad_norm": 1.125,
      "learning_rate": 0.00046694962312066046,
      "loss": 5.1475,
      "mean_token_accuracy": 0.19905100166797637,
      "num_tokens": 41686787.0,
      "step": 20085
    },
    {
      "entropy": 5.545039892196655,
      "epoch": 1.822387518142235,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046693273975600036,
      "loss": 4.944,
      "mean_token_accuracy": 0.20888380259275435,
      "num_tokens": 41697168.0,
      "step": 20090
    },
    {
      "entropy": 5.503949499130249,
      "epoch": 1.8228410740203191,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004669158524221222,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20895278602838516,
      "num_tokens": 41707524.0,
      "step": 20095
    },
    {
      "entropy": 5.587649965286255,
      "epoch": 1.8232946298984034,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046689896111937537,
      "loss": 5.0626,
      "mean_token_accuracy": 0.2100888043642044,
      "num_tokens": 41718109.0,
      "step": 20100
    },
    {
      "entropy": 5.64175763130188,
      "epoch": 1.8237481857764877,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004668820658481091,
      "loss": 5.0354,
      "mean_token_accuracy": 0.20424672663211824,
      "num_tokens": 41728862.0,
      "step": 20105
    },
    {
      "entropy": 5.652068376541138,
      "epoch": 1.824201741654572,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046686516660867283,
      "loss": 5.1023,
      "mean_token_accuracy": 0.1998430848121643,
      "num_tokens": 41739024.0,
      "step": 20110
    },
    {
      "entropy": 5.565501117706299,
      "epoch": 1.824655297532656,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004668482634014159,
      "loss": 5.0277,
      "mean_token_accuracy": 0.20229787528514862,
      "num_tokens": 41748979.0,
      "step": 20115
    },
    {
      "entropy": 5.510527801513672,
      "epoch": 1.8251088534107403,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004668313562266882,
      "loss": 5.036,
      "mean_token_accuracy": 0.208828841149807,
      "num_tokens": 41759972.0,
      "step": 20120
    },
    {
      "entropy": 5.5156959056854244,
      "epoch": 1.8255624092888243,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000466814445084839,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21409412175416948,
      "num_tokens": 41770150.0,
      "step": 20125
    },
    {
      "entropy": 5.549961090087891,
      "epoch": 1.8260159651669086,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004667975299762183,
      "loss": 4.9707,
      "mean_token_accuracy": 0.21211618036031724,
      "num_tokens": 41780405.0,
      "step": 20130
    },
    {
      "entropy": 5.5162074089050295,
      "epoch": 1.8264695210449928,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046678061090117576,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21304504275321962,
      "num_tokens": 41791269.0,
      "step": 20135
    },
    {
      "entropy": 5.442075777053833,
      "epoch": 1.8269230769230769,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004667636878600613,
      "loss": 4.917,
      "mean_token_accuracy": 0.20943055897951127,
      "num_tokens": 41802513.0,
      "step": 20140
    },
    {
      "entropy": 5.578953790664673,
      "epoch": 1.8273766328011611,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004667467608532249,
      "loss": 5.0589,
      "mean_token_accuracy": 0.20094809979200362,
      "num_tokens": 41814149.0,
      "step": 20145
    },
    {
      "entropy": 5.572923946380615,
      "epoch": 1.8278301886792452,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004667298298810167,
      "loss": 5.035,
      "mean_token_accuracy": 0.20357026308774948,
      "num_tokens": 41825763.0,
      "step": 20150
    },
    {
      "entropy": 5.522440624237061,
      "epoch": 1.8282837445573294,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004667128949437867,
      "loss": 5.0115,
      "mean_token_accuracy": 0.20984289944171905,
      "num_tokens": 41834543.0,
      "step": 20155
    },
    {
      "entropy": 5.519170665740967,
      "epoch": 1.8287373004354137,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004666959560418852,
      "loss": 4.9166,
      "mean_token_accuracy": 0.21681906580924987,
      "num_tokens": 41844940.0,
      "step": 20160
    },
    {
      "entropy": 5.498330211639404,
      "epoch": 1.829190856313498,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046667901317566245,
      "loss": 4.9249,
      "mean_token_accuracy": 0.20911678224802016,
      "num_tokens": 41855532.0,
      "step": 20165
    },
    {
      "entropy": 5.592134714126587,
      "epoch": 1.829644412191582,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004666620663454689,
      "loss": 5.0242,
      "mean_token_accuracy": 0.20335020124912262,
      "num_tokens": 41865661.0,
      "step": 20170
    },
    {
      "entropy": 5.5471094131469725,
      "epoch": 1.830097968069666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046664511555165496,
      "loss": 5.0347,
      "mean_token_accuracy": 0.2055257648229599,
      "num_tokens": 41875662.0,
      "step": 20175
    },
    {
      "entropy": 5.522885942459107,
      "epoch": 1.8305515239477503,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046662816079457114,
      "loss": 4.9624,
      "mean_token_accuracy": 0.21466411501169205,
      "num_tokens": 41886211.0,
      "step": 20180
    },
    {
      "entropy": 5.623113489151001,
      "epoch": 1.8310050798258346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004666112020745682,
      "loss": 5.1163,
      "mean_token_accuracy": 0.19894250482320786,
      "num_tokens": 41896737.0,
      "step": 20185
    },
    {
      "entropy": 5.552207946777344,
      "epoch": 1.8314586357039189,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004665942393919967,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20558253079652786,
      "num_tokens": 41907768.0,
      "step": 20190
    },
    {
      "entropy": 5.5546571731567385,
      "epoch": 1.831912191582003,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046657727274720755,
      "loss": 4.9649,
      "mean_token_accuracy": 0.20418288111686705,
      "num_tokens": 41917587.0,
      "step": 20195
    },
    {
      "entropy": 5.475027942657471,
      "epoch": 1.832365747460087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004665603021405516,
      "loss": 4.9196,
      "mean_token_accuracy": 0.21469545364379883,
      "num_tokens": 41927842.0,
      "step": 20200
    },
    {
      "entropy": 5.485632419586182,
      "epoch": 1.8328193033381712,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004665433275723798,
      "loss": 4.9552,
      "mean_token_accuracy": 0.21441649943590163,
      "num_tokens": 41939899.0,
      "step": 20205
    },
    {
      "entropy": 5.527506351470947,
      "epoch": 1.8332728592162555,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004665263490430431,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20336755961179734,
      "num_tokens": 41950932.0,
      "step": 20210
    },
    {
      "entropy": 5.5579382419586185,
      "epoch": 1.8337264150943398,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004665093665528928,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20819393843412398,
      "num_tokens": 41963512.0,
      "step": 20215
    },
    {
      "entropy": 5.636223602294922,
      "epoch": 1.8341799709724238,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046649238010227983,
      "loss": 5.0341,
      "mean_token_accuracy": 0.21151686012744902,
      "num_tokens": 41974677.0,
      "step": 20220
    },
    {
      "entropy": 5.485407495498658,
      "epoch": 1.8346335268505078,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004664753896915558,
      "loss": 4.9522,
      "mean_token_accuracy": 0.21072100549936296,
      "num_tokens": 41984582.0,
      "step": 20225
    },
    {
      "entropy": 5.543138790130615,
      "epoch": 1.8350870827285921,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004664583953210719,
      "loss": 4.9368,
      "mean_token_accuracy": 0.20556853264570235,
      "num_tokens": 41994624.0,
      "step": 20230
    },
    {
      "entropy": 5.525238418579102,
      "epoch": 1.8355406386066764,
      "grad_norm": 1.125,
      "learning_rate": 0.0004664413969911795,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21315260231494904,
      "num_tokens": 42004835.0,
      "step": 20235
    },
    {
      "entropy": 5.474159622192383,
      "epoch": 1.8359941944847606,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004664243947022303,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20505469739437104,
      "num_tokens": 42014727.0,
      "step": 20240
    },
    {
      "entropy": 5.56663556098938,
      "epoch": 1.8364477503628447,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046640738845457586,
      "loss": 5.0238,
      "mean_token_accuracy": 0.20703188925981522,
      "num_tokens": 42023628.0,
      "step": 20245
    },
    {
      "entropy": 5.59652886390686,
      "epoch": 1.8369013062409287,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004663903782485678,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20446922183036803,
      "num_tokens": 42033587.0,
      "step": 20250
    },
    {
      "entropy": 5.5812348365783695,
      "epoch": 1.837354862119013,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046637336408455793,
      "loss": 5.0331,
      "mean_token_accuracy": 0.20150297582149507,
      "num_tokens": 42043765.0,
      "step": 20255
    },
    {
      "entropy": 5.608831453323364,
      "epoch": 1.8378084179970973,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046635634596289816,
      "loss": 5.0815,
      "mean_token_accuracy": 0.20444309413433076,
      "num_tokens": 42054205.0,
      "step": 20260
    },
    {
      "entropy": 5.487960290908814,
      "epoch": 1.8382619738751815,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004663393238839404,
      "loss": 4.9104,
      "mean_token_accuracy": 0.20541877448558807,
      "num_tokens": 42064512.0,
      "step": 20265
    },
    {
      "entropy": 5.507561206817627,
      "epoch": 1.8387155297532656,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004663222978480366,
      "loss": 4.9937,
      "mean_token_accuracy": 0.205137300491333,
      "num_tokens": 42074954.0,
      "step": 20270
    },
    {
      "entropy": 5.568202161788941,
      "epoch": 1.8391690856313496,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000466305267855539,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20168593227863313,
      "num_tokens": 42084914.0,
      "step": 20275
    },
    {
      "entropy": 5.576024961471558,
      "epoch": 1.8396226415094339,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004662882339067996,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20538800060749055,
      "num_tokens": 42095454.0,
      "step": 20280
    },
    {
      "entropy": 5.587305212020874,
      "epoch": 1.8400761973875182,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004662711960021709,
      "loss": 5.0882,
      "mean_token_accuracy": 0.1997831478714943,
      "num_tokens": 42106960.0,
      "step": 20285
    },
    {
      "entropy": 5.525085353851319,
      "epoch": 1.8405297532656024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046625415414200513,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20938224345445633,
      "num_tokens": 42117009.0,
      "step": 20290
    },
    {
      "entropy": 5.577379465103149,
      "epoch": 1.8409833091436865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004662371083266546,
      "loss": 5.002,
      "mean_token_accuracy": 0.2110588639974594,
      "num_tokens": 42126371.0,
      "step": 20295
    },
    {
      "entropy": 5.58545708656311,
      "epoch": 1.8414368650217707,
      "grad_norm": 1.125,
      "learning_rate": 0.000466220058556472,
      "loss": 5.031,
      "mean_token_accuracy": 0.20728013813495635,
      "num_tokens": 42135890.0,
      "step": 20300
    },
    {
      "entropy": 5.6177905082702635,
      "epoch": 1.8418904208998548,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046620300483180986,
      "loss": 5.1298,
      "mean_token_accuracy": 0.19759823381900787,
      "num_tokens": 42145696.0,
      "step": 20305
    },
    {
      "entropy": 5.548390865325928,
      "epoch": 1.842343976777939,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004661859471530209,
      "loss": 5.0381,
      "mean_token_accuracy": 0.21023260056972504,
      "num_tokens": 42156236.0,
      "step": 20310
    },
    {
      "entropy": 5.597132444381714,
      "epoch": 1.8427975326560233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004661688855204578,
      "loss": 5.0606,
      "mean_token_accuracy": 0.20116368681192398,
      "num_tokens": 42166179.0,
      "step": 20315
    },
    {
      "entropy": 5.510843658447266,
      "epoch": 1.8432510885341074,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046615181993447344,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20522765815258026,
      "num_tokens": 42176643.0,
      "step": 20320
    },
    {
      "entropy": 5.596554136276245,
      "epoch": 1.8437046444121916,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004661347503954207,
      "loss": 5.1961,
      "mean_token_accuracy": 0.19032192081212998,
      "num_tokens": 42188328.0,
      "step": 20325
    },
    {
      "entropy": 5.601656866073609,
      "epoch": 1.8441582002902757,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046611767690365265,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21157871037721634,
      "num_tokens": 42198110.0,
      "step": 20330
    },
    {
      "entropy": 5.527664279937744,
      "epoch": 1.84461175616836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004661005994595223,
      "loss": 4.9823,
      "mean_token_accuracy": 0.19972015023231507,
      "num_tokens": 42208438.0,
      "step": 20335
    },
    {
      "entropy": 5.560026597976685,
      "epoch": 1.8450653120464442,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004660835180633829,
      "loss": 4.9784,
      "mean_token_accuracy": 0.2038367673754692,
      "num_tokens": 42220052.0,
      "step": 20340
    },
    {
      "entropy": 5.538200902938843,
      "epoch": 1.8455188679245285,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004660664327155876,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20943403989076614,
      "num_tokens": 42229497.0,
      "step": 20345
    },
    {
      "entropy": 5.496162986755371,
      "epoch": 1.8459724238026125,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004660493434164898,
      "loss": 5.0363,
      "mean_token_accuracy": 0.20182014554738997,
      "num_tokens": 42240725.0,
      "step": 20350
    },
    {
      "entropy": 5.565572261810303,
      "epoch": 1.8464259796806965,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004660322501664429,
      "loss": 5.0526,
      "mean_token_accuracy": 0.20267361104488374,
      "num_tokens": 42251114.0,
      "step": 20355
    },
    {
      "entropy": 5.619518995285034,
      "epoch": 1.8468795355587808,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004660151529658004,
      "loss": 5.1897,
      "mean_token_accuracy": 0.19703874588012696,
      "num_tokens": 42262825.0,
      "step": 20360
    },
    {
      "entropy": 5.617018222808838,
      "epoch": 1.847333091436865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004659980518149158,
      "loss": 5.0151,
      "mean_token_accuracy": 0.21149738729000092,
      "num_tokens": 42272176.0,
      "step": 20365
    },
    {
      "entropy": 5.602189350128174,
      "epoch": 1.8477866473149493,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046598094671414285,
      "loss": 5.066,
      "mean_token_accuracy": 0.2062876984477043,
      "num_tokens": 42281934.0,
      "step": 20370
    },
    {
      "entropy": 5.540198755264282,
      "epoch": 1.8482402031930334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046596383766383533,
      "loss": 4.9589,
      "mean_token_accuracy": 0.20870660990476608,
      "num_tokens": 42291927.0,
      "step": 20375
    },
    {
      "entropy": 5.5219872951507565,
      "epoch": 1.8486937590711174,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004659467246643469,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20699819028377534,
      "num_tokens": 42301407.0,
      "step": 20380
    },
    {
      "entropy": 5.455481624603271,
      "epoch": 1.8491473149492017,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046592960771603154,
      "loss": 4.9019,
      "mean_token_accuracy": 0.21629468202590943,
      "num_tokens": 42311199.0,
      "step": 20385
    },
    {
      "entropy": 5.533198833465576,
      "epoch": 1.849600870827286,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004659124868192433,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20851015746593476,
      "num_tokens": 42320725.0,
      "step": 20390
    },
    {
      "entropy": 5.621488285064697,
      "epoch": 1.8500544267053702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004658953619743361,
      "loss": 5.0941,
      "mean_token_accuracy": 0.20408866852521895,
      "num_tokens": 42330857.0,
      "step": 20395
    },
    {
      "entropy": 5.526382637023926,
      "epoch": 1.8505079825834543,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046587823318166427,
      "loss": 4.9196,
      "mean_token_accuracy": 0.20995397865772247,
      "num_tokens": 42340241.0,
      "step": 20400
    },
    {
      "entropy": 5.619599676132202,
      "epoch": 1.8509615384615383,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046586110044158184,
      "loss": 5.1337,
      "mean_token_accuracy": 0.19478118121623994,
      "num_tokens": 42351756.0,
      "step": 20405
    },
    {
      "entropy": 5.573098421096802,
      "epoch": 1.8514150943396226,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046584396375444335,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21466975510120392,
      "num_tokens": 42361649.0,
      "step": 20410
    },
    {
      "entropy": 5.54986982345581,
      "epoch": 1.8518686502177069,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046582682312060295,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20996458381414412,
      "num_tokens": 42371273.0,
      "step": 20415
    },
    {
      "entropy": 5.4714539527893065,
      "epoch": 1.8523222060957911,
      "grad_norm": 1.0,
      "learning_rate": 0.0004658096785404152,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20906607210636138,
      "num_tokens": 42382275.0,
      "step": 20420
    },
    {
      "entropy": 5.61240701675415,
      "epoch": 1.8527757619738752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046579253001423473,
      "loss": 5.0182,
      "mean_token_accuracy": 0.201275971531868,
      "num_tokens": 42392360.0,
      "step": 20425
    },
    {
      "entropy": 5.576451110839844,
      "epoch": 1.8532293178519592,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046577537754241604,
      "loss": 5.0072,
      "mean_token_accuracy": 0.19999832063913345,
      "num_tokens": 42403695.0,
      "step": 20430
    },
    {
      "entropy": 5.557355308532715,
      "epoch": 1.8536828737300435,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000465758221125314,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20686184316873552,
      "num_tokens": 42414199.0,
      "step": 20435
    },
    {
      "entropy": 5.566457271575928,
      "epoch": 1.8541364296081277,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046574106076328333,
      "loss": 5.0809,
      "mean_token_accuracy": 0.2030840277671814,
      "num_tokens": 42423655.0,
      "step": 20440
    },
    {
      "entropy": 5.526126050949097,
      "epoch": 1.854589985486212,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004657238964566789,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21088251173496247,
      "num_tokens": 42434235.0,
      "step": 20445
    },
    {
      "entropy": 5.480642986297608,
      "epoch": 1.855043541364296,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004657067282058557,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20678504556417465,
      "num_tokens": 42444701.0,
      "step": 20450
    },
    {
      "entropy": 5.4687940120697025,
      "epoch": 1.85549709724238,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004656895560111687,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20691045373678207,
      "num_tokens": 42454499.0,
      "step": 20455
    },
    {
      "entropy": 5.5942786693572994,
      "epoch": 1.8559506531204644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046567237987297323,
      "loss": 4.9735,
      "mean_token_accuracy": 0.1980825424194336,
      "num_tokens": 42465391.0,
      "step": 20460
    },
    {
      "entropy": 5.624168586730957,
      "epoch": 1.8564042089985486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004656551997916243,
      "loss": 4.9827,
      "mean_token_accuracy": 0.1953688681125641,
      "num_tokens": 42476126.0,
      "step": 20465
    },
    {
      "entropy": 5.570189476013184,
      "epoch": 1.856857764876633,
      "grad_norm": 1.0,
      "learning_rate": 0.00046563801576747716,
      "loss": 5.0589,
      "mean_token_accuracy": 0.2046019032597542,
      "num_tokens": 42486833.0,
      "step": 20470
    },
    {
      "entropy": 5.648507928848266,
      "epoch": 1.857311320754717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046562082780088733,
      "loss": 5.0745,
      "mean_token_accuracy": 0.20140296071767808,
      "num_tokens": 42496986.0,
      "step": 20475
    },
    {
      "entropy": 5.571720695495605,
      "epoch": 1.8577648766328012,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004656036358922103,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20405839681625365,
      "num_tokens": 42507352.0,
      "step": 20480
    },
    {
      "entropy": 5.565200138092041,
      "epoch": 1.8582184325108853,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004655864400418014,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20897690653800965,
      "num_tokens": 42517655.0,
      "step": 20485
    },
    {
      "entropy": 5.513050222396851,
      "epoch": 1.8586719883889695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004655692402500163,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21601160764694213,
      "num_tokens": 42527502.0,
      "step": 20490
    },
    {
      "entropy": 5.529283952713013,
      "epoch": 1.8591255442670538,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046555203651721085,
      "loss": 4.9749,
      "mean_token_accuracy": 0.2061847746372223,
      "num_tokens": 42538893.0,
      "step": 20495
    },
    {
      "entropy": 5.429849100112915,
      "epoch": 1.8595791001451378,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004655348288437407,
      "loss": 4.861,
      "mean_token_accuracy": 0.21934590935707093,
      "num_tokens": 42550372.0,
      "step": 20500
    },
    {
      "entropy": 5.584519672393799,
      "epoch": 1.860032656023222,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004655176172299618,
      "loss": 5.0942,
      "mean_token_accuracy": 0.19986526370048524,
      "num_tokens": 42560814.0,
      "step": 20505
    },
    {
      "entropy": 5.634581851959228,
      "epoch": 1.8604862119013061,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046550040167622986,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19995513409376145,
      "num_tokens": 42570993.0,
      "step": 20510
    },
    {
      "entropy": 5.559375667572022,
      "epoch": 1.8609397677793904,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046548318218290124,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20816790759563447,
      "num_tokens": 42581357.0,
      "step": 20515
    },
    {
      "entropy": 5.523706865310669,
      "epoch": 1.8613933236574747,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004654659587503318,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19584809392690658,
      "num_tokens": 42591653.0,
      "step": 20520
    },
    {
      "entropy": 5.612924957275391,
      "epoch": 1.861846879535559,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004654487313788778,
      "loss": 5.1077,
      "mean_token_accuracy": 0.19749281257390977,
      "num_tokens": 42602232.0,
      "step": 20525
    },
    {
      "entropy": 5.571268939971924,
      "epoch": 1.862300435413643,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046543150006889546,
      "loss": 5.0063,
      "mean_token_accuracy": 0.20954685807228088,
      "num_tokens": 42613283.0,
      "step": 20530
    },
    {
      "entropy": 5.507944822311401,
      "epoch": 1.862753991291727,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004654142648207412,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21491098999977112,
      "num_tokens": 42623594.0,
      "step": 20535
    },
    {
      "entropy": 5.565047264099121,
      "epoch": 1.8632075471698113,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004653970256347715,
      "loss": 5.0551,
      "mean_token_accuracy": 0.20384258925914764,
      "num_tokens": 42633587.0,
      "step": 20540
    },
    {
      "entropy": 5.593127250671387,
      "epoch": 1.8636611030478956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046537978251134265,
      "loss": 4.9958,
      "mean_token_accuracy": 0.20397452265024185,
      "num_tokens": 42643426.0,
      "step": 20545
    },
    {
      "entropy": 5.48998236656189,
      "epoch": 1.8641146589259798,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004653625354508115,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20894977897405625,
      "num_tokens": 42653521.0,
      "step": 20550
    },
    {
      "entropy": 5.5964569568634035,
      "epoch": 1.8645682148040639,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004653452844535346,
      "loss": 5.0268,
      "mean_token_accuracy": 0.2043809026479721,
      "num_tokens": 42664818.0,
      "step": 20555
    },
    {
      "entropy": 5.667501926422119,
      "epoch": 1.865021770682148,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046532802951986863,
      "loss": 5.0684,
      "mean_token_accuracy": 0.20821174383163452,
      "num_tokens": 42674738.0,
      "step": 20560
    },
    {
      "entropy": 5.529953861236573,
      "epoch": 1.8654753265602322,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004653107706501705,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21852142363786697,
      "num_tokens": 42685003.0,
      "step": 20565
    },
    {
      "entropy": 5.484201097488404,
      "epoch": 1.8659288824383164,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004652935078447973,
      "loss": 5.0168,
      "mean_token_accuracy": 0.20671058297157288,
      "num_tokens": 42695428.0,
      "step": 20570
    },
    {
      "entropy": 5.577590703964233,
      "epoch": 1.8663824383164007,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046527624110410574,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20736818611621857,
      "num_tokens": 42704901.0,
      "step": 20575
    },
    {
      "entropy": 5.609286212921143,
      "epoch": 1.8668359941944848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000465258970428453,
      "loss": 5.0598,
      "mean_token_accuracy": 0.20195237398147584,
      "num_tokens": 42714379.0,
      "step": 20580
    },
    {
      "entropy": 5.624383544921875,
      "epoch": 1.8672895500725688,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046524169581819633,
      "loss": 5.1681,
      "mean_token_accuracy": 0.19813774675130844,
      "num_tokens": 42724813.0,
      "step": 20585
    },
    {
      "entropy": 5.512622976303101,
      "epoch": 1.867743105950653,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004652244172736929,
      "loss": 5.0003,
      "mean_token_accuracy": 0.20564983189105987,
      "num_tokens": 42736590.0,
      "step": 20590
    },
    {
      "entropy": 5.616318225860596,
      "epoch": 1.8681966618287373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004652071347953,
      "loss": 5.0857,
      "mean_token_accuracy": 0.1994020715355873,
      "num_tokens": 42746718.0,
      "step": 20595
    },
    {
      "entropy": 5.560640478134156,
      "epoch": 1.8686502177068216,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046518984838337513,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20223457664251326,
      "num_tokens": 42756484.0,
      "step": 20600
    },
    {
      "entropy": 5.579528045654297,
      "epoch": 1.8691037735849056,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004651725580382757,
      "loss": 5.0752,
      "mean_token_accuracy": 0.20161826014518738,
      "num_tokens": 42768512.0,
      "step": 20605
    },
    {
      "entropy": 5.533065414428711,
      "epoch": 1.8695573294629897,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046515526376035934,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21082388907670974,
      "num_tokens": 42779812.0,
      "step": 20610
    },
    {
      "entropy": 5.4643162250518795,
      "epoch": 1.870010885341074,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004651379655499837,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21543058305978774,
      "num_tokens": 42791163.0,
      "step": 20615
    },
    {
      "entropy": 5.540845632553101,
      "epoch": 1.8704644412191582,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046512066340750636,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20470986813306807,
      "num_tokens": 42801810.0,
      "step": 20620
    },
    {
      "entropy": 5.603699636459351,
      "epoch": 1.8709179970972425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004651033573332853,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20591593086719512,
      "num_tokens": 42813358.0,
      "step": 20625
    },
    {
      "entropy": 5.460213851928711,
      "epoch": 1.8713715529753265,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004650860473276784,
      "loss": 4.8519,
      "mean_token_accuracy": 0.21921395212411882,
      "num_tokens": 42823519.0,
      "step": 20630
    },
    {
      "entropy": 5.571874475479126,
      "epoch": 1.8718251088534106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004650687333910436,
      "loss": 5.0334,
      "mean_token_accuracy": 0.20522348582744598,
      "num_tokens": 42833257.0,
      "step": 20635
    },
    {
      "entropy": 5.595556211471558,
      "epoch": 1.8722786647314948,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046505141552373897,
      "loss": 5.0203,
      "mean_token_accuracy": 0.21454567611217498,
      "num_tokens": 42843536.0,
      "step": 20640
    },
    {
      "entropy": 5.54931116104126,
      "epoch": 1.8727322206095791,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004650340937261226,
      "loss": 5.0055,
      "mean_token_accuracy": 0.2033155769109726,
      "num_tokens": 42853911.0,
      "step": 20645
    },
    {
      "entropy": 5.641363668441772,
      "epoch": 1.8731857764876634,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004650167679985528,
      "loss": 5.1005,
      "mean_token_accuracy": 0.19384931772947311,
      "num_tokens": 42864173.0,
      "step": 20650
    },
    {
      "entropy": 5.5625217914581295,
      "epoch": 1.8736393323657474,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004649994383413877,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20902465730905534,
      "num_tokens": 42873470.0,
      "step": 20655
    },
    {
      "entropy": 5.509566593170166,
      "epoch": 1.8740928882438317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046498210475498595,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21467747837305068,
      "num_tokens": 42882496.0,
      "step": 20660
    },
    {
      "entropy": 5.505311059951782,
      "epoch": 1.8745464441219157,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046496476723970574,
      "loss": 4.9549,
      "mean_token_accuracy": 0.20503194034099578,
      "num_tokens": 42893163.0,
      "step": 20665
    },
    {
      "entropy": 5.522883796691895,
      "epoch": 1.875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004649474257959058,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20853859633207322,
      "num_tokens": 42903406.0,
      "step": 20670
    },
    {
      "entropy": 5.593277215957642,
      "epoch": 1.8754535558780843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004649300804239447,
      "loss": 4.9718,
      "mean_token_accuracy": 0.20735423564910888,
      "num_tokens": 42913702.0,
      "step": 20675
    },
    {
      "entropy": 5.506184911727905,
      "epoch": 1.8759071117561683,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046491273112418105,
      "loss": 4.9302,
      "mean_token_accuracy": 0.2134859025478363,
      "num_tokens": 42923877.0,
      "step": 20680
    },
    {
      "entropy": 5.5144805908203125,
      "epoch": 1.8763606676342526,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046489537789697386,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20901622474193574,
      "num_tokens": 42934393.0,
      "step": 20685
    },
    {
      "entropy": 5.567466974258423,
      "epoch": 1.8768142235123366,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046487802074268177,
      "loss": 5.0404,
      "mean_token_accuracy": 0.21161091774702073,
      "num_tokens": 42945803.0,
      "step": 20690
    },
    {
      "entropy": 5.5185154438018795,
      "epoch": 1.8772677793904209,
      "grad_norm": 1.125,
      "learning_rate": 0.00046486065966166377,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21167248189449311,
      "num_tokens": 42955829.0,
      "step": 20695
    },
    {
      "entropy": 5.543769025802613,
      "epoch": 1.8777213352685052,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004648432946542791,
      "loss": 5.0168,
      "mean_token_accuracy": 0.20266976952552795,
      "num_tokens": 42967080.0,
      "step": 20700
    },
    {
      "entropy": 5.512664699554444,
      "epoch": 1.8781748911465894,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046482592572088654,
      "loss": 4.9818,
      "mean_token_accuracy": 0.2005089059472084,
      "num_tokens": 42978082.0,
      "step": 20705
    },
    {
      "entropy": 5.572058773040771,
      "epoch": 1.8786284470246735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046480855286184565,
      "loss": 5.0521,
      "mean_token_accuracy": 0.20274498760700227,
      "num_tokens": 42988982.0,
      "step": 20710
    },
    {
      "entropy": 5.5833902835845945,
      "epoch": 1.8790820029027575,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046479117607751535,
      "loss": 4.9986,
      "mean_token_accuracy": 0.1983781024813652,
      "num_tokens": 42998982.0,
      "step": 20715
    },
    {
      "entropy": 5.55176157951355,
      "epoch": 1.8795355587808418,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046477379536825534,
      "loss": 4.9359,
      "mean_token_accuracy": 0.20915919542312622,
      "num_tokens": 43009574.0,
      "step": 20720
    },
    {
      "entropy": 5.560165309906006,
      "epoch": 1.879989114658926,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004647564107344247,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20352196842432022,
      "num_tokens": 43019335.0,
      "step": 20725
    },
    {
      "entropy": 5.481369352340698,
      "epoch": 1.8804426705370103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046473902217638327,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20774939954280852,
      "num_tokens": 43029264.0,
      "step": 20730
    },
    {
      "entropy": 5.516702032089233,
      "epoch": 1.8808962264150944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046472162969449037,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2099528893828392,
      "num_tokens": 43040519.0,
      "step": 20735
    },
    {
      "entropy": 5.604666137695313,
      "epoch": 1.8813497822931784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046470423328910594,
      "loss": 5.0098,
      "mean_token_accuracy": 0.20703690648078918,
      "num_tokens": 43051146.0,
      "step": 20740
    },
    {
      "entropy": 5.6006255626678465,
      "epoch": 1.8818033381712627,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046468683296058957,
      "loss": 5.0863,
      "mean_token_accuracy": 0.20637607127428054,
      "num_tokens": 43061339.0,
      "step": 20745
    },
    {
      "entropy": 5.504037857055664,
      "epoch": 1.882256894049347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046466942870930126,
      "loss": 5.0038,
      "mean_token_accuracy": 0.20894727855920792,
      "num_tokens": 43072624.0,
      "step": 20750
    },
    {
      "entropy": 5.590467262268066,
      "epoch": 1.8827104499274312,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004646520205356007,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2081509128212929,
      "num_tokens": 43082120.0,
      "step": 20755
    },
    {
      "entropy": 5.649250793457031,
      "epoch": 1.8831640058055152,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046463460843984805,
      "loss": 5.0907,
      "mean_token_accuracy": 0.20072863548994063,
      "num_tokens": 43093202.0,
      "step": 20760
    },
    {
      "entropy": 5.665876722335815,
      "epoch": 1.8836175616835993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004646171924224034,
      "loss": 5.1642,
      "mean_token_accuracy": 0.2048816204071045,
      "num_tokens": 43104801.0,
      "step": 20765
    },
    {
      "entropy": 5.5828657150268555,
      "epoch": 1.8840711175616836,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046459977248362687,
      "loss": 4.9738,
      "mean_token_accuracy": 0.2094803050160408,
      "num_tokens": 43114855.0,
      "step": 20770
    },
    {
      "entropy": 5.570040464401245,
      "epoch": 1.8845246734397678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004645823486238788,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20368480086326599,
      "num_tokens": 43124759.0,
      "step": 20775
    },
    {
      "entropy": 5.521081876754761,
      "epoch": 1.884978229317852,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004645649208435193,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21389394551515578,
      "num_tokens": 43135526.0,
      "step": 20780
    },
    {
      "entropy": 5.5732262134552,
      "epoch": 1.8854317851959361,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046454748914290906,
      "loss": 5.0718,
      "mean_token_accuracy": 0.20253242701292037,
      "num_tokens": 43145605.0,
      "step": 20785
    },
    {
      "entropy": 5.642473983764648,
      "epoch": 1.8858853410740202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046453005352240835,
      "loss": 5.1235,
      "mean_token_accuracy": 0.1973668545484543,
      "num_tokens": 43156578.0,
      "step": 20790
    },
    {
      "entropy": 5.60530800819397,
      "epoch": 1.8863388969521044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000464512613982378,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20926489681005478,
      "num_tokens": 43166870.0,
      "step": 20795
    },
    {
      "entropy": 5.562167406082153,
      "epoch": 1.8867924528301887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004644951705231783,
      "loss": 4.9538,
      "mean_token_accuracy": 0.2154717341065407,
      "num_tokens": 43176406.0,
      "step": 20800
    },
    {
      "entropy": 5.656374311447143,
      "epoch": 1.887246008708273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046447772314517024,
      "loss": 5.1511,
      "mean_token_accuracy": 0.2020910695195198,
      "num_tokens": 43187502.0,
      "step": 20805
    },
    {
      "entropy": 5.501628065109253,
      "epoch": 1.887699564586357,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046446027184871464,
      "loss": 4.9177,
      "mean_token_accuracy": 0.21230237782001496,
      "num_tokens": 43197928.0,
      "step": 20810
    },
    {
      "entropy": 5.533857202529907,
      "epoch": 1.888153120464441,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046444281663417236,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21144766956567765,
      "num_tokens": 43209083.0,
      "step": 20815
    },
    {
      "entropy": 5.538638877868652,
      "epoch": 1.8886066763425253,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004644253575019043,
      "loss": 4.9852,
      "mean_token_accuracy": 0.2015242889523506,
      "num_tokens": 43219089.0,
      "step": 20820
    },
    {
      "entropy": 5.602917671203613,
      "epoch": 1.8890602322206096,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004644078944522716,
      "loss": 5.051,
      "mean_token_accuracy": 0.20531389713287354,
      "num_tokens": 43229440.0,
      "step": 20825
    },
    {
      "entropy": 5.597049617767334,
      "epoch": 1.8895137880986939,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004643904274856353,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20621335208415986,
      "num_tokens": 43239995.0,
      "step": 20830
    },
    {
      "entropy": 5.554989004135132,
      "epoch": 1.889967343976778,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046437295660235674,
      "loss": 4.977,
      "mean_token_accuracy": 0.21172338724136353,
      "num_tokens": 43249722.0,
      "step": 20835
    },
    {
      "entropy": 5.4992584705352785,
      "epoch": 1.8904208998548622,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004643554818027972,
      "loss": 5.0197,
      "mean_token_accuracy": 0.2075279548764229,
      "num_tokens": 43260029.0,
      "step": 20840
    },
    {
      "entropy": 5.554016780853272,
      "epoch": 1.8908744557329462,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004643380030873181,
      "loss": 4.9851,
      "mean_token_accuracy": 0.20793743878602983,
      "num_tokens": 43270571.0,
      "step": 20845
    },
    {
      "entropy": 5.570363712310791,
      "epoch": 1.8913280116110305,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004643205204562808,
      "loss": 4.9316,
      "mean_token_accuracy": 0.21176953464746476,
      "num_tokens": 43281310.0,
      "step": 20850
    },
    {
      "entropy": 5.55697021484375,
      "epoch": 1.8917815674891147,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046430303391004687,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20941325575113295,
      "num_tokens": 43292904.0,
      "step": 20855
    },
    {
      "entropy": 5.554282379150391,
      "epoch": 1.8922351233671988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004642855434489779,
      "loss": 4.9897,
      "mean_token_accuracy": 0.2037817120552063,
      "num_tokens": 43303745.0,
      "step": 20860
    },
    {
      "entropy": 5.586295175552368,
      "epoch": 1.892688679245283,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046426804907343574,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20697399526834487,
      "num_tokens": 43314652.0,
      "step": 20865
    },
    {
      "entropy": 5.527583217620849,
      "epoch": 1.893142235123367,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046425055078378206,
      "loss": 4.8744,
      "mean_token_accuracy": 0.21875638663768768,
      "num_tokens": 43324242.0,
      "step": 20870
    },
    {
      "entropy": 5.523820161819458,
      "epoch": 1.8935957910014514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046423304858037885,
      "loss": 4.9859,
      "mean_token_accuracy": 0.21595541387796402,
      "num_tokens": 43334054.0,
      "step": 20875
    },
    {
      "entropy": 5.469158792495728,
      "epoch": 1.8940493468795356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004642155424635878,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2215532347559929,
      "num_tokens": 43343389.0,
      "step": 20880
    },
    {
      "entropy": 5.477652502059937,
      "epoch": 1.89450290275762,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046419803243377125,
      "loss": 4.9423,
      "mean_token_accuracy": 0.2153505265712738,
      "num_tokens": 43353892.0,
      "step": 20885
    },
    {
      "entropy": 5.501353168487549,
      "epoch": 1.894956458635704,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046418051849129115,
      "loss": 4.972,
      "mean_token_accuracy": 0.214233835041523,
      "num_tokens": 43363912.0,
      "step": 20890
    },
    {
      "entropy": 5.545388412475586,
      "epoch": 1.895410014513788,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004641630006365097,
      "loss": 5.0428,
      "mean_token_accuracy": 0.20356074273586272,
      "num_tokens": 43374719.0,
      "step": 20895
    },
    {
      "entropy": 5.580736112594605,
      "epoch": 1.8958635703918723,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046414547886978926,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2027682676911354,
      "num_tokens": 43385549.0,
      "step": 20900
    },
    {
      "entropy": 5.600714111328125,
      "epoch": 1.8963171262699565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046412795319149204,
      "loss": 5.0252,
      "mean_token_accuracy": 0.21062333434820174,
      "num_tokens": 43395503.0,
      "step": 20905
    },
    {
      "entropy": 5.538871383666992,
      "epoch": 1.8967706821480408,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046411042360198053,
      "loss": 5.0386,
      "mean_token_accuracy": 0.2010958045721054,
      "num_tokens": 43406913.0,
      "step": 20910
    },
    {
      "entropy": 5.562172079086304,
      "epoch": 1.8972242380261248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004640928901016173,
      "loss": 4.947,
      "mean_token_accuracy": 0.20708273500204086,
      "num_tokens": 43416459.0,
      "step": 20915
    },
    {
      "entropy": 5.53361029624939,
      "epoch": 1.8976777939042089,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046407535269076493,
      "loss": 5.0172,
      "mean_token_accuracy": 0.20667988210916519,
      "num_tokens": 43426280.0,
      "step": 20920
    },
    {
      "entropy": 5.610696172714233,
      "epoch": 1.8981313497822931,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046405781136978606,
      "loss": 5.0492,
      "mean_token_accuracy": 0.20656325072050094,
      "num_tokens": 43436676.0,
      "step": 20925
    },
    {
      "entropy": 5.660415744781494,
      "epoch": 1.8985849056603774,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046404026613904357,
      "loss": 5.0354,
      "mean_token_accuracy": 0.2016023129224777,
      "num_tokens": 43447019.0,
      "step": 20930
    },
    {
      "entropy": 5.605124664306641,
      "epoch": 1.8990384615384617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004640227169989001,
      "loss": 5.0048,
      "mean_token_accuracy": 0.2014377772808075,
      "num_tokens": 43456609.0,
      "step": 20935
    },
    {
      "entropy": 5.576879262924194,
      "epoch": 1.8994920174165457,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004640051639497187,
      "loss": 5.0459,
      "mean_token_accuracy": 0.2026502937078476,
      "num_tokens": 43467616.0,
      "step": 20940
    },
    {
      "entropy": 5.485981082916259,
      "epoch": 1.8999455732946298,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004639876069918624,
      "loss": 5.0186,
      "mean_token_accuracy": 0.20398564338684083,
      "num_tokens": 43478209.0,
      "step": 20945
    },
    {
      "entropy": 5.527855396270752,
      "epoch": 1.900399129172714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046397004612569416,
      "loss": 4.9072,
      "mean_token_accuracy": 0.20824503153562546,
      "num_tokens": 43487278.0,
      "step": 20950
    },
    {
      "entropy": 5.507301187515258,
      "epoch": 1.9008526850507983,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046395248135157715,
      "loss": 4.9297,
      "mean_token_accuracy": 0.2197753131389618,
      "num_tokens": 43497457.0,
      "step": 20955
    },
    {
      "entropy": 5.562612533569336,
      "epoch": 1.9013062409288826,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046393491266987484,
      "loss": 5.0487,
      "mean_token_accuracy": 0.20178602188825606,
      "num_tokens": 43507761.0,
      "step": 20960
    },
    {
      "entropy": 5.497231721878052,
      "epoch": 1.9017597968069666,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004639173400809502,
      "loss": 4.9672,
      "mean_token_accuracy": 0.20882880091667175,
      "num_tokens": 43517198.0,
      "step": 20965
    },
    {
      "entropy": 5.540993881225586,
      "epoch": 1.9022133526850507,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004638997635851669,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20050388872623442,
      "num_tokens": 43527429.0,
      "step": 20970
    },
    {
      "entropy": 5.612675666809082,
      "epoch": 1.902666908563135,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046388218318288844,
      "loss": 5.0583,
      "mean_token_accuracy": 0.20635149627923965,
      "num_tokens": 43537867.0,
      "step": 20975
    },
    {
      "entropy": 5.600853681564331,
      "epoch": 1.9031204644412192,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004638645988744782,
      "loss": 4.995,
      "mean_token_accuracy": 0.20810700356960296,
      "num_tokens": 43547284.0,
      "step": 20980
    },
    {
      "entropy": 5.544989919662475,
      "epoch": 1.9035740203193035,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004638470106603,
      "loss": 5.0496,
      "mean_token_accuracy": 0.20433826893568038,
      "num_tokens": 43557693.0,
      "step": 20985
    },
    {
      "entropy": 5.580530118942261,
      "epoch": 1.9040275761973875,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004638294185407174,
      "loss": 4.9906,
      "mean_token_accuracy": 0.21300598680973054,
      "num_tokens": 43566909.0,
      "step": 20990
    },
    {
      "entropy": 5.587985372543335,
      "epoch": 1.9044811320754715,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004638118225160944,
      "loss": 5.0365,
      "mean_token_accuracy": 0.21214412152767181,
      "num_tokens": 43576963.0,
      "step": 20995
    },
    {
      "entropy": 5.497220659255982,
      "epoch": 1.9049346879535558,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046379422258679477,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20666191428899766,
      "num_tokens": 43586649.0,
      "step": 21000
    },
    {
      "epoch": 1.9049346879535558,
      "eval_entropy": 5.2827021436993356,
      "eval_loss": 5.041497230529785,
      "eval_mean_token_accuracy": 0.21340517185426136,
      "eval_num_tokens": 43586649.0,
      "eval_runtime": 20.6472,
      "eval_samples_per_second": 1712.58,
      "eval_steps_per_second": 214.072,
      "step": 21000
    },
    {
      "entropy": 5.486915922164917,
      "epoch": 1.90538824383164,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004637766187531825,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21056651920080185,
      "num_tokens": 43596506.0,
      "step": 21005
    },
    {
      "entropy": 5.500951766967773,
      "epoch": 1.9058417997097243,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046375901101562163,
      "loss": 5.0464,
      "mean_token_accuracy": 0.20540109127759934,
      "num_tokens": 43607188.0,
      "step": 21010
    },
    {
      "entropy": 5.562234878540039,
      "epoch": 1.9062953555878084,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004637413993744763,
      "loss": 4.9637,
      "mean_token_accuracy": 0.21434430927038192,
      "num_tokens": 43617465.0,
      "step": 21015
    },
    {
      "entropy": 5.558161783218384,
      "epoch": 1.9067489114658926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046372378383011077,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20854316800832748,
      "num_tokens": 43627692.0,
      "step": 21020
    },
    {
      "entropy": 5.5884510517120365,
      "epoch": 1.9072024673439767,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046370616438288933,
      "loss": 5.0386,
      "mean_token_accuracy": 0.20229142010211945,
      "num_tokens": 43637350.0,
      "step": 21025
    },
    {
      "entropy": 5.564635181427002,
      "epoch": 1.907656023222061,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046368854103317624,
      "loss": 5.0286,
      "mean_token_accuracy": 0.2014291226863861,
      "num_tokens": 43647092.0,
      "step": 21030
    },
    {
      "entropy": 5.542835474014282,
      "epoch": 1.9081095791001452,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046367091378133615,
      "loss": 4.986,
      "mean_token_accuracy": 0.20898470431566238,
      "num_tokens": 43657237.0,
      "step": 21035
    },
    {
      "entropy": 5.577055358886719,
      "epoch": 1.9085631349782293,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004636532826277333,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20737866908311844,
      "num_tokens": 43667966.0,
      "step": 21040
    },
    {
      "entropy": 5.530750703811646,
      "epoch": 1.9090166908563135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046363564757273264,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2092711329460144,
      "num_tokens": 43678059.0,
      "step": 21045
    },
    {
      "entropy": 5.525274276733398,
      "epoch": 1.9094702467343976,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046361800861669867,
      "loss": 4.9845,
      "mean_token_accuracy": 0.2132541298866272,
      "num_tokens": 43687556.0,
      "step": 21050
    },
    {
      "entropy": 5.587990045547485,
      "epoch": 1.9099238026124818,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046360036575999623,
      "loss": 5.0244,
      "mean_token_accuracy": 0.20550597310066224,
      "num_tokens": 43697791.0,
      "step": 21055
    },
    {
      "entropy": 5.578096628189087,
      "epoch": 1.9103773584905661,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004635827190029902,
      "loss": 4.9358,
      "mean_token_accuracy": 0.20452708899974822,
      "num_tokens": 43707390.0,
      "step": 21060
    },
    {
      "entropy": 5.6091104507446286,
      "epoch": 1.9108309143686504,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004635650683460454,
      "loss": 5.062,
      "mean_token_accuracy": 0.20398611426353455,
      "num_tokens": 43716842.0,
      "step": 21065
    },
    {
      "entropy": 5.482925319671631,
      "epoch": 1.9112844702467344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000463547413789527,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20588076561689378,
      "num_tokens": 43726967.0,
      "step": 21070
    },
    {
      "entropy": 5.621833944320679,
      "epoch": 1.9117380261248185,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046352975533379997,
      "loss": 4.9941,
      "mean_token_accuracy": 0.21100803166627885,
      "num_tokens": 43737425.0,
      "step": 21075
    },
    {
      "entropy": 5.5032435894012455,
      "epoch": 1.9121915820029027,
      "grad_norm": 1.125,
      "learning_rate": 0.00046351209297922964,
      "loss": 4.9277,
      "mean_token_accuracy": 0.215091273188591,
      "num_tokens": 43747395.0,
      "step": 21080
    },
    {
      "entropy": 5.527018260955811,
      "epoch": 1.912645137880987,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046349442672618114,
      "loss": 5.0381,
      "mean_token_accuracy": 0.2003874033689499,
      "num_tokens": 43756348.0,
      "step": 21085
    },
    {
      "entropy": 5.550302171707154,
      "epoch": 1.9130986937590713,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046347675657501986,
      "loss": 4.9845,
      "mean_token_accuracy": 0.20703417658805848,
      "num_tokens": 43767629.0,
      "step": 21090
    },
    {
      "entropy": 5.559140729904175,
      "epoch": 1.9135522496371553,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004634590825261113,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21757467538118364,
      "num_tokens": 43777398.0,
      "step": 21095
    },
    {
      "entropy": 5.564572525024414,
      "epoch": 1.9140058055152394,
      "grad_norm": 1.125,
      "learning_rate": 0.00046344140457982075,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20926016867160796,
      "num_tokens": 43787605.0,
      "step": 21100
    },
    {
      "entropy": 5.522140550613403,
      "epoch": 1.9144593613933236,
      "grad_norm": 1.109375,
      "learning_rate": 0.000463423722736514,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20708602368831636,
      "num_tokens": 43796989.0,
      "step": 21105
    },
    {
      "entropy": 5.550064373016357,
      "epoch": 1.9149129172714079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004634060369965567,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20980975776910782,
      "num_tokens": 43806653.0,
      "step": 21110
    },
    {
      "entropy": 5.5119102001190186,
      "epoch": 1.9153664731494922,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046338834736031446,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21025218218564987,
      "num_tokens": 43817152.0,
      "step": 21115
    },
    {
      "entropy": 5.545574283599853,
      "epoch": 1.9158200290275762,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046337065382815324,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20765405297279357,
      "num_tokens": 43827341.0,
      "step": 21120
    },
    {
      "entropy": 5.510351037979126,
      "epoch": 1.9162735849056602,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046335295640043887,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21740228086709976,
      "num_tokens": 43837195.0,
      "step": 21125
    },
    {
      "entropy": 5.531359481811523,
      "epoch": 1.9167271407837445,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046333525507753743,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21550391167402266,
      "num_tokens": 43847473.0,
      "step": 21130
    },
    {
      "entropy": 5.509198904037476,
      "epoch": 1.9171806966618288,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004633175498598148,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20922050327062608,
      "num_tokens": 43857590.0,
      "step": 21135
    },
    {
      "entropy": 5.563808107376099,
      "epoch": 1.917634252539913,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046329984074763733,
      "loss": 5.1241,
      "mean_token_accuracy": 0.1958355963230133,
      "num_tokens": 43867085.0,
      "step": 21140
    },
    {
      "entropy": 5.573216676712036,
      "epoch": 1.918087808417997,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004632821277413711,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20524263232946396,
      "num_tokens": 43877153.0,
      "step": 21145
    },
    {
      "entropy": 5.632125520706177,
      "epoch": 1.9185413642960811,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004632644108413825,
      "loss": 5.0833,
      "mean_token_accuracy": 0.19107387959957123,
      "num_tokens": 43887977.0,
      "step": 21150
    },
    {
      "entropy": 5.476994419097901,
      "epoch": 1.9189949201741654,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046324669004803803,
      "loss": 4.9789,
      "mean_token_accuracy": 0.20683950334787368,
      "num_tokens": 43898649.0,
      "step": 21155
    },
    {
      "entropy": 5.6364538192749025,
      "epoch": 1.9194484760522497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000463228965361704,
      "loss": 5.0881,
      "mean_token_accuracy": 0.2023804023861885,
      "num_tokens": 43909369.0,
      "step": 21160
    },
    {
      "entropy": 5.600297927856445,
      "epoch": 1.919902031930334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004632112367827469,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20831674337387085,
      "num_tokens": 43919941.0,
      "step": 21165
    },
    {
      "entropy": 5.576290559768677,
      "epoch": 1.920355587808418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004631935043115335,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20568810999393464,
      "num_tokens": 43929710.0,
      "step": 21170
    },
    {
      "entropy": 5.52694091796875,
      "epoch": 1.920809143686502,
      "grad_norm": 1.125,
      "learning_rate": 0.00046317576794843056,
      "loss": 4.9718,
      "mean_token_accuracy": 0.20809148699045182,
      "num_tokens": 43939850.0,
      "step": 21175
    },
    {
      "entropy": 5.530804395675659,
      "epoch": 1.9212626995645863,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046315802769380474,
      "loss": 4.9089,
      "mean_token_accuracy": 0.2124344989657402,
      "num_tokens": 43950067.0,
      "step": 21180
    },
    {
      "entropy": 5.560138988494873,
      "epoch": 1.9217162554426706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004631402835480229,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21008018553256988,
      "num_tokens": 43960029.0,
      "step": 21185
    },
    {
      "entropy": 5.544347286224365,
      "epoch": 1.9221698113207548,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046312253551145213,
      "loss": 5.0084,
      "mean_token_accuracy": 0.20792224556207656,
      "num_tokens": 43971221.0,
      "step": 21190
    },
    {
      "entropy": 5.570632648468018,
      "epoch": 1.9226233671988389,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046310478358445935,
      "loss": 5.0277,
      "mean_token_accuracy": 0.2045084208250046,
      "num_tokens": 43981579.0,
      "step": 21195
    },
    {
      "entropy": 5.587553834915161,
      "epoch": 1.9230769230769231,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004630870277674117,
      "loss": 5.123,
      "mean_token_accuracy": 0.19563973993062972,
      "num_tokens": 43991132.0,
      "step": 21200
    },
    {
      "entropy": 5.548740816116333,
      "epoch": 1.9235304789550072,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004630692680606764,
      "loss": 4.9196,
      "mean_token_accuracy": 0.20882577300071717,
      "num_tokens": 44001255.0,
      "step": 21205
    },
    {
      "entropy": 5.490432262420654,
      "epoch": 1.9239840348330914,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004630515044646207,
      "loss": 4.836,
      "mean_token_accuracy": 0.21653300970792771,
      "num_tokens": 44011275.0,
      "step": 21210
    },
    {
      "entropy": 5.536353349685669,
      "epoch": 1.9244375907111757,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000463033736979612,
      "loss": 5.0134,
      "mean_token_accuracy": 0.20539278090000151,
      "num_tokens": 44022524.0,
      "step": 21215
    },
    {
      "entropy": 5.524604558944702,
      "epoch": 1.9248911465892597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046301596560601767,
      "loss": 4.918,
      "mean_token_accuracy": 0.20774720907211303,
      "num_tokens": 44033955.0,
      "step": 21220
    },
    {
      "entropy": 5.55246868133545,
      "epoch": 1.925344702467344,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004629981903442052,
      "loss": 5.0268,
      "mean_token_accuracy": 0.2021826833486557,
      "num_tokens": 44045149.0,
      "step": 21225
    },
    {
      "entropy": 5.6093236923217775,
      "epoch": 1.925798258345428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004629804111945424,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20155078619718553,
      "num_tokens": 44055280.0,
      "step": 21230
    },
    {
      "entropy": 5.553763055801392,
      "epoch": 1.9262518142235123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004629626281573966,
      "loss": 4.9593,
      "mean_token_accuracy": 0.21317025125026703,
      "num_tokens": 44065578.0,
      "step": 21235
    },
    {
      "entropy": 5.47047119140625,
      "epoch": 1.9267053701015966,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004629448412331359,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21356330811977386,
      "num_tokens": 44075454.0,
      "step": 21240
    },
    {
      "entropy": 5.518563747406006,
      "epoch": 1.9271589259796809,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004629270504221279,
      "loss": 5.0179,
      "mean_token_accuracy": 0.2049138143658638,
      "num_tokens": 44085680.0,
      "step": 21245
    },
    {
      "entropy": 5.4879231452941895,
      "epoch": 1.927612481857765,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004629092557247407,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21290490329265593,
      "num_tokens": 44097177.0,
      "step": 21250
    },
    {
      "entropy": 5.588929128646851,
      "epoch": 1.928066037735849,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004628914571413422,
      "loss": 5.05,
      "mean_token_accuracy": 0.20613875538110732,
      "num_tokens": 44107251.0,
      "step": 21255
    },
    {
      "entropy": 5.552894973754883,
      "epoch": 1.9285195936139332,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046287365467230035,
      "loss": 5.0305,
      "mean_token_accuracy": 0.20274046510457994,
      "num_tokens": 44117943.0,
      "step": 21260
    },
    {
      "entropy": 5.620401859283447,
      "epoch": 1.9289731494920175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046285584831798354,
      "loss": 5.0251,
      "mean_token_accuracy": 0.21080010682344436,
      "num_tokens": 44128608.0,
      "step": 21265
    },
    {
      "entropy": 5.584255838394165,
      "epoch": 1.9294267053701017,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046283803807875996,
      "loss": 4.9946,
      "mean_token_accuracy": 0.21257734894752503,
      "num_tokens": 44138809.0,
      "step": 21270
    },
    {
      "entropy": 5.531477785110473,
      "epoch": 1.9298802612481858,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046282022395499785,
      "loss": 5.0136,
      "mean_token_accuracy": 0.2101381629705429,
      "num_tokens": 44148532.0,
      "step": 21275
    },
    {
      "entropy": 5.580505990982056,
      "epoch": 1.9303338171262698,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004628024059470656,
      "loss": 5.0275,
      "mean_token_accuracy": 0.20669040381908416,
      "num_tokens": 44158628.0,
      "step": 21280
    },
    {
      "entropy": 5.5577842712402346,
      "epoch": 1.930787373004354,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004627845840553318,
      "loss": 4.8918,
      "mean_token_accuracy": 0.21354376524686813,
      "num_tokens": 44168574.0,
      "step": 21285
    },
    {
      "entropy": 5.599446964263916,
      "epoch": 1.9312409288824384,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046276675828016493,
      "loss": 5.0614,
      "mean_token_accuracy": 0.20820466727018355,
      "num_tokens": 44180207.0,
      "step": 21290
    },
    {
      "entropy": 5.609013271331787,
      "epoch": 1.9316944847605226,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004627489286219336,
      "loss": 5.1058,
      "mean_token_accuracy": 0.1978868678212166,
      "num_tokens": 44190473.0,
      "step": 21295
    },
    {
      "entropy": 5.6146135330200195,
      "epoch": 1.9321480406386067,
      "grad_norm": 1.0,
      "learning_rate": 0.0004627310950810066,
      "loss": 5.0343,
      "mean_token_accuracy": 0.20951875001192094,
      "num_tokens": 44201150.0,
      "step": 21300
    },
    {
      "entropy": 5.520677757263184,
      "epoch": 1.9326015965166907,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004627132576577528,
      "loss": 4.9987,
      "mean_token_accuracy": 0.19714128077030182,
      "num_tokens": 44211272.0,
      "step": 21305
    },
    {
      "entropy": 5.577132272720337,
      "epoch": 1.933055152394775,
      "grad_norm": 0.96875,
      "learning_rate": 0.000462695416352541,
      "loss": 5.0724,
      "mean_token_accuracy": 0.20158825367689132,
      "num_tokens": 44223556.0,
      "step": 21310
    },
    {
      "entropy": 5.546791362762451,
      "epoch": 1.9335087082728593,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046267757116574,
      "loss": 4.9681,
      "mean_token_accuracy": 0.2087857723236084,
      "num_tokens": 44233911.0,
      "step": 21315
    },
    {
      "entropy": 5.603171873092651,
      "epoch": 1.9339622641509435,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004626597220977191,
      "loss": 5.0477,
      "mean_token_accuracy": 0.21053002923727035,
      "num_tokens": 44244108.0,
      "step": 21320
    },
    {
      "entropy": 5.486507225036621,
      "epoch": 1.9344158200290276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046264186914884737,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2079022005200386,
      "num_tokens": 44253604.0,
      "step": 21325
    },
    {
      "entropy": 5.514385032653808,
      "epoch": 1.9348693759071116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046262401231949386,
      "loss": 5.0501,
      "mean_token_accuracy": 0.199178808927536,
      "num_tokens": 44263573.0,
      "step": 21330
    },
    {
      "entropy": 5.542400693893432,
      "epoch": 1.9353229317851959,
      "grad_norm": 1.0,
      "learning_rate": 0.00046260615161002803,
      "loss": 4.9609,
      "mean_token_accuracy": 0.2072915554046631,
      "num_tokens": 44274071.0,
      "step": 21335
    },
    {
      "entropy": 5.650053405761719,
      "epoch": 1.9357764876632801,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004625882870208192,
      "loss": 5.0094,
      "mean_token_accuracy": 0.20556609183549882,
      "num_tokens": 44283997.0,
      "step": 21340
    },
    {
      "entropy": 5.605504035949707,
      "epoch": 1.9362300435413644,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046257041855223674,
      "loss": 4.9875,
      "mean_token_accuracy": 0.20844724476337434,
      "num_tokens": 44295024.0,
      "step": 21345
    },
    {
      "entropy": 5.49817156791687,
      "epoch": 1.9366835994194485,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046255254620465033,
      "loss": 4.9489,
      "mean_token_accuracy": 0.21543079614639282,
      "num_tokens": 44305476.0,
      "step": 21350
    },
    {
      "entropy": 5.509833717346192,
      "epoch": 1.9371371552975325,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046253466997842936,
      "loss": 4.9126,
      "mean_token_accuracy": 0.21113743185997008,
      "num_tokens": 44316101.0,
      "step": 21355
    },
    {
      "entropy": 5.503489017486572,
      "epoch": 1.9375907111756168,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004625167898739436,
      "loss": 4.8728,
      "mean_token_accuracy": 0.2139115959405899,
      "num_tokens": 44325058.0,
      "step": 21360
    },
    {
      "entropy": 5.665496063232422,
      "epoch": 1.938044267053701,
      "grad_norm": 1.015625,
      "learning_rate": 0.000462498905891563,
      "loss": 5.1797,
      "mean_token_accuracy": 0.18986935168504715,
      "num_tokens": 44336585.0,
      "step": 21365
    },
    {
      "entropy": 5.543269491195678,
      "epoch": 1.9384978229317853,
      "grad_norm": 1.125,
      "learning_rate": 0.0004624810180316571,
      "loss": 4.9843,
      "mean_token_accuracy": 0.2091907352209091,
      "num_tokens": 44346911.0,
      "step": 21370
    },
    {
      "entropy": 5.515858936309814,
      "epoch": 1.9389513788098693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046246312629459596,
      "loss": 4.9494,
      "mean_token_accuracy": 0.207963527739048,
      "num_tokens": 44358040.0,
      "step": 21375
    },
    {
      "entropy": 5.564346837997436,
      "epoch": 1.9394049346879536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004624452306807497,
      "loss": 4.9829,
      "mean_token_accuracy": 0.21364262849092483,
      "num_tokens": 44368781.0,
      "step": 21380
    },
    {
      "entropy": 5.488349914550781,
      "epoch": 1.9398584905660377,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004624273311904882,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21095094680786133,
      "num_tokens": 44378809.0,
      "step": 21385
    },
    {
      "entropy": 5.605105686187744,
      "epoch": 1.940312046444122,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046240942782418184,
      "loss": 5.0824,
      "mean_token_accuracy": 0.20438190251588823,
      "num_tokens": 44388517.0,
      "step": 21390
    },
    {
      "entropy": 5.631004095077515,
      "epoch": 1.9407656023222062,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004623915205822007,
      "loss": 5.0344,
      "mean_token_accuracy": 0.20357698798179627,
      "num_tokens": 44399364.0,
      "step": 21395
    },
    {
      "entropy": 5.540302753448486,
      "epoch": 1.9412191582002902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046237360946491514,
      "loss": 4.9337,
      "mean_token_accuracy": 0.20093414485454558,
      "num_tokens": 44409506.0,
      "step": 21400
    },
    {
      "entropy": 5.5960570812225345,
      "epoch": 1.9416727140783745,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004623556944726956,
      "loss": 5.0113,
      "mean_token_accuracy": 0.20793051570653914,
      "num_tokens": 44419446.0,
      "step": 21405
    },
    {
      "entropy": 5.544157934188843,
      "epoch": 1.9421262699564585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046233777560591256,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20520949959754944,
      "num_tokens": 44429560.0,
      "step": 21410
    },
    {
      "entropy": 5.546304750442505,
      "epoch": 1.9425798258345428,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046231985286493654,
      "loss": 4.975,
      "mean_token_accuracy": 0.20405519604682923,
      "num_tokens": 44439640.0,
      "step": 21415
    },
    {
      "entropy": 5.578876399993897,
      "epoch": 1.943033381712627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046230192625013836,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20514049381017685,
      "num_tokens": 44450480.0,
      "step": 21420
    },
    {
      "entropy": 5.479057550430298,
      "epoch": 1.9434869375907113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046228399576188844,
      "loss": 4.892,
      "mean_token_accuracy": 0.20939820855855942,
      "num_tokens": 44460633.0,
      "step": 21425
    },
    {
      "entropy": 5.556435871124267,
      "epoch": 1.9439404934687954,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046226606140055784,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2110549733042717,
      "num_tokens": 44470917.0,
      "step": 21430
    },
    {
      "entropy": 5.555643892288208,
      "epoch": 1.9443940493468794,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046224812316651736,
      "loss": 5.0144,
      "mean_token_accuracy": 0.2043233558535576,
      "num_tokens": 44482538.0,
      "step": 21435
    },
    {
      "entropy": 5.5876607418060305,
      "epoch": 1.9448476052249637,
      "grad_norm": 1.0625,
      "learning_rate": 0.000462230181060138,
      "loss": 5.0392,
      "mean_token_accuracy": 0.20266881585121155,
      "num_tokens": 44493518.0,
      "step": 21440
    },
    {
      "entropy": 5.587052583694458,
      "epoch": 1.945301161103048,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046221223508179077,
      "loss": 5.0227,
      "mean_token_accuracy": 0.20541617274284363,
      "num_tokens": 44503685.0,
      "step": 21445
    },
    {
      "entropy": 5.538943290710449,
      "epoch": 1.9457547169811322,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046219428523184685,
      "loss": 4.956,
      "mean_token_accuracy": 0.21194851249456406,
      "num_tokens": 44514015.0,
      "step": 21450
    },
    {
      "entropy": 5.5282660007476805,
      "epoch": 1.9462082728592163,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004621763315106774,
      "loss": 4.9838,
      "mean_token_accuracy": 0.2088385656476021,
      "num_tokens": 44525187.0,
      "step": 21455
    },
    {
      "entropy": 5.470666694641113,
      "epoch": 1.9466618287373003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046215837391865364,
      "loss": 4.9605,
      "mean_token_accuracy": 0.2128278523683548,
      "num_tokens": 44535964.0,
      "step": 21460
    },
    {
      "entropy": 5.503521203994751,
      "epoch": 1.9471153846153846,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046214041245614713,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2047021880745888,
      "num_tokens": 44546985.0,
      "step": 21465
    },
    {
      "entropy": 5.594559001922607,
      "epoch": 1.9475689404934688,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004621224471235291,
      "loss": 5.0258,
      "mean_token_accuracy": 0.19286321550607682,
      "num_tokens": 44556888.0,
      "step": 21470
    },
    {
      "entropy": 5.569971132278442,
      "epoch": 1.9480224963715531,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046210447792117114,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20596435964107512,
      "num_tokens": 44566857.0,
      "step": 21475
    },
    {
      "entropy": 5.581793975830078,
      "epoch": 1.9484760522496372,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000462086504849445,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20321155339479446,
      "num_tokens": 44578152.0,
      "step": 21480
    },
    {
      "entropy": 5.5432069301605225,
      "epoch": 1.9489296081277212,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004620685279087222,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21598812192678452,
      "num_tokens": 44588814.0,
      "step": 21485
    },
    {
      "entropy": 5.535352325439453,
      "epoch": 1.9493831640058055,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046205054709937474,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20474015027284623,
      "num_tokens": 44599366.0,
      "step": 21490
    },
    {
      "entropy": 5.661087274551392,
      "epoch": 1.9498367198838897,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004620325624217741,
      "loss": 5.0773,
      "mean_token_accuracy": 0.1957104668021202,
      "num_tokens": 44609962.0,
      "step": 21495
    },
    {
      "entropy": 5.627064371109009,
      "epoch": 1.950290275761974,
      "grad_norm": 1.125,
      "learning_rate": 0.0004620145738762925,
      "loss": 5.0339,
      "mean_token_accuracy": 0.20886536687612534,
      "num_tokens": 44619917.0,
      "step": 21500
    },
    {
      "entropy": 5.508686161041259,
      "epoch": 1.950743831640058,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004619965814633019,
      "loss": 4.982,
      "mean_token_accuracy": 0.21039706617593765,
      "num_tokens": 44629723.0,
      "step": 21505
    },
    {
      "entropy": 5.548584842681885,
      "epoch": 1.951197387518142,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004619785851831742,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20363247990608216,
      "num_tokens": 44641011.0,
      "step": 21510
    },
    {
      "entropy": 5.573685359954834,
      "epoch": 1.9516509433962264,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046196058503628176,
      "loss": 4.963,
      "mean_token_accuracy": 0.20982809662818908,
      "num_tokens": 44651154.0,
      "step": 21515
    },
    {
      "entropy": 5.5663313388824465,
      "epoch": 1.9521044992743106,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004619425810229968,
      "loss": 5.0547,
      "mean_token_accuracy": 0.2016923740506172,
      "num_tokens": 44661672.0,
      "step": 21520
    },
    {
      "entropy": 5.506641054153443,
      "epoch": 1.952558055152395,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004619245731436917,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21523777395486832,
      "num_tokens": 44671449.0,
      "step": 21525
    },
    {
      "entropy": 5.567077016830444,
      "epoch": 1.953011611030479,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046190656139873863,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21322334855794906,
      "num_tokens": 44681977.0,
      "step": 21530
    },
    {
      "entropy": 5.548317098617554,
      "epoch": 1.953465166908563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004618885457885103,
      "loss": 4.9549,
      "mean_token_accuracy": 0.21130643784999847,
      "num_tokens": 44691946.0,
      "step": 21535
    },
    {
      "entropy": 5.526917743682861,
      "epoch": 1.9539187227866472,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004618705263133793,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20774240046739578,
      "num_tokens": 44702012.0,
      "step": 21540
    },
    {
      "entropy": 5.534388828277588,
      "epoch": 1.9543722786647315,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004618525029737182,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21082423031330108,
      "num_tokens": 44711605.0,
      "step": 21545
    },
    {
      "entropy": 5.611813592910766,
      "epoch": 1.9548258345428158,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046183447576989966,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20988747030496596,
      "num_tokens": 44722765.0,
      "step": 21550
    },
    {
      "entropy": 5.5836060523986815,
      "epoch": 1.9552793904208998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046181644470229656,
      "loss": 4.9701,
      "mean_token_accuracy": 0.208165480196476,
      "num_tokens": 44732082.0,
      "step": 21555
    },
    {
      "entropy": 5.461510801315308,
      "epoch": 1.955732946298984,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004617984097712817,
      "loss": 4.8977,
      "mean_token_accuracy": 0.21089304983615875,
      "num_tokens": 44742054.0,
      "step": 21560
    },
    {
      "entropy": 5.486566877365112,
      "epoch": 1.9561865021770681,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004617803709772282,
      "loss": 5.0066,
      "mean_token_accuracy": 0.2106909289956093,
      "num_tokens": 44752954.0,
      "step": 21565
    },
    {
      "entropy": 5.573259687423706,
      "epoch": 1.9566400580551524,
      "grad_norm": 1.0,
      "learning_rate": 0.0004617623283205091,
      "loss": 5.0292,
      "mean_token_accuracy": 0.20604926198720933,
      "num_tokens": 44764277.0,
      "step": 21570
    },
    {
      "entropy": 5.6039666652679445,
      "epoch": 1.9570936139332367,
      "grad_norm": 1.125,
      "learning_rate": 0.00046174428180149733,
      "loss": 5.057,
      "mean_token_accuracy": 0.20239367187023163,
      "num_tokens": 44775429.0,
      "step": 21575
    },
    {
      "entropy": 5.613869333267212,
      "epoch": 1.9575471698113207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004617262314205663,
      "loss": 5.0372,
      "mean_token_accuracy": 0.20427386611700057,
      "num_tokens": 44786390.0,
      "step": 21580
    },
    {
      "entropy": 5.583804893493652,
      "epoch": 1.958000725689405,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004617081771780893,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20702233165502548,
      "num_tokens": 44796473.0,
      "step": 21585
    },
    {
      "entropy": 5.519788503646851,
      "epoch": 1.958454281567489,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046169011907443944,
      "loss": 4.9429,
      "mean_token_accuracy": 0.20918683409690858,
      "num_tokens": 44806379.0,
      "step": 21590
    },
    {
      "entropy": 5.5597998142242435,
      "epoch": 1.9589078374455733,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004616720571099904,
      "loss": 5.0267,
      "mean_token_accuracy": 0.20871667712926864,
      "num_tokens": 44817659.0,
      "step": 21595
    },
    {
      "entropy": 5.581849050521851,
      "epoch": 1.9593613933236576,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004616539912851157,
      "loss": 4.9463,
      "mean_token_accuracy": 0.20679994970560073,
      "num_tokens": 44827073.0,
      "step": 21600
    },
    {
      "entropy": 5.507561159133911,
      "epoch": 1.9598149492017418,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046163592160018886,
      "loss": 4.9576,
      "mean_token_accuracy": 0.20796812921762467,
      "num_tokens": 44837019.0,
      "step": 21605
    },
    {
      "entropy": 5.568103218078614,
      "epoch": 1.9602685050798259,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004616178480555836,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21116245090961455,
      "num_tokens": 44846670.0,
      "step": 21610
    },
    {
      "entropy": 5.479678535461426,
      "epoch": 1.96072206095791,
      "grad_norm": 1.125,
      "learning_rate": 0.0004615997706516737,
      "loss": 4.9048,
      "mean_token_accuracy": 0.21223325580358504,
      "num_tokens": 44857016.0,
      "step": 21615
    },
    {
      "entropy": 5.618633842468261,
      "epoch": 1.9611756168359942,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000461581689388833,
      "loss": 5.0739,
      "mean_token_accuracy": 0.19679954797029495,
      "num_tokens": 44868180.0,
      "step": 21620
    },
    {
      "entropy": 5.514395332336425,
      "epoch": 1.9616291727140784,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046156360426743536,
      "loss": 4.9259,
      "mean_token_accuracy": 0.2127016082406044,
      "num_tokens": 44877864.0,
      "step": 21625
    },
    {
      "entropy": 5.584148025512695,
      "epoch": 1.9620827285921627,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004615455152878549,
      "loss": 4.9796,
      "mean_token_accuracy": 0.2177381232380867,
      "num_tokens": 44887737.0,
      "step": 21630
    },
    {
      "entropy": 5.482257509231568,
      "epoch": 1.9625362844702468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004615274224504657,
      "loss": 4.9875,
      "mean_token_accuracy": 0.20395839661359788,
      "num_tokens": 44898669.0,
      "step": 21635
    },
    {
      "entropy": 5.585608720779419,
      "epoch": 1.9629898403483308,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046150932575564173,
      "loss": 5.0484,
      "mean_token_accuracy": 0.21015833765268327,
      "num_tokens": 44909074.0,
      "step": 21640
    },
    {
      "entropy": 5.535607528686524,
      "epoch": 1.963443396226415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046149122520375746,
      "loss": 4.9414,
      "mean_token_accuracy": 0.20969487279653548,
      "num_tokens": 44919468.0,
      "step": 21645
    },
    {
      "entropy": 5.601579999923706,
      "epoch": 1.9638969521044993,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004614731207951871,
      "loss": 5.0555,
      "mean_token_accuracy": 0.19970683753490448,
      "num_tokens": 44930470.0,
      "step": 21650
    },
    {
      "entropy": 5.551943969726563,
      "epoch": 1.9643505079825836,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004614550125303051,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20635733604431153,
      "num_tokens": 44940611.0,
      "step": 21655
    },
    {
      "entropy": 5.551587009429932,
      "epoch": 1.9648040638606676,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004614369004094859,
      "loss": 4.9948,
      "mean_token_accuracy": 0.20839242786169052,
      "num_tokens": 44950985.0,
      "step": 21660
    },
    {
      "entropy": 5.5767988681793215,
      "epoch": 1.9652576197387517,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046141878443310415,
      "loss": 5.0377,
      "mean_token_accuracy": 0.20357128083705903,
      "num_tokens": 44961858.0,
      "step": 21665
    },
    {
      "entropy": 5.585680866241455,
      "epoch": 1.965711175616836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046140066460153444,
      "loss": 4.9962,
      "mean_token_accuracy": 0.2094133287668228,
      "num_tokens": 44972048.0,
      "step": 21670
    },
    {
      "entropy": 5.550045347213745,
      "epoch": 1.9661647314949202,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046138254091515136,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2133963719010353,
      "num_tokens": 44981835.0,
      "step": 21675
    },
    {
      "entropy": 5.527277660369873,
      "epoch": 1.9666182873730045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046136441337432996,
      "loss": 4.902,
      "mean_token_accuracy": 0.22058788537979127,
      "num_tokens": 44992108.0,
      "step": 21680
    },
    {
      "entropy": 5.657939195632935,
      "epoch": 1.9670718432510885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046134628197944496,
      "loss": 5.068,
      "mean_token_accuracy": 0.20214518755674363,
      "num_tokens": 45002105.0,
      "step": 21685
    },
    {
      "entropy": 5.52752161026001,
      "epoch": 1.9675253991291726,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046132814673087143,
      "loss": 4.9957,
      "mean_token_accuracy": 0.20533880591392517,
      "num_tokens": 45013934.0,
      "step": 21690
    },
    {
      "entropy": 5.514544820785522,
      "epoch": 1.9679789550072568,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004613100076289842,
      "loss": 4.9806,
      "mean_token_accuracy": 0.2086989015340805,
      "num_tokens": 45023851.0,
      "step": 21695
    },
    {
      "entropy": 5.508292245864868,
      "epoch": 1.968432510885341,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004612918646741586,
      "loss": 4.9118,
      "mean_token_accuracy": 0.20739118456840516,
      "num_tokens": 45034140.0,
      "step": 21700
    },
    {
      "entropy": 5.541641807556152,
      "epoch": 1.9688860667634254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004612737178667698,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20653848052024842,
      "num_tokens": 45044452.0,
      "step": 21705
    },
    {
      "entropy": 5.540272903442383,
      "epoch": 1.9693396226415094,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046125556720719304,
      "loss": 5.0117,
      "mean_token_accuracy": 0.2012146830558777,
      "num_tokens": 45053845.0,
      "step": 21710
    },
    {
      "entropy": 5.551776456832886,
      "epoch": 1.9697931785195935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004612374126958037,
      "loss": 4.9341,
      "mean_token_accuracy": 0.20701974779367446,
      "num_tokens": 45063973.0,
      "step": 21715
    },
    {
      "entropy": 5.608538961410522,
      "epoch": 1.9702467343976777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004612192543329771,
      "loss": 5.0878,
      "mean_token_accuracy": 0.20605275481939317,
      "num_tokens": 45074879.0,
      "step": 21720
    },
    {
      "entropy": 5.523556232452393,
      "epoch": 1.970700290275762,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000461201092119089,
      "loss": 4.9229,
      "mean_token_accuracy": 0.21459337770938874,
      "num_tokens": 45085161.0,
      "step": 21725
    },
    {
      "entropy": 5.583820486068726,
      "epoch": 1.9711538461538463,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004611829260545148,
      "loss": 5.063,
      "mean_token_accuracy": 0.20150579512119293,
      "num_tokens": 45095862.0,
      "step": 21730
    },
    {
      "entropy": 5.590506267547608,
      "epoch": 1.9716074020319303,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046116475613963026,
      "loss": 4.9857,
      "mean_token_accuracy": 0.2055190920829773,
      "num_tokens": 45106396.0,
      "step": 21735
    },
    {
      "entropy": 5.557551670074463,
      "epoch": 1.9720609579100146,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046114658237481114,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21439230144023896,
      "num_tokens": 45117178.0,
      "step": 21740
    },
    {
      "entropy": 5.4734259128570555,
      "epoch": 1.9725145137880986,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004611284047604332,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21700416952371598,
      "num_tokens": 45126571.0,
      "step": 21745
    },
    {
      "entropy": 5.553644704818725,
      "epoch": 1.9729680696661829,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046111022329687246,
      "loss": 5.0282,
      "mean_token_accuracy": 0.21026648730039596,
      "num_tokens": 45137247.0,
      "step": 21750
    },
    {
      "entropy": 5.633781290054321,
      "epoch": 1.9734216255442671,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046109203798450483,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2073136553168297,
      "num_tokens": 45147531.0,
      "step": 21755
    },
    {
      "entropy": 5.5364891529083256,
      "epoch": 1.9738751814223512,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004610738488237065,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2094478502869606,
      "num_tokens": 45158312.0,
      "step": 21760
    },
    {
      "entropy": 5.513592100143432,
      "epoch": 1.9743287373004355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004610556558148535,
      "loss": 5.0161,
      "mean_token_accuracy": 0.2071317106485367,
      "num_tokens": 45169754.0,
      "step": 21765
    },
    {
      "entropy": 5.587693548202514,
      "epoch": 1.9747822931785195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004610374589583222,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20725469440221786,
      "num_tokens": 45180428.0,
      "step": 21770
    },
    {
      "entropy": 5.55584135055542,
      "epoch": 1.9752358490566038,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004610192582544888,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2135287344455719,
      "num_tokens": 45191558.0,
      "step": 21775
    },
    {
      "entropy": 5.486757230758667,
      "epoch": 1.975689404934688,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004610010537037297,
      "loss": 4.9375,
      "mean_token_accuracy": 0.21148865520954133,
      "num_tokens": 45201567.0,
      "step": 21780
    },
    {
      "entropy": 5.618310594558716,
      "epoch": 1.9761429608127723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004609828453064214,
      "loss": 5.0631,
      "mean_token_accuracy": 0.20185273587703706,
      "num_tokens": 45211359.0,
      "step": 21785
    },
    {
      "entropy": 5.617689752578736,
      "epoch": 1.9765965166908563,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046096463306294057,
      "loss": 4.98,
      "mean_token_accuracy": 0.20351368337869644,
      "num_tokens": 45222106.0,
      "step": 21790
    },
    {
      "entropy": 5.480583572387696,
      "epoch": 1.9770500725689404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046094641697366366,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20901469141244888,
      "num_tokens": 45233169.0,
      "step": 21795
    },
    {
      "entropy": 5.605236577987671,
      "epoch": 1.9775036284470247,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004609281970389675,
      "loss": 5.064,
      "mean_token_accuracy": 0.19933681339025497,
      "num_tokens": 45243627.0,
      "step": 21800
    },
    {
      "entropy": 5.594206094741821,
      "epoch": 1.977957184325109,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046090997325922887,
      "loss": 4.922,
      "mean_token_accuracy": 0.21322551965713502,
      "num_tokens": 45253730.0,
      "step": 21805
    },
    {
      "entropy": 5.533895969390869,
      "epoch": 1.9784107402031932,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046089174563482456,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20635027885437013,
      "num_tokens": 45263443.0,
      "step": 21810
    },
    {
      "entropy": 5.496335220336914,
      "epoch": 1.9788642960812772,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004608735141661316,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21626696288585662,
      "num_tokens": 45275298.0,
      "step": 21815
    },
    {
      "entropy": 5.553706550598145,
      "epoch": 1.9793178519593613,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046085527885352694,
      "loss": 5.0437,
      "mean_token_accuracy": 0.20689027458429338,
      "num_tokens": 45285628.0,
      "step": 21820
    },
    {
      "entropy": 5.535992097854614,
      "epoch": 1.9797714078374455,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004608370396973878,
      "loss": 4.9098,
      "mean_token_accuracy": 0.21273459494113922,
      "num_tokens": 45295559.0,
      "step": 21825
    },
    {
      "entropy": 5.507489633560181,
      "epoch": 1.9802249637155298,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046081879669809133,
      "loss": 4.9325,
      "mean_token_accuracy": 0.20965660661458968,
      "num_tokens": 45305395.0,
      "step": 21830
    },
    {
      "entropy": 5.523382043838501,
      "epoch": 1.980678519593614,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004608005498560148,
      "loss": 4.9574,
      "mean_token_accuracy": 0.20966893136501313,
      "num_tokens": 45316763.0,
      "step": 21835
    },
    {
      "entropy": 5.581750822067261,
      "epoch": 1.9811320754716981,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004607822991715356,
      "loss": 5.095,
      "mean_token_accuracy": 0.20398128926753997,
      "num_tokens": 45326484.0,
      "step": 21840
    },
    {
      "entropy": 5.575086307525635,
      "epoch": 1.9815856313497822,
      "grad_norm": 1.21875,
      "learning_rate": 0.000460764044645031,
      "loss": 5.0174,
      "mean_token_accuracy": 0.2053569659590721,
      "num_tokens": 45335792.0,
      "step": 21845
    },
    {
      "entropy": 5.5670332431793215,
      "epoch": 1.9820391872278664,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004607457862768787,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2041582301259041,
      "num_tokens": 45347247.0,
      "step": 21850
    },
    {
      "entropy": 5.608470106124878,
      "epoch": 1.9824927431059507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004607275240674562,
      "loss": 5.063,
      "mean_token_accuracy": 0.20145944505929947,
      "num_tokens": 45357994.0,
      "step": 21855
    },
    {
      "entropy": 5.502914667129517,
      "epoch": 1.982946298984035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004607092580171411,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20760190933942796,
      "num_tokens": 45368170.0,
      "step": 21860
    },
    {
      "entropy": 5.592855215072632,
      "epoch": 1.983399854862119,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004606909881263113,
      "loss": 5.0437,
      "mean_token_accuracy": 0.20415429025888443,
      "num_tokens": 45378696.0,
      "step": 21865
    },
    {
      "entropy": 5.6079676151275635,
      "epoch": 1.983853410740203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046067271439534445,
      "loss": 5.0706,
      "mean_token_accuracy": 0.20656584650278093,
      "num_tokens": 45388578.0,
      "step": 21870
    },
    {
      "entropy": 5.5629298210144045,
      "epoch": 1.9843069666182873,
      "grad_norm": 1.125,
      "learning_rate": 0.00046065443682461864,
      "loss": 5.0487,
      "mean_token_accuracy": 0.20561712235212326,
      "num_tokens": 45398835.0,
      "step": 21875
    },
    {
      "entropy": 5.595419931411743,
      "epoch": 1.9847605224963716,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046063615541451177,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20945125967264175,
      "num_tokens": 45409032.0,
      "step": 21880
    },
    {
      "entropy": 5.625519227981568,
      "epoch": 1.9852140783744558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046061787016540184,
      "loss": 5.1106,
      "mean_token_accuracy": 0.1957710400223732,
      "num_tokens": 45420661.0,
      "step": 21885
    },
    {
      "entropy": 5.5140228271484375,
      "epoch": 1.98566763425254,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046059958107766706,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22318844944238664,
      "num_tokens": 45429751.0,
      "step": 21890
    },
    {
      "entropy": 5.571798849105835,
      "epoch": 1.986121190130624,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004605812881516857,
      "loss": 4.9878,
      "mean_token_accuracy": 0.2143436923623085,
      "num_tokens": 45439551.0,
      "step": 21895
    },
    {
      "entropy": 5.482573986053467,
      "epoch": 1.9865747460087082,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046056299138783595,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22369229346513747,
      "num_tokens": 45449543.0,
      "step": 21900
    },
    {
      "entropy": 5.5848533630371096,
      "epoch": 1.9870283018867925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004605446907864963,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20833965241909028,
      "num_tokens": 45459069.0,
      "step": 21905
    },
    {
      "entropy": 5.585950422286987,
      "epoch": 1.9874818577648767,
      "grad_norm": 1.0,
      "learning_rate": 0.0004605263863480452,
      "loss": 5.074,
      "mean_token_accuracy": 0.20062010884284973,
      "num_tokens": 45468987.0,
      "step": 21910
    },
    {
      "entropy": 5.604973936080933,
      "epoch": 1.9879354136429608,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046050807807286105,
      "loss": 4.9927,
      "mean_token_accuracy": 0.20938576012849808,
      "num_tokens": 45479781.0,
      "step": 21915
    },
    {
      "entropy": 5.662037038803101,
      "epoch": 1.988388969521045,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046048976596132265,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19111909568309784,
      "num_tokens": 45490318.0,
      "step": 21920
    },
    {
      "entropy": 5.551020193099975,
      "epoch": 1.988842525399129,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046047145001380856,
      "loss": 4.9609,
      "mean_token_accuracy": 0.20929338186979293,
      "num_tokens": 45501514.0,
      "step": 21925
    },
    {
      "entropy": 5.570407772064209,
      "epoch": 1.9892960812772134,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046045313023069764,
      "loss": 4.9726,
      "mean_token_accuracy": 0.20886309146881105,
      "num_tokens": 45511727.0,
      "step": 21930
    },
    {
      "entropy": 5.60915699005127,
      "epoch": 1.9897496371552976,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004604348066123688,
      "loss": 5.0612,
      "mean_token_accuracy": 0.20543305873870848,
      "num_tokens": 45521807.0,
      "step": 21935
    },
    {
      "entropy": 5.625822973251343,
      "epoch": 1.9902031930333817,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004604164791592008,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19708407819271087,
      "num_tokens": 45532033.0,
      "step": 21940
    },
    {
      "entropy": 5.567453861236572,
      "epoch": 1.990656748911466,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046039814787157283,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20609915554523467,
      "num_tokens": 45542806.0,
      "step": 21945
    },
    {
      "entropy": 5.53896164894104,
      "epoch": 1.99111030478955,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004603798127498639,
      "loss": 4.89,
      "mean_token_accuracy": 0.22646187990903854,
      "num_tokens": 45551982.0,
      "step": 21950
    },
    {
      "entropy": 5.554815864562988,
      "epoch": 1.9915638606676342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004603614737944532,
      "loss": 5.002,
      "mean_token_accuracy": 0.20761286318302155,
      "num_tokens": 45561998.0,
      "step": 21955
    },
    {
      "entropy": 5.531924295425415,
      "epoch": 1.9920174165457185,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046034313100571994,
      "loss": 5.0012,
      "mean_token_accuracy": 0.20836948603391647,
      "num_tokens": 45572246.0,
      "step": 21960
    },
    {
      "entropy": 5.587581634521484,
      "epoch": 1.9924709724238028,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004603247843840436,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20864470452070236,
      "num_tokens": 45582251.0,
      "step": 21965
    },
    {
      "entropy": 5.560734987258911,
      "epoch": 1.9929245283018868,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004603064339298034,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20998208373785018,
      "num_tokens": 45592749.0,
      "step": 21970
    },
    {
      "entropy": 5.526675510406494,
      "epoch": 1.9933780841799709,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004602880796433789,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21031934171915054,
      "num_tokens": 45602412.0,
      "step": 21975
    },
    {
      "entropy": 5.597551918029785,
      "epoch": 1.9938316400580551,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046026972152514976,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20619491785764693,
      "num_tokens": 45612736.0,
      "step": 21980
    },
    {
      "entropy": 5.570060729980469,
      "epoch": 1.9942851959361394,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004602513595754956,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20696091502904893,
      "num_tokens": 45622952.0,
      "step": 21985
    },
    {
      "entropy": 5.615493488311768,
      "epoch": 1.9947387518142237,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046023299379479606,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20682632327079772,
      "num_tokens": 45633569.0,
      "step": 21990
    },
    {
      "entropy": 5.508515644073486,
      "epoch": 1.9951923076923077,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046021462418343104,
      "loss": 4.9777,
      "mean_token_accuracy": 0.21096506118774414,
      "num_tokens": 45644347.0,
      "step": 21995
    },
    {
      "entropy": 5.6098042011260985,
      "epoch": 1.9956458635703918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004601962507417804,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20653981119394302,
      "num_tokens": 45655268.0,
      "step": 22000
    },
    {
      "entropy": 5.545614004135132,
      "epoch": 1.996099419448476,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046017787347022403,
      "loss": 4.9611,
      "mean_token_accuracy": 0.20893973112106323,
      "num_tokens": 45664900.0,
      "step": 22005
    },
    {
      "entropy": 5.592525959014893,
      "epoch": 1.9965529753265603,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046015949236914217,
      "loss": 5.0392,
      "mean_token_accuracy": 0.21572151631116868,
      "num_tokens": 45675252.0,
      "step": 22010
    },
    {
      "entropy": 5.647371912002564,
      "epoch": 1.9970065312046446,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004601411074389147,
      "loss": 4.9595,
      "mean_token_accuracy": 0.2131752163171768,
      "num_tokens": 45686027.0,
      "step": 22015
    },
    {
      "entropy": 5.564004421234131,
      "epoch": 1.9974600870827286,
      "grad_norm": 1.125,
      "learning_rate": 0.000460122718679922,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20969758331775665,
      "num_tokens": 45696436.0,
      "step": 22020
    },
    {
      "entropy": 5.594499158859253,
      "epoch": 1.9979136429608126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004601043260925443,
      "loss": 5.0641,
      "mean_token_accuracy": 0.21029372215270997,
      "num_tokens": 45706509.0,
      "step": 22025
    },
    {
      "entropy": 5.629207944869995,
      "epoch": 1.998367198838897,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046008592967716197,
      "loss": 5.0873,
      "mean_token_accuracy": 0.20248838514089584,
      "num_tokens": 45717933.0,
      "step": 22030
    },
    {
      "entropy": 5.560246419906616,
      "epoch": 1.9988207547169812,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004600675294341554,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19791806489229202,
      "num_tokens": 45727966.0,
      "step": 22035
    },
    {
      "entropy": 5.554030799865723,
      "epoch": 1.9992743105950654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046004912536390526,
      "loss": 5.0008,
      "mean_token_accuracy": 0.20639860033988952,
      "num_tokens": 45737649.0,
      "step": 22040
    },
    {
      "entropy": 5.598674964904785,
      "epoch": 1.9997278664731495,
      "grad_norm": 1.15625,
      "learning_rate": 0.000460030717466792,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2102483779191971,
      "num_tokens": 45747561.0,
      "step": 22045
    },
    {
      "entropy": 5.536699819564819,
      "epoch": 2.0001814223512335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004600123057431963,
      "loss": 4.9111,
      "mean_token_accuracy": 0.21040636897087098,
      "num_tokens": 45758183.0,
      "step": 22050
    },
    {
      "entropy": 5.535328960418701,
      "epoch": 2.000634978229318,
      "grad_norm": 1.171875,
      "learning_rate": 0.000459993890193499,
      "loss": 4.8764,
      "mean_token_accuracy": 0.218867327272892,
      "num_tokens": 45768006.0,
      "step": 22055
    },
    {
      "entropy": 5.5945357322692875,
      "epoch": 2.001088534107402,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004599754708180809,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20445465445518493,
      "num_tokens": 45777981.0,
      "step": 22060
    },
    {
      "entropy": 5.567545747756958,
      "epoch": 2.0015420899854863,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000459957047617323,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2058892861008644,
      "num_tokens": 45789083.0,
      "step": 22065
    },
    {
      "entropy": 5.577370500564575,
      "epoch": 2.0019956458635706,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045993862059160617,
      "loss": 4.8965,
      "mean_token_accuracy": 0.20997656732797623,
      "num_tokens": 45799750.0,
      "step": 22070
    },
    {
      "entropy": 5.539130353927613,
      "epoch": 2.0024492017416544,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004599201897413116,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21443042457103728,
      "num_tokens": 45810321.0,
      "step": 22075
    },
    {
      "entropy": 5.4724390506744385,
      "epoch": 2.0029027576197387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045990175506682024,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22037743180990219,
      "num_tokens": 45820581.0,
      "step": 22080
    },
    {
      "entropy": 5.513353395462036,
      "epoch": 2.003356313497823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045988331656851355,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20550288558006286,
      "num_tokens": 45831504.0,
      "step": 22085
    },
    {
      "entropy": 5.612738418579101,
      "epoch": 2.003809869375907,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045986487424677284,
      "loss": 4.991,
      "mean_token_accuracy": 0.2069953367114067,
      "num_tokens": 45841100.0,
      "step": 22090
    },
    {
      "entropy": 5.600549650192261,
      "epoch": 2.0042634252539915,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004598464281019793,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21853204518556596,
      "num_tokens": 45851647.0,
      "step": 22095
    },
    {
      "entropy": 5.486761808395386,
      "epoch": 2.0047169811320753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045982797813451463,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22009832113981248,
      "num_tokens": 45862995.0,
      "step": 22100
    },
    {
      "entropy": 5.495511436462403,
      "epoch": 2.0051705370101596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045980952434476014,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21044859141111374,
      "num_tokens": 45872360.0,
      "step": 22105
    },
    {
      "entropy": 5.545560312271118,
      "epoch": 2.005624092888244,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045979106673309765,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21053716242313386,
      "num_tokens": 45883440.0,
      "step": 22110
    },
    {
      "entropy": 5.516869068145752,
      "epoch": 2.006077648766328,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045977260529990883,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20772037357091905,
      "num_tokens": 45894323.0,
      "step": 22115
    },
    {
      "entropy": 5.561671257019043,
      "epoch": 2.0065312046444124,
      "grad_norm": 1.125,
      "learning_rate": 0.0004597541400455755,
      "loss": 4.9718,
      "mean_token_accuracy": 0.20316219478845596,
      "num_tokens": 45904544.0,
      "step": 22120
    },
    {
      "entropy": 5.622522163391113,
      "epoch": 2.006984760522496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045973567097047933,
      "loss": 4.8478,
      "mean_token_accuracy": 0.23042139708995818,
      "num_tokens": 45913225.0,
      "step": 22125
    },
    {
      "entropy": 5.55759711265564,
      "epoch": 2.0074383164005805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045971719807500253,
      "loss": 4.9422,
      "mean_token_accuracy": 0.21341023445129395,
      "num_tokens": 45923306.0,
      "step": 22130
    },
    {
      "entropy": 5.517950534820557,
      "epoch": 2.0078918722786647,
      "grad_norm": 1.078125,
      "learning_rate": 0.000459698721359527,
      "loss": 4.921,
      "mean_token_accuracy": 0.2100364997982979,
      "num_tokens": 45933243.0,
      "step": 22135
    },
    {
      "entropy": 5.517447280883789,
      "epoch": 2.008345428156749,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004596802408244348,
      "loss": 4.9912,
      "mean_token_accuracy": 0.21013824194669722,
      "num_tokens": 45944498.0,
      "step": 22140
    },
    {
      "entropy": 5.495066976547241,
      "epoch": 2.0087989840348333,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004596617564701081,
      "loss": 4.8275,
      "mean_token_accuracy": 0.2200333520770073,
      "num_tokens": 45954550.0,
      "step": 22145
    },
    {
      "entropy": 5.603550148010254,
      "epoch": 2.009252539912917,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045964326829692925,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20771233290433883,
      "num_tokens": 45965425.0,
      "step": 22150
    },
    {
      "entropy": 5.553044176101684,
      "epoch": 2.0097060957910013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045962477630528064,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22131365537643433,
      "num_tokens": 45975637.0,
      "step": 22155
    },
    {
      "entropy": 5.54318413734436,
      "epoch": 2.0101596516690856,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045960628049554453,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2027548834681511,
      "num_tokens": 45985268.0,
      "step": 22160
    },
    {
      "entropy": 5.509607362747192,
      "epoch": 2.01061320754717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045958778086810354,
      "loss": 4.8561,
      "mean_token_accuracy": 0.21449377685785292,
      "num_tokens": 45995335.0,
      "step": 22165
    },
    {
      "entropy": 5.540615558624268,
      "epoch": 2.011066763425254,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004595692774233402,
      "loss": 4.9489,
      "mean_token_accuracy": 0.20536104440689087,
      "num_tokens": 46006037.0,
      "step": 22170
    },
    {
      "entropy": 5.570735120773316,
      "epoch": 2.011520319303338,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045955077016163713,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21552205085754395,
      "num_tokens": 46017016.0,
      "step": 22175
    },
    {
      "entropy": 5.614500570297241,
      "epoch": 2.0119738751814222,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045953225908337717,
      "loss": 5.0132,
      "mean_token_accuracy": 0.20326677411794664,
      "num_tokens": 46026797.0,
      "step": 22180
    },
    {
      "entropy": 5.510417938232422,
      "epoch": 2.0124274310595065,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045951374418894304,
      "loss": 4.848,
      "mean_token_accuracy": 0.2150076538324356,
      "num_tokens": 46037799.0,
      "step": 22185
    },
    {
      "entropy": 5.550413417816162,
      "epoch": 2.0128809869375908,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045949522547871765,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2123866006731987,
      "num_tokens": 46048617.0,
      "step": 22190
    },
    {
      "entropy": 5.554560995101928,
      "epoch": 2.013334542815675,
      "grad_norm": 1.0625,
      "learning_rate": 0.000459476702953084,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21581512242555617,
      "num_tokens": 46060122.0,
      "step": 22195
    },
    {
      "entropy": 5.525198078155517,
      "epoch": 2.013788098693759,
      "grad_norm": 1.046875,
      "learning_rate": 0.000459458176612425,
      "loss": 4.9455,
      "mean_token_accuracy": 0.21504294276237487,
      "num_tokens": 46071488.0,
      "step": 22200
    },
    {
      "entropy": 5.601229619979859,
      "epoch": 2.014241654571843,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045943964645712407,
      "loss": 4.9768,
      "mean_token_accuracy": 0.20384662449359894,
      "num_tokens": 46081954.0,
      "step": 22205
    },
    {
      "entropy": 5.561482715606689,
      "epoch": 2.0146952104499274,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004594211124875642,
      "loss": 4.8906,
      "mean_token_accuracy": 0.21491264104843139,
      "num_tokens": 46092488.0,
      "step": 22210
    },
    {
      "entropy": 5.512816715240478,
      "epoch": 2.0151487663280117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004594025747041287,
      "loss": 4.8885,
      "mean_token_accuracy": 0.22213494777679443,
      "num_tokens": 46102451.0,
      "step": 22215
    },
    {
      "entropy": 5.5255286693573,
      "epoch": 2.015602322206096,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000459384033107201,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21453291177749634,
      "num_tokens": 46113862.0,
      "step": 22220
    },
    {
      "entropy": 5.624523496627807,
      "epoch": 2.01605587808418,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004593654876971645,
      "loss": 5.0044,
      "mean_token_accuracy": 0.20698038190603257,
      "num_tokens": 46124517.0,
      "step": 22225
    },
    {
      "entropy": 5.494170188903809,
      "epoch": 2.016509433962264,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004593469384744027,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21207363903522491,
      "num_tokens": 46134221.0,
      "step": 22230
    },
    {
      "entropy": 5.55197434425354,
      "epoch": 2.0169629898403483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045932838543929924,
      "loss": 4.999,
      "mean_token_accuracy": 0.2088332384824753,
      "num_tokens": 46144512.0,
      "step": 22235
    },
    {
      "entropy": 5.592677450180053,
      "epoch": 2.0174165457184325,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004593098285922378,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21223630905151367,
      "num_tokens": 46154373.0,
      "step": 22240
    },
    {
      "entropy": 5.619633913040161,
      "epoch": 2.017870101596517,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004592912679336021,
      "loss": 5.021,
      "mean_token_accuracy": 0.20819658786058426,
      "num_tokens": 46164834.0,
      "step": 22245
    },
    {
      "entropy": 5.5458661079406735,
      "epoch": 2.018323657474601,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004592727034637761,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2142745301127434,
      "num_tokens": 46174815.0,
      "step": 22250
    },
    {
      "entropy": 5.522609376907349,
      "epoch": 2.018777213352685,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045925413518314363,
      "loss": 4.9128,
      "mean_token_accuracy": 0.2084043338894844,
      "num_tokens": 46185155.0,
      "step": 22255
    },
    {
      "entropy": 5.575503015518189,
      "epoch": 2.019230769230769,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004592355630920886,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21316245347261428,
      "num_tokens": 46194537.0,
      "step": 22260
    },
    {
      "entropy": 5.62617506980896,
      "epoch": 2.0196843251088534,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045921698719099527,
      "loss": 4.9839,
      "mean_token_accuracy": 0.20964009761810304,
      "num_tokens": 46205494.0,
      "step": 22265
    },
    {
      "entropy": 5.507011365890503,
      "epoch": 2.0201378809869377,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045919840748024764,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21459405422210692,
      "num_tokens": 46216399.0,
      "step": 22270
    },
    {
      "entropy": 5.4886754035949705,
      "epoch": 2.020591436865022,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045917982396023,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2169172540307045,
      "num_tokens": 46226515.0,
      "step": 22275
    },
    {
      "entropy": 5.583468389511109,
      "epoch": 2.021044992743106,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045916123663132676,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20524397492408752,
      "num_tokens": 46236688.0,
      "step": 22280
    },
    {
      "entropy": 5.549386024475098,
      "epoch": 2.02149854862119,
      "grad_norm": 1.125,
      "learning_rate": 0.00045914264549392215,
      "loss": 4.9033,
      "mean_token_accuracy": 0.2174192637205124,
      "num_tokens": 46247166.0,
      "step": 22285
    },
    {
      "entropy": 5.583451938629151,
      "epoch": 2.0219521044992743,
      "grad_norm": 1.125,
      "learning_rate": 0.0004591240505484007,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2175706371665001,
      "num_tokens": 46258902.0,
      "step": 22290
    },
    {
      "entropy": 5.581418418884278,
      "epoch": 2.0224056603773586,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045910545179514694,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20716777294874192,
      "num_tokens": 46269819.0,
      "step": 22295
    },
    {
      "entropy": 5.639160013198852,
      "epoch": 2.022859216255443,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004590868492345455,
      "loss": 4.9928,
      "mean_token_accuracy": 0.2072971522808075,
      "num_tokens": 46280715.0,
      "step": 22300
    },
    {
      "entropy": 5.602681875228882,
      "epoch": 2.0233127721335267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004590682428669813,
      "loss": 4.9922,
      "mean_token_accuracy": 0.2026600271463394,
      "num_tokens": 46291079.0,
      "step": 22305
    },
    {
      "entropy": 5.590440082550049,
      "epoch": 2.023766328011611,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004590496326928387,
      "loss": 4.9328,
      "mean_token_accuracy": 0.21221930533647537,
      "num_tokens": 46301362.0,
      "step": 22310
    },
    {
      "entropy": 5.5843523979187015,
      "epoch": 2.024219883889695,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045903101871250295,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21649962663650513,
      "num_tokens": 46311820.0,
      "step": 22315
    },
    {
      "entropy": 5.631623840332031,
      "epoch": 2.0246734397677795,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004590124009263587,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20982985496520995,
      "num_tokens": 46321992.0,
      "step": 22320
    },
    {
      "entropy": 5.550974369049072,
      "epoch": 2.0251269956458637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004589937793347912,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20854606479406357,
      "num_tokens": 46331537.0,
      "step": 22325
    },
    {
      "entropy": 5.574235677719116,
      "epoch": 2.0255805515239476,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045897515393818547,
      "loss": 4.9232,
      "mean_token_accuracy": 0.2099696949124336,
      "num_tokens": 46342214.0,
      "step": 22330
    },
    {
      "entropy": 5.5319356441497805,
      "epoch": 2.026034107402032,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004589565247369266,
      "loss": 4.9246,
      "mean_token_accuracy": 0.20641277283430098,
      "num_tokens": 46352471.0,
      "step": 22335
    },
    {
      "entropy": 5.530709600448608,
      "epoch": 2.026487663280116,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004589378917314,
      "loss": 4.949,
      "mean_token_accuracy": 0.21187900006771088,
      "num_tokens": 46363221.0,
      "step": 22340
    },
    {
      "entropy": 5.510725355148315,
      "epoch": 2.0269412191582004,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045891925492199087,
      "loss": 4.8199,
      "mean_token_accuracy": 0.22246098071336745,
      "num_tokens": 46373454.0,
      "step": 22345
    },
    {
      "entropy": 5.676409912109375,
      "epoch": 2.0273947750362846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004589006143090847,
      "loss": 5.0993,
      "mean_token_accuracy": 0.2050787091255188,
      "num_tokens": 46384403.0,
      "step": 22350
    },
    {
      "entropy": 5.518602800369263,
      "epoch": 2.0278483309143684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045888196989306697,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22396107017993927,
      "num_tokens": 46393362.0,
      "step": 22355
    },
    {
      "entropy": 5.625599002838134,
      "epoch": 2.0283018867924527,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004588633216743232,
      "loss": 5.0268,
      "mean_token_accuracy": 0.21074841171503067,
      "num_tokens": 46405100.0,
      "step": 22360
    },
    {
      "entropy": 5.484832143783569,
      "epoch": 2.028755442670537,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045884466965323914,
      "loss": 4.86,
      "mean_token_accuracy": 0.21490903645753862,
      "num_tokens": 46415058.0,
      "step": 22365
    },
    {
      "entropy": 5.550878620147705,
      "epoch": 2.0292089985486212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045882601383020045,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21324272751808165,
      "num_tokens": 46425863.0,
      "step": 22370
    },
    {
      "entropy": 5.562830114364624,
      "epoch": 2.0296625544267055,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004588073542055929,
      "loss": 4.9422,
      "mean_token_accuracy": 0.21125178188085555,
      "num_tokens": 46436172.0,
      "step": 22375
    },
    {
      "entropy": 5.621502637863159,
      "epoch": 2.0301161103047893,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045878869077980246,
      "loss": 4.9414,
      "mean_token_accuracy": 0.20946284085512162,
      "num_tokens": 46446717.0,
      "step": 22380
    },
    {
      "entropy": 5.596818637847901,
      "epoch": 2.0305696661828736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045877002355321504,
      "loss": 4.9694,
      "mean_token_accuracy": 0.20303172022104263,
      "num_tokens": 46456944.0,
      "step": 22385
    },
    {
      "entropy": 5.57465672492981,
      "epoch": 2.031023222060958,
      "grad_norm": 1.125,
      "learning_rate": 0.0004587513525262168,
      "loss": 5.0337,
      "mean_token_accuracy": 0.2074401006102562,
      "num_tokens": 46467231.0,
      "step": 22390
    },
    {
      "entropy": 5.548603296279907,
      "epoch": 2.031476777939042,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045873267769919364,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21052717566490173,
      "num_tokens": 46477107.0,
      "step": 22395
    },
    {
      "entropy": 5.531497287750244,
      "epoch": 2.0319303338171264,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004587139990725319,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21285859644412994,
      "num_tokens": 46487351.0,
      "step": 22400
    },
    {
      "entropy": 5.566560363769531,
      "epoch": 2.0323838896952107,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004586953166466179,
      "loss": 4.956,
      "mean_token_accuracy": 0.2106267601251602,
      "num_tokens": 46497685.0,
      "step": 22405
    },
    {
      "entropy": 5.556187582015991,
      "epoch": 2.0328374455732945,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004586766304218379,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20468230098485946,
      "num_tokens": 46507866.0,
      "step": 22410
    },
    {
      "entropy": 5.490482902526855,
      "epoch": 2.0332910014513788,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045865794039857846,
      "loss": 4.8451,
      "mean_token_accuracy": 0.21900141686201097,
      "num_tokens": 46518797.0,
      "step": 22415
    },
    {
      "entropy": 5.539396524429321,
      "epoch": 2.033744557329463,
      "grad_norm": 1.125,
      "learning_rate": 0.0004586392465772259,
      "loss": 4.906,
      "mean_token_accuracy": 0.21572514176368712,
      "num_tokens": 46528759.0,
      "step": 22420
    },
    {
      "entropy": 5.615487670898437,
      "epoch": 2.0341981132075473,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045862054895816703,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20383737683296205,
      "num_tokens": 46539014.0,
      "step": 22425
    },
    {
      "entropy": 5.518242406845093,
      "epoch": 2.0346516690856316,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004586018475417884,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21100321561098098,
      "num_tokens": 46548856.0,
      "step": 22430
    },
    {
      "entropy": 5.5146091938018795,
      "epoch": 2.0351052249637154,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004585831423284766,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2158732146024704,
      "num_tokens": 46559164.0,
      "step": 22435
    },
    {
      "entropy": 5.5795269966125485,
      "epoch": 2.0355587808417996,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004585644333186188,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20651134997606277,
      "num_tokens": 46569793.0,
      "step": 22440
    },
    {
      "entropy": 5.577578020095825,
      "epoch": 2.036012336719884,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045854572051260166,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21044086813926696,
      "num_tokens": 46580257.0,
      "step": 22445
    },
    {
      "entropy": 5.554258012771607,
      "epoch": 2.036465892597968,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045852700391081234,
      "loss": 4.9359,
      "mean_token_accuracy": 0.2066876173019409,
      "num_tokens": 46590297.0,
      "step": 22450
    },
    {
      "entropy": 5.534800577163696,
      "epoch": 2.0369194484760524,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004585082835136378,
      "loss": 4.8971,
      "mean_token_accuracy": 0.21658104062080383,
      "num_tokens": 46600647.0,
      "step": 22455
    },
    {
      "entropy": 5.5505164623260494,
      "epoch": 2.0373730043541363,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004584895593214652,
      "loss": 4.9035,
      "mean_token_accuracy": 0.20982402712106704,
      "num_tokens": 46610478.0,
      "step": 22460
    },
    {
      "entropy": 5.550243902206421,
      "epoch": 2.0378265602322205,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045847083133468164,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20916029214859008,
      "num_tokens": 46620159.0,
      "step": 22465
    },
    {
      "entropy": 5.575983047485352,
      "epoch": 2.038280116110305,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045845209955367464,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21891459673643113,
      "num_tokens": 46630555.0,
      "step": 22470
    },
    {
      "entropy": 5.559483385086059,
      "epoch": 2.038733671988389,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045843336397883146,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20618917942047119,
      "num_tokens": 46640188.0,
      "step": 22475
    },
    {
      "entropy": 5.5388209342956545,
      "epoch": 2.0391872278664733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004584146246105396,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2146706461906433,
      "num_tokens": 46650284.0,
      "step": 22480
    },
    {
      "entropy": 5.616133403778076,
      "epoch": 2.039640783744557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045839588144918643,
      "loss": 4.9239,
      "mean_token_accuracy": 0.213815775513649,
      "num_tokens": 46661466.0,
      "step": 22485
    },
    {
      "entropy": 5.587265300750732,
      "epoch": 2.0400943396226414,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004583771344951598,
      "loss": 4.9139,
      "mean_token_accuracy": 0.2136480137705803,
      "num_tokens": 46671683.0,
      "step": 22490
    },
    {
      "entropy": 5.5348492622375485,
      "epoch": 2.0405478955007257,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045835838374884733,
      "loss": 4.967,
      "mean_token_accuracy": 0.20732931941747665,
      "num_tokens": 46681787.0,
      "step": 22495
    },
    {
      "entropy": 5.57278995513916,
      "epoch": 2.04100145137881,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045833962921063665,
      "loss": 4.9065,
      "mean_token_accuracy": 0.21543323397636413,
      "num_tokens": 46692752.0,
      "step": 22500
    },
    {
      "entropy": 5.590698528289795,
      "epoch": 2.041455007256894,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045832087088091574,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2086937367916107,
      "num_tokens": 46702980.0,
      "step": 22505
    },
    {
      "entropy": 5.530666971206665,
      "epoch": 2.041908563134978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045830210876007257,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21580173969268798,
      "num_tokens": 46712281.0,
      "step": 22510
    },
    {
      "entropy": 5.538294315338135,
      "epoch": 2.0423621190130623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000458283342848495,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21018820703029634,
      "num_tokens": 46722097.0,
      "step": 22515
    },
    {
      "entropy": 5.492975091934204,
      "epoch": 2.0428156748911466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004582645731465713,
      "loss": 4.8293,
      "mean_token_accuracy": 0.21653183400630951,
      "num_tokens": 46733056.0,
      "step": 22520
    },
    {
      "entropy": 5.4882568359375,
      "epoch": 2.043269230769231,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045824579965468944,
      "loss": 4.884,
      "mean_token_accuracy": 0.21330133825540543,
      "num_tokens": 46742608.0,
      "step": 22525
    },
    {
      "entropy": 5.492119073867798,
      "epoch": 2.043722786647315,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045822702237323777,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22310970425605775,
      "num_tokens": 46751762.0,
      "step": 22530
    },
    {
      "entropy": 5.488946914672852,
      "epoch": 2.044176342525399,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045820824130260457,
      "loss": 4.8463,
      "mean_token_accuracy": 0.2245909482240677,
      "num_tokens": 46762252.0,
      "step": 22535
    },
    {
      "entropy": 5.533684301376343,
      "epoch": 2.044629898403483,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004581894564431783,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20734742134809495,
      "num_tokens": 46772172.0,
      "step": 22540
    },
    {
      "entropy": 5.528283929824829,
      "epoch": 2.0450834542815675,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004581706677953473,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21463873088359833,
      "num_tokens": 46782894.0,
      "step": 22545
    },
    {
      "entropy": 5.561836528778076,
      "epoch": 2.0455370101596517,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045815187535950034,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21210812479257585,
      "num_tokens": 46793097.0,
      "step": 22550
    },
    {
      "entropy": 5.611650180816651,
      "epoch": 2.045990566037736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045813307913602586,
      "loss": 5.0305,
      "mean_token_accuracy": 0.200439915060997,
      "num_tokens": 46804196.0,
      "step": 22555
    },
    {
      "entropy": 5.569736814498901,
      "epoch": 2.04644412191582,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004581142791253126,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2087482213973999,
      "num_tokens": 46813985.0,
      "step": 22560
    },
    {
      "entropy": 5.493589544296265,
      "epoch": 2.046897677793904,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045809547532774943,
      "loss": 4.8426,
      "mean_token_accuracy": 0.20691220760345458,
      "num_tokens": 46824125.0,
      "step": 22565
    },
    {
      "entropy": 5.514066123962403,
      "epoch": 2.0473512336719883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045807666774372514,
      "loss": 4.8747,
      "mean_token_accuracy": 0.2152392864227295,
      "num_tokens": 46834982.0,
      "step": 22570
    },
    {
      "entropy": 5.515033960342407,
      "epoch": 2.0478047895500726,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045805785637362874,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22229514569044112,
      "num_tokens": 46844669.0,
      "step": 22575
    },
    {
      "entropy": 5.5233604431152346,
      "epoch": 2.048258345428157,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004580390412178492,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21486603766679763,
      "num_tokens": 46855108.0,
      "step": 22580
    },
    {
      "entropy": 5.532042169570923,
      "epoch": 2.048711901306241,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045802022227677566,
      "loss": 4.885,
      "mean_token_accuracy": 0.2107235759496689,
      "num_tokens": 46864878.0,
      "step": 22585
    },
    {
      "entropy": 5.54194221496582,
      "epoch": 2.049165457184325,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045800139955079727,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2083554208278656,
      "num_tokens": 46874916.0,
      "step": 22590
    },
    {
      "entropy": 5.603750705718994,
      "epoch": 2.0496190130624092,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004579825730403033,
      "loss": 5.0055,
      "mean_token_accuracy": 0.20589129775762557,
      "num_tokens": 46885167.0,
      "step": 22595
    },
    {
      "entropy": 5.590600776672363,
      "epoch": 2.0500725689404935,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004579637427456831,
      "loss": 4.9481,
      "mean_token_accuracy": 0.20603059381246566,
      "num_tokens": 46895526.0,
      "step": 22600
    },
    {
      "entropy": 5.523321104049683,
      "epoch": 2.0505261248185778,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045794490866732607,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21477593183517457,
      "num_tokens": 46905426.0,
      "step": 22605
    },
    {
      "entropy": 5.528275632858277,
      "epoch": 2.050979680696662,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004579260708056217,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21035714596509933,
      "num_tokens": 46915728.0,
      "step": 22610
    },
    {
      "entropy": 5.515733766555786,
      "epoch": 2.051433236574746,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004579072291609596,
      "loss": 4.8976,
      "mean_token_accuracy": 0.211222904920578,
      "num_tokens": 46925972.0,
      "step": 22615
    },
    {
      "entropy": 5.529329919815064,
      "epoch": 2.05188679245283,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045788838373372935,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2146576702594757,
      "num_tokens": 46937108.0,
      "step": 22620
    },
    {
      "entropy": 5.509870147705078,
      "epoch": 2.0523403483309144,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045786953452432076,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21134349703788757,
      "num_tokens": 46947684.0,
      "step": 22625
    },
    {
      "entropy": 5.615060329437256,
      "epoch": 2.0527939042089987,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004578506815331236,
      "loss": 4.999,
      "mean_token_accuracy": 0.20833959579467773,
      "num_tokens": 46958089.0,
      "step": 22630
    },
    {
      "entropy": 5.541087007522583,
      "epoch": 2.053247460087083,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004578318247605277,
      "loss": 4.8883,
      "mean_token_accuracy": 0.21928365379571915,
      "num_tokens": 46968618.0,
      "step": 22635
    },
    {
      "entropy": 5.5500030517578125,
      "epoch": 2.0537010159651667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004578129642069231,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21412350684404374,
      "num_tokens": 46978791.0,
      "step": 22640
    },
    {
      "entropy": 5.551610231399536,
      "epoch": 2.054154571843251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004577940998726998,
      "loss": 4.8697,
      "mean_token_accuracy": 0.21991531401872635,
      "num_tokens": 46988706.0,
      "step": 22645
    },
    {
      "entropy": 5.626473760604858,
      "epoch": 2.0546081277213353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045777523175824796,
      "loss": 5.0538,
      "mean_token_accuracy": 0.19763527065515518,
      "num_tokens": 47000140.0,
      "step": 22650
    },
    {
      "entropy": 5.540607929229736,
      "epoch": 2.0550616835994195,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004577563598639577,
      "loss": 4.8813,
      "mean_token_accuracy": 0.21221189200878143,
      "num_tokens": 47009773.0,
      "step": 22655
    },
    {
      "entropy": 5.593835735321045,
      "epoch": 2.055515239477504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045773748419021944,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20747489333152772,
      "num_tokens": 47020326.0,
      "step": 22660
    },
    {
      "entropy": 5.50357141494751,
      "epoch": 2.0559687953555876,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045771860473742347,
      "loss": 4.898,
      "mean_token_accuracy": 0.20945340842008592,
      "num_tokens": 47030410.0,
      "step": 22665
    },
    {
      "entropy": 5.569175815582275,
      "epoch": 2.056422351233672,
      "grad_norm": 1.125,
      "learning_rate": 0.00045769972150596006,
      "loss": 4.9419,
      "mean_token_accuracy": 0.211944343149662,
      "num_tokens": 47040546.0,
      "step": 22670
    },
    {
      "entropy": 5.5609148979187015,
      "epoch": 2.056875907111756,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045768083449621987,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22088221311569214,
      "num_tokens": 47050210.0,
      "step": 22675
    },
    {
      "entropy": 5.6539835929870605,
      "epoch": 2.0573294629898404,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045766194370859347,
      "loss": 5.0634,
      "mean_token_accuracy": 0.19811297506093978,
      "num_tokens": 47060641.0,
      "step": 22680
    },
    {
      "entropy": 5.566586685180664,
      "epoch": 2.0577830188679247,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004576430491434716,
      "loss": 4.9672,
      "mean_token_accuracy": 0.2138103500008583,
      "num_tokens": 47070737.0,
      "step": 22685
    },
    {
      "entropy": 5.616622066497802,
      "epoch": 2.0582365747460085,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045762415080124487,
      "loss": 4.9319,
      "mean_token_accuracy": 0.21003361493349076,
      "num_tokens": 47080884.0,
      "step": 22690
    },
    {
      "entropy": 5.548874759674073,
      "epoch": 2.058690130624093,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045760524868230417,
      "loss": 4.9835,
      "mean_token_accuracy": 0.20848450362682341,
      "num_tokens": 47090865.0,
      "step": 22695
    },
    {
      "entropy": 5.560907363891602,
      "epoch": 2.059143686502177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004575863427870404,
      "loss": 4.9163,
      "mean_token_accuracy": 0.21426442712545396,
      "num_tokens": 47100570.0,
      "step": 22700
    },
    {
      "entropy": 5.5511678695678714,
      "epoch": 2.0595972423802613,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004575674331158446,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21494615375995635,
      "num_tokens": 47111199.0,
      "step": 22705
    },
    {
      "entropy": 5.543911790847778,
      "epoch": 2.0600507982583456,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004575485196691076,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21214178055524827,
      "num_tokens": 47121191.0,
      "step": 22710
    },
    {
      "entropy": 5.604065036773681,
      "epoch": 2.0605043541364294,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004575296024472209,
      "loss": 4.975,
      "mean_token_accuracy": 0.21015238910913467,
      "num_tokens": 47131608.0,
      "step": 22715
    },
    {
      "entropy": 5.618283414840699,
      "epoch": 2.0609579100145137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004575106814505753,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20580857396125793,
      "num_tokens": 47141492.0,
      "step": 22720
    },
    {
      "entropy": 5.503385400772094,
      "epoch": 2.061411465892598,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004574917566795624,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2223336100578308,
      "num_tokens": 47152576.0,
      "step": 22725
    },
    {
      "entropy": 5.550375938415527,
      "epoch": 2.061865021770682,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045747282813457345,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20989041775465012,
      "num_tokens": 47163863.0,
      "step": 22730
    },
    {
      "entropy": 5.534601259231567,
      "epoch": 2.0623185776487665,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000457453895816,
      "loss": 4.9076,
      "mean_token_accuracy": 0.2132399782538414,
      "num_tokens": 47174140.0,
      "step": 22735
    },
    {
      "entropy": 5.581269454956055,
      "epoch": 2.0627721335268503,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045743495972423346,
      "loss": 4.9705,
      "mean_token_accuracy": 0.20813938081264496,
      "num_tokens": 47184708.0,
      "step": 22740
    },
    {
      "entropy": 5.5300328731536865,
      "epoch": 2.0632256894049346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004574160198596653,
      "loss": 4.93,
      "mean_token_accuracy": 0.2126962959766388,
      "num_tokens": 47196286.0,
      "step": 22745
    },
    {
      "entropy": 5.450542259216308,
      "epoch": 2.063679245283019,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045739707622268757,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22261857241392136,
      "num_tokens": 47206774.0,
      "step": 22750
    },
    {
      "entropy": 5.497993469238281,
      "epoch": 2.064132801161103,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004573781288136917,
      "loss": 4.8038,
      "mean_token_accuracy": 0.21777256727218627,
      "num_tokens": 47217009.0,
      "step": 22755
    },
    {
      "entropy": 5.552196025848389,
      "epoch": 2.0645863570391874,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045735917763306967,
      "loss": 4.944,
      "mean_token_accuracy": 0.21327149868011475,
      "num_tokens": 47226718.0,
      "step": 22760
    },
    {
      "entropy": 5.496644496917725,
      "epoch": 2.065039912917271,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004573402226812134,
      "loss": 4.826,
      "mean_token_accuracy": 0.2083118140697479,
      "num_tokens": 47237666.0,
      "step": 22765
    },
    {
      "entropy": 5.5309202671051025,
      "epoch": 2.0654934687953554,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045732126395851483,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20627781599760056,
      "num_tokens": 47249230.0,
      "step": 22770
    },
    {
      "entropy": 5.517955160140991,
      "epoch": 2.0659470246734397,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004573023014653661,
      "loss": 4.885,
      "mean_token_accuracy": 0.2155887544155121,
      "num_tokens": 47259232.0,
      "step": 22775
    },
    {
      "entropy": 5.521569013595581,
      "epoch": 2.066400580551524,
      "grad_norm": 1.125,
      "learning_rate": 0.00045728333520215933,
      "loss": 4.939,
      "mean_token_accuracy": 0.2105821266770363,
      "num_tokens": 47269006.0,
      "step": 22780
    },
    {
      "entropy": 5.5902228355407715,
      "epoch": 2.0668541364296082,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004572643651692867,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20120707750320435,
      "num_tokens": 47280048.0,
      "step": 22785
    },
    {
      "entropy": 5.550869750976562,
      "epoch": 2.0673076923076925,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045724539136714063,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2146774187684059,
      "num_tokens": 47290019.0,
      "step": 22790
    },
    {
      "entropy": 5.550041007995605,
      "epoch": 2.0677612481857763,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004572264137961133,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21964723169803618,
      "num_tokens": 47300573.0,
      "step": 22795
    },
    {
      "entropy": 5.49859585762024,
      "epoch": 2.0682148040638606,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045720743245659734,
      "loss": 4.905,
      "mean_token_accuracy": 0.21301880180835725,
      "num_tokens": 47311795.0,
      "step": 22800
    },
    {
      "entropy": 5.470848321914673,
      "epoch": 2.068668359941945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004571884473489853,
      "loss": 4.8148,
      "mean_token_accuracy": 0.2240484967827797,
      "num_tokens": 47321768.0,
      "step": 22805
    },
    {
      "entropy": 5.553780889511108,
      "epoch": 2.069121915820029,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004571694584736696,
      "loss": 4.9909,
      "mean_token_accuracy": 0.2161708205938339,
      "num_tokens": 47332662.0,
      "step": 22810
    },
    {
      "entropy": 5.542087602615356,
      "epoch": 2.0695754716981134,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004571504658310432,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2147950202226639,
      "num_tokens": 47344355.0,
      "step": 22815
    },
    {
      "entropy": 5.545966148376465,
      "epoch": 2.0700290275761972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004571314694214987,
      "loss": 4.8736,
      "mean_token_accuracy": 0.21230216473340988,
      "num_tokens": 47354228.0,
      "step": 22820
    },
    {
      "entropy": 5.560914754867554,
      "epoch": 2.0704825834542815,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004571124692454291,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22051794081926346,
      "num_tokens": 47364477.0,
      "step": 22825
    },
    {
      "entropy": 5.556259107589722,
      "epoch": 2.0709361393323658,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045709346530322715,
      "loss": 4.9892,
      "mean_token_accuracy": 0.21238750964403152,
      "num_tokens": 47375745.0,
      "step": 22830
    },
    {
      "entropy": 5.588777732849121,
      "epoch": 2.07138969521045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004570744575952859,
      "loss": 5.0073,
      "mean_token_accuracy": 0.21455432325601578,
      "num_tokens": 47385341.0,
      "step": 22835
    },
    {
      "entropy": 5.580568599700928,
      "epoch": 2.0718432510885343,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004570554461219985,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21542086750268935,
      "num_tokens": 47396266.0,
      "step": 22840
    },
    {
      "entropy": 5.57943058013916,
      "epoch": 2.072296806966618,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045703643088375806,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20775123983621596,
      "num_tokens": 47406772.0,
      "step": 22845
    },
    {
      "entropy": 5.596522045135498,
      "epoch": 2.0727503628447024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045701741188095793,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20698040276765822,
      "num_tokens": 47416833.0,
      "step": 22850
    },
    {
      "entropy": 5.519446086883545,
      "epoch": 2.0732039187227866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004569983891139912,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21298438608646392,
      "num_tokens": 47426954.0,
      "step": 22855
    },
    {
      "entropy": 5.498295164108276,
      "epoch": 2.073657474600871,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045697936258325155,
      "loss": 4.869,
      "mean_token_accuracy": 0.22404805421829224,
      "num_tokens": 47437520.0,
      "step": 22860
    },
    {
      "entropy": 5.554725551605225,
      "epoch": 2.074111030478955,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045696033228913226,
      "loss": 4.8891,
      "mean_token_accuracy": 0.20843892097473143,
      "num_tokens": 47447977.0,
      "step": 22865
    },
    {
      "entropy": 5.582836627960205,
      "epoch": 2.074564586357039,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045694129823202684,
      "loss": 4.978,
      "mean_token_accuracy": 0.20482730567455293,
      "num_tokens": 47458458.0,
      "step": 22870
    },
    {
      "entropy": 5.559929847717285,
      "epoch": 2.0750181422351233,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004569222604123291,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2129221186041832,
      "num_tokens": 47469048.0,
      "step": 22875
    },
    {
      "entropy": 5.6153302669525145,
      "epoch": 2.0754716981132075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004569032188304326,
      "loss": 4.9845,
      "mean_token_accuracy": 0.20533379912376404,
      "num_tokens": 47480353.0,
      "step": 22880
    },
    {
      "entropy": 5.511811351776123,
      "epoch": 2.075925253991292,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004568841734867313,
      "loss": 4.8144,
      "mean_token_accuracy": 0.21520548313856125,
      "num_tokens": 47489951.0,
      "step": 22885
    },
    {
      "entropy": 5.480315065383911,
      "epoch": 2.076378809869376,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004568651243816188,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21955198049545288,
      "num_tokens": 47499015.0,
      "step": 22890
    },
    {
      "entropy": 5.561438846588135,
      "epoch": 2.07683236574746,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045684607151548923,
      "loss": 4.9169,
      "mean_token_accuracy": 0.21274450570344924,
      "num_tokens": 47508694.0,
      "step": 22895
    },
    {
      "entropy": 5.6556556224823,
      "epoch": 2.077285921625544,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045682701488873655,
      "loss": 5.0717,
      "mean_token_accuracy": 0.2010905608534813,
      "num_tokens": 47521336.0,
      "step": 22900
    },
    {
      "entropy": 5.606842470169068,
      "epoch": 2.0777394775036284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004568079545017549,
      "loss": 5.0475,
      "mean_token_accuracy": 0.20524126589298247,
      "num_tokens": 47531857.0,
      "step": 22905
    },
    {
      "entropy": 5.6149121761322025,
      "epoch": 2.0781930333817127,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004567888903549385,
      "loss": 4.9674,
      "mean_token_accuracy": 0.2053648605942726,
      "num_tokens": 47542224.0,
      "step": 22910
    },
    {
      "entropy": 5.579704284667969,
      "epoch": 2.078646589259797,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004567698224486813,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21154584139585494,
      "num_tokens": 47552370.0,
      "step": 22915
    },
    {
      "entropy": 5.645527410507202,
      "epoch": 2.0791001451378808,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045675075078337803,
      "loss": 4.904,
      "mean_token_accuracy": 0.2167135238647461,
      "num_tokens": 47561997.0,
      "step": 22920
    },
    {
      "entropy": 5.5685585021972654,
      "epoch": 2.079553701015965,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004567316753594229,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21692643761634828,
      "num_tokens": 47571968.0,
      "step": 22925
    },
    {
      "entropy": 5.562157154083252,
      "epoch": 2.0800072568940493,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045671259617721037,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20979191064834596,
      "num_tokens": 47583459.0,
      "step": 22930
    },
    {
      "entropy": 5.519590950012207,
      "epoch": 2.0804608127721336,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000456693513237135,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2219153270125389,
      "num_tokens": 47594026.0,
      "step": 22935
    },
    {
      "entropy": 5.617216777801514,
      "epoch": 2.080914368650218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004566744265395916,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21273531317710875,
      "num_tokens": 47605246.0,
      "step": 22940
    },
    {
      "entropy": 5.607788515090943,
      "epoch": 2.081367924528302,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045665533608497467,
      "loss": 5.0235,
      "mean_token_accuracy": 0.19994515925645828,
      "num_tokens": 47616702.0,
      "step": 22945
    },
    {
      "entropy": 5.601177740097046,
      "epoch": 2.081821480406386,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045663624187367914,
      "loss": 4.9846,
      "mean_token_accuracy": 0.2063418596982956,
      "num_tokens": 47627616.0,
      "step": 22950
    },
    {
      "entropy": 5.563710880279541,
      "epoch": 2.08227503628447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004566171439060998,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21134505420923233,
      "num_tokens": 47637287.0,
      "step": 22955
    },
    {
      "entropy": 5.487324285507202,
      "epoch": 2.0827285921625545,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004565980421826317,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21375772953033448,
      "num_tokens": 47647474.0,
      "step": 22960
    },
    {
      "entropy": 5.589184713363648,
      "epoch": 2.0831821480406387,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004565789367036697,
      "loss": 4.9651,
      "mean_token_accuracy": 0.21286556273698806,
      "num_tokens": 47657036.0,
      "step": 22965
    },
    {
      "entropy": 5.522977304458618,
      "epoch": 2.083635703918723,
      "grad_norm": 1.125,
      "learning_rate": 0.0004565598274696091,
      "loss": 4.8472,
      "mean_token_accuracy": 0.21326051652431488,
      "num_tokens": 47668026.0,
      "step": 22970
    },
    {
      "entropy": 5.616479015350341,
      "epoch": 2.084089259796807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045654071448084506,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20693727433681489,
      "num_tokens": 47679508.0,
      "step": 22975
    },
    {
      "entropy": 5.580051517486572,
      "epoch": 2.084542815674891,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045652159773777273,
      "loss": 4.967,
      "mean_token_accuracy": 0.21123090833425523,
      "num_tokens": 47688474.0,
      "step": 22980
    },
    {
      "entropy": 5.688438034057617,
      "epoch": 2.0849963715529753,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004565024772407875,
      "loss": 5.094,
      "mean_token_accuracy": 0.19593742191791536,
      "num_tokens": 47698675.0,
      "step": 22985
    },
    {
      "entropy": 5.52439513206482,
      "epoch": 2.0854499274310596,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004564833529902848,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22284892052412034,
      "num_tokens": 47709214.0,
      "step": 22990
    },
    {
      "entropy": 5.5863415718078615,
      "epoch": 2.085903483309144,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045646422498666,
      "loss": 5.0213,
      "mean_token_accuracy": 0.21298460066318511,
      "num_tokens": 47721000.0,
      "step": 22995
    },
    {
      "entropy": 5.508289813995361,
      "epoch": 2.0863570391872277,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045644509323030893,
      "loss": 4.9241,
      "mean_token_accuracy": 0.21045969277620316,
      "num_tokens": 47731218.0,
      "step": 23000
    },
    {
      "entropy": 5.584042310714722,
      "epoch": 2.086810595065312,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045642595772162704,
      "loss": 4.918,
      "mean_token_accuracy": 0.21153358221054078,
      "num_tokens": 47742339.0,
      "step": 23005
    },
    {
      "entropy": 5.617961692810058,
      "epoch": 2.0872641509433962,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045640681846101014,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21020923554897308,
      "num_tokens": 47752858.0,
      "step": 23010
    },
    {
      "entropy": 5.517202377319336,
      "epoch": 2.0877177068214805,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004563876754488539,
      "loss": 4.8849,
      "mean_token_accuracy": 0.21729450225830077,
      "num_tokens": 47763303.0,
      "step": 23015
    },
    {
      "entropy": 5.56376805305481,
      "epoch": 2.0881712626995648,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004563685286855544,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2116104394197464,
      "num_tokens": 47773212.0,
      "step": 23020
    },
    {
      "entropy": 5.612044334411621,
      "epoch": 2.0886248185776486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004563493781715075,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20824681967496872,
      "num_tokens": 47783376.0,
      "step": 23025
    },
    {
      "entropy": 5.500726747512817,
      "epoch": 2.089078374455733,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004563302239071093,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21655123680830002,
      "num_tokens": 47794859.0,
      "step": 23030
    },
    {
      "entropy": 5.621681451797485,
      "epoch": 2.089531930333817,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045631106589275576,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21255283206701278,
      "num_tokens": 47804721.0,
      "step": 23035
    },
    {
      "entropy": 5.539343643188476,
      "epoch": 2.0899854862119014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045629190412884324,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22014436423778533,
      "num_tokens": 47814701.0,
      "step": 23040
    },
    {
      "entropy": 5.550324535369873,
      "epoch": 2.0904390420899857,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045627273861576786,
      "loss": 4.897,
      "mean_token_accuracy": 0.214018777012825,
      "num_tokens": 47823941.0,
      "step": 23045
    },
    {
      "entropy": 5.532428026199341,
      "epoch": 2.0908925979680695,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045625356935392614,
      "loss": 4.9408,
      "mean_token_accuracy": 0.21294294595718383,
      "num_tokens": 47835793.0,
      "step": 23050
    },
    {
      "entropy": 5.62921404838562,
      "epoch": 2.0913461538461537,
      "grad_norm": 1.125,
      "learning_rate": 0.0004562343963437144,
      "loss": 4.989,
      "mean_token_accuracy": 0.21092495769262315,
      "num_tokens": 47845677.0,
      "step": 23055
    },
    {
      "entropy": 5.574399471282959,
      "epoch": 2.091799709724238,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004562152195855291,
      "loss": 4.8915,
      "mean_token_accuracy": 0.22070108354091644,
      "num_tokens": 47855611.0,
      "step": 23060
    },
    {
      "entropy": 5.498503684997559,
      "epoch": 2.0922532656023223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004561960390797669,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21368443369865417,
      "num_tokens": 47865115.0,
      "step": 23065
    },
    {
      "entropy": 5.627487516403198,
      "epoch": 2.0927068214804065,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045617685482682454,
      "loss": 5.0087,
      "mean_token_accuracy": 0.2100916847586632,
      "num_tokens": 47876690.0,
      "step": 23070
    },
    {
      "entropy": 5.541545295715332,
      "epoch": 2.0931603773584904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004561576668270986,
      "loss": 4.9537,
      "mean_token_accuracy": 0.2084503874182701,
      "num_tokens": 47887709.0,
      "step": 23075
    },
    {
      "entropy": 5.53898401260376,
      "epoch": 2.0936139332365746,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045613847508098587,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21091461032629014,
      "num_tokens": 47898209.0,
      "step": 23080
    },
    {
      "entropy": 5.537871313095093,
      "epoch": 2.094067489114659,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045611927958888335,
      "loss": 4.8277,
      "mean_token_accuracy": 0.2170143648982048,
      "num_tokens": 47907494.0,
      "step": 23085
    },
    {
      "entropy": 5.56508584022522,
      "epoch": 2.094521044992743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045610008035118797,
      "loss": 4.9727,
      "mean_token_accuracy": 0.20951011329889296,
      "num_tokens": 47916923.0,
      "step": 23090
    },
    {
      "entropy": 5.49456295967102,
      "epoch": 2.0949746008708274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045608087736829687,
      "loss": 4.8571,
      "mean_token_accuracy": 0.21550212651491166,
      "num_tokens": 47927043.0,
      "step": 23095
    },
    {
      "entropy": 5.5701109886169435,
      "epoch": 2.0954281567489113,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000456061670640607,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20838372707366942,
      "num_tokens": 47937296.0,
      "step": 23100
    },
    {
      "entropy": 5.561228179931641,
      "epoch": 2.0958817126269955,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004560424601685157,
      "loss": 4.963,
      "mean_token_accuracy": 0.21140010803937911,
      "num_tokens": 47948268.0,
      "step": 23105
    },
    {
      "entropy": 5.587416791915894,
      "epoch": 2.09633526850508,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004560232459524201,
      "loss": 4.8873,
      "mean_token_accuracy": 0.20976995974779128,
      "num_tokens": 47958375.0,
      "step": 23110
    },
    {
      "entropy": 5.544656419754029,
      "epoch": 2.096788824383164,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045600402799271766,
      "loss": 4.9835,
      "mean_token_accuracy": 0.21522742062807082,
      "num_tokens": 47968469.0,
      "step": 23115
    },
    {
      "entropy": 5.519022512435913,
      "epoch": 2.0972423802612483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004559848062898058,
      "loss": 4.8586,
      "mean_token_accuracy": 0.2184354066848755,
      "num_tokens": 47979630.0,
      "step": 23120
    },
    {
      "entropy": 5.56188588142395,
      "epoch": 2.097695936139332,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004559655808440821,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2151502102613449,
      "num_tokens": 47989828.0,
      "step": 23125
    },
    {
      "entropy": 5.618375301361084,
      "epoch": 2.0981494920174164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045594635165594404,
      "loss": 5.048,
      "mean_token_accuracy": 0.20327326655387878,
      "num_tokens": 47999219.0,
      "step": 23130
    },
    {
      "entropy": 5.575772428512574,
      "epoch": 2.0986030478955007,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045592711872578925,
      "loss": 4.916,
      "mean_token_accuracy": 0.2209403082728386,
      "num_tokens": 48009244.0,
      "step": 23135
    },
    {
      "entropy": 5.583085727691651,
      "epoch": 2.099056603773585,
      "grad_norm": 1.125,
      "learning_rate": 0.00045590788205401557,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21605250835418702,
      "num_tokens": 48019066.0,
      "step": 23140
    },
    {
      "entropy": 5.491763639450073,
      "epoch": 2.099510159651669,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045588864164102086,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21540116965770723,
      "num_tokens": 48028704.0,
      "step": 23145
    },
    {
      "entropy": 5.520301723480225,
      "epoch": 2.0999637155297535,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045586939748720284,
      "loss": 4.9058,
      "mean_token_accuracy": 0.215412475168705,
      "num_tokens": 48039210.0,
      "step": 23150
    },
    {
      "entropy": 5.631437540054321,
      "epoch": 2.1004172714078373,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045585014959295965,
      "loss": 5.0383,
      "mean_token_accuracy": 0.206656414270401,
      "num_tokens": 48048446.0,
      "step": 23155
    },
    {
      "entropy": 5.579017972946167,
      "epoch": 2.1008708272859216,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045583089795868934,
      "loss": 4.935,
      "mean_token_accuracy": 0.2172452911734581,
      "num_tokens": 48058880.0,
      "step": 23160
    },
    {
      "entropy": 5.5533143997192385,
      "epoch": 2.101324383164006,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004558116425847899,
      "loss": 4.9833,
      "mean_token_accuracy": 0.21692482084035875,
      "num_tokens": 48070224.0,
      "step": 23165
    },
    {
      "entropy": 5.5385700225830075,
      "epoch": 2.10177793904209,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004557923834716596,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20600900650024415,
      "num_tokens": 48080914.0,
      "step": 23170
    },
    {
      "entropy": 5.532124996185303,
      "epoch": 2.1022314949201744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045577312061969677,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20475602895021439,
      "num_tokens": 48090873.0,
      "step": 23175
    },
    {
      "entropy": 5.679980802536011,
      "epoch": 2.102685050798258,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004557538540292997,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2085855036973953,
      "num_tokens": 48101885.0,
      "step": 23180
    },
    {
      "entropy": 5.607133913040161,
      "epoch": 2.1031386066763424,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004557345837008669,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20910215377807617,
      "num_tokens": 48111472.0,
      "step": 23185
    },
    {
      "entropy": 5.484545183181763,
      "epoch": 2.1035921625544267,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045571530963479684,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22218698263168335,
      "num_tokens": 48121351.0,
      "step": 23190
    },
    {
      "entropy": 5.587153577804566,
      "epoch": 2.104045718432511,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045569603183148814,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2215775027871132,
      "num_tokens": 48131104.0,
      "step": 23195
    },
    {
      "entropy": 5.60333948135376,
      "epoch": 2.1044992743105952,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004556767502913395,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20779150277376174,
      "num_tokens": 48142895.0,
      "step": 23200
    },
    {
      "entropy": 5.4921972274780275,
      "epoch": 2.104952830188679,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045565746501474946,
      "loss": 4.8109,
      "mean_token_accuracy": 0.21743903458118438,
      "num_tokens": 48153294.0,
      "step": 23205
    },
    {
      "entropy": 5.5781573295593265,
      "epoch": 2.1054063860667633,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045563817600211714,
      "loss": 4.9845,
      "mean_token_accuracy": 0.206672765314579,
      "num_tokens": 48164319.0,
      "step": 23210
    },
    {
      "entropy": 5.580169200897217,
      "epoch": 2.1058599419448476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045561888325384124,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21305814683437346,
      "num_tokens": 48173918.0,
      "step": 23215
    },
    {
      "entropy": 5.615588712692261,
      "epoch": 2.106313497822932,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004555995867703209,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2084985077381134,
      "num_tokens": 48184127.0,
      "step": 23220
    },
    {
      "entropy": 5.520053958892822,
      "epoch": 2.106767053701016,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045558028655195504,
      "loss": 4.8951,
      "mean_token_accuracy": 0.21235456466674804,
      "num_tokens": 48194198.0,
      "step": 23225
    },
    {
      "entropy": 5.468355131149292,
      "epoch": 2.1072206095791,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045556098259914293,
      "loss": 4.9041,
      "mean_token_accuracy": 0.21232093423604964,
      "num_tokens": 48204654.0,
      "step": 23230
    },
    {
      "entropy": 5.588650321960449,
      "epoch": 2.1076741654571842,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045554167491228357,
      "loss": 5.0538,
      "mean_token_accuracy": 0.19435366839170456,
      "num_tokens": 48215322.0,
      "step": 23235
    },
    {
      "entropy": 5.589576196670532,
      "epoch": 2.1081277213352685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045552236349177637,
      "loss": 4.8267,
      "mean_token_accuracy": 0.22067209631204604,
      "num_tokens": 48226142.0,
      "step": 23240
    },
    {
      "entropy": 5.512128734588623,
      "epoch": 2.1085812772133528,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045550304833802066,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21464934051036835,
      "num_tokens": 48236896.0,
      "step": 23245
    },
    {
      "entropy": 5.46775712966919,
      "epoch": 2.109034833091437,
      "grad_norm": 1.09375,
      "learning_rate": 0.000455483729451416,
      "loss": 4.8861,
      "mean_token_accuracy": 0.21393498480319978,
      "num_tokens": 48247425.0,
      "step": 23250
    },
    {
      "entropy": 5.542385149002075,
      "epoch": 2.109488388969521,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045546440683236173,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20822241604328157,
      "num_tokens": 48258560.0,
      "step": 23255
    },
    {
      "entropy": 5.650876474380493,
      "epoch": 2.109941944847605,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045544508048125763,
      "loss": 5.0357,
      "mean_token_accuracy": 0.2067191407084465,
      "num_tokens": 48269469.0,
      "step": 23260
    },
    {
      "entropy": 5.60484209060669,
      "epoch": 2.1103955007256894,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004554257503985032,
      "loss": 4.9354,
      "mean_token_accuracy": 0.2131854325532913,
      "num_tokens": 48278847.0,
      "step": 23265
    },
    {
      "entropy": 5.486921310424805,
      "epoch": 2.1108490566037736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045540641658449834,
      "loss": 4.8604,
      "mean_token_accuracy": 0.21766035705804826,
      "num_tokens": 48288091.0,
      "step": 23270
    },
    {
      "entropy": 5.487744569778442,
      "epoch": 2.111302612481858,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045538707903964275,
      "loss": 4.8656,
      "mean_token_accuracy": 0.21312690526247025,
      "num_tokens": 48297946.0,
      "step": 23275
    },
    {
      "entropy": 5.534839630126953,
      "epoch": 2.1117561683599417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045536773776433647,
      "loss": 4.9356,
      "mean_token_accuracy": 0.21512040346860886,
      "num_tokens": 48307650.0,
      "step": 23280
    },
    {
      "entropy": 5.582086658477783,
      "epoch": 2.112209724238026,
      "grad_norm": 1.125,
      "learning_rate": 0.00045534839275897937,
      "loss": 5.0283,
      "mean_token_accuracy": 0.20173343420028686,
      "num_tokens": 48318824.0,
      "step": 23285
    },
    {
      "entropy": 5.536384391784668,
      "epoch": 2.1126632801161103,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045532904402397157,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21404924392700195,
      "num_tokens": 48329909.0,
      "step": 23290
    },
    {
      "entropy": 5.5862708568573,
      "epoch": 2.1131168359941945,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004553096915597132,
      "loss": 5.0504,
      "mean_token_accuracy": 0.20401450842618943,
      "num_tokens": 48340142.0,
      "step": 23295
    },
    {
      "entropy": 5.566872692108154,
      "epoch": 2.113570391872279,
      "grad_norm": 1.125,
      "learning_rate": 0.0004552903353666044,
      "loss": 4.892,
      "mean_token_accuracy": 0.21543981432914733,
      "num_tokens": 48350540.0,
      "step": 23300
    },
    {
      "entropy": 5.531789922714234,
      "epoch": 2.114023947750363,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045527097544504556,
      "loss": 4.872,
      "mean_token_accuracy": 0.2165081039071083,
      "num_tokens": 48360153.0,
      "step": 23305
    },
    {
      "entropy": 5.497782611846924,
      "epoch": 2.114477503628447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000455251611795437,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21449376791715621,
      "num_tokens": 48370088.0,
      "step": 23310
    },
    {
      "entropy": 5.582926845550537,
      "epoch": 2.114931059506531,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004552322444181792,
      "loss": 4.9555,
      "mean_token_accuracy": 0.20889388918876647,
      "num_tokens": 48381325.0,
      "step": 23315
    },
    {
      "entropy": 5.611449670791626,
      "epoch": 2.1153846153846154,
      "grad_norm": 1.0,
      "learning_rate": 0.0004552128733136726,
      "loss": 4.9383,
      "mean_token_accuracy": 0.20794569104909896,
      "num_tokens": 48393077.0,
      "step": 23320
    },
    {
      "entropy": 5.54364161491394,
      "epoch": 2.1158381712626997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004551934984823179,
      "loss": 4.921,
      "mean_token_accuracy": 0.21392431557178498,
      "num_tokens": 48403260.0,
      "step": 23325
    },
    {
      "entropy": 5.47652792930603,
      "epoch": 2.116291727140784,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004551741199245157,
      "loss": 4.8837,
      "mean_token_accuracy": 0.21594650596380233,
      "num_tokens": 48413616.0,
      "step": 23330
    },
    {
      "entropy": 5.547682046890259,
      "epoch": 2.1167452830188678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045515473764066685,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21258217990398406,
      "num_tokens": 48424189.0,
      "step": 23335
    },
    {
      "entropy": 5.576747989654541,
      "epoch": 2.117198838896952,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045513535163117205,
      "loss": 4.9804,
      "mean_token_accuracy": 0.2100134789943695,
      "num_tokens": 48433603.0,
      "step": 23340
    },
    {
      "entropy": 5.584131240844727,
      "epoch": 2.1176523947750363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004551159618964323,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21004039496183396,
      "num_tokens": 48443043.0,
      "step": 23345
    },
    {
      "entropy": 5.494337797164917,
      "epoch": 2.1181059506531206,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004550965684368486,
      "loss": 4.8283,
      "mean_token_accuracy": 0.2251942738890648,
      "num_tokens": 48452765.0,
      "step": 23350
    },
    {
      "entropy": 5.531205320358277,
      "epoch": 2.118559506531205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004550771712528219,
      "loss": 4.8794,
      "mean_token_accuracy": 0.22011577785015107,
      "num_tokens": 48463191.0,
      "step": 23355
    },
    {
      "entropy": 5.548751163482666,
      "epoch": 2.1190130624092887,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004550577703447534,
      "loss": 4.8854,
      "mean_token_accuracy": 0.21713743656873702,
      "num_tokens": 48473280.0,
      "step": 23360
    },
    {
      "entropy": 5.517544317245483,
      "epoch": 2.119466618287373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045503836571304443,
      "loss": 4.829,
      "mean_token_accuracy": 0.2206776559352875,
      "num_tokens": 48483465.0,
      "step": 23365
    },
    {
      "entropy": 5.500027513504028,
      "epoch": 2.119920174165457,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004550189573580961,
      "loss": 4.8575,
      "mean_token_accuracy": 0.2173631399869919,
      "num_tokens": 48494497.0,
      "step": 23370
    },
    {
      "entropy": 5.540167760848999,
      "epoch": 2.1203737300435415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004549995452803098,
      "loss": 4.9593,
      "mean_token_accuracy": 0.21660860776901245,
      "num_tokens": 48504106.0,
      "step": 23375
    },
    {
      "entropy": 5.61021842956543,
      "epoch": 2.1208272859216257,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004549801294800872,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21148616075515747,
      "num_tokens": 48515729.0,
      "step": 23380
    },
    {
      "entropy": 5.598976898193359,
      "epoch": 2.1212808417997095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004549607099578295,
      "loss": 5.0031,
      "mean_token_accuracy": 0.2067449375987053,
      "num_tokens": 48526611.0,
      "step": 23385
    },
    {
      "entropy": 5.519251966476441,
      "epoch": 2.121734397677794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004549412867139386,
      "loss": 4.8778,
      "mean_token_accuracy": 0.2188269853591919,
      "num_tokens": 48537016.0,
      "step": 23390
    },
    {
      "entropy": 5.564614009857178,
      "epoch": 2.122187953555878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045492185974881587,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20828444957733155,
      "num_tokens": 48546786.0,
      "step": 23395
    },
    {
      "entropy": 5.59769926071167,
      "epoch": 2.1226415094339623,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004549024290628634,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21260345876216888,
      "num_tokens": 48556306.0,
      "step": 23400
    },
    {
      "entropy": 5.582816505432129,
      "epoch": 2.1230950653120466,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004548829946564827,
      "loss": 4.9756,
      "mean_token_accuracy": 0.21278643161058425,
      "num_tokens": 48566303.0,
      "step": 23405
    },
    {
      "entropy": 5.57006721496582,
      "epoch": 2.1235486211901304,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004548635565300759,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21467667669057847,
      "num_tokens": 48576313.0,
      "step": 23410
    },
    {
      "entropy": 5.523672008514405,
      "epoch": 2.1240021770682147,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004548441146840449,
      "loss": 4.935,
      "mean_token_accuracy": 0.21271440833806993,
      "num_tokens": 48587133.0,
      "step": 23415
    },
    {
      "entropy": 5.589933681488037,
      "epoch": 2.124455732946299,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045482466911879184,
      "loss": 5.0569,
      "mean_token_accuracy": 0.2045355036854744,
      "num_tokens": 48598524.0,
      "step": 23420
    },
    {
      "entropy": 5.527286243438721,
      "epoch": 2.1249092888243832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045480521983471875,
      "loss": 4.8935,
      "mean_token_accuracy": 0.21761407554149628,
      "num_tokens": 48609870.0,
      "step": 23425
    },
    {
      "entropy": 5.547309160232544,
      "epoch": 2.1253628447024675,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004547857668322279,
      "loss": 4.9556,
      "mean_token_accuracy": 0.2139271765947342,
      "num_tokens": 48619828.0,
      "step": 23430
    },
    {
      "entropy": 5.548403358459472,
      "epoch": 2.1258164005805513,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004547663101117215,
      "loss": 4.8687,
      "mean_token_accuracy": 0.21930610984563828,
      "num_tokens": 48630361.0,
      "step": 23435
    },
    {
      "entropy": 5.569103813171386,
      "epoch": 2.1262699564586356,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045474684967360216,
      "loss": 4.9307,
      "mean_token_accuracy": 0.20342365950345992,
      "num_tokens": 48641113.0,
      "step": 23440
    },
    {
      "entropy": 5.58072943687439,
      "epoch": 2.12672351233672,
      "grad_norm": 1.078125,
      "learning_rate": 0.000454727385518272,
      "loss": 5.0156,
      "mean_token_accuracy": 0.19991778135299682,
      "num_tokens": 48651777.0,
      "step": 23445
    },
    {
      "entropy": 5.620652103424073,
      "epoch": 2.127177068214804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004547079176461338,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2037887990474701,
      "num_tokens": 48663237.0,
      "step": 23450
    },
    {
      "entropy": 5.54860782623291,
      "epoch": 2.1276306240928884,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004546884460575901,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2187935620546341,
      "num_tokens": 48673932.0,
      "step": 23455
    },
    {
      "entropy": 5.490437364578247,
      "epoch": 2.1280841799709727,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004546689707530434,
      "loss": 4.8626,
      "mean_token_accuracy": 0.21403494775295256,
      "num_tokens": 48684304.0,
      "step": 23460
    },
    {
      "entropy": 5.485919523239136,
      "epoch": 2.1285377358490565,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045464949173289676,
      "loss": 4.8308,
      "mean_token_accuracy": 0.21719606965780258,
      "num_tokens": 48695569.0,
      "step": 23465
    },
    {
      "entropy": 5.472944831848144,
      "epoch": 2.1289912917271407,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004546300089975528,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21602316349744796,
      "num_tokens": 48706550.0,
      "step": 23470
    },
    {
      "entropy": 5.574724197387695,
      "epoch": 2.129444847605225,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045461052254741445,
      "loss": 4.9903,
      "mean_token_accuracy": 0.20163908302783967,
      "num_tokens": 48716852.0,
      "step": 23475
    },
    {
      "entropy": 5.538242483139038,
      "epoch": 2.1298984034833093,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045459103238288474,
      "loss": 4.8617,
      "mean_token_accuracy": 0.21364195495843888,
      "num_tokens": 48726607.0,
      "step": 23480
    },
    {
      "entropy": 5.5888269424438475,
      "epoch": 2.130351959361393,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004545715385043667,
      "loss": 5.0809,
      "mean_token_accuracy": 0.20001124292612077,
      "num_tokens": 48738011.0,
      "step": 23485
    },
    {
      "entropy": 5.56582202911377,
      "epoch": 2.1308055152394774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004545520409122635,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2065620392560959,
      "num_tokens": 48747899.0,
      "step": 23490
    },
    {
      "entropy": 5.623044347763061,
      "epoch": 2.1312590711175616,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004545325396069783,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20890477001667024,
      "num_tokens": 48758719.0,
      "step": 23495
    },
    {
      "entropy": 5.595681190490723,
      "epoch": 2.131712626995646,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004545130345889145,
      "loss": 4.9914,
      "mean_token_accuracy": 0.21252372711896897,
      "num_tokens": 48769677.0,
      "step": 23500
    },
    {
      "entropy": 5.579863166809082,
      "epoch": 2.13216618287373,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004544935258584754,
      "loss": 4.9313,
      "mean_token_accuracy": 0.21335101127624512,
      "num_tokens": 48778440.0,
      "step": 23505
    },
    {
      "entropy": 5.610200357437134,
      "epoch": 2.1326197387518144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004544740134160644,
      "loss": 5.0227,
      "mean_token_accuracy": 0.2070917695760727,
      "num_tokens": 48788328.0,
      "step": 23510
    },
    {
      "entropy": 5.606459283828736,
      "epoch": 2.1330732946298983,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004544544972620851,
      "loss": 4.9925,
      "mean_token_accuracy": 0.20363520532846452,
      "num_tokens": 48799218.0,
      "step": 23515
    },
    {
      "entropy": 5.616060352325439,
      "epoch": 2.1335268505079825,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004544349773969411,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20226793438196183,
      "num_tokens": 48809648.0,
      "step": 23520
    },
    {
      "entropy": 5.560133981704712,
      "epoch": 2.133980406386067,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004544154538210359,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20985978841781616,
      "num_tokens": 48819965.0,
      "step": 23525
    },
    {
      "entropy": 5.6355616569519045,
      "epoch": 2.134433962264151,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004543959265347735,
      "loss": 5.0093,
      "mean_token_accuracy": 0.21467642337083817,
      "num_tokens": 48829862.0,
      "step": 23530
    },
    {
      "entropy": 5.609732627868652,
      "epoch": 2.1348875181422353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045437639553855757,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21068114787340164,
      "num_tokens": 48840551.0,
      "step": 23535
    },
    {
      "entropy": 5.525649309158325,
      "epoch": 2.135341074020319,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004543568608327921,
      "loss": 4.9468,
      "mean_token_accuracy": 0.21656380146741866,
      "num_tokens": 48851032.0,
      "step": 23540
    },
    {
      "entropy": 5.539538955688476,
      "epoch": 2.1357946298984034,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045433732241788105,
      "loss": 5.0047,
      "mean_token_accuracy": 0.20535550713539125,
      "num_tokens": 48861773.0,
      "step": 23545
    },
    {
      "entropy": 5.617505979537964,
      "epoch": 2.1362481857764877,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045431778029422846,
      "loss": 5.0529,
      "mean_token_accuracy": 0.2073085844516754,
      "num_tokens": 48872430.0,
      "step": 23550
    },
    {
      "entropy": 5.575256824493408,
      "epoch": 2.136701741654572,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004542982344622385,
      "loss": 4.8317,
      "mean_token_accuracy": 0.21494340896606445,
      "num_tokens": 48882575.0,
      "step": 23555
    },
    {
      "entropy": 5.5381001949310305,
      "epoch": 2.137155297532656,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045427868492231533,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2106066793203354,
      "num_tokens": 48893337.0,
      "step": 23560
    },
    {
      "entropy": 5.649920129776001,
      "epoch": 2.13760885341074,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045425913167486333,
      "loss": 5.0761,
      "mean_token_accuracy": 0.20791659653186798,
      "num_tokens": 48904675.0,
      "step": 23565
    },
    {
      "entropy": 5.524318504333496,
      "epoch": 2.1380624092888243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045423957472028677,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22435435503721238,
      "num_tokens": 48914831.0,
      "step": 23570
    },
    {
      "entropy": 5.476833629608154,
      "epoch": 2.1385159651669086,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004542200140589901,
      "loss": 4.8389,
      "mean_token_accuracy": 0.21513169705867768,
      "num_tokens": 48924574.0,
      "step": 23575
    },
    {
      "entropy": 5.561999893188476,
      "epoch": 2.138969521044993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045420044969137787,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2093830093741417,
      "num_tokens": 48933921.0,
      "step": 23580
    },
    {
      "entropy": 5.5500823020935055,
      "epoch": 2.139423076923077,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004541808816178547,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21145492494106294,
      "num_tokens": 48944571.0,
      "step": 23585
    },
    {
      "entropy": 5.566286087036133,
      "epoch": 2.139876632801161,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045416130983882524,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2070055663585663,
      "num_tokens": 48954569.0,
      "step": 23590
    },
    {
      "entropy": 5.570382690429687,
      "epoch": 2.140330188679245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045414173435469427,
      "loss": 4.924,
      "mean_token_accuracy": 0.21110861897468566,
      "num_tokens": 48965886.0,
      "step": 23595
    },
    {
      "entropy": 5.597574377059937,
      "epoch": 2.1407837445573294,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045412215516586654,
      "loss": 4.9425,
      "mean_token_accuracy": 0.21023617088794708,
      "num_tokens": 48976219.0,
      "step": 23600
    },
    {
      "entropy": 5.507729434967041,
      "epoch": 2.1412373004354137,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000454102572272747,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21244221925735474,
      "num_tokens": 48986322.0,
      "step": 23605
    },
    {
      "entropy": 5.546385765075684,
      "epoch": 2.141690856313498,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004540829856757406,
      "loss": 4.8756,
      "mean_token_accuracy": 0.2178581491112709,
      "num_tokens": 48996297.0,
      "step": 23610
    },
    {
      "entropy": 5.5631684303283695,
      "epoch": 2.142144412191582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004540633953752525,
      "loss": 4.9392,
      "mean_token_accuracy": 0.2150893673300743,
      "num_tokens": 49006165.0,
      "step": 23615
    },
    {
      "entropy": 5.526386499404907,
      "epoch": 2.142597968069666,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004540438013716876,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2047666236758232,
      "num_tokens": 49017278.0,
      "step": 23620
    },
    {
      "entropy": 5.5054583072662355,
      "epoch": 2.1430515239477503,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004540242036654514,
      "loss": 4.8807,
      "mean_token_accuracy": 0.21305353343486785,
      "num_tokens": 49027088.0,
      "step": 23625
    },
    {
      "entropy": 5.555301380157471,
      "epoch": 2.1435050798258346,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000454004602256949,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21166100054979325,
      "num_tokens": 49038156.0,
      "step": 23630
    },
    {
      "entropy": 5.602395725250244,
      "epoch": 2.143958635703919,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004539849971465858,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20847307592630387,
      "num_tokens": 49047480.0,
      "step": 23635
    },
    {
      "entropy": 5.614607524871826,
      "epoch": 2.1444121915820027,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045396538833476725,
      "loss": 4.9708,
      "mean_token_accuracy": 0.21530205011367798,
      "num_tokens": 49057373.0,
      "step": 23640
    },
    {
      "entropy": 5.558904933929443,
      "epoch": 2.144865747460087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004539457758218989,
      "loss": 4.9691,
      "mean_token_accuracy": 0.2142976775765419,
      "num_tokens": 49068380.0,
      "step": 23645
    },
    {
      "entropy": 5.540809392929077,
      "epoch": 2.1453193033381712,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004539261596083863,
      "loss": 4.9392,
      "mean_token_accuracy": 0.21224504560232163,
      "num_tokens": 49078646.0,
      "step": 23650
    },
    {
      "entropy": 5.550167655944824,
      "epoch": 2.1457728592162555,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004539065396946351,
      "loss": 4.88,
      "mean_token_accuracy": 0.2214643895626068,
      "num_tokens": 49089281.0,
      "step": 23655
    },
    {
      "entropy": 5.573638916015625,
      "epoch": 2.1462264150943398,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004538869160810511,
      "loss": 4.9537,
      "mean_token_accuracy": 0.20815649181604384,
      "num_tokens": 49099514.0,
      "step": 23660
    },
    {
      "entropy": 5.573917293548584,
      "epoch": 2.146679970972424,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045386728876804,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20969745069742202,
      "num_tokens": 49110247.0,
      "step": 23665
    },
    {
      "entropy": 5.619055366516113,
      "epoch": 2.147133526850508,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045384765775600787,
      "loss": 5.0286,
      "mean_token_accuracy": 0.20521268248558044,
      "num_tokens": 49122492.0,
      "step": 23670
    },
    {
      "entropy": 5.594598436355591,
      "epoch": 2.147587082728592,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004538280230453606,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2154301866889,
      "num_tokens": 49132395.0,
      "step": 23675
    },
    {
      "entropy": 5.611228895187378,
      "epoch": 2.1480406386066764,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004538083846365042,
      "loss": 4.9674,
      "mean_token_accuracy": 0.20996081531047822,
      "num_tokens": 49142909.0,
      "step": 23680
    },
    {
      "entropy": 5.562023687362671,
      "epoch": 2.1484941944847606,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004537887425298448,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21113996356725692,
      "num_tokens": 49152933.0,
      "step": 23685
    },
    {
      "entropy": 5.5440997123718265,
      "epoch": 2.1489477503628445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045376909672578875,
      "loss": 4.9325,
      "mean_token_accuracy": 0.2081219121813774,
      "num_tokens": 49163595.0,
      "step": 23690
    },
    {
      "entropy": 5.492967176437378,
      "epoch": 2.1494013062409287,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004537494472247422,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21840045154094695,
      "num_tokens": 49174664.0,
      "step": 23695
    },
    {
      "entropy": 5.581111717224121,
      "epoch": 2.149854862119013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004537297940271115,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21579648852348327,
      "num_tokens": 49185105.0,
      "step": 23700
    },
    {
      "entropy": 5.547584915161133,
      "epoch": 2.1503084179970973,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004537101371333031,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21742856353521348,
      "num_tokens": 49194679.0,
      "step": 23705
    },
    {
      "entropy": 5.57237753868103,
      "epoch": 2.1507619738751815,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045369047654372356,
      "loss": 4.8511,
      "mean_token_accuracy": 0.21498076766729354,
      "num_tokens": 49204563.0,
      "step": 23710
    },
    {
      "entropy": 5.598416185379028,
      "epoch": 2.151215529753266,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004536708122587795,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21903055608272554,
      "num_tokens": 49213759.0,
      "step": 23715
    },
    {
      "entropy": 5.504739713668823,
      "epoch": 2.1516690856313496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004536511442788774,
      "loss": 4.8407,
      "mean_token_accuracy": 0.21756711155176162,
      "num_tokens": 49223604.0,
      "step": 23720
    },
    {
      "entropy": 5.545053482055664,
      "epoch": 2.152122641509434,
      "grad_norm": 1.125,
      "learning_rate": 0.0004536314726044241,
      "loss": 4.957,
      "mean_token_accuracy": 0.2128641873598099,
      "num_tokens": 49234000.0,
      "step": 23725
    },
    {
      "entropy": 5.592650985717773,
      "epoch": 2.152576197387518,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045361179723582654,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20984700918197632,
      "num_tokens": 49244874.0,
      "step": 23730
    },
    {
      "entropy": 5.665120935440063,
      "epoch": 2.1530297532656024,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004535921181734914,
      "loss": 5.0996,
      "mean_token_accuracy": 0.20604943633079528,
      "num_tokens": 49255495.0,
      "step": 23735
    },
    {
      "entropy": 5.5517720699310305,
      "epoch": 2.1534833091436867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004535724354178257,
      "loss": 4.8858,
      "mean_token_accuracy": 0.21245681643486022,
      "num_tokens": 49265357.0,
      "step": 23740
    },
    {
      "entropy": 5.604379177093506,
      "epoch": 2.1539368650217705,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045355274896923664,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20246616750955582,
      "num_tokens": 49276396.0,
      "step": 23745
    },
    {
      "entropy": 5.5582743167877195,
      "epoch": 2.1543904208998548,
      "grad_norm": 1.125,
      "learning_rate": 0.0004535330588281312,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21818099468946456,
      "num_tokens": 49285925.0,
      "step": 23750
    },
    {
      "entropy": 5.509740018844605,
      "epoch": 2.154843976777939,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004535133649949166,
      "loss": 4.8968,
      "mean_token_accuracy": 0.2197282150387764,
      "num_tokens": 49296493.0,
      "step": 23755
    },
    {
      "entropy": 5.493279123306275,
      "epoch": 2.1552975326560233,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004534936674700001,
      "loss": 4.8728,
      "mean_token_accuracy": 0.21899235099554062,
      "num_tokens": 49307273.0,
      "step": 23760
    },
    {
      "entropy": 5.558998394012451,
      "epoch": 2.1557510885341076,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045347396625378914,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22031073719263078,
      "num_tokens": 49317055.0,
      "step": 23765
    },
    {
      "entropy": 5.583795595169067,
      "epoch": 2.1562046444121914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045345426134669103,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2076543837785721,
      "num_tokens": 49327420.0,
      "step": 23770
    },
    {
      "entropy": 5.564263010025025,
      "epoch": 2.1566582002902757,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045343455274911325,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20900900959968566,
      "num_tokens": 49338323.0,
      "step": 23775
    },
    {
      "entropy": 5.584456729888916,
      "epoch": 2.15711175616836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004534148404614635,
      "loss": 4.966,
      "mean_token_accuracy": 0.2122632309794426,
      "num_tokens": 49348364.0,
      "step": 23780
    },
    {
      "entropy": 5.641426658630371,
      "epoch": 2.157565312046444,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045339512448414943,
      "loss": 4.9978,
      "mean_token_accuracy": 0.21000421792268753,
      "num_tokens": 49358274.0,
      "step": 23785
    },
    {
      "entropy": 5.5278363704681395,
      "epoch": 2.1580188679245285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004533754048175787,
      "loss": 4.8781,
      "mean_token_accuracy": 0.21102921068668365,
      "num_tokens": 49368651.0,
      "step": 23790
    },
    {
      "entropy": 5.525886058807373,
      "epoch": 2.1584724238026123,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004533556814621591,
      "loss": 4.8891,
      "mean_token_accuracy": 0.20612184703350067,
      "num_tokens": 49378363.0,
      "step": 23795
    },
    {
      "entropy": 5.562664365768432,
      "epoch": 2.1589259796806965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045333595441829865,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2144285872578621,
      "num_tokens": 49388241.0,
      "step": 23800
    },
    {
      "entropy": 5.56508150100708,
      "epoch": 2.159379535558781,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004533162236864051,
      "loss": 4.9281,
      "mean_token_accuracy": 0.2083830565214157,
      "num_tokens": 49397678.0,
      "step": 23805
    },
    {
      "entropy": 5.620884561538697,
      "epoch": 2.159833091436865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004532964892668867,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20939994752407073,
      "num_tokens": 49407368.0,
      "step": 23810
    },
    {
      "entropy": 5.5370481491088865,
      "epoch": 2.1602866473149493,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045327675116015146,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21844192892313002,
      "num_tokens": 49417926.0,
      "step": 23815
    },
    {
      "entropy": 5.649501085281372,
      "epoch": 2.1607402031930336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004532570093666075,
      "loss": 5.0501,
      "mean_token_accuracy": 0.19813580811023712,
      "num_tokens": 49427899.0,
      "step": 23820
    },
    {
      "entropy": 5.547929573059082,
      "epoch": 2.1611937590711174,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004532372638866633,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22209731489419937,
      "num_tokens": 49439102.0,
      "step": 23825
    },
    {
      "entropy": 5.551114797592163,
      "epoch": 2.1616473149492017,
      "grad_norm": 1.25,
      "learning_rate": 0.00045321751472072693,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22399413585662842,
      "num_tokens": 49448013.0,
      "step": 23830
    },
    {
      "entropy": 5.476213884353638,
      "epoch": 2.162100870827286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045319776186920695,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22196878045797347,
      "num_tokens": 49457620.0,
      "step": 23835
    },
    {
      "entropy": 5.486994743347168,
      "epoch": 2.1625544267053702,
      "grad_norm": 1.125,
      "learning_rate": 0.00045317800533251193,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2092035233974457,
      "num_tokens": 49468038.0,
      "step": 23840
    },
    {
      "entropy": 5.511063575744629,
      "epoch": 2.163007982583454,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045315824511105027,
      "loss": 4.8703,
      "mean_token_accuracy": 0.21102674603462218,
      "num_tokens": 49477714.0,
      "step": 23845
    },
    {
      "entropy": 5.595931816101074,
      "epoch": 2.1634615384615383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045313848120523073,
      "loss": 4.8559,
      "mean_token_accuracy": 0.219837549328804,
      "num_tokens": 49488214.0,
      "step": 23850
    },
    {
      "entropy": 5.57024393081665,
      "epoch": 2.1639150943396226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045311871361546205,
      "loss": 4.9818,
      "mean_token_accuracy": 0.21644212901592255,
      "num_tokens": 49498444.0,
      "step": 23855
    },
    {
      "entropy": 5.478400087356567,
      "epoch": 2.164368650217707,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004530989423421529,
      "loss": 4.8408,
      "mean_token_accuracy": 0.21539229452610015,
      "num_tokens": 49508759.0,
      "step": 23860
    },
    {
      "entropy": 5.462951278686523,
      "epoch": 2.164822206095791,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045307916738571223,
      "loss": 4.8137,
      "mean_token_accuracy": 0.22646919190883635,
      "num_tokens": 49518787.0,
      "step": 23865
    },
    {
      "entropy": 5.5541462898254395,
      "epoch": 2.1652757619738754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004530593887465491,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21947762221097947,
      "num_tokens": 49528130.0,
      "step": 23870
    },
    {
      "entropy": 5.621574974060058,
      "epoch": 2.165729317851959,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045303960642507235,
      "loss": 5.0673,
      "mean_token_accuracy": 0.2061009705066681,
      "num_tokens": 49539128.0,
      "step": 23875
    },
    {
      "entropy": 5.498931550979615,
      "epoch": 2.1661828737300435,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045301982042169123,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21534635871648788,
      "num_tokens": 49550571.0,
      "step": 23880
    },
    {
      "entropy": 5.532294464111328,
      "epoch": 2.1666364296081277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045300003073681485,
      "loss": 4.9278,
      "mean_token_accuracy": 0.2141397401690483,
      "num_tokens": 49561106.0,
      "step": 23885
    },
    {
      "entropy": 5.519245576858521,
      "epoch": 2.167089985486212,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045298023737085243,
      "loss": 4.8295,
      "mean_token_accuracy": 0.21391812115907669,
      "num_tokens": 49571389.0,
      "step": 23890
    },
    {
      "entropy": 5.542738008499145,
      "epoch": 2.1675435413642963,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004529604403242134,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2099479153752327,
      "num_tokens": 49582224.0,
      "step": 23895
    },
    {
      "entropy": 5.548014879226685,
      "epoch": 2.16799709724238,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045294063959730707,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20937033742666245,
      "num_tokens": 49592413.0,
      "step": 23900
    },
    {
      "entropy": 5.5853612422943115,
      "epoch": 2.1684506531204644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045292083519054297,
      "loss": 5.0179,
      "mean_token_accuracy": 0.19779800474643708,
      "num_tokens": 49602796.0,
      "step": 23905
    },
    {
      "entropy": 5.516379737854004,
      "epoch": 2.1689042089985486,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004529010271043307,
      "loss": 4.8884,
      "mean_token_accuracy": 0.21849001199007034,
      "num_tokens": 49613625.0,
      "step": 23910
    },
    {
      "entropy": 5.6649391651153564,
      "epoch": 2.169357764876633,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045288121533907977,
      "loss": 5.0185,
      "mean_token_accuracy": 0.208993361890316,
      "num_tokens": 49624719.0,
      "step": 23915
    },
    {
      "entropy": 5.614285802841186,
      "epoch": 2.169811320754717,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045286139989520006,
      "loss": 4.988,
      "mean_token_accuracy": 0.20872919261455536,
      "num_tokens": 49634916.0,
      "step": 23920
    },
    {
      "entropy": 5.578924083709717,
      "epoch": 2.170264876632801,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004528415807731012,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21143622994422911,
      "num_tokens": 49645215.0,
      "step": 23925
    },
    {
      "entropy": 5.541259670257569,
      "epoch": 2.1707184325108853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045282175797319317,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21639792621135712,
      "num_tokens": 49655805.0,
      "step": 23930
    },
    {
      "entropy": 5.569958972930908,
      "epoch": 2.1711719883889695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004528019314958859,
      "loss": 4.9323,
      "mean_token_accuracy": 0.2147790864109993,
      "num_tokens": 49665264.0,
      "step": 23935
    },
    {
      "entropy": 5.5874114513397215,
      "epoch": 2.171625544267054,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045278210134158935,
      "loss": 4.989,
      "mean_token_accuracy": 0.20953418165445328,
      "num_tokens": 49676152.0,
      "step": 23940
    },
    {
      "entropy": 5.58998851776123,
      "epoch": 2.172079100145138,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004527622675107137,
      "loss": 4.9423,
      "mean_token_accuracy": 0.20832472294569016,
      "num_tokens": 49686762.0,
      "step": 23945
    },
    {
      "entropy": 5.613668918609619,
      "epoch": 2.172532656023222,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045274243000366897,
      "loss": 4.9671,
      "mean_token_accuracy": 0.21186559349298478,
      "num_tokens": 49697504.0,
      "step": 23950
    },
    {
      "entropy": 5.5716845989227295,
      "epoch": 2.172986211901306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004527225888208655,
      "loss": 5.0054,
      "mean_token_accuracy": 0.20693681091070176,
      "num_tokens": 49709197.0,
      "step": 23955
    },
    {
      "entropy": 5.57248010635376,
      "epoch": 2.1734397677793904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004527027439627137,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20444791167974471,
      "num_tokens": 49719764.0,
      "step": 23960
    },
    {
      "entropy": 5.5693206787109375,
      "epoch": 2.1738933236574747,
      "grad_norm": 1.125,
      "learning_rate": 0.00045268289542962377,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20790381133556365,
      "num_tokens": 49730363.0,
      "step": 23965
    },
    {
      "entropy": 5.575118541717529,
      "epoch": 2.174346879535559,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004526630432220063,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20046700537204742,
      "num_tokens": 49741037.0,
      "step": 23970
    },
    {
      "entropy": 5.541949987411499,
      "epoch": 2.1748004354136428,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004526431873402719,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2153545692563057,
      "num_tokens": 49750127.0,
      "step": 23975
    },
    {
      "entropy": 5.653799819946289,
      "epoch": 2.175253991291727,
      "grad_norm": 1.109375,
      "learning_rate": 0.000452623327784831,
      "loss": 5.0475,
      "mean_token_accuracy": 0.20731836259365083,
      "num_tokens": 49760298.0,
      "step": 23980
    },
    {
      "entropy": 5.575394868850708,
      "epoch": 2.1757075471698113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004526034645560945,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20924535840749742,
      "num_tokens": 49770841.0,
      "step": 23985
    },
    {
      "entropy": 5.515012645721436,
      "epoch": 2.1761611030478956,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045258359765447306,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21334540694952012,
      "num_tokens": 49781000.0,
      "step": 23990
    },
    {
      "entropy": 5.541502332687378,
      "epoch": 2.17661465892598,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004525637270803776,
      "loss": 4.9125,
      "mean_token_accuracy": 0.215426667034626,
      "num_tokens": 49790886.0,
      "step": 23995
    },
    {
      "entropy": 5.552193069458008,
      "epoch": 2.1770682148040637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045254385283421896,
      "loss": 4.9144,
      "mean_token_accuracy": 0.21511530727148057,
      "num_tokens": 49801477.0,
      "step": 24000
    },
    {
      "epoch": 2.1770682148040637,
      "eval_entropy": 5.3124205934515905,
      "eval_loss": 5.021059513092041,
      "eval_mean_token_accuracy": 0.21675180812193529,
      "eval_num_tokens": 49801477.0,
      "eval_runtime": 20.6759,
      "eval_samples_per_second": 1710.202,
      "eval_steps_per_second": 213.775,
      "step": 24000
    },
    {
      "entropy": 5.567465543746948,
      "epoch": 2.177521770682148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004525239749164082,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21617337167263032,
      "num_tokens": 49812044.0,
      "step": 24005
    },
    {
      "entropy": 5.517443132400513,
      "epoch": 2.177975326560232,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004525040933273564,
      "loss": 4.866,
      "mean_token_accuracy": 0.21279383152723313,
      "num_tokens": 49821559.0,
      "step": 24010
    },
    {
      "entropy": 5.611775588989258,
      "epoch": 2.1784288824383164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045248420806747483,
      "loss": 5.035,
      "mean_token_accuracy": 0.2034580260515213,
      "num_tokens": 49832937.0,
      "step": 24015
    },
    {
      "entropy": 5.610374784469604,
      "epoch": 2.1788824383164007,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004524643191371745,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21035014986991882,
      "num_tokens": 49843176.0,
      "step": 24020
    },
    {
      "entropy": 5.558076333999634,
      "epoch": 2.179335994194485,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045244442653686684,
      "loss": 4.8717,
      "mean_token_accuracy": 0.22338457107543946,
      "num_tokens": 49852522.0,
      "step": 24025
    },
    {
      "entropy": 5.5005677223205565,
      "epoch": 2.179789550072569,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045242453026696325,
      "loss": 4.8727,
      "mean_token_accuracy": 0.2199748158454895,
      "num_tokens": 49863143.0,
      "step": 24030
    },
    {
      "entropy": 5.598064994812011,
      "epoch": 2.180243105950653,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004524046303278752,
      "loss": 4.9663,
      "mean_token_accuracy": 0.211096827685833,
      "num_tokens": 49873465.0,
      "step": 24035
    },
    {
      "entropy": 5.659970378875732,
      "epoch": 2.1806966618287373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004523847267200141,
      "loss": 5.042,
      "mean_token_accuracy": 0.20355025976896285,
      "num_tokens": 49883690.0,
      "step": 24040
    },
    {
      "entropy": 5.613793897628784,
      "epoch": 2.1811502177068216,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045236481944379165,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20588492155075072,
      "num_tokens": 49894418.0,
      "step": 24045
    },
    {
      "entropy": 5.574164962768554,
      "epoch": 2.1816037735849054,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004523449084996196,
      "loss": 4.8774,
      "mean_token_accuracy": 0.21884071081876755,
      "num_tokens": 49903927.0,
      "step": 24050
    },
    {
      "entropy": 5.557644462585449,
      "epoch": 2.1820573294629897,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004523249938879096,
      "loss": 4.9373,
      "mean_token_accuracy": 0.21283381432294846,
      "num_tokens": 49914203.0,
      "step": 24055
    },
    {
      "entropy": 5.572479343414306,
      "epoch": 2.182510885341074,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045230507560907354,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21755848675966263,
      "num_tokens": 49924320.0,
      "step": 24060
    },
    {
      "entropy": 5.510980892181396,
      "epoch": 2.1829644412191582,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045228515366352334,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2151745319366455,
      "num_tokens": 49934634.0,
      "step": 24065
    },
    {
      "entropy": 5.628230476379395,
      "epoch": 2.1834179970972425,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045226522805167095,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20996217131614686,
      "num_tokens": 49944586.0,
      "step": 24070
    },
    {
      "entropy": 5.573509883880615,
      "epoch": 2.1838715529753268,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004522452987739286,
      "loss": 4.939,
      "mean_token_accuracy": 0.21171447187662124,
      "num_tokens": 49955441.0,
      "step": 24075
    },
    {
      "entropy": 5.625335645675659,
      "epoch": 2.1843251088534106,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045222536583070823,
      "loss": 5.0501,
      "mean_token_accuracy": 0.21236481964588166,
      "num_tokens": 49966718.0,
      "step": 24080
    },
    {
      "entropy": 5.623920345306397,
      "epoch": 2.184778664731495,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045220542922242205,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20676001459360122,
      "num_tokens": 49979483.0,
      "step": 24085
    },
    {
      "entropy": 5.569506120681763,
      "epoch": 2.185232220609579,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045218548894948254,
      "loss": 4.9226,
      "mean_token_accuracy": 0.2171776846051216,
      "num_tokens": 49989976.0,
      "step": 24090
    },
    {
      "entropy": 5.5316712856292725,
      "epoch": 2.1856857764876634,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045216554501230187,
      "loss": 4.8031,
      "mean_token_accuracy": 0.21656409054994583,
      "num_tokens": 50000069.0,
      "step": 24095
    },
    {
      "entropy": 5.537426471710205,
      "epoch": 2.1861393323657476,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004521455974112927,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21228833943605424,
      "num_tokens": 50009970.0,
      "step": 24100
    },
    {
      "entropy": 5.587401485443115,
      "epoch": 2.1865928882438315,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045212564614686735,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21030141711235045,
      "num_tokens": 50019590.0,
      "step": 24105
    },
    {
      "entropy": 5.57425594329834,
      "epoch": 2.1870464441219157,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004521056912194386,
      "loss": 4.941,
      "mean_token_accuracy": 0.2013477236032486,
      "num_tokens": 50030092.0,
      "step": 24110
    },
    {
      "entropy": 5.510309839248658,
      "epoch": 2.1875,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004520857326294189,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21099548786878586,
      "num_tokens": 50040318.0,
      "step": 24115
    },
    {
      "entropy": 5.581630992889404,
      "epoch": 2.1879535558780843,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045206577037722114,
      "loss": 5.0615,
      "mean_token_accuracy": 0.19968541115522384,
      "num_tokens": 50050507.0,
      "step": 24120
    },
    {
      "entropy": 5.713084602355957,
      "epoch": 2.1884071117561685,
      "grad_norm": 1.125,
      "learning_rate": 0.0004520458044632582,
      "loss": 5.0184,
      "mean_token_accuracy": 0.20956000536680222,
      "num_tokens": 50059727.0,
      "step": 24125
    },
    {
      "entropy": 5.605908966064453,
      "epoch": 2.1888606676342524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045202583488794285,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21035483330488206,
      "num_tokens": 50070812.0,
      "step": 24130
    },
    {
      "entropy": 5.614676666259766,
      "epoch": 2.1893142235123366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004520058616516881,
      "loss": 5.0271,
      "mean_token_accuracy": 0.2030799001455307,
      "num_tokens": 50081175.0,
      "step": 24135
    },
    {
      "entropy": 5.563994359970093,
      "epoch": 2.189767779390421,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045198588475490716,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21069299876689912,
      "num_tokens": 50092246.0,
      "step": 24140
    },
    {
      "entropy": 5.591595077514649,
      "epoch": 2.190221335268505,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004519659041980129,
      "loss": 4.987,
      "mean_token_accuracy": 0.2111479938030243,
      "num_tokens": 50101934.0,
      "step": 24145
    },
    {
      "entropy": 5.58695707321167,
      "epoch": 2.1906748911465894,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004519459199814187,
      "loss": 4.9896,
      "mean_token_accuracy": 0.2108437404036522,
      "num_tokens": 50112159.0,
      "step": 24150
    },
    {
      "entropy": 5.547046375274658,
      "epoch": 2.1911284470246732,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004519259321055377,
      "loss": 4.97,
      "mean_token_accuracy": 0.207854525744915,
      "num_tokens": 50123079.0,
      "step": 24155
    },
    {
      "entropy": 5.56029863357544,
      "epoch": 2.1915820029027575,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004519059405707834,
      "loss": 4.9458,
      "mean_token_accuracy": 0.20935097485780715,
      "num_tokens": 50133097.0,
      "step": 24160
    },
    {
      "entropy": 5.640683174133301,
      "epoch": 2.1920355587808418,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045188594537756916,
      "loss": 5.0188,
      "mean_token_accuracy": 0.21063667833805083,
      "num_tokens": 50143062.0,
      "step": 24165
    },
    {
      "entropy": 5.625964641571045,
      "epoch": 2.192489114658926,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045186594652630853,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21432217061519623,
      "num_tokens": 50154482.0,
      "step": 24170
    },
    {
      "entropy": 5.566651487350464,
      "epoch": 2.1929426705370103,
      "grad_norm": 1.125,
      "learning_rate": 0.000451845944017415,
      "loss": 4.8989,
      "mean_token_accuracy": 0.21672665178775788,
      "num_tokens": 50164880.0,
      "step": 24175
    },
    {
      "entropy": 5.5295075416564945,
      "epoch": 2.1933962264150946,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045182593785130225,
      "loss": 4.8864,
      "mean_token_accuracy": 0.21461732685565948,
      "num_tokens": 50173896.0,
      "step": 24180
    },
    {
      "entropy": 5.587013864517212,
      "epoch": 2.1938497822931784,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004518059280283842,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20583032071590424,
      "num_tokens": 50185078.0,
      "step": 24185
    },
    {
      "entropy": 5.597308826446533,
      "epoch": 2.1943033381712627,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045178591454907434,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21808431446552276,
      "num_tokens": 50195500.0,
      "step": 24190
    },
    {
      "entropy": 5.514723110198974,
      "epoch": 2.194756894049347,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045176589741378675,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21505440771579742,
      "num_tokens": 50206103.0,
      "step": 24195
    },
    {
      "entropy": 5.494816303253174,
      "epoch": 2.195210449927431,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004517458766229354,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21155305355787277,
      "num_tokens": 50216522.0,
      "step": 24200
    },
    {
      "entropy": 5.617796611785889,
      "epoch": 2.195664005805515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045172585217693426,
      "loss": 4.9519,
      "mean_token_accuracy": 0.2140292927622795,
      "num_tokens": 50226721.0,
      "step": 24205
    },
    {
      "entropy": 5.660838317871094,
      "epoch": 2.1961175616835993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045170582407619747,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20557497888803483,
      "num_tokens": 50236327.0,
      "step": 24210
    },
    {
      "entropy": 5.515146970748901,
      "epoch": 2.1965711175616836,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004516857923211392,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2197105437517166,
      "num_tokens": 50245729.0,
      "step": 24215
    },
    {
      "entropy": 5.580048704147339,
      "epoch": 2.197024673439768,
      "grad_norm": 1.125,
      "learning_rate": 0.00045166575691217375,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20498782992362977,
      "num_tokens": 50257024.0,
      "step": 24220
    },
    {
      "entropy": 5.585898065567017,
      "epoch": 2.197478229317852,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045164571784971536,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21638111472129823,
      "num_tokens": 50266910.0,
      "step": 24225
    },
    {
      "entropy": 5.564283275604248,
      "epoch": 2.1979317851959363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045162567513417864,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21388950496912001,
      "num_tokens": 50278233.0,
      "step": 24230
    },
    {
      "entropy": 5.54777045249939,
      "epoch": 2.19838534107402,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045160562876597776,
      "loss": 4.9073,
      "mean_token_accuracy": 0.20627465546131135,
      "num_tokens": 50289276.0,
      "step": 24235
    },
    {
      "entropy": 5.553523683547974,
      "epoch": 2.1988388969521044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004515855787455276,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21545596271753312,
      "num_tokens": 50299336.0,
      "step": 24240
    },
    {
      "entropy": 5.560890579223633,
      "epoch": 2.1992924528301887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004515655250732426,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2169785499572754,
      "num_tokens": 50308308.0,
      "step": 24245
    },
    {
      "entropy": 5.527594089508057,
      "epoch": 2.199746008708273,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045154546774953764,
      "loss": 4.9408,
      "mean_token_accuracy": 0.21240466088056564,
      "num_tokens": 50318295.0,
      "step": 24250
    },
    {
      "entropy": 5.504431247711182,
      "epoch": 2.2001995645863572,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004515254067748274,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21872513592243195,
      "num_tokens": 50329296.0,
      "step": 24255
    },
    {
      "entropy": 5.55621371269226,
      "epoch": 2.200653120464441,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004515053421495267,
      "loss": 5.001,
      "mean_token_accuracy": 0.20430677384138107,
      "num_tokens": 50339389.0,
      "step": 24260
    },
    {
      "entropy": 5.650838375091553,
      "epoch": 2.2011066763425253,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045148527387405056,
      "loss": 5.0359,
      "mean_token_accuracy": 0.20117291063070297,
      "num_tokens": 50349647.0,
      "step": 24265
    },
    {
      "entropy": 5.642973470687866,
      "epoch": 2.2015602322206096,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004514652019488141,
      "loss": 5.0029,
      "mean_token_accuracy": 0.2057658925652504,
      "num_tokens": 50361292.0,
      "step": 24270
    },
    {
      "entropy": 5.57130913734436,
      "epoch": 2.202013788098694,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004514451263742321,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21088439375162124,
      "num_tokens": 50372293.0,
      "step": 24275
    },
    {
      "entropy": 5.51547155380249,
      "epoch": 2.202467343976778,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004514250471507201,
      "loss": 4.9004,
      "mean_token_accuracy": 0.21744245737791063,
      "num_tokens": 50382462.0,
      "step": 24280
    },
    {
      "entropy": 5.591727590560913,
      "epoch": 2.202920899854862,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045140496427869306,
      "loss": 4.9739,
      "mean_token_accuracy": 0.2198596939444542,
      "num_tokens": 50393157.0,
      "step": 24285
    },
    {
      "entropy": 5.487828397750855,
      "epoch": 2.203374455732946,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004513848777585664,
      "loss": 4.847,
      "mean_token_accuracy": 0.21846867501735687,
      "num_tokens": 50403474.0,
      "step": 24290
    },
    {
      "entropy": 5.560375642776489,
      "epoch": 2.2038280116110305,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004513647875907556,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2093304619193077,
      "num_tokens": 50414337.0,
      "step": 24295
    },
    {
      "entropy": 5.637948083877563,
      "epoch": 2.2042815674891147,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045134469377567597,
      "loss": 5.0628,
      "mean_token_accuracy": 0.19925625026226043,
      "num_tokens": 50424863.0,
      "step": 24300
    },
    {
      "entropy": 5.61402997970581,
      "epoch": 2.204735123367199,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004513245963137432,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20893265604972838,
      "num_tokens": 50435982.0,
      "step": 24305
    },
    {
      "entropy": 5.5586028575897215,
      "epoch": 2.205188679245283,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004513044952053728,
      "loss": 4.9096,
      "mean_token_accuracy": 0.21000106036663055,
      "num_tokens": 50445112.0,
      "step": 24310
    },
    {
      "entropy": 5.49434266090393,
      "epoch": 2.205642235123367,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045128439045098045,
      "loss": 4.8892,
      "mean_token_accuracy": 0.2223430946469307,
      "num_tokens": 50455266.0,
      "step": 24315
    },
    {
      "entropy": 5.593818855285645,
      "epoch": 2.2060957910014514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045126428205098207,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20155433416366578,
      "num_tokens": 50466930.0,
      "step": 24320
    },
    {
      "entropy": 5.599624061584473,
      "epoch": 2.2065493468795356,
      "grad_norm": 1.125,
      "learning_rate": 0.0004512441700057934,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21479129493236543,
      "num_tokens": 50477534.0,
      "step": 24325
    },
    {
      "entropy": 5.580485439300537,
      "epoch": 2.20700290275762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004512240543158305,
      "loss": 4.9292,
      "mean_token_accuracy": 0.21126864701509476,
      "num_tokens": 50487352.0,
      "step": 24330
    },
    {
      "entropy": 5.515688037872314,
      "epoch": 2.2074564586357037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045120393498150905,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22341122329235077,
      "num_tokens": 50496227.0,
      "step": 24335
    },
    {
      "entropy": 5.569360017776489,
      "epoch": 2.207910014513788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045118381200324537,
      "loss": 4.9655,
      "mean_token_accuracy": 0.20855049341917037,
      "num_tokens": 50507536.0,
      "step": 24340
    },
    {
      "entropy": 5.61978759765625,
      "epoch": 2.2083635703918723,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004511636853814556,
      "loss": 4.9499,
      "mean_token_accuracy": 0.2121044009923935,
      "num_tokens": 50517659.0,
      "step": 24345
    },
    {
      "entropy": 5.570859956741333,
      "epoch": 2.2088171262699565,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045114355511655595,
      "loss": 4.935,
      "mean_token_accuracy": 0.2064822569489479,
      "num_tokens": 50528717.0,
      "step": 24350
    },
    {
      "entropy": 5.504991149902343,
      "epoch": 2.209270682148041,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045112342120896267,
      "loss": 4.8521,
      "mean_token_accuracy": 0.215097114443779,
      "num_tokens": 50538449.0,
      "step": 24355
    },
    {
      "entropy": 5.598603582382202,
      "epoch": 2.2097242380261246,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045110328365909215,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20462184697389602,
      "num_tokens": 50549098.0,
      "step": 24360
    },
    {
      "entropy": 5.603333568572998,
      "epoch": 2.210177793904209,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004510831424673609,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20936270505189897,
      "num_tokens": 50560047.0,
      "step": 24365
    },
    {
      "entropy": 5.609273386001587,
      "epoch": 2.210631349782293,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004510629976341853,
      "loss": 4.9174,
      "mean_token_accuracy": 0.20975878089666367,
      "num_tokens": 50569704.0,
      "step": 24370
    },
    {
      "entropy": 5.578531885147095,
      "epoch": 2.2110849056603774,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045104284915998215,
      "loss": 4.9763,
      "mean_token_accuracy": 0.21189524382352828,
      "num_tokens": 50580006.0,
      "step": 24375
    },
    {
      "entropy": 5.522437000274659,
      "epoch": 2.2115384615384617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045102269704516793,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21835915595293046,
      "num_tokens": 50589593.0,
      "step": 24380
    },
    {
      "entropy": 5.56684250831604,
      "epoch": 2.211992017416546,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004510025412901595,
      "loss": 4.9066,
      "mean_token_accuracy": 0.2102399706840515,
      "num_tokens": 50598774.0,
      "step": 24385
    },
    {
      "entropy": 5.5032249927520756,
      "epoch": 2.2124455732946298,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045098238189537367,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21406116038560868,
      "num_tokens": 50608815.0,
      "step": 24390
    },
    {
      "entropy": 5.494785213470459,
      "epoch": 2.212899129172714,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045096221886122733,
      "loss": 5.0285,
      "mean_token_accuracy": 0.21701135933399202,
      "num_tokens": 50619255.0,
      "step": 24395
    },
    {
      "entropy": 5.5682470321655275,
      "epoch": 2.2133526850507983,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004509420521881375,
      "loss": 4.9253,
      "mean_token_accuracy": 0.21815359443426133,
      "num_tokens": 50629483.0,
      "step": 24400
    },
    {
      "entropy": 5.548216533660889,
      "epoch": 2.2138062409288826,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004509218818765211,
      "loss": 4.8681,
      "mean_token_accuracy": 0.2216430589556694,
      "num_tokens": 50638552.0,
      "step": 24405
    },
    {
      "entropy": 5.578045082092285,
      "epoch": 2.2142597968069664,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045090170792679544,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21352774649858475,
      "num_tokens": 50648752.0,
      "step": 24410
    },
    {
      "entropy": 5.580213165283203,
      "epoch": 2.2147133526850507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045088153033937765,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21593828201293946,
      "num_tokens": 50658373.0,
      "step": 24415
    },
    {
      "entropy": 5.544181299209595,
      "epoch": 2.215166908563135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045086134911468496,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21816433668136598,
      "num_tokens": 50668529.0,
      "step": 24420
    },
    {
      "entropy": 5.570469760894776,
      "epoch": 2.215620464441219,
      "grad_norm": 1.25,
      "learning_rate": 0.0004508411642531347,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21682122796773912,
      "num_tokens": 50678038.0,
      "step": 24425
    },
    {
      "entropy": 5.5718841552734375,
      "epoch": 2.2160740203193035,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004508209757551445,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21935876458883286,
      "num_tokens": 50687788.0,
      "step": 24430
    },
    {
      "entropy": 5.594445133209229,
      "epoch": 2.2165275761973877,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045080078362113153,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2059182718396187,
      "num_tokens": 50697748.0,
      "step": 24435
    },
    {
      "entropy": 5.504906702041626,
      "epoch": 2.2169811320754715,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045078058785151363,
      "loss": 4.8443,
      "mean_token_accuracy": 0.21902491003274918,
      "num_tokens": 50707776.0,
      "step": 24440
    },
    {
      "entropy": 5.534537315368652,
      "epoch": 2.217434687953556,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004507603884467085,
      "loss": 4.9992,
      "mean_token_accuracy": 0.20853210538625716,
      "num_tokens": 50717547.0,
      "step": 24445
    },
    {
      "entropy": 5.510666036605835,
      "epoch": 2.21788824383164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004507401854071336,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22325698137283326,
      "num_tokens": 50728446.0,
      "step": 24450
    },
    {
      "entropy": 5.55709867477417,
      "epoch": 2.2183417997097243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045071997873320687,
      "loss": 4.8459,
      "mean_token_accuracy": 0.21365610361099244,
      "num_tokens": 50737909.0,
      "step": 24455
    },
    {
      "entropy": 5.510608863830567,
      "epoch": 2.2187953555878086,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045069976842534635,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21610202342271806,
      "num_tokens": 50749745.0,
      "step": 24460
    },
    {
      "entropy": 5.58241982460022,
      "epoch": 2.2192489114658924,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004506795544839697,
      "loss": 4.9127,
      "mean_token_accuracy": 0.20932943969964982,
      "num_tokens": 50760073.0,
      "step": 24465
    },
    {
      "entropy": 5.608687257766723,
      "epoch": 2.2197024673439767,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004506593369094951,
      "loss": 4.9105,
      "mean_token_accuracy": 0.20908024162054062,
      "num_tokens": 50769261.0,
      "step": 24470
    },
    {
      "entropy": 5.613238573074341,
      "epoch": 2.220156023222061,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045063911570234074,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20674648582935334,
      "num_tokens": 50779572.0,
      "step": 24475
    },
    {
      "entropy": 5.5163829803466795,
      "epoch": 2.2206095791001452,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004506188908629247,
      "loss": 4.8853,
      "mean_token_accuracy": 0.21855538785457612,
      "num_tokens": 50789893.0,
      "step": 24480
    },
    {
      "entropy": 5.482339859008789,
      "epoch": 2.2210631349782295,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004505986623916652,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2160807654261589,
      "num_tokens": 50799177.0,
      "step": 24485
    },
    {
      "entropy": 5.515284681320191,
      "epoch": 2.2215166908563133,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004505784302889806,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22159485816955565,
      "num_tokens": 50810093.0,
      "step": 24490
    },
    {
      "entropy": 5.64695177078247,
      "epoch": 2.2219702467343976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045055819455528935,
      "loss": 5.0639,
      "mean_token_accuracy": 0.20666398406028746,
      "num_tokens": 50820272.0,
      "step": 24495
    },
    {
      "entropy": 5.598090410232544,
      "epoch": 2.222423802612482,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004505379551910099,
      "loss": 4.9955,
      "mean_token_accuracy": 0.21202658116817474,
      "num_tokens": 50830301.0,
      "step": 24500
    },
    {
      "entropy": 5.6368327140808105,
      "epoch": 2.222877358490566,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004505177121965608,
      "loss": 4.9907,
      "mean_token_accuracy": 0.21438877284526825,
      "num_tokens": 50839937.0,
      "step": 24505
    },
    {
      "entropy": 5.54852991104126,
      "epoch": 2.2233309143686504,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045049746557236067,
      "loss": 4.9042,
      "mean_token_accuracy": 0.22185829281806946,
      "num_tokens": 50850295.0,
      "step": 24510
    },
    {
      "entropy": 5.605092287063599,
      "epoch": 2.223784470246734,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045047721531882827,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21104562729597093,
      "num_tokens": 50861889.0,
      "step": 24515
    },
    {
      "entropy": 5.591089344024658,
      "epoch": 2.2242380261248185,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004504569614363823,
      "loss": 4.917,
      "mean_token_accuracy": 0.21555061042308807,
      "num_tokens": 50872011.0,
      "step": 24520
    },
    {
      "entropy": 5.61662449836731,
      "epoch": 2.2246915820029027,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004504367039254416,
      "loss": 4.9594,
      "mean_token_accuracy": 0.20999687314033508,
      "num_tokens": 50882559.0,
      "step": 24525
    },
    {
      "entropy": 5.5419602394104,
      "epoch": 2.225145137880987,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045041644278642523,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20472318381071092,
      "num_tokens": 50894132.0,
      "step": 24530
    },
    {
      "entropy": 5.621619796752929,
      "epoch": 2.2255986937590713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045039617801975214,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2130837544798851,
      "num_tokens": 50904337.0,
      "step": 24535
    },
    {
      "entropy": 5.63733286857605,
      "epoch": 2.2260522496371555,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045037590962584143,
      "loss": 4.9846,
      "mean_token_accuracy": 0.2069944381713867,
      "num_tokens": 50914961.0,
      "step": 24540
    },
    {
      "entropy": 5.613501644134521,
      "epoch": 2.2265058055152394,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004503556376051121,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21068182587623596,
      "num_tokens": 50925682.0,
      "step": 24545
    },
    {
      "entropy": 5.535657167434692,
      "epoch": 2.2269593613933236,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045033536195798355,
      "loss": 4.851,
      "mean_token_accuracy": 0.22736817598342896,
      "num_tokens": 50935809.0,
      "step": 24550
    },
    {
      "entropy": 5.619937372207642,
      "epoch": 2.227412917271408,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000450315082684875,
      "loss": 5.047,
      "mean_token_accuracy": 0.20100729316473007,
      "num_tokens": 50946098.0,
      "step": 24555
    },
    {
      "entropy": 5.615520238876343,
      "epoch": 2.227866473149492,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004502947997862059,
      "loss": 5.0035,
      "mean_token_accuracy": 0.21025876849889755,
      "num_tokens": 50956240.0,
      "step": 24560
    },
    {
      "entropy": 5.579733180999756,
      "epoch": 2.228320029027576,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004502745132623957,
      "loss": 4.9102,
      "mean_token_accuracy": 0.21890990436077118,
      "num_tokens": 50966870.0,
      "step": 24565
    },
    {
      "entropy": 5.560952377319336,
      "epoch": 2.2287735849056602,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004502542231138639,
      "loss": 4.9208,
      "mean_token_accuracy": 0.20925770401954652,
      "num_tokens": 50977528.0,
      "step": 24570
    },
    {
      "entropy": 5.581137466430664,
      "epoch": 2.2292271407837445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045023392934103005,
      "loss": 4.9437,
      "mean_token_accuracy": 0.20734782069921492,
      "num_tokens": 50988565.0,
      "step": 24575
    },
    {
      "entropy": 5.579733753204346,
      "epoch": 2.2296806966618288,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045021363194431397,
      "loss": 4.992,
      "mean_token_accuracy": 0.21358343511819838,
      "num_tokens": 50998410.0,
      "step": 24580
    },
    {
      "entropy": 5.603824758529663,
      "epoch": 2.230134252539913,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004501933309241352,
      "loss": 4.9775,
      "mean_token_accuracy": 0.2056296244263649,
      "num_tokens": 51008890.0,
      "step": 24585
    },
    {
      "entropy": 5.516703844070435,
      "epoch": 2.2305878084179973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004501730262809138,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2169082432985306,
      "num_tokens": 51020264.0,
      "step": 24590
    },
    {
      "entropy": 5.5458611965179445,
      "epoch": 2.231041364296081,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004501527180150696,
      "loss": 4.8414,
      "mean_token_accuracy": 0.21861355155706405,
      "num_tokens": 51029892.0,
      "step": 24595
    },
    {
      "entropy": 5.640199327468872,
      "epoch": 2.2314949201741654,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004501324061270224,
      "loss": 5.0587,
      "mean_token_accuracy": 0.20373291969299318,
      "num_tokens": 51040499.0,
      "step": 24600
    },
    {
      "entropy": 5.482412433624267,
      "epoch": 2.2319484760522497,
      "grad_norm": 1.125,
      "learning_rate": 0.0004501120906171926,
      "loss": 4.8561,
      "mean_token_accuracy": 0.21426650732755662,
      "num_tokens": 51050395.0,
      "step": 24605
    },
    {
      "entropy": 5.545130729675293,
      "epoch": 2.232402031930334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045009177148600004,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21900064647197723,
      "num_tokens": 51060316.0,
      "step": 24610
    },
    {
      "entropy": 5.520968961715698,
      "epoch": 2.232855587808418,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045007144873386514,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21584844291210176,
      "num_tokens": 51070032.0,
      "step": 24615
    },
    {
      "entropy": 5.623284912109375,
      "epoch": 2.233309143686502,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045005112236120795,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20193691551685333,
      "num_tokens": 51080738.0,
      "step": 24620
    },
    {
      "entropy": 5.575223064422607,
      "epoch": 2.2337626995645863,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000450030792368449,
      "loss": 4.8939,
      "mean_token_accuracy": 0.2171100065112114,
      "num_tokens": 51090287.0,
      "step": 24625
    },
    {
      "entropy": 5.5425323963165285,
      "epoch": 2.2342162554426706,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004500104587560086,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21029032915830612,
      "num_tokens": 51101188.0,
      "step": 24630
    },
    {
      "entropy": 5.523554563522339,
      "epoch": 2.234669811320755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004499901215243074,
      "loss": 4.8496,
      "mean_token_accuracy": 0.21467358469963074,
      "num_tokens": 51112310.0,
      "step": 24635
    },
    {
      "entropy": 5.584855222702027,
      "epoch": 2.235123367198839,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044996978067376575,
      "loss": 4.9789,
      "mean_token_accuracy": 0.20993858724832534,
      "num_tokens": 51122481.0,
      "step": 24640
    },
    {
      "entropy": 5.536005973815918,
      "epoch": 2.235576923076923,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044994943620480453,
      "loss": 4.9226,
      "mean_token_accuracy": 0.20727061331272126,
      "num_tokens": 51131918.0,
      "step": 24645
    },
    {
      "entropy": 5.561415958404541,
      "epoch": 2.236030478955007,
      "grad_norm": 1.0,
      "learning_rate": 0.00044992908811784446,
      "loss": 5.0119,
      "mean_token_accuracy": 0.19910293072462082,
      "num_tokens": 51142861.0,
      "step": 24650
    },
    {
      "entropy": 5.587594652175904,
      "epoch": 2.2364840348330914,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044990873641330615,
      "loss": 4.9828,
      "mean_token_accuracy": 0.20912633687257767,
      "num_tokens": 51153956.0,
      "step": 24655
    },
    {
      "entropy": 5.579059219360351,
      "epoch": 2.2369375907111757,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004498883810916107,
      "loss": 4.8496,
      "mean_token_accuracy": 0.22160615921020507,
      "num_tokens": 51164499.0,
      "step": 24660
    },
    {
      "entropy": 5.60390830039978,
      "epoch": 2.23739114658926,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004498680221531789,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20653716325759888,
      "num_tokens": 51175722.0,
      "step": 24665
    },
    {
      "entropy": 5.558173799514771,
      "epoch": 2.237844702467344,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004498476595984319,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21457604616880416,
      "num_tokens": 51184597.0,
      "step": 24670
    },
    {
      "entropy": 5.50778603553772,
      "epoch": 2.238298258345428,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004498272934277906,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21343494057655335,
      "num_tokens": 51194751.0,
      "step": 24675
    },
    {
      "entropy": 5.573074197769165,
      "epoch": 2.2387518142235123,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044980692364167646,
      "loss": 4.923,
      "mean_token_accuracy": 0.2212190091609955,
      "num_tokens": 51204184.0,
      "step": 24680
    },
    {
      "entropy": 5.582958459854126,
      "epoch": 2.2392053701015966,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004497865502405105,
      "loss": 4.9859,
      "mean_token_accuracy": 0.2145560324192047,
      "num_tokens": 51213761.0,
      "step": 24685
    },
    {
      "entropy": 5.572753095626831,
      "epoch": 2.239658925979681,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044976617322471424,
      "loss": 4.9087,
      "mean_token_accuracy": 0.2164181038737297,
      "num_tokens": 51224150.0,
      "step": 24690
    },
    {
      "entropy": 5.610567235946656,
      "epoch": 2.2401124818577647,
      "grad_norm": 1.171875,
      "learning_rate": 0.000449745792594709,
      "loss": 5.0188,
      "mean_token_accuracy": 0.21368137151002883,
      "num_tokens": 51235133.0,
      "step": 24695
    },
    {
      "entropy": 5.604644346237182,
      "epoch": 2.240566037735849,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044972540835091617,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21150542795658112,
      "num_tokens": 51245632.0,
      "step": 24700
    },
    {
      "entropy": 5.56734790802002,
      "epoch": 2.241019593613933,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044970502049375736,
      "loss": 4.968,
      "mean_token_accuracy": 0.21566979140043258,
      "num_tokens": 51255747.0,
      "step": 24705
    },
    {
      "entropy": 5.56240406036377,
      "epoch": 2.2414731494920175,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004496846290236542,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21618969440460206,
      "num_tokens": 51265531.0,
      "step": 24710
    },
    {
      "entropy": 5.555978298187256,
      "epoch": 2.2419267053701017,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044966423394102845,
      "loss": 4.9472,
      "mean_token_accuracy": 0.21604196578264237,
      "num_tokens": 51275319.0,
      "step": 24715
    },
    {
      "entropy": 5.575274419784546,
      "epoch": 2.2423802612481856,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044964383524630176,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2166734054684639,
      "num_tokens": 51285655.0,
      "step": 24720
    },
    {
      "entropy": 5.512102842330933,
      "epoch": 2.24283381712627,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004496234329398961,
      "loss": 4.8978,
      "mean_token_accuracy": 0.20599962621927262,
      "num_tokens": 51295938.0,
      "step": 24725
    },
    {
      "entropy": 5.594706106185913,
      "epoch": 2.243287373004354,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004496030270222333,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21997081786394118,
      "num_tokens": 51306483.0,
      "step": 24730
    },
    {
      "entropy": 5.577207040786743,
      "epoch": 2.2437409288824384,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044958261749373547,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2139873504638672,
      "num_tokens": 51316446.0,
      "step": 24735
    },
    {
      "entropy": 5.533952760696411,
      "epoch": 2.2441944847605226,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044956220435482465,
      "loss": 4.9607,
      "mean_token_accuracy": 0.20779209733009338,
      "num_tokens": 51326840.0,
      "step": 24740
    },
    {
      "entropy": 5.570036602020264,
      "epoch": 2.244648040638607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044954178760592283,
      "loss": 4.9931,
      "mean_token_accuracy": 0.2070079118013382,
      "num_tokens": 51337573.0,
      "step": 24745
    },
    {
      "entropy": 5.621497583389282,
      "epoch": 2.2451015965166907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044952136724745254,
      "loss": 4.9649,
      "mean_token_accuracy": 0.2153195932507515,
      "num_tokens": 51347827.0,
      "step": 24750
    },
    {
      "entropy": 5.55088438987732,
      "epoch": 2.245555152394775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004495009432798358,
      "loss": 4.925,
      "mean_token_accuracy": 0.2125857651233673,
      "num_tokens": 51358396.0,
      "step": 24755
    },
    {
      "entropy": 5.608096933364868,
      "epoch": 2.2460087082728593,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044948051570349503,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20708179622888565,
      "num_tokens": 51368882.0,
      "step": 24760
    },
    {
      "entropy": 5.584758567810058,
      "epoch": 2.2464622641509435,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004494600845188528,
      "loss": 4.9998,
      "mean_token_accuracy": 0.2085038587450981,
      "num_tokens": 51378717.0,
      "step": 24765
    },
    {
      "entropy": 5.562956714630127,
      "epoch": 2.2469158200290273,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044943964972633154,
      "loss": 4.9331,
      "mean_token_accuracy": 0.2069125398993492,
      "num_tokens": 51389270.0,
      "step": 24770
    },
    {
      "entropy": 5.55361909866333,
      "epoch": 2.2473693759071116,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004494192113263539,
      "loss": 4.952,
      "mean_token_accuracy": 0.21096873581409453,
      "num_tokens": 51400025.0,
      "step": 24775
    },
    {
      "entropy": 5.5834801197052,
      "epoch": 2.247822931785196,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044939876931934255,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2143678143620491,
      "num_tokens": 51410903.0,
      "step": 24780
    },
    {
      "entropy": 5.543683385848999,
      "epoch": 2.24827648766328,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004493783237057201,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2178712785243988,
      "num_tokens": 51420246.0,
      "step": 24785
    },
    {
      "entropy": 5.6401002407073975,
      "epoch": 2.2487300435413644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004493578744859095,
      "loss": 5.0649,
      "mean_token_accuracy": 0.2057617723941803,
      "num_tokens": 51431309.0,
      "step": 24790
    },
    {
      "entropy": 5.633981227874756,
      "epoch": 2.2491835994194487,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004493374216603338,
      "loss": 5.008,
      "mean_token_accuracy": 0.2052177146077156,
      "num_tokens": 51443142.0,
      "step": 24795
    },
    {
      "entropy": 5.68134036064148,
      "epoch": 2.2496371552975325,
      "grad_norm": 1.5,
      "learning_rate": 0.0004493169652294156,
      "loss": 5.158,
      "mean_token_accuracy": 0.19841213375329972,
      "num_tokens": 51454499.0,
      "step": 24800
    },
    {
      "entropy": 5.557585763931274,
      "epoch": 2.2500907111756168,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044929650519357823,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21476107835769653,
      "num_tokens": 51464667.0,
      "step": 24805
    },
    {
      "entropy": 5.621891021728516,
      "epoch": 2.250544267053701,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004492760415532447,
      "loss": 4.9671,
      "mean_token_accuracy": 0.21079372763633727,
      "num_tokens": 51473627.0,
      "step": 24810
    },
    {
      "entropy": 5.571342897415161,
      "epoch": 2.2509978229317853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004492555743088382,
      "loss": 4.815,
      "mean_token_accuracy": 0.21659711450338365,
      "num_tokens": 51483052.0,
      "step": 24815
    },
    {
      "entropy": 5.5308513164520265,
      "epoch": 2.2514513788098696,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000449235103460782,
      "loss": 4.9526,
      "mean_token_accuracy": 0.205110627412796,
      "num_tokens": 51492422.0,
      "step": 24820
    },
    {
      "entropy": 5.516646051406861,
      "epoch": 2.2519049346879534,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004492146290094995,
      "loss": 4.9061,
      "mean_token_accuracy": 0.21152358651161193,
      "num_tokens": 51503566.0,
      "step": 24825
    },
    {
      "entropy": 5.539857292175293,
      "epoch": 2.2523584905660377,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044919415095541406,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21697582751512529,
      "num_tokens": 51513699.0,
      "step": 24830
    },
    {
      "entropy": 5.566490268707275,
      "epoch": 2.252812046444122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044917366929894914,
      "loss": 4.9397,
      "mean_token_accuracy": 0.21021585911512375,
      "num_tokens": 51524745.0,
      "step": 24835
    },
    {
      "entropy": 5.557812213897705,
      "epoch": 2.253265602322206,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044915318404052846,
      "loss": 4.8814,
      "mean_token_accuracy": 0.2129119500517845,
      "num_tokens": 51536516.0,
      "step": 24840
    },
    {
      "entropy": 5.608424949645996,
      "epoch": 2.2537191582002905,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044913269518057547,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20357286930084229,
      "num_tokens": 51547063.0,
      "step": 24845
    },
    {
      "entropy": 5.580503749847412,
      "epoch": 2.2541727140783743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000449112202719514,
      "loss": 4.9343,
      "mean_token_accuracy": 0.20828542709350586,
      "num_tokens": 51557244.0,
      "step": 24850
    },
    {
      "entropy": 5.6066258430480955,
      "epoch": 2.2546262699564585,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004490917066577678,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2110108882188797,
      "num_tokens": 51567630.0,
      "step": 24855
    },
    {
      "entropy": 5.545233631134034,
      "epoch": 2.255079825834543,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044907120699576075,
      "loss": 4.9776,
      "mean_token_accuracy": 0.21206414699554443,
      "num_tokens": 51577951.0,
      "step": 24860
    },
    {
      "entropy": 5.606132650375367,
      "epoch": 2.255533381712627,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044905070373391684,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21519958674907685,
      "num_tokens": 51587533.0,
      "step": 24865
    },
    {
      "entropy": 5.653391981124878,
      "epoch": 2.2559869375907113,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004490301968726599,
      "loss": 4.9249,
      "mean_token_accuracy": 0.22210706323385238,
      "num_tokens": 51597212.0,
      "step": 24870
    },
    {
      "entropy": 5.57596230506897,
      "epoch": 2.256440493468795,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044900968641241426,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21647840589284897,
      "num_tokens": 51608520.0,
      "step": 24875
    },
    {
      "entropy": 5.585930585861206,
      "epoch": 2.2568940493468794,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004489891723536039,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20756766498088836,
      "num_tokens": 51619651.0,
      "step": 24880
    },
    {
      "entropy": 5.539716196060181,
      "epoch": 2.2573476052249637,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004489686546966531,
      "loss": 4.849,
      "mean_token_accuracy": 0.21233907490968704,
      "num_tokens": 51630184.0,
      "step": 24885
    },
    {
      "entropy": 5.499330997467041,
      "epoch": 2.257801161103048,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044894813344198626,
      "loss": 4.8481,
      "mean_token_accuracy": 0.21994235813617707,
      "num_tokens": 51641293.0,
      "step": 24890
    },
    {
      "entropy": 5.581316947937012,
      "epoch": 2.2582547169811322,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004489276085900276,
      "loss": 4.945,
      "mean_token_accuracy": 0.21121911704540253,
      "num_tokens": 51652002.0,
      "step": 24895
    },
    {
      "entropy": 5.49250864982605,
      "epoch": 2.2587082728592165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044890708014120174,
      "loss": 4.8225,
      "mean_token_accuracy": 0.227053464949131,
      "num_tokens": 51661668.0,
      "step": 24900
    },
    {
      "entropy": 5.619518613815307,
      "epoch": 2.2591618287373003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004488865480959331,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20568533092737198,
      "num_tokens": 51672879.0,
      "step": 24905
    },
    {
      "entropy": 5.578426885604858,
      "epoch": 2.2596153846153846,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004488660124546464,
      "loss": 4.9223,
      "mean_token_accuracy": 0.21344517469406127,
      "num_tokens": 51683878.0,
      "step": 24910
    },
    {
      "entropy": 5.5278290748596195,
      "epoch": 2.260068940493469,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004488454732177662,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2105911359190941,
      "num_tokens": 51693971.0,
      "step": 24915
    },
    {
      "entropy": 5.551317501068115,
      "epoch": 2.260522496371553,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044882493038571737,
      "loss": 4.8475,
      "mean_token_accuracy": 0.22081339955329896,
      "num_tokens": 51703685.0,
      "step": 24920
    },
    {
      "entropy": 5.635507583618164,
      "epoch": 2.260976052249637,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044880438395892457,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2090151622891426,
      "num_tokens": 51714469.0,
      "step": 24925
    },
    {
      "entropy": 5.596588516235352,
      "epoch": 2.261429608127721,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004487838339378129,
      "loss": 5.0244,
      "mean_token_accuracy": 0.20091243386268615,
      "num_tokens": 51725283.0,
      "step": 24930
    },
    {
      "entropy": 5.5576558113098145,
      "epoch": 2.2618831640058055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004487632803228072,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20955396443605423,
      "num_tokens": 51735628.0,
      "step": 24935
    },
    {
      "entropy": 5.463111448287964,
      "epoch": 2.2623367198838897,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044874272311433265,
      "loss": 4.8174,
      "mean_token_accuracy": 0.22551854848861694,
      "num_tokens": 51745991.0,
      "step": 24940
    },
    {
      "entropy": 5.517724275588989,
      "epoch": 2.262790275761974,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004487221623128143,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2159314587712288,
      "num_tokens": 51756306.0,
      "step": 24945
    },
    {
      "entropy": 5.593941879272461,
      "epoch": 2.2632438316400583,
      "grad_norm": 1.125,
      "learning_rate": 0.00044870159791867743,
      "loss": 4.9835,
      "mean_token_accuracy": 0.21535757929086685,
      "num_tokens": 51765466.0,
      "step": 24950
    },
    {
      "entropy": 5.592212200164795,
      "epoch": 2.263697387518142,
      "grad_norm": 1.125,
      "learning_rate": 0.00044868102993234716,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21097905188798904,
      "num_tokens": 51775526.0,
      "step": 24955
    },
    {
      "entropy": 5.590959644317627,
      "epoch": 2.2641509433962264,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044866045835424896,
      "loss": 4.9512,
      "mean_token_accuracy": 0.20720784664154052,
      "num_tokens": 51785434.0,
      "step": 24960
    },
    {
      "entropy": 5.5903346061706545,
      "epoch": 2.2646044992743106,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004486398831848082,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2029450610280037,
      "num_tokens": 51796047.0,
      "step": 24965
    },
    {
      "entropy": 5.624813604354858,
      "epoch": 2.265058055152395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044861930442445037,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20542479157447815,
      "num_tokens": 51806173.0,
      "step": 24970
    },
    {
      "entropy": 5.6734966278076175,
      "epoch": 2.2655116110304787,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004485987220736012,
      "loss": 5.1028,
      "mean_token_accuracy": 0.20142967998981476,
      "num_tokens": 51817644.0,
      "step": 24975
    },
    {
      "entropy": 5.6251873016357425,
      "epoch": 2.265965166908563,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004485781361326861,
      "loss": 5.0719,
      "mean_token_accuracy": 0.20615520179271699,
      "num_tokens": 51829746.0,
      "step": 24980
    },
    {
      "entropy": 5.575164556503296,
      "epoch": 2.2664187227866472,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000448557546602131,
      "loss": 5.0051,
      "mean_token_accuracy": 0.20462606251239776,
      "num_tokens": 51840868.0,
      "step": 24985
    },
    {
      "entropy": 5.597962856292725,
      "epoch": 2.2668722786647315,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044853695348236156,
      "loss": 4.8825,
      "mean_token_accuracy": 0.2149131953716278,
      "num_tokens": 51851385.0,
      "step": 24990
    },
    {
      "entropy": 5.585072469711304,
      "epoch": 2.2673258345428158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044851635677380374,
      "loss": 4.915,
      "mean_token_accuracy": 0.20616335272789002,
      "num_tokens": 51861021.0,
      "step": 24995
    },
    {
      "entropy": 5.59707498550415,
      "epoch": 2.2677793904209,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004484957564768834,
      "loss": 5.0089,
      "mean_token_accuracy": 0.2096236154437065,
      "num_tokens": 51870688.0,
      "step": 25000
    },
    {
      "entropy": 5.583831310272217,
      "epoch": 2.268232946298984,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044847515259202657,
      "loss": 4.9627,
      "mean_token_accuracy": 0.20856474786996843,
      "num_tokens": 51880439.0,
      "step": 25005
    },
    {
      "entropy": 5.505197811126709,
      "epoch": 2.268686502177068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044845454511965947,
      "loss": 4.9315,
      "mean_token_accuracy": 0.2147160291671753,
      "num_tokens": 51890861.0,
      "step": 25010
    },
    {
      "entropy": 5.592146205902099,
      "epoch": 2.2691400580551524,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044843393406020823,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21316009610891343,
      "num_tokens": 51900687.0,
      "step": 25015
    },
    {
      "entropy": 5.636794376373291,
      "epoch": 2.2695936139332367,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044841331941409884,
      "loss": 5.0809,
      "mean_token_accuracy": 0.19589564800262452,
      "num_tokens": 51911805.0,
      "step": 25020
    },
    {
      "entropy": 5.581126880645752,
      "epoch": 2.270047169811321,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044839270118175793,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2211558312177658,
      "num_tokens": 51922520.0,
      "step": 25025
    },
    {
      "entropy": 5.6436036109924315,
      "epoch": 2.2705007256894048,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004483720793636118,
      "loss": 5.0192,
      "mean_token_accuracy": 0.2037653461098671,
      "num_tokens": 51934056.0,
      "step": 25030
    },
    {
      "entropy": 5.606632852554322,
      "epoch": 2.270954281567489,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044835145396008685,
      "loss": 5.0042,
      "mean_token_accuracy": 0.2117276072502136,
      "num_tokens": 51944696.0,
      "step": 25035
    },
    {
      "entropy": 5.610119152069092,
      "epoch": 2.2714078374455733,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004483308249716097,
      "loss": 5.0026,
      "mean_token_accuracy": 0.21197018772363663,
      "num_tokens": 51955111.0,
      "step": 25040
    },
    {
      "entropy": 5.563672399520874,
      "epoch": 2.2718613933236576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004483101923986068,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21111331433057784,
      "num_tokens": 51966036.0,
      "step": 25045
    },
    {
      "entropy": 5.5305970191955565,
      "epoch": 2.272314949201742,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044828955624150505,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21775335520505906,
      "num_tokens": 51976060.0,
      "step": 25050
    },
    {
      "entropy": 5.594537305831909,
      "epoch": 2.2727685050798256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004482689165007311,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21135286688804628,
      "num_tokens": 51986016.0,
      "step": 25055
    },
    {
      "entropy": 5.552269363403321,
      "epoch": 2.27322206095791,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044824827317671177,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22400961071252823,
      "num_tokens": 51994794.0,
      "step": 25060
    },
    {
      "entropy": 5.499716854095459,
      "epoch": 2.273675616835994,
      "grad_norm": 1.109375,
      "learning_rate": 0.000448227626269874,
      "loss": 4.8706,
      "mean_token_accuracy": 0.2188450112938881,
      "num_tokens": 52004976.0,
      "step": 25065
    },
    {
      "entropy": 5.589347171783447,
      "epoch": 2.2741291727140784,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044820697578064485,
      "loss": 4.9807,
      "mean_token_accuracy": 0.20299387872219085,
      "num_tokens": 52016380.0,
      "step": 25070
    },
    {
      "entropy": 5.6079377174377445,
      "epoch": 2.2745827285921627,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004481863217094513,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21439362019300462,
      "num_tokens": 52026439.0,
      "step": 25075
    },
    {
      "entropy": 5.564224910736084,
      "epoch": 2.2750362844702465,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004481656640567204,
      "loss": 4.8869,
      "mean_token_accuracy": 0.21509212106466294,
      "num_tokens": 52037111.0,
      "step": 25080
    },
    {
      "entropy": 5.531151485443115,
      "epoch": 2.275489840348331,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004481450028228795,
      "loss": 4.944,
      "mean_token_accuracy": 0.21160723865032197,
      "num_tokens": 52047463.0,
      "step": 25085
    },
    {
      "entropy": 5.566102647781372,
      "epoch": 2.275943396226415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004481243380083558,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20788409560918808,
      "num_tokens": 52058081.0,
      "step": 25090
    },
    {
      "entropy": 5.564450120925903,
      "epoch": 2.2763969521044993,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004481036696135767,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21422482430934905,
      "num_tokens": 52069367.0,
      "step": 25095
    },
    {
      "entropy": 5.593443250656128,
      "epoch": 2.2768505079825836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004480829976389695,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21186336874961853,
      "num_tokens": 52080023.0,
      "step": 25100
    },
    {
      "entropy": 5.5127521514892575,
      "epoch": 2.277304063860668,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044806232208496194,
      "loss": 4.965,
      "mean_token_accuracy": 0.21391810327768326,
      "num_tokens": 52089184.0,
      "step": 25105
    },
    {
      "entropy": 5.615357303619385,
      "epoch": 2.2777576197387517,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004480416429519814,
      "loss": 5.0102,
      "mean_token_accuracy": 0.20993676781654358,
      "num_tokens": 52098711.0,
      "step": 25110
    },
    {
      "entropy": 5.608211612701416,
      "epoch": 2.278211175616836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004480209602404556,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21644964814186096,
      "num_tokens": 52108826.0,
      "step": 25115
    },
    {
      "entropy": 5.5315389156341555,
      "epoch": 2.27866473149492,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004480002739508122,
      "loss": 4.8865,
      "mean_token_accuracy": 0.22185250818729402,
      "num_tokens": 52118934.0,
      "step": 25120
    },
    {
      "entropy": 5.5315416812896725,
      "epoch": 2.2791182873730045,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004479795840834791,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22009823471307755,
      "num_tokens": 52129803.0,
      "step": 25125
    },
    {
      "entropy": 5.482956600189209,
      "epoch": 2.2795718432510883,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044795889063888413,
      "loss": 4.8142,
      "mean_token_accuracy": 0.22065597474575044,
      "num_tokens": 52140070.0,
      "step": 25130
    },
    {
      "entropy": 5.682845640182495,
      "epoch": 2.2800253991291726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004479381936174552,
      "loss": 5.0843,
      "mean_token_accuracy": 0.20483203381299972,
      "num_tokens": 52150462.0,
      "step": 25135
    },
    {
      "entropy": 5.65266170501709,
      "epoch": 2.280478955007257,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004479174930196204,
      "loss": 4.9609,
      "mean_token_accuracy": 0.2097917303442955,
      "num_tokens": 52160552.0,
      "step": 25140
    },
    {
      "entropy": 5.487208604812622,
      "epoch": 2.280932510885341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044789678884580767,
      "loss": 4.8523,
      "mean_token_accuracy": 0.21823084056377412,
      "num_tokens": 52169780.0,
      "step": 25145
    },
    {
      "entropy": 5.578747701644898,
      "epoch": 2.2813860667634254,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044787608109644536,
      "loss": 4.9735,
      "mean_token_accuracy": 0.21678511202335357,
      "num_tokens": 52180055.0,
      "step": 25150
    },
    {
      "entropy": 5.577519845962525,
      "epoch": 2.2818396226415096,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004478553697719616,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21233561933040618,
      "num_tokens": 52189772.0,
      "step": 25155
    },
    {
      "entropy": 5.558465003967285,
      "epoch": 2.2822931785195935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004478346548727848,
      "loss": 4.8662,
      "mean_token_accuracy": 0.22545824497938155,
      "num_tokens": 52199608.0,
      "step": 25160
    },
    {
      "entropy": 5.523789119720459,
      "epoch": 2.2827467343976777,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004478139363993433,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2116127207875252,
      "num_tokens": 52208236.0,
      "step": 25165
    },
    {
      "entropy": 5.589986658096313,
      "epoch": 2.283200290275762,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004477932143520655,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21422459483146666,
      "num_tokens": 52218403.0,
      "step": 25170
    },
    {
      "entropy": 5.614393663406372,
      "epoch": 2.2836538461538463,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044777248873138005,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21445739418268203,
      "num_tokens": 52228255.0,
      "step": 25175
    },
    {
      "entropy": 5.5787577629089355,
      "epoch": 2.2841074020319305,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044775175953771547,
      "loss": 4.9651,
      "mean_token_accuracy": 0.20945563316345214,
      "num_tokens": 52237732.0,
      "step": 25180
    },
    {
      "entropy": 5.664718961715698,
      "epoch": 2.2845609579100143,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004477310267715005,
      "loss": 5.1079,
      "mean_token_accuracy": 0.2007945492863655,
      "num_tokens": 52249226.0,
      "step": 25185
    },
    {
      "entropy": 5.631059408187866,
      "epoch": 2.2850145137880986,
      "grad_norm": 1.125,
      "learning_rate": 0.0004477102904331639,
      "loss": 4.937,
      "mean_token_accuracy": 0.2161140561103821,
      "num_tokens": 52259550.0,
      "step": 25190
    },
    {
      "entropy": 5.632875394821167,
      "epoch": 2.285468069666183,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044768955052313443,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20975142866373062,
      "num_tokens": 52269670.0,
      "step": 25195
    },
    {
      "entropy": 5.618857431411743,
      "epoch": 2.285921625544267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044766880704184104,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2161665678024292,
      "num_tokens": 52280135.0,
      "step": 25200
    },
    {
      "entropy": 5.545126342773438,
      "epoch": 2.2863751814223514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004476480599897128,
      "loss": 4.8653,
      "mean_token_accuracy": 0.21941546797752381,
      "num_tokens": 52291806.0,
      "step": 25205
    },
    {
      "entropy": 5.500838422775269,
      "epoch": 2.2868287373004352,
      "grad_norm": 1.125,
      "learning_rate": 0.0004476273093671786,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2143415242433548,
      "num_tokens": 52302118.0,
      "step": 25210
    },
    {
      "entropy": 5.600105094909668,
      "epoch": 2.2872822931785195,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044760655517466766,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2031088277697563,
      "num_tokens": 52312762.0,
      "step": 25215
    },
    {
      "entropy": 5.611731958389282,
      "epoch": 2.2877358490566038,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044758579741260916,
      "loss": 4.9785,
      "mean_token_accuracy": 0.2034713312983513,
      "num_tokens": 52323367.0,
      "step": 25220
    },
    {
      "entropy": 5.64713225364685,
      "epoch": 2.288189404934688,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044756503608143247,
      "loss": 5.0154,
      "mean_token_accuracy": 0.20669188350439072,
      "num_tokens": 52335231.0,
      "step": 25225
    },
    {
      "entropy": 5.631257104873657,
      "epoch": 2.2886429608127723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044754427118156675,
      "loss": 4.895,
      "mean_token_accuracy": 0.21595149636268615,
      "num_tokens": 52345066.0,
      "step": 25230
    },
    {
      "entropy": 5.583748435974121,
      "epoch": 2.289096516690856,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004475235027134416,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21480559706687927,
      "num_tokens": 52356383.0,
      "step": 25235
    },
    {
      "entropy": 5.626114797592163,
      "epoch": 2.2895500725689404,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004475027306774865,
      "loss": 5.058,
      "mean_token_accuracy": 0.21025279760360718,
      "num_tokens": 52366013.0,
      "step": 25240
    },
    {
      "entropy": 5.597537517547607,
      "epoch": 2.2900036284470247,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044748195507413095,
      "loss": 4.9891,
      "mean_token_accuracy": 0.21081256568431855,
      "num_tokens": 52376298.0,
      "step": 25245
    },
    {
      "entropy": 5.483092641830444,
      "epoch": 2.290457184325109,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044746117590380456,
      "loss": 4.8581,
      "mean_token_accuracy": 0.214636367559433,
      "num_tokens": 52387134.0,
      "step": 25250
    },
    {
      "entropy": 5.540257787704467,
      "epoch": 2.290910740203193,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004474403931669371,
      "loss": 4.8492,
      "mean_token_accuracy": 0.21721140146255494,
      "num_tokens": 52397389.0,
      "step": 25255
    },
    {
      "entropy": 5.5425032615661625,
      "epoch": 2.2913642960812775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004474196068639584,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2169995740056038,
      "num_tokens": 52407456.0,
      "step": 25260
    },
    {
      "entropy": 5.502102088928223,
      "epoch": 2.2918178519593613,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004473988169952983,
      "loss": 4.9571,
      "mean_token_accuracy": 0.21231054067611693,
      "num_tokens": 52417967.0,
      "step": 25265
    },
    {
      "entropy": 5.617834186553955,
      "epoch": 2.2922714078374455,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044737802356138674,
      "loss": 5.0111,
      "mean_token_accuracy": 0.21403928399085997,
      "num_tokens": 52428757.0,
      "step": 25270
    },
    {
      "entropy": 5.58148341178894,
      "epoch": 2.29272496371553,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004473572265626537,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2188520073890686,
      "num_tokens": 52438973.0,
      "step": 25275
    },
    {
      "entropy": 5.55060076713562,
      "epoch": 2.293178519593614,
      "grad_norm": 1.125,
      "learning_rate": 0.0004473364259995293,
      "loss": 4.8986,
      "mean_token_accuracy": 0.21128974258899688,
      "num_tokens": 52449323.0,
      "step": 25280
    },
    {
      "entropy": 5.576638221740723,
      "epoch": 2.293632075471698,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004473156218724436,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21800605207681656,
      "num_tokens": 52459299.0,
      "step": 25285
    },
    {
      "entropy": 5.477514743804932,
      "epoch": 2.294085631349782,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044729481418182704,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21173993051052092,
      "num_tokens": 52470026.0,
      "step": 25290
    },
    {
      "entropy": 5.584178495407104,
      "epoch": 2.2945391872278664,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044727400292810985,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21646443158388137,
      "num_tokens": 52479975.0,
      "step": 25295
    },
    {
      "entropy": 5.638037490844726,
      "epoch": 2.2949927431059507,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044725318811172227,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20517731010913848,
      "num_tokens": 52491067.0,
      "step": 25300
    },
    {
      "entropy": 5.529987907409668,
      "epoch": 2.295446298984035,
      "grad_norm": 1.109375,
      "learning_rate": 0.000447232369733095,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22039924412965775,
      "num_tokens": 52500671.0,
      "step": 25305
    },
    {
      "entropy": 5.567880916595459,
      "epoch": 2.2958998548621192,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004472115477926583,
      "loss": 4.8843,
      "mean_token_accuracy": 0.21178943365812303,
      "num_tokens": 52509910.0,
      "step": 25310
    },
    {
      "entropy": 5.56077823638916,
      "epoch": 2.296353410740203,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044719072229084295,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22036605179309846,
      "num_tokens": 52520517.0,
      "step": 25315
    },
    {
      "entropy": 5.552681875228882,
      "epoch": 2.2968069666182873,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004471698932280796,
      "loss": 4.9811,
      "mean_token_accuracy": 0.21290714740753175,
      "num_tokens": 52530760.0,
      "step": 25320
    },
    {
      "entropy": 5.6255382061004635,
      "epoch": 2.2972605224963716,
      "grad_norm": 1.1875,
      "learning_rate": 0.000447149060604799,
      "loss": 5.0208,
      "mean_token_accuracy": 0.20467558205127717,
      "num_tokens": 52541791.0,
      "step": 25325
    },
    {
      "entropy": 5.683467864990234,
      "epoch": 2.297714078374456,
      "grad_norm": 1.125,
      "learning_rate": 0.00044712822442143194,
      "loss": 4.9881,
      "mean_token_accuracy": 0.2148816168308258,
      "num_tokens": 52551885.0,
      "step": 25330
    },
    {
      "entropy": 5.559752464294434,
      "epoch": 2.2981676342525397,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044710738467840927,
      "loss": 4.89,
      "mean_token_accuracy": 0.214259734749794,
      "num_tokens": 52563080.0,
      "step": 25335
    },
    {
      "entropy": 5.5173670768737795,
      "epoch": 2.298621190130624,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004470865413761621,
      "loss": 4.8557,
      "mean_token_accuracy": 0.22494282871484755,
      "num_tokens": 52573564.0,
      "step": 25340
    },
    {
      "entropy": 5.560221719741821,
      "epoch": 2.299074746008708,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044706569451512143,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21242703199386598,
      "num_tokens": 52583801.0,
      "step": 25345
    },
    {
      "entropy": 5.5933174133300785,
      "epoch": 2.2995283018867925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044704484409571825,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2079051613807678,
      "num_tokens": 52594908.0,
      "step": 25350
    },
    {
      "entropy": 5.555398559570312,
      "epoch": 2.2999818577648767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044702399011838393,
      "loss": 4.8865,
      "mean_token_accuracy": 0.20889928340911865,
      "num_tokens": 52604861.0,
      "step": 25355
    },
    {
      "entropy": 5.5551429271698,
      "epoch": 2.300435413642961,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004470031325835497,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2133139193058014,
      "num_tokens": 52616203.0,
      "step": 25360
    },
    {
      "entropy": 5.527337265014649,
      "epoch": 2.300888969521045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004469822714916468,
      "loss": 4.8687,
      "mean_token_accuracy": 0.21441444605588914,
      "num_tokens": 52625838.0,
      "step": 25365
    },
    {
      "entropy": 5.484132194519043,
      "epoch": 2.301342525399129,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004469614068431067,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2139444500207901,
      "num_tokens": 52635741.0,
      "step": 25370
    },
    {
      "entropy": 5.520980930328369,
      "epoch": 2.3017960812772134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004469405386383608,
      "loss": 4.8594,
      "mean_token_accuracy": 0.21493380516767502,
      "num_tokens": 52645948.0,
      "step": 25375
    },
    {
      "entropy": 5.585295009613037,
      "epoch": 2.3022496371552976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044691966687784087,
      "loss": 4.9343,
      "mean_token_accuracy": 0.21457646936178207,
      "num_tokens": 52655073.0,
      "step": 25380
    },
    {
      "entropy": 5.547964763641358,
      "epoch": 2.302703193033382,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044689879156197833,
      "loss": 4.904,
      "mean_token_accuracy": 0.2276657536625862,
      "num_tokens": 52664429.0,
      "step": 25385
    },
    {
      "entropy": 5.62707109451294,
      "epoch": 2.3031567489114657,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044687791269120504,
      "loss": 5.0123,
      "mean_token_accuracy": 0.2093101218342781,
      "num_tokens": 52674890.0,
      "step": 25390
    },
    {
      "entropy": 5.681874513626099,
      "epoch": 2.30361030478955,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044685703026595266,
      "loss": 5.0301,
      "mean_token_accuracy": 0.2142328441143036,
      "num_tokens": 52685673.0,
      "step": 25395
    },
    {
      "entropy": 5.667135286331177,
      "epoch": 2.3040638606676342,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044683614428665314,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20860655307769777,
      "num_tokens": 52696696.0,
      "step": 25400
    },
    {
      "entropy": 5.587316799163818,
      "epoch": 2.3045174165457185,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044681525475373843,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20681025832891464,
      "num_tokens": 52707662.0,
      "step": 25405
    },
    {
      "entropy": 5.542032146453858,
      "epoch": 2.3049709724238028,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004467943616676403,
      "loss": 4.9186,
      "mean_token_accuracy": 0.217025026679039,
      "num_tokens": 52718438.0,
      "step": 25410
    },
    {
      "entropy": 5.538061571121216,
      "epoch": 2.3054245283018866,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004467734650287911,
      "loss": 4.9106,
      "mean_token_accuracy": 0.21316201984882355,
      "num_tokens": 52729267.0,
      "step": 25415
    },
    {
      "entropy": 5.536345815658569,
      "epoch": 2.305878084179971,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004467525648376229,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22299905866384506,
      "num_tokens": 52738754.0,
      "step": 25420
    },
    {
      "entropy": 5.513626670837402,
      "epoch": 2.306331640058055,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004467316610945679,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21875788271427155,
      "num_tokens": 52749086.0,
      "step": 25425
    },
    {
      "entropy": 5.627924585342408,
      "epoch": 2.3067851959361394,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044671075380005825,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20999130755662918,
      "num_tokens": 52758961.0,
      "step": 25430
    },
    {
      "entropy": 5.613173151016236,
      "epoch": 2.3072387518142237,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004466898429545266,
      "loss": 4.9584,
      "mean_token_accuracy": 0.21569590419530868,
      "num_tokens": 52769447.0,
      "step": 25435
    },
    {
      "entropy": 5.646428394317627,
      "epoch": 2.3076923076923075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044666892855840513,
      "loss": 5.0137,
      "mean_token_accuracy": 0.2048293337225914,
      "num_tokens": 52780306.0,
      "step": 25440
    },
    {
      "entropy": 5.551281499862671,
      "epoch": 2.3081458635703918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004466480106121266,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2122344419360161,
      "num_tokens": 52791263.0,
      "step": 25445
    },
    {
      "entropy": 5.594593191146851,
      "epoch": 2.308599419448476,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004466270891161234,
      "loss": 4.9433,
      "mean_token_accuracy": 0.2077154964208603,
      "num_tokens": 52802194.0,
      "step": 25450
    },
    {
      "entropy": 5.5721752643585205,
      "epoch": 2.3090529753265603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044660616407082827,
      "loss": 4.9162,
      "mean_token_accuracy": 0.22245496213436128,
      "num_tokens": 52812391.0,
      "step": 25455
    },
    {
      "entropy": 5.605469846725464,
      "epoch": 2.3095065312046446,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044658523547667385,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20990120321512223,
      "num_tokens": 52823540.0,
      "step": 25460
    },
    {
      "entropy": 5.529719972610474,
      "epoch": 2.309960087082729,
      "grad_norm": 1.125,
      "learning_rate": 0.0004465643033340931,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21658777296543122,
      "num_tokens": 52833238.0,
      "step": 25465
    },
    {
      "entropy": 5.609346723556518,
      "epoch": 2.3104136429608126,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044654336764351886,
      "loss": 5.0096,
      "mean_token_accuracy": 0.20375905334949493,
      "num_tokens": 52844510.0,
      "step": 25470
    },
    {
      "entropy": 5.580329370498657,
      "epoch": 2.310867198838897,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044652242840538406,
      "loss": 4.9805,
      "mean_token_accuracy": 0.21358157694339752,
      "num_tokens": 52856311.0,
      "step": 25475
    },
    {
      "entropy": 5.528303480148315,
      "epoch": 2.311320754716981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004465014856201218,
      "loss": 4.8753,
      "mean_token_accuracy": 0.22072956562042237,
      "num_tokens": 52866295.0,
      "step": 25480
    },
    {
      "entropy": 5.583780813217163,
      "epoch": 2.3117743105950654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000446480539288165,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21221563816070557,
      "num_tokens": 52876818.0,
      "step": 25485
    },
    {
      "entropy": 5.623132562637329,
      "epoch": 2.3122278664731493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004464595894099469,
      "loss": 4.896,
      "mean_token_accuracy": 0.21328379809856415,
      "num_tokens": 52886173.0,
      "step": 25490
    },
    {
      "entropy": 5.517612457275391,
      "epoch": 2.3126814223512335,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044643863598590094,
      "loss": 4.9182,
      "mean_token_accuracy": 0.20874154418706894,
      "num_tokens": 52895875.0,
      "step": 25495
    },
    {
      "entropy": 5.586219549179077,
      "epoch": 2.313134978229318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044641767901646025,
      "loss": 5.0206,
      "mean_token_accuracy": 0.20557665228843688,
      "num_tokens": 52906465.0,
      "step": 25500
    },
    {
      "entropy": 5.626094198226928,
      "epoch": 2.313588534107402,
      "grad_norm": 1.25,
      "learning_rate": 0.0004463967185020583,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21509349197149277,
      "num_tokens": 52916378.0,
      "step": 25505
    },
    {
      "entropy": 5.608964347839356,
      "epoch": 2.3140420899854863,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004463757544431285,
      "loss": 4.9886,
      "mean_token_accuracy": 0.21702294945716857,
      "num_tokens": 52926698.0,
      "step": 25510
    },
    {
      "entropy": 5.637996673583984,
      "epoch": 2.3144956458635706,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004463547868401044,
      "loss": 4.9439,
      "mean_token_accuracy": 0.20812619775533675,
      "num_tokens": 52936469.0,
      "step": 25515
    },
    {
      "entropy": 5.615101003646851,
      "epoch": 2.3149492017416544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044633381569341964,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21329115629196166,
      "num_tokens": 52947006.0,
      "step": 25520
    },
    {
      "entropy": 5.484480094909668,
      "epoch": 2.3154027576197387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004463128410035079,
      "loss": 4.8253,
      "mean_token_accuracy": 0.2190695196390152,
      "num_tokens": 52956663.0,
      "step": 25525
    },
    {
      "entropy": 5.544159984588623,
      "epoch": 2.315856313497823,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044629186277080307,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2122292175889015,
      "num_tokens": 52967383.0,
      "step": 25530
    },
    {
      "entropy": 5.564082908630371,
      "epoch": 2.316309869375907,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044627088099573884,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22225950062274932,
      "num_tokens": 52976567.0,
      "step": 25535
    },
    {
      "entropy": 5.544712543487549,
      "epoch": 2.3167634252539915,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044624989567874907,
      "loss": 4.938,
      "mean_token_accuracy": 0.21366299390792848,
      "num_tokens": 52986159.0,
      "step": 25540
    },
    {
      "entropy": 5.560599517822266,
      "epoch": 2.3172169811320753,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044622890682026787,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20261744856834413,
      "num_tokens": 52996028.0,
      "step": 25545
    },
    {
      "entropy": 5.6129392147064205,
      "epoch": 2.3176705370101596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004462079144207292,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21037234514951705,
      "num_tokens": 53006446.0,
      "step": 25550
    },
    {
      "entropy": 5.630120658874512,
      "epoch": 2.318124092888244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044618691848056726,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21305860728025436,
      "num_tokens": 53016497.0,
      "step": 25555
    },
    {
      "entropy": 5.611555194854736,
      "epoch": 2.318577648766328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004461659190002162,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21871512085199357,
      "num_tokens": 53027811.0,
      "step": 25560
    },
    {
      "entropy": 5.590132474899292,
      "epoch": 2.3190312046444124,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044614491598011027,
      "loss": 4.9358,
      "mean_token_accuracy": 0.20619187206029893,
      "num_tokens": 53038206.0,
      "step": 25565
    },
    {
      "entropy": 5.620518922805786,
      "epoch": 2.319484760522496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004461239094206839,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21192284673452377,
      "num_tokens": 53047974.0,
      "step": 25570
    },
    {
      "entropy": 5.613166236877442,
      "epoch": 2.3199383164005805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044610289932237154,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21265901774168014,
      "num_tokens": 53057904.0,
      "step": 25575
    },
    {
      "entropy": 5.526574945449829,
      "epoch": 2.3203918722786647,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044608188568560756,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21119341105222703,
      "num_tokens": 53068447.0,
      "step": 25580
    },
    {
      "entropy": 5.556252574920654,
      "epoch": 2.320845428156749,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004460608685108266,
      "loss": 4.9762,
      "mean_token_accuracy": 0.2042188212275505,
      "num_tokens": 53079369.0,
      "step": 25585
    },
    {
      "entropy": 5.6675385475158695,
      "epoch": 2.3212989840348333,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044603984779846334,
      "loss": 4.9963,
      "mean_token_accuracy": 0.2049856588244438,
      "num_tokens": 53090794.0,
      "step": 25590
    },
    {
      "entropy": 5.603202867507934,
      "epoch": 2.321752539912917,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004460188235489524,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21291725039482118,
      "num_tokens": 53100802.0,
      "step": 25595
    },
    {
      "entropy": 5.505622053146363,
      "epoch": 2.3222060957910013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004459977957627285,
      "loss": 4.9722,
      "mean_token_accuracy": 0.2128172129392624,
      "num_tokens": 53111950.0,
      "step": 25600
    },
    {
      "entropy": 5.447553920745849,
      "epoch": 2.3226596516690856,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044597676444022683,
      "loss": 4.7931,
      "mean_token_accuracy": 0.22066829800605775,
      "num_tokens": 53121908.0,
      "step": 25605
    },
    {
      "entropy": 5.579293870925904,
      "epoch": 2.32311320754717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044595572958188194,
      "loss": 4.955,
      "mean_token_accuracy": 0.21176907867193223,
      "num_tokens": 53133065.0,
      "step": 25610
    },
    {
      "entropy": 5.490471887588501,
      "epoch": 2.323566763425254,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004459346911881291,
      "loss": 4.8891,
      "mean_token_accuracy": 0.2105861097574234,
      "num_tokens": 53144135.0,
      "step": 25615
    },
    {
      "entropy": 5.559974908828735,
      "epoch": 2.3240203193033384,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004459136492594032,
      "loss": 4.857,
      "mean_token_accuracy": 0.2246379464864731,
      "num_tokens": 53153916.0,
      "step": 25620
    },
    {
      "entropy": 5.522878265380859,
      "epoch": 2.3244738751814222,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004458926037961394,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22637065798044204,
      "num_tokens": 53163937.0,
      "step": 25625
    },
    {
      "entropy": 5.617217588424682,
      "epoch": 2.3249274310595065,
      "grad_norm": 1.125,
      "learning_rate": 0.0004458715547987732,
      "loss": 5.0086,
      "mean_token_accuracy": 0.2043337643146515,
      "num_tokens": 53174327.0,
      "step": 25630
    },
    {
      "entropy": 5.584192562103271,
      "epoch": 2.3253809869375908,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004458505022677396,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21107761561870575,
      "num_tokens": 53185946.0,
      "step": 25635
    },
    {
      "entropy": 5.554040098190308,
      "epoch": 2.325834542815675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004458294462034741,
      "loss": 4.8991,
      "mean_token_accuracy": 0.21663629561662673,
      "num_tokens": 53195944.0,
      "step": 25640
    },
    {
      "entropy": 5.509247446060181,
      "epoch": 2.326288098693759,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004458083866064121,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21436946243047714,
      "num_tokens": 53206054.0,
      "step": 25645
    },
    {
      "entropy": 5.595161628723145,
      "epoch": 2.326741654571843,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004457873234769891,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21010857969522476,
      "num_tokens": 53216923.0,
      "step": 25650
    },
    {
      "entropy": 5.568327903747559,
      "epoch": 2.3271952104499274,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044576625681564086,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21277860254049302,
      "num_tokens": 53226788.0,
      "step": 25655
    },
    {
      "entropy": 5.60441951751709,
      "epoch": 2.3276487663280117,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044574518662280276,
      "loss": 4.9673,
      "mean_token_accuracy": 0.2090952455997467,
      "num_tokens": 53237573.0,
      "step": 25660
    },
    {
      "entropy": 5.585796451568603,
      "epoch": 2.328102322206096,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004457241128989108,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20564925223588942,
      "num_tokens": 53247724.0,
      "step": 25665
    },
    {
      "entropy": 5.584307098388672,
      "epoch": 2.32855587808418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004457030356444006,
      "loss": 4.9289,
      "mean_token_accuracy": 0.21314208954572678,
      "num_tokens": 53258774.0,
      "step": 25670
    },
    {
      "entropy": 5.581422567367554,
      "epoch": 2.329009433962264,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004456819548597082,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2158464729785919,
      "num_tokens": 53268550.0,
      "step": 25675
    },
    {
      "entropy": 5.497116804122925,
      "epoch": 2.3294629898403483,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044566087054526946,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21785036027431487,
      "num_tokens": 53277975.0,
      "step": 25680
    },
    {
      "entropy": 5.621309804916382,
      "epoch": 2.3299165457184325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004456397827015204,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21554505378007888,
      "num_tokens": 53287939.0,
      "step": 25685
    },
    {
      "entropy": 5.613113832473755,
      "epoch": 2.330370101596517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004456186913288971,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2131609320640564,
      "num_tokens": 53298355.0,
      "step": 25690
    },
    {
      "entropy": 5.664609336853028,
      "epoch": 2.3308236574746006,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004455975964278359,
      "loss": 5.036,
      "mean_token_accuracy": 0.21076706498861314,
      "num_tokens": 53308513.0,
      "step": 25695
    },
    {
      "entropy": 5.647288131713867,
      "epoch": 2.331277213352685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044557649799877285,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21256288439035415,
      "num_tokens": 53318249.0,
      "step": 25700
    },
    {
      "entropy": 5.5144682884216305,
      "epoch": 2.331730769230769,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004455553960421443,
      "loss": 4.8763,
      "mean_token_accuracy": 0.2187486171722412,
      "num_tokens": 53328460.0,
      "step": 25705
    },
    {
      "entropy": 5.5543145656585695,
      "epoch": 2.3321843251088534,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044553429055838685,
      "loss": 4.933,
      "mean_token_accuracy": 0.22022654116153717,
      "num_tokens": 53338984.0,
      "step": 25710
    },
    {
      "entropy": 5.611541986465454,
      "epoch": 2.3326378809869377,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044551318154793676,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20399201065301895,
      "num_tokens": 53348092.0,
      "step": 25715
    },
    {
      "entropy": 5.538498067855835,
      "epoch": 2.333091436865022,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044549206901123056,
      "loss": 4.8956,
      "mean_token_accuracy": 0.2126649484038353,
      "num_tokens": 53357343.0,
      "step": 25720
    },
    {
      "entropy": 5.53882098197937,
      "epoch": 2.333544992743106,
      "grad_norm": 1.078125,
      "learning_rate": 0.000445470952948705,
      "loss": 4.8499,
      "mean_token_accuracy": 0.21637628674507142,
      "num_tokens": 53367343.0,
      "step": 25725
    },
    {
      "entropy": 5.643903064727783,
      "epoch": 2.33399854862119,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044544983336079665,
      "loss": 5.0417,
      "mean_token_accuracy": 0.21098848730325698,
      "num_tokens": 53379093.0,
      "step": 25730
    },
    {
      "entropy": 5.555362224578857,
      "epoch": 2.3344521044992743,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004454287102479423,
      "loss": 4.8974,
      "mean_token_accuracy": 0.21512557417154313,
      "num_tokens": 53388883.0,
      "step": 25735
    },
    {
      "entropy": 5.5754283428192135,
      "epoch": 2.3349056603773586,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004454075836105788,
      "loss": 4.9047,
      "mean_token_accuracy": 0.21789999902248383,
      "num_tokens": 53399393.0,
      "step": 25740
    },
    {
      "entropy": 5.552889013290406,
      "epoch": 2.335359216255443,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044538645344914307,
      "loss": 4.8749,
      "mean_token_accuracy": 0.21663407981395721,
      "num_tokens": 53408545.0,
      "step": 25745
    },
    {
      "entropy": 5.5533253192901615,
      "epoch": 2.3358127721335267,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004453653197640721,
      "loss": 4.976,
      "mean_token_accuracy": 0.21037684231996537,
      "num_tokens": 53418325.0,
      "step": 25750
    },
    {
      "entropy": 5.573678302764892,
      "epoch": 2.336266328011611,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044534418255580286,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20263057351112365,
      "num_tokens": 53427984.0,
      "step": 25755
    },
    {
      "entropy": 5.64089035987854,
      "epoch": 2.336719883889695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044532304182477245,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20502175986766816,
      "num_tokens": 53438765.0,
      "step": 25760
    },
    {
      "entropy": 5.646881151199341,
      "epoch": 2.3371734397677795,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004453018975714182,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21376003324985504,
      "num_tokens": 53448792.0,
      "step": 25765
    },
    {
      "entropy": 5.648402404785156,
      "epoch": 2.3376269956458637,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004452807497961773,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20995502769947053,
      "num_tokens": 53459334.0,
      "step": 25770
    },
    {
      "entropy": 5.590705156326294,
      "epoch": 2.3380805515239476,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044525959849948704,
      "loss": 5.0056,
      "mean_token_accuracy": 0.21176755279302598,
      "num_tokens": 53469841.0,
      "step": 25775
    },
    {
      "entropy": 5.569129753112793,
      "epoch": 2.338534107402032,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044523844368178503,
      "loss": 4.9615,
      "mean_token_accuracy": 0.21363317519426345,
      "num_tokens": 53480660.0,
      "step": 25780
    },
    {
      "entropy": 5.604327297210693,
      "epoch": 2.338987663280116,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004452172853435085,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21227259188890457,
      "num_tokens": 53490276.0,
      "step": 25785
    },
    {
      "entropy": 5.484948587417603,
      "epoch": 2.3394412191582004,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004451961234850953,
      "loss": 4.854,
      "mean_token_accuracy": 0.21953050494194032,
      "num_tokens": 53501100.0,
      "step": 25790
    },
    {
      "entropy": 5.518176174163818,
      "epoch": 2.3398947750362846,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004451749581069827,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21475230157375336,
      "num_tokens": 53511680.0,
      "step": 25795
    },
    {
      "entropy": 5.597425174713135,
      "epoch": 2.3403483309143684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004451537892096087,
      "loss": 5.0064,
      "mean_token_accuracy": 0.21020373851060867,
      "num_tokens": 53521298.0,
      "step": 25800
    },
    {
      "entropy": 5.572571229934693,
      "epoch": 2.3408018867924527,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044513261679341094,
      "loss": 4.8332,
      "mean_token_accuracy": 0.22635883688926697,
      "num_tokens": 53530413.0,
      "step": 25805
    },
    {
      "entropy": 5.541699361801148,
      "epoch": 2.341255442670537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004451114408588273,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21247395426034926,
      "num_tokens": 53540959.0,
      "step": 25810
    },
    {
      "entropy": 5.57380428314209,
      "epoch": 2.3417089985486212,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044509026140629584,
      "loss": 4.9376,
      "mean_token_accuracy": 0.21315263956785202,
      "num_tokens": 53551265.0,
      "step": 25815
    },
    {
      "entropy": 5.615974426269531,
      "epoch": 2.3421625544267055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044506907843625427,
      "loss": 4.9587,
      "mean_token_accuracy": 0.20951335728168488,
      "num_tokens": 53562666.0,
      "step": 25820
    },
    {
      "entropy": 5.568785715103149,
      "epoch": 2.34261611030479,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044504789194914086,
      "loss": 5.0182,
      "mean_token_accuracy": 0.2091977760195732,
      "num_tokens": 53573688.0,
      "step": 25825
    },
    {
      "entropy": 5.594330787658691,
      "epoch": 2.3430696661828736,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044502670194539374,
      "loss": 4.9062,
      "mean_token_accuracy": 0.21697243452072143,
      "num_tokens": 53584055.0,
      "step": 25830
    },
    {
      "entropy": 5.515743541717529,
      "epoch": 2.343523222060958,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044500550842545104,
      "loss": 4.8124,
      "mean_token_accuracy": 0.22413455843925476,
      "num_tokens": 53593765.0,
      "step": 25835
    },
    {
      "entropy": 5.498024892807007,
      "epoch": 2.343976777939042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004449843113897512,
      "loss": 4.8786,
      "mean_token_accuracy": 0.22074564099311828,
      "num_tokens": 53603480.0,
      "step": 25840
    },
    {
      "entropy": 5.560669851303101,
      "epoch": 2.3444303338171264,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004449631108387325,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21525682210922242,
      "num_tokens": 53613797.0,
      "step": 25845
    },
    {
      "entropy": 5.551551389694214,
      "epoch": 2.34488388969521,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004449419067728333,
      "loss": 4.875,
      "mean_token_accuracy": 0.217022804915905,
      "num_tokens": 53623646.0,
      "step": 25850
    },
    {
      "entropy": 5.54360671043396,
      "epoch": 2.3453374455732945,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044492069919249206,
      "loss": 4.917,
      "mean_token_accuracy": 0.2137106642127037,
      "num_tokens": 53634118.0,
      "step": 25855
    },
    {
      "entropy": 5.596779251098633,
      "epoch": 2.3457910014513788,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004448994880981476,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21488005816936492,
      "num_tokens": 53644370.0,
      "step": 25860
    },
    {
      "entropy": 5.608159160614013,
      "epoch": 2.346244557329463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044487827349023833,
      "loss": 4.9784,
      "mean_token_accuracy": 0.20647551417350768,
      "num_tokens": 53655560.0,
      "step": 25865
    },
    {
      "entropy": 5.576163816452026,
      "epoch": 2.3466981132075473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044485705536920315,
      "loss": 5.0071,
      "mean_token_accuracy": 0.21219031810760497,
      "num_tokens": 53666401.0,
      "step": 25870
    },
    {
      "entropy": 5.66574068069458,
      "epoch": 2.3471516690856316,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044483583373548075,
      "loss": 4.98,
      "mean_token_accuracy": 0.22362128794193267,
      "num_tokens": 53676133.0,
      "step": 25875
    },
    {
      "entropy": 5.569415044784546,
      "epoch": 2.3476052249637154,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004448146085895101,
      "loss": 4.9314,
      "mean_token_accuracy": 0.20970430672168733,
      "num_tokens": 53687683.0,
      "step": 25880
    },
    {
      "entropy": 5.5365321159362795,
      "epoch": 2.3480587808417996,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004447933799317299,
      "loss": 5.016,
      "mean_token_accuracy": 0.20889658778905867,
      "num_tokens": 53698343.0,
      "step": 25885
    },
    {
      "entropy": 5.577454376220703,
      "epoch": 2.348512336719884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004447721477625795,
      "loss": 4.9249,
      "mean_token_accuracy": 0.21522369235754013,
      "num_tokens": 53709014.0,
      "step": 25890
    },
    {
      "entropy": 5.5477344512939455,
      "epoch": 2.348965892597968,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044475091208249774,
      "loss": 4.9343,
      "mean_token_accuracy": 0.21616510450839996,
      "num_tokens": 53718838.0,
      "step": 25895
    },
    {
      "entropy": 5.506769418716431,
      "epoch": 2.3494194484760524,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004447296728919239,
      "loss": 4.893,
      "mean_token_accuracy": 0.21399504840373992,
      "num_tokens": 53728526.0,
      "step": 25900
    },
    {
      "entropy": 5.607238864898681,
      "epoch": 2.3498730043541363,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004447084301912972,
      "loss": 5.0108,
      "mean_token_accuracy": 0.21239181607961655,
      "num_tokens": 53739261.0,
      "step": 25905
    },
    {
      "entropy": 5.605481147766113,
      "epoch": 2.3503265602322205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004446871839810568,
      "loss": 4.9869,
      "mean_token_accuracy": 0.212431001663208,
      "num_tokens": 53749599.0,
      "step": 25910
    },
    {
      "entropy": 5.585039186477661,
      "epoch": 2.350780116110305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044466593426164224,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20691626369953156,
      "num_tokens": 53759898.0,
      "step": 25915
    },
    {
      "entropy": 5.534364414215088,
      "epoch": 2.351233671988389,
      "grad_norm": 1.078125,
      "learning_rate": 0.000444644681033493,
      "loss": 4.9034,
      "mean_token_accuracy": 0.20887214094400405,
      "num_tokens": 53770561.0,
      "step": 25920
    },
    {
      "entropy": 5.546531629562378,
      "epoch": 2.3516872278664733,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044462342429704845,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21344873458147048,
      "num_tokens": 53780408.0,
      "step": 25925
    },
    {
      "entropy": 5.485399580001831,
      "epoch": 2.352140783744557,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044460216405274827,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22735902070999145,
      "num_tokens": 53789125.0,
      "step": 25930
    },
    {
      "entropy": 5.536479234695435,
      "epoch": 2.3525943396226414,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044458090030103216,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2195194274187088,
      "num_tokens": 53799164.0,
      "step": 25935
    },
    {
      "entropy": 5.549217414855957,
      "epoch": 2.3530478955007257,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044455963304233976,
      "loss": 4.871,
      "mean_token_accuracy": 0.2197946310043335,
      "num_tokens": 53810710.0,
      "step": 25940
    },
    {
      "entropy": 5.509092664718628,
      "epoch": 2.35350145137881,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000444538362277111,
      "loss": 4.9183,
      "mean_token_accuracy": 0.2079112008213997,
      "num_tokens": 53820692.0,
      "step": 25945
    },
    {
      "entropy": 5.617864179611206,
      "epoch": 2.353955007256894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004445170880057857,
      "loss": 4.9774,
      "mean_token_accuracy": 0.2089620739221573,
      "num_tokens": 53831082.0,
      "step": 25950
    },
    {
      "entropy": 5.635778188705444,
      "epoch": 2.354408563134978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044449581022880375,
      "loss": 5.0121,
      "mean_token_accuracy": 0.19856876879930496,
      "num_tokens": 53842602.0,
      "step": 25955
    },
    {
      "entropy": 5.548915004730224,
      "epoch": 2.3548621190130623,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004444745289466053,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21658931374549867,
      "num_tokens": 53853784.0,
      "step": 25960
    },
    {
      "entropy": 5.522101163864136,
      "epoch": 2.3553156748911466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044445324415963043,
      "loss": 4.8197,
      "mean_token_accuracy": 0.22228266745805741,
      "num_tokens": 53863526.0,
      "step": 25965
    },
    {
      "entropy": 5.514787483215332,
      "epoch": 2.355769230769231,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004444319558683192,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21255785822868348,
      "num_tokens": 53874594.0,
      "step": 25970
    },
    {
      "entropy": 5.575469303131103,
      "epoch": 2.356222786647315,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044441066407311205,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21542281061410903,
      "num_tokens": 53884771.0,
      "step": 25975
    },
    {
      "entropy": 5.605532455444336,
      "epoch": 2.3566763425253994,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004443893687744492,
      "loss": 4.9253,
      "mean_token_accuracy": 0.21457685381174088,
      "num_tokens": 53895181.0,
      "step": 25980
    },
    {
      "entropy": 5.514532136917114,
      "epoch": 2.357129898403483,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044436806997277096,
      "loss": 4.8244,
      "mean_token_accuracy": 0.22391154766082763,
      "num_tokens": 53904153.0,
      "step": 25985
    },
    {
      "entropy": 5.499745178222656,
      "epoch": 2.3575834542815675,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044434676766851787,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22061283886432648,
      "num_tokens": 53914064.0,
      "step": 25990
    },
    {
      "entropy": 5.56377010345459,
      "epoch": 2.3580370101596517,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044432546186213054,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2109364777803421,
      "num_tokens": 53924240.0,
      "step": 25995
    },
    {
      "entropy": 5.593591117858887,
      "epoch": 2.358490566037736,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044430415255404944,
      "loss": 4.9747,
      "mean_token_accuracy": 0.2133777320384979,
      "num_tokens": 53935395.0,
      "step": 26000
    },
    {
      "entropy": 5.577879428863525,
      "epoch": 2.35894412191582,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044428283974471533,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21419965773820876,
      "num_tokens": 53945768.0,
      "step": 26005
    },
    {
      "entropy": 5.498639488220215,
      "epoch": 2.359397677793904,
      "grad_norm": 1.15625,
      "learning_rate": 0.000444261523434569,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22037704586982726,
      "num_tokens": 53956467.0,
      "step": 26010
    },
    {
      "entropy": 5.513798713684082,
      "epoch": 2.3598512336719883,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044424020362405126,
      "loss": 4.8569,
      "mean_token_accuracy": 0.21677091717720032,
      "num_tokens": 53967587.0,
      "step": 26015
    },
    {
      "entropy": 5.534829187393188,
      "epoch": 2.3603047895500726,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004442188803136029,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2128376007080078,
      "num_tokens": 53977691.0,
      "step": 26020
    },
    {
      "entropy": 5.54969973564148,
      "epoch": 2.360758345428157,
      "grad_norm": 1.21875,
      "learning_rate": 0.000444197553503665,
      "loss": 4.9756,
      "mean_token_accuracy": 0.20649277418851852,
      "num_tokens": 53987608.0,
      "step": 26025
    },
    {
      "entropy": 5.599653005599976,
      "epoch": 2.361211901306241,
      "grad_norm": 1.125,
      "learning_rate": 0.00044417622319467857,
      "loss": 5.06,
      "mean_token_accuracy": 0.20780419409275055,
      "num_tokens": 53998010.0,
      "step": 26030
    },
    {
      "entropy": 5.565209770202637,
      "epoch": 2.361665457184325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044415488938708475,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2175135225057602,
      "num_tokens": 54009852.0,
      "step": 26035
    },
    {
      "entropy": 5.563492345809936,
      "epoch": 2.3621190130624092,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004441335520813247,
      "loss": 4.8692,
      "mean_token_accuracy": 0.21580897271633148,
      "num_tokens": 54019635.0,
      "step": 26040
    },
    {
      "entropy": 5.539868545532227,
      "epoch": 2.3625725689404935,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044411221127783966,
      "loss": 4.9258,
      "mean_token_accuracy": 0.2137252390384674,
      "num_tokens": 54030046.0,
      "step": 26045
    },
    {
      "entropy": 5.5821287631988525,
      "epoch": 2.3630261248185778,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004440908669770711,
      "loss": 4.9133,
      "mean_token_accuracy": 0.21307014524936677,
      "num_tokens": 54041197.0,
      "step": 26050
    },
    {
      "entropy": 5.579737138748169,
      "epoch": 2.3634796806966616,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004440695191794602,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2196875348687172,
      "num_tokens": 54051592.0,
      "step": 26055
    },
    {
      "entropy": 5.5798334121704105,
      "epoch": 2.363933236574746,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004440481678854487,
      "loss": 4.9805,
      "mean_token_accuracy": 0.2075364723801613,
      "num_tokens": 54062893.0,
      "step": 26060
    },
    {
      "entropy": 5.627664184570312,
      "epoch": 2.36438679245283,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044402681309547786,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20873741209506988,
      "num_tokens": 54073651.0,
      "step": 26065
    },
    {
      "entropy": 5.5339478015899655,
      "epoch": 2.3648403483309144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044400545480998956,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22097087204456328,
      "num_tokens": 54084889.0,
      "step": 26070
    },
    {
      "entropy": 5.5441789627075195,
      "epoch": 2.3652939042089987,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044398409302942533,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2119978666305542,
      "num_tokens": 54095724.0,
      "step": 26075
    },
    {
      "entropy": 5.56972017288208,
      "epoch": 2.365747460087083,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000443962727754227,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21552262902259828,
      "num_tokens": 54106581.0,
      "step": 26080
    },
    {
      "entropy": 5.6068250179290775,
      "epoch": 2.3662010159651667,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004439413589848364,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20996494740247726,
      "num_tokens": 54116074.0,
      "step": 26085
    },
    {
      "entropy": 5.548000335693359,
      "epoch": 2.366654571843251,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004439199867216955,
      "loss": 4.9625,
      "mean_token_accuracy": 0.20807096362113953,
      "num_tokens": 54126902.0,
      "step": 26090
    },
    {
      "entropy": 5.5538010597229,
      "epoch": 2.3671081277213353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004438986109652462,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21629703044891357,
      "num_tokens": 54137554.0,
      "step": 26095
    },
    {
      "entropy": 5.5640634059906,
      "epoch": 2.3675616835994195,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004438772317159306,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21244510710239412,
      "num_tokens": 54148035.0,
      "step": 26100
    },
    {
      "entropy": 5.536005544662475,
      "epoch": 2.368015239477504,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044385584897419075,
      "loss": 4.9207,
      "mean_token_accuracy": 0.22129922360181808,
      "num_tokens": 54157417.0,
      "step": 26105
    },
    {
      "entropy": 5.594117927551269,
      "epoch": 2.3684687953555876,
      "grad_norm": 1.125,
      "learning_rate": 0.000443834462740469,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20927917808294297,
      "num_tokens": 54168353.0,
      "step": 26110
    },
    {
      "entropy": 5.554742050170899,
      "epoch": 2.368922351233672,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004438130730152074,
      "loss": 5.0115,
      "mean_token_accuracy": 0.2028622180223465,
      "num_tokens": 54180182.0,
      "step": 26115
    },
    {
      "entropy": 5.606842660903931,
      "epoch": 2.369375907111756,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044379167979884847,
      "loss": 4.9568,
      "mean_token_accuracy": 0.2098795086145401,
      "num_tokens": 54190973.0,
      "step": 26120
    },
    {
      "entropy": 5.609960985183716,
      "epoch": 2.3698294629898404,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004437702830918347,
      "loss": 5.0338,
      "mean_token_accuracy": 0.20531612187623977,
      "num_tokens": 54202077.0,
      "step": 26125
    },
    {
      "entropy": 5.572955703735351,
      "epoch": 2.3702830188679247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004437488828946084,
      "loss": 4.9428,
      "mean_token_accuracy": 0.20869537591934204,
      "num_tokens": 54211778.0,
      "step": 26130
    },
    {
      "entropy": 5.566894578933716,
      "epoch": 2.3707365747460085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004437274792076122,
      "loss": 4.8134,
      "mean_token_accuracy": 0.23084428906440735,
      "num_tokens": 54220941.0,
      "step": 26135
    },
    {
      "entropy": 5.588488435745239,
      "epoch": 2.371190130624093,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044370607203128874,
      "loss": 5.0337,
      "mean_token_accuracy": 0.20585249811410905,
      "num_tokens": 54231602.0,
      "step": 26140
    },
    {
      "entropy": 5.5495302200317385,
      "epoch": 2.371643686502177,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004436846613660807,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2138413593173027,
      "num_tokens": 54241797.0,
      "step": 26145
    },
    {
      "entropy": 5.629802227020264,
      "epoch": 2.3720972423802613,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004436632472124308,
      "loss": 5.0088,
      "mean_token_accuracy": 0.21093651354312898,
      "num_tokens": 54253085.0,
      "step": 26150
    },
    {
      "entropy": 5.655661010742188,
      "epoch": 2.3725507982583456,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004436418295707821,
      "loss": 4.9764,
      "mean_token_accuracy": 0.2152244508266449,
      "num_tokens": 54262164.0,
      "step": 26155
    },
    {
      "entropy": 5.574744415283203,
      "epoch": 2.3730043541364294,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004436204084415774,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21065271645784378,
      "num_tokens": 54271922.0,
      "step": 26160
    },
    {
      "entropy": 5.567212820053101,
      "epoch": 2.3734579100145137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004435989838252596,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22161255031824112,
      "num_tokens": 54281738.0,
      "step": 26165
    },
    {
      "entropy": 5.570029592514038,
      "epoch": 2.373911465892598,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004435775557222719,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21278369724750518,
      "num_tokens": 54292252.0,
      "step": 26170
    },
    {
      "entropy": 5.570502758026123,
      "epoch": 2.374365021770682,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044355612413305736,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2120476931333542,
      "num_tokens": 54302518.0,
      "step": 26175
    },
    {
      "entropy": 5.595395278930664,
      "epoch": 2.3748185776487665,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004435346890580593,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2197017937898636,
      "num_tokens": 54312917.0,
      "step": 26180
    },
    {
      "entropy": 5.613895893096924,
      "epoch": 2.3752721335268507,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044351325049772085,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2203422635793686,
      "num_tokens": 54322552.0,
      "step": 26185
    },
    {
      "entropy": 5.571087455749511,
      "epoch": 2.3757256894049346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044349180845248536,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21754180192947387,
      "num_tokens": 54332917.0,
      "step": 26190
    },
    {
      "entropy": 5.502777910232544,
      "epoch": 2.376179245283019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004434703629227964,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21895778328180313,
      "num_tokens": 54342791.0,
      "step": 26195
    },
    {
      "entropy": 5.514581489562988,
      "epoch": 2.376632801161103,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004434489139090975,
      "loss": 4.8011,
      "mean_token_accuracy": 0.22824004888534546,
      "num_tokens": 54352821.0,
      "step": 26200
    },
    {
      "entropy": 5.634966659545898,
      "epoch": 2.3770863570391874,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044342746141183206,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20839511901140212,
      "num_tokens": 54363095.0,
      "step": 26205
    },
    {
      "entropy": 5.56240291595459,
      "epoch": 2.377539912917271,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004434060054314438,
      "loss": 4.9056,
      "mean_token_accuracy": 0.2107112556695938,
      "num_tokens": 54373841.0,
      "step": 26210
    },
    {
      "entropy": 5.509205007553101,
      "epoch": 2.3779934687953554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044338454596837637,
      "loss": 4.8809,
      "mean_token_accuracy": 0.216173654794693,
      "num_tokens": 54385012.0,
      "step": 26215
    },
    {
      "entropy": 5.5585497379302975,
      "epoch": 2.3784470246734397,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004433630830230737,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21192312687635423,
      "num_tokens": 54395298.0,
      "step": 26220
    },
    {
      "entropy": 5.568934106826783,
      "epoch": 2.378900580551524,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004433416165959795,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21015683114528655,
      "num_tokens": 54405752.0,
      "step": 26225
    },
    {
      "entropy": 5.590658378601074,
      "epoch": 2.3793541364296082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004433201466875378,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21067356169223786,
      "num_tokens": 54416607.0,
      "step": 26230
    },
    {
      "entropy": 5.585775995254517,
      "epoch": 2.3798076923076925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004432986732981926,
      "loss": 4.914,
      "mean_token_accuracy": 0.2157020777463913,
      "num_tokens": 54426873.0,
      "step": 26235
    },
    {
      "entropy": 5.492963695526123,
      "epoch": 2.3802612481857763,
      "grad_norm": 1.0625,
      "learning_rate": 0.000443277196428388,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22379397302865983,
      "num_tokens": 54437611.0,
      "step": 26240
    },
    {
      "entropy": 5.611445760726928,
      "epoch": 2.3807148040638606,
      "grad_norm": 1.046875,
      "learning_rate": 0.000443255716078568,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21650339365005494,
      "num_tokens": 54448992.0,
      "step": 26245
    },
    {
      "entropy": 5.575338125228882,
      "epoch": 2.381168359941945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004432342322491769,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22124216109514236,
      "num_tokens": 54459172.0,
      "step": 26250
    },
    {
      "entropy": 5.499160575866699,
      "epoch": 2.381621915820029,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044321274494065903,
      "loss": 4.8903,
      "mean_token_accuracy": 0.21635547727346421,
      "num_tokens": 54469660.0,
      "step": 26255
    },
    {
      "entropy": 5.596979665756225,
      "epoch": 2.3820754716981134,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044319125415345886,
      "loss": 5.0191,
      "mean_token_accuracy": 0.2071585014462471,
      "num_tokens": 54479567.0,
      "step": 26260
    },
    {
      "entropy": 5.565261697769165,
      "epoch": 2.3825290275761972,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004431697598880205,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20945729017257692,
      "num_tokens": 54489690.0,
      "step": 26265
    },
    {
      "entropy": 5.585064554214478,
      "epoch": 2.3829825834542815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044314826214478877,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2184809312224388,
      "num_tokens": 54500876.0,
      "step": 26270
    },
    {
      "entropy": 5.606135892868042,
      "epoch": 2.3834361393323658,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044312676092420804,
      "loss": 4.9813,
      "mean_token_accuracy": 0.2080484375357628,
      "num_tokens": 54510988.0,
      "step": 26275
    },
    {
      "entropy": 5.565968656539917,
      "epoch": 2.38388969521045,
      "grad_norm": 1.0,
      "learning_rate": 0.0004431052562267231,
      "loss": 4.8853,
      "mean_token_accuracy": 0.223616386950016,
      "num_tokens": 54522149.0,
      "step": 26280
    },
    {
      "entropy": 5.637773895263672,
      "epoch": 2.3843432510885343,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044308374805277867,
      "loss": 5.0566,
      "mean_token_accuracy": 0.20029253214597703,
      "num_tokens": 54533482.0,
      "step": 26285
    },
    {
      "entropy": 5.562591981887818,
      "epoch": 2.384796806966618,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004430622364028194,
      "loss": 4.9738,
      "mean_token_accuracy": 0.20799656063318253,
      "num_tokens": 54543058.0,
      "step": 26290
    },
    {
      "entropy": 5.578580474853515,
      "epoch": 2.3852503628447024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044304072127729035,
      "loss": 4.9324,
      "mean_token_accuracy": 0.2148285612463951,
      "num_tokens": 54552655.0,
      "step": 26295
    },
    {
      "entropy": 5.569213390350342,
      "epoch": 2.3857039187227866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044301920267663633,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20749802887439728,
      "num_tokens": 54562696.0,
      "step": 26300
    },
    {
      "entropy": 5.5724883556365965,
      "epoch": 2.386157474600871,
      "grad_norm": 1.125,
      "learning_rate": 0.0004429976806013024,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21777765154838563,
      "num_tokens": 54572803.0,
      "step": 26305
    },
    {
      "entropy": 5.521192264556885,
      "epoch": 2.386611030478955,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004429761550517336,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21111124455928804,
      "num_tokens": 54582450.0,
      "step": 26310
    },
    {
      "entropy": 5.478409290313721,
      "epoch": 2.387064586357039,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044295462602837504,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21607948392629622,
      "num_tokens": 54591868.0,
      "step": 26315
    },
    {
      "entropy": 5.5848370552062985,
      "epoch": 2.3875181422351233,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004429330935316721,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21125050783157348,
      "num_tokens": 54601858.0,
      "step": 26320
    },
    {
      "entropy": 5.571154832839966,
      "epoch": 2.3879716981132075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044291155756206996,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21169865280389785,
      "num_tokens": 54612243.0,
      "step": 26325
    },
    {
      "entropy": 5.570634555816651,
      "epoch": 2.388425253991292,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044289001812001404,
      "loss": 4.8407,
      "mean_token_accuracy": 0.22116668671369552,
      "num_tokens": 54622365.0,
      "step": 26330
    },
    {
      "entropy": 5.552473688125611,
      "epoch": 2.388878809869376,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004428684752059497,
      "loss": 4.8622,
      "mean_token_accuracy": 0.22066307663917542,
      "num_tokens": 54632356.0,
      "step": 26335
    },
    {
      "entropy": 5.6089372634887695,
      "epoch": 2.3893323657474603,
      "grad_norm": 1.125,
      "learning_rate": 0.0004428469288203226,
      "loss": 4.9569,
      "mean_token_accuracy": 0.20489265620708466,
      "num_tokens": 54642976.0,
      "step": 26340
    },
    {
      "entropy": 5.657524251937867,
      "epoch": 2.389785921625544,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004428253789635783,
      "loss": 4.9738,
      "mean_token_accuracy": 0.21573778241872787,
      "num_tokens": 54652597.0,
      "step": 26345
    },
    {
      "entropy": 5.560487127304077,
      "epoch": 2.3902394775036284,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004428038256361622,
      "loss": 5.0598,
      "mean_token_accuracy": 0.207101708650589,
      "num_tokens": 54662930.0,
      "step": 26350
    },
    {
      "entropy": 5.689995527267456,
      "epoch": 2.3906930333817127,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004427822688385203,
      "loss": 5.1682,
      "mean_token_accuracy": 0.20526119619607924,
      "num_tokens": 54674562.0,
      "step": 26355
    },
    {
      "entropy": 5.630713653564453,
      "epoch": 2.391146589259797,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004427607085710984,
      "loss": 4.951,
      "mean_token_accuracy": 0.2108895167708397,
      "num_tokens": 54684432.0,
      "step": 26360
    },
    {
      "entropy": 5.632601737976074,
      "epoch": 2.3916001451378808,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044273914483434226,
      "loss": 4.9766,
      "mean_token_accuracy": 0.20941840261220931,
      "num_tokens": 54694060.0,
      "step": 26365
    },
    {
      "entropy": 5.551342439651489,
      "epoch": 2.392053701015965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044271757762869777,
      "loss": 4.9579,
      "mean_token_accuracy": 0.2113799899816513,
      "num_tokens": 54704411.0,
      "step": 26370
    },
    {
      "entropy": 5.565278053283691,
      "epoch": 2.3925072568940493,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004426960069546112,
      "loss": 4.9725,
      "mean_token_accuracy": 0.21328798830509185,
      "num_tokens": 54715185.0,
      "step": 26375
    },
    {
      "entropy": 5.562286853790283,
      "epoch": 2.3929608127721336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004426744328125284,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22069818377494813,
      "num_tokens": 54725529.0,
      "step": 26380
    },
    {
      "entropy": 5.648525619506836,
      "epoch": 2.393414368650218,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004426528552028955,
      "loss": 5.0024,
      "mean_token_accuracy": 0.2120383784174919,
      "num_tokens": 54736992.0,
      "step": 26385
    },
    {
      "entropy": 5.593166875839233,
      "epoch": 2.393867924528302,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044263127412615897,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2161580428481102,
      "num_tokens": 54746951.0,
      "step": 26390
    },
    {
      "entropy": 5.52850022315979,
      "epoch": 2.394321480406386,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004426096895827649,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22522672712802888,
      "num_tokens": 54757272.0,
      "step": 26395
    },
    {
      "entropy": 5.565531539916992,
      "epoch": 2.39477503628447,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044258810157315977,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20992569923400878,
      "num_tokens": 54767835.0,
      "step": 26400
    },
    {
      "entropy": 5.588186407089234,
      "epoch": 2.3952285921625545,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044256651009779,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21384909749031067,
      "num_tokens": 54778432.0,
      "step": 26405
    },
    {
      "entropy": 5.637002038955688,
      "epoch": 2.3956821480406387,
      "grad_norm": 1.125,
      "learning_rate": 0.00044254491515710213,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20795969665050507,
      "num_tokens": 54788582.0,
      "step": 26410
    },
    {
      "entropy": 5.6249936580657955,
      "epoch": 2.3961357039187225,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004425233167515426,
      "loss": 4.9645,
      "mean_token_accuracy": 0.21718416959047318,
      "num_tokens": 54798848.0,
      "step": 26415
    },
    {
      "entropy": 5.541302633285523,
      "epoch": 2.396589259796807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004425017148815583,
      "loss": 4.8537,
      "mean_token_accuracy": 0.21915711760520934,
      "num_tokens": 54809867.0,
      "step": 26420
    },
    {
      "entropy": 5.5384095191955565,
      "epoch": 2.397042815674891,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044248010954759575,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2144401639699936,
      "num_tokens": 54819905.0,
      "step": 26425
    },
    {
      "entropy": 5.6020440578460695,
      "epoch": 2.3974963715529753,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044245850075010193,
      "loss": 4.9437,
      "mean_token_accuracy": 0.21297521591186525,
      "num_tokens": 54829723.0,
      "step": 26430
    },
    {
      "entropy": 5.551946640014648,
      "epoch": 2.3979499274310596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044243688848952364,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2183666169643402,
      "num_tokens": 54840008.0,
      "step": 26435
    },
    {
      "entropy": 5.564555835723877,
      "epoch": 2.398403483309144,
      "grad_norm": 1.125,
      "learning_rate": 0.0004424152727663078,
      "loss": 4.9441,
      "mean_token_accuracy": 0.20714806020259857,
      "num_tokens": 54851693.0,
      "step": 26440
    },
    {
      "entropy": 5.597462797164917,
      "epoch": 2.3988570391872277,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004423936535809015,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21012354642152786,
      "num_tokens": 54862086.0,
      "step": 26445
    },
    {
      "entropy": 5.687418413162232,
      "epoch": 2.399310595065312,
      "grad_norm": 1.125,
      "learning_rate": 0.0004423720309337517,
      "loss": 5.0297,
      "mean_token_accuracy": 0.20929399728775025,
      "num_tokens": 54872883.0,
      "step": 26450
    },
    {
      "entropy": 5.622654771804809,
      "epoch": 2.3997641509433962,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044235040482530576,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2101919338107109,
      "num_tokens": 54882815.0,
      "step": 26455
    },
    {
      "entropy": 5.431966876983642,
      "epoch": 2.4002177068214805,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004423287752560108,
      "loss": 4.7813,
      "mean_token_accuracy": 0.22615410685539244,
      "num_tokens": 54892480.0,
      "step": 26460
    },
    {
      "entropy": 5.460355615615844,
      "epoch": 2.4006712626995648,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044230714222631397,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22623133063316345,
      "num_tokens": 54901607.0,
      "step": 26465
    },
    {
      "entropy": 5.678241014480591,
      "epoch": 2.4011248185776486,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044228550573666294,
      "loss": 5.0553,
      "mean_token_accuracy": 0.21256614476442337,
      "num_tokens": 54913545.0,
      "step": 26470
    },
    {
      "entropy": 5.628347158432007,
      "epoch": 2.401578374455733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000442263865787505,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21294753700494767,
      "num_tokens": 54924392.0,
      "step": 26475
    },
    {
      "entropy": 5.581802845001221,
      "epoch": 2.402031930333817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004422422223792877,
      "loss": 4.9217,
      "mean_token_accuracy": 0.21415944993495942,
      "num_tokens": 54934114.0,
      "step": 26480
    },
    {
      "entropy": 5.615803384780884,
      "epoch": 2.4024854862119014,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004422205755124586,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2141389861702919,
      "num_tokens": 54944268.0,
      "step": 26485
    },
    {
      "entropy": 5.608943605422974,
      "epoch": 2.4029390420899857,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044219892518746536,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21248486638069153,
      "num_tokens": 54954877.0,
      "step": 26490
    },
    {
      "entropy": 5.542152738571167,
      "epoch": 2.4033925979680695,
      "grad_norm": 1.125,
      "learning_rate": 0.00044217727140475585,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21532075703144074,
      "num_tokens": 54964745.0,
      "step": 26495
    },
    {
      "entropy": 5.565544939041137,
      "epoch": 2.4038461538461537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044215561416477764,
      "loss": 4.8404,
      "mean_token_accuracy": 0.21421201080083846,
      "num_tokens": 54975256.0,
      "step": 26500
    },
    {
      "entropy": 5.649453830718994,
      "epoch": 2.404299709724238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004421339534679788,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20731568038463594,
      "num_tokens": 54986226.0,
      "step": 26505
    },
    {
      "entropy": 5.620152568817138,
      "epoch": 2.4047532656023223,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004421122893148072,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20383986830711365,
      "num_tokens": 54997350.0,
      "step": 26510
    },
    {
      "entropy": 5.603748464584351,
      "epoch": 2.4052068214804065,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044209062170571086,
      "loss": 4.9188,
      "mean_token_accuracy": 0.21386370807886124,
      "num_tokens": 55007947.0,
      "step": 26515
    },
    {
      "entropy": 5.586966896057129,
      "epoch": 2.4056603773584904,
      "grad_norm": 1.125,
      "learning_rate": 0.0004420689506411379,
      "loss": 4.9634,
      "mean_token_accuracy": 0.21214287132024764,
      "num_tokens": 55018789.0,
      "step": 26520
    },
    {
      "entropy": 5.553619337081909,
      "epoch": 2.4061139332365746,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044204727612153643,
      "loss": 4.881,
      "mean_token_accuracy": 0.21062198877334595,
      "num_tokens": 55028934.0,
      "step": 26525
    },
    {
      "entropy": 5.553179025650024,
      "epoch": 2.406567489114659,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044202559814735475,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22172596901655198,
      "num_tokens": 55039395.0,
      "step": 26530
    },
    {
      "entropy": 5.515214729309082,
      "epoch": 2.407021044992743,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004420039167190411,
      "loss": 4.8549,
      "mean_token_accuracy": 0.21208732426166535,
      "num_tokens": 55049932.0,
      "step": 26535
    },
    {
      "entropy": 5.583847188949585,
      "epoch": 2.4074746008708274,
      "grad_norm": 1.296875,
      "learning_rate": 0.000441982231837044,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20788389146327974,
      "num_tokens": 55061532.0,
      "step": 26540
    },
    {
      "entropy": 5.64016637802124,
      "epoch": 2.4079281567489117,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004419605435018117,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20780168771743773,
      "num_tokens": 55072368.0,
      "step": 26545
    },
    {
      "entropy": 5.589163970947266,
      "epoch": 2.4083817126269955,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004419388517137928,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21165295392274858,
      "num_tokens": 55083128.0,
      "step": 26550
    },
    {
      "entropy": 5.5810487270355225,
      "epoch": 2.40883526850508,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044191715647343594,
      "loss": 4.9368,
      "mean_token_accuracy": 0.2091229736804962,
      "num_tokens": 55094071.0,
      "step": 26555
    },
    {
      "entropy": 5.589597702026367,
      "epoch": 2.409288824383164,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044189545778118973,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21490462124347687,
      "num_tokens": 55104504.0,
      "step": 26560
    },
    {
      "entropy": 5.570869112014771,
      "epoch": 2.4097423802612483,
      "grad_norm": 1.109375,
      "learning_rate": 0.000441873755637503,
      "loss": 4.9477,
      "mean_token_accuracy": 0.21151310801506043,
      "num_tokens": 55115722.0,
      "step": 26565
    },
    {
      "entropy": 5.536609172821045,
      "epoch": 2.410195936139332,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004418520500428244,
      "loss": 4.8922,
      "mean_token_accuracy": 0.2122745156288147,
      "num_tokens": 55126628.0,
      "step": 26570
    },
    {
      "entropy": 5.540182781219483,
      "epoch": 2.4106494920174164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044183034099760287,
      "loss": 4.936,
      "mean_token_accuracy": 0.21147914379835128,
      "num_tokens": 55137352.0,
      "step": 26575
    },
    {
      "entropy": 5.6101806640625,
      "epoch": 2.4111030478955007,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004418086285022874,
      "loss": 4.9256,
      "mean_token_accuracy": 0.22326893955469132,
      "num_tokens": 55147145.0,
      "step": 26580
    },
    {
      "entropy": 5.630547618865966,
      "epoch": 2.411556603773585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044178691255732707,
      "loss": 4.9437,
      "mean_token_accuracy": 0.21231580674648284,
      "num_tokens": 55157386.0,
      "step": 26585
    },
    {
      "entropy": 5.61014232635498,
      "epoch": 2.412010159651669,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004417651931631708,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20744775086641312,
      "num_tokens": 55168969.0,
      "step": 26590
    },
    {
      "entropy": 5.550246000289917,
      "epoch": 2.4124637155297535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004417434703202678,
      "loss": 4.9375,
      "mean_token_accuracy": 0.2032379239797592,
      "num_tokens": 55180123.0,
      "step": 26595
    },
    {
      "entropy": 5.546901369094849,
      "epoch": 2.4129172714078373,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004417217440290673,
      "loss": 4.8927,
      "mean_token_accuracy": 0.2159874141216278,
      "num_tokens": 55190668.0,
      "step": 26600
    },
    {
      "entropy": 5.527334308624267,
      "epoch": 2.4133708272859216,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004417000142900188,
      "loss": 4.8643,
      "mean_token_accuracy": 0.21732689291238785,
      "num_tokens": 55200852.0,
      "step": 26605
    },
    {
      "entropy": 5.70292558670044,
      "epoch": 2.413824383164006,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004416782811035715,
      "loss": 5.0804,
      "mean_token_accuracy": 0.20385408997535706,
      "num_tokens": 55211174.0,
      "step": 26610
    },
    {
      "entropy": 5.5272753715515135,
      "epoch": 2.41427793904209,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044165654447017485,
      "loss": 4.8274,
      "mean_token_accuracy": 0.22203521877527238,
      "num_tokens": 55222042.0,
      "step": 26615
    },
    {
      "entropy": 5.579027938842773,
      "epoch": 2.4147314949201744,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044163480439027845,
      "loss": 4.967,
      "mean_token_accuracy": 0.20782914459705354,
      "num_tokens": 55231688.0,
      "step": 26620
    },
    {
      "entropy": 5.581716680526734,
      "epoch": 2.415185050798258,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004416130608643317,
      "loss": 4.9613,
      "mean_token_accuracy": 0.21451929211616516,
      "num_tokens": 55241089.0,
      "step": 26625
    },
    {
      "entropy": 5.622543573379517,
      "epoch": 2.4156386066763424,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004415913138927844,
      "loss": 4.9023,
      "mean_token_accuracy": 0.21910626143217088,
      "num_tokens": 55251511.0,
      "step": 26630
    },
    {
      "entropy": 5.543213653564453,
      "epoch": 2.4160921625544267,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004415695634760864,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2101625144481659,
      "num_tokens": 55262855.0,
      "step": 26635
    },
    {
      "entropy": 5.511327409744263,
      "epoch": 2.416545718432511,
      "grad_norm": 1.125,
      "learning_rate": 0.0004415478096146873,
      "loss": 4.8254,
      "mean_token_accuracy": 0.2197219431400299,
      "num_tokens": 55273035.0,
      "step": 26640
    },
    {
      "entropy": 5.556900215148926,
      "epoch": 2.4169992743105952,
      "grad_norm": 1.078125,
      "learning_rate": 0.000441526052309037,
      "loss": 4.9108,
      "mean_token_accuracy": 0.22204774171113967,
      "num_tokens": 55284033.0,
      "step": 26645
    },
    {
      "entropy": 5.592622947692871,
      "epoch": 2.417452830188679,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004415042915595856,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2190182015299797,
      "num_tokens": 55294290.0,
      "step": 26650
    },
    {
      "entropy": 5.625838899612427,
      "epoch": 2.4179063860667633,
      "grad_norm": 1.0625,
      "learning_rate": 0.000441482527366783,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21190671920776366,
      "num_tokens": 55305681.0,
      "step": 26655
    },
    {
      "entropy": 5.533696269989013,
      "epoch": 2.4183599419448476,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004414607597310793,
      "loss": 4.907,
      "mean_token_accuracy": 0.2169225424528122,
      "num_tokens": 55316647.0,
      "step": 26660
    },
    {
      "entropy": 5.619781112670898,
      "epoch": 2.418813497822932,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004414389886529247,
      "loss": 5.0531,
      "mean_token_accuracy": 0.20682642012834548,
      "num_tokens": 55328318.0,
      "step": 26665
    },
    {
      "entropy": 5.645016288757324,
      "epoch": 2.419267053701016,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004414172141327693,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21090358942747117,
      "num_tokens": 55338483.0,
      "step": 26670
    },
    {
      "entropy": 5.60189962387085,
      "epoch": 2.4197206095791,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004413954361710636,
      "loss": 5.0211,
      "mean_token_accuracy": 0.21040410697460174,
      "num_tokens": 55348648.0,
      "step": 26675
    },
    {
      "entropy": 5.540328168869019,
      "epoch": 2.4201741654571842,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044137365476825785,
      "loss": 4.9061,
      "mean_token_accuracy": 0.21204569935798645,
      "num_tokens": 55358599.0,
      "step": 26680
    },
    {
      "entropy": 5.634864091873169,
      "epoch": 2.4206277213352685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044135186992480253,
      "loss": 4.9466,
      "mean_token_accuracy": 0.21380085200071336,
      "num_tokens": 55369847.0,
      "step": 26685
    },
    {
      "entropy": 5.568818044662476,
      "epoch": 2.4210812772133528,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044133008164114814,
      "loss": 4.941,
      "mean_token_accuracy": 0.2139912351965904,
      "num_tokens": 55381235.0,
      "step": 26690
    },
    {
      "entropy": 5.564904975891113,
      "epoch": 2.421534833091437,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044130828991774526,
      "loss": 4.9743,
      "mean_token_accuracy": 0.2082348197698593,
      "num_tokens": 55391527.0,
      "step": 26695
    },
    {
      "entropy": 5.620860910415649,
      "epoch": 2.4219883889695213,
      "grad_norm": 1.125,
      "learning_rate": 0.00044128649475504454,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20783987045288085,
      "num_tokens": 55401528.0,
      "step": 26700
    },
    {
      "entropy": 5.638621282577515,
      "epoch": 2.422441944847605,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004412646961534968,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20500356405973436,
      "num_tokens": 55411411.0,
      "step": 26705
    },
    {
      "entropy": 5.496764421463013,
      "epoch": 2.4228955007256894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004412428941135528,
      "loss": 4.7909,
      "mean_token_accuracy": 0.22079988569021225,
      "num_tokens": 55423164.0,
      "step": 26710
    },
    {
      "entropy": 5.562102031707764,
      "epoch": 2.4233490566037736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004412210886356633,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22214736938476562,
      "num_tokens": 55433397.0,
      "step": 26715
    },
    {
      "entropy": 5.551472425460815,
      "epoch": 2.423802612481858,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044119927972027936,
      "loss": 4.9297,
      "mean_token_accuracy": 0.2118145391345024,
      "num_tokens": 55442898.0,
      "step": 26720
    },
    {
      "entropy": 5.5450035572052006,
      "epoch": 2.4242561683599417,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044117746736785195,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21898261159658433,
      "num_tokens": 55452226.0,
      "step": 26725
    },
    {
      "entropy": 5.585405063629151,
      "epoch": 2.424709724238026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004411556515788322,
      "loss": 4.935,
      "mean_token_accuracy": 0.21154498159885407,
      "num_tokens": 55462503.0,
      "step": 26730
    },
    {
      "entropy": 5.639528274536133,
      "epoch": 2.4251632801161103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004411338323536712,
      "loss": 5.0724,
      "mean_token_accuracy": 0.2096162259578705,
      "num_tokens": 55472789.0,
      "step": 26735
    },
    {
      "entropy": 5.54489369392395,
      "epoch": 2.4256168359941945,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004411120096928203,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2140708476305008,
      "num_tokens": 55482572.0,
      "step": 26740
    },
    {
      "entropy": 5.596988201141357,
      "epoch": 2.426070391872279,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004410901835967306,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21821867376565934,
      "num_tokens": 55493033.0,
      "step": 26745
    },
    {
      "entropy": 5.57719464302063,
      "epoch": 2.426523947750363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044106835406585354,
      "loss": 4.8818,
      "mean_token_accuracy": 0.21200233548879624,
      "num_tokens": 55503574.0,
      "step": 26750
    },
    {
      "entropy": 5.626631355285644,
      "epoch": 2.426977503628447,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044104652110064067,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20684119313955307,
      "num_tokens": 55514119.0,
      "step": 26755
    },
    {
      "entropy": 5.5654267311096195,
      "epoch": 2.427431059506531,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004410246847015435,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21291468888521195,
      "num_tokens": 55524973.0,
      "step": 26760
    },
    {
      "entropy": 5.560578012466431,
      "epoch": 2.4278846153846154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004410028448690134,
      "loss": 4.9826,
      "mean_token_accuracy": 0.21115875989198685,
      "num_tokens": 55534847.0,
      "step": 26765
    },
    {
      "entropy": 5.625048875808716,
      "epoch": 2.4283381712626997,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004409810016035023,
      "loss": 4.9555,
      "mean_token_accuracy": 0.20490000545978546,
      "num_tokens": 55544886.0,
      "step": 26770
    },
    {
      "entropy": 5.583855676651001,
      "epoch": 2.4287917271407835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004409591549054617,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20886138081550598,
      "num_tokens": 55554534.0,
      "step": 26775
    },
    {
      "entropy": 5.600439214706421,
      "epoch": 2.4292452830188678,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044093730477534355,
      "loss": 4.9783,
      "mean_token_accuracy": 0.206478089094162,
      "num_tokens": 55564302.0,
      "step": 26780
    },
    {
      "entropy": 5.533636617660522,
      "epoch": 2.429698838896952,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004409154512135996,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21612302958965302,
      "num_tokens": 55574543.0,
      "step": 26785
    },
    {
      "entropy": 5.534697389602661,
      "epoch": 2.4301523947750363,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004408935942206818,
      "loss": 4.9004,
      "mean_token_accuracy": 0.21744074523448945,
      "num_tokens": 55583849.0,
      "step": 26790
    },
    {
      "entropy": 5.5638738632202145,
      "epoch": 2.4306059506531206,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004408717337970423,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20987699925899506,
      "num_tokens": 55594294.0,
      "step": 26795
    },
    {
      "entropy": 5.579269886016846,
      "epoch": 2.431059506531205,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044084986994313304,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21099089533090593,
      "num_tokens": 55604185.0,
      "step": 26800
    },
    {
      "entropy": 5.560730695724487,
      "epoch": 2.4315130624092887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004408280026594062,
      "loss": 4.9014,
      "mean_token_accuracy": 0.214804445207119,
      "num_tokens": 55615203.0,
      "step": 26805
    },
    {
      "entropy": 5.561976432800293,
      "epoch": 2.431966618287373,
      "grad_norm": 1.125,
      "learning_rate": 0.000440806131946314,
      "loss": 4.9852,
      "mean_token_accuracy": 0.21126452535390855,
      "num_tokens": 55626157.0,
      "step": 26810
    },
    {
      "entropy": 5.628096294403076,
      "epoch": 2.432420174165457,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044078425780430877,
      "loss": 4.956,
      "mean_token_accuracy": 0.20642924904823304,
      "num_tokens": 55636862.0,
      "step": 26815
    },
    {
      "entropy": 5.566771030426025,
      "epoch": 2.4328737300435415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004407623802338428,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2200087621808052,
      "num_tokens": 55647646.0,
      "step": 26820
    },
    {
      "entropy": 5.568132448196411,
      "epoch": 2.4333272859216257,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044074049923536855,
      "loss": 4.9186,
      "mean_token_accuracy": 0.2192446321249008,
      "num_tokens": 55659847.0,
      "step": 26825
    },
    {
      "entropy": 5.514598560333252,
      "epoch": 2.4337808417997095,
      "grad_norm": 1.125,
      "learning_rate": 0.0004407186148093386,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21441839039325714,
      "num_tokens": 55670626.0,
      "step": 26830
    },
    {
      "entropy": 5.5989141941070555,
      "epoch": 2.434234397677794,
      "grad_norm": 1.125,
      "learning_rate": 0.00044069672695620543,
      "loss": 4.992,
      "mean_token_accuracy": 0.21164102256298065,
      "num_tokens": 55681368.0,
      "step": 26835
    },
    {
      "entropy": 5.559159183502198,
      "epoch": 2.434687953555878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004406748356764217,
      "loss": 4.8726,
      "mean_token_accuracy": 0.20950720459222794,
      "num_tokens": 55691606.0,
      "step": 26840
    },
    {
      "entropy": 5.6074470520019535,
      "epoch": 2.4351415094339623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044065294097044017,
      "loss": 5.0474,
      "mean_token_accuracy": 0.21100991666316987,
      "num_tokens": 55702223.0,
      "step": 26845
    },
    {
      "entropy": 5.5065570831298825,
      "epoch": 2.4355950653120466,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004406310428387136,
      "loss": 4.8449,
      "mean_token_accuracy": 0.21619075089693068,
      "num_tokens": 55712729.0,
      "step": 26850
    },
    {
      "entropy": 5.5781244277954105,
      "epoch": 2.4360486211901304,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004406091412816949,
      "loss": 4.9549,
      "mean_token_accuracy": 0.21669239848852156,
      "num_tokens": 55722863.0,
      "step": 26855
    },
    {
      "entropy": 5.645908784866333,
      "epoch": 2.4365021770682147,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044058723629983694,
      "loss": 5.0198,
      "mean_token_accuracy": 0.21438144147396088,
      "num_tokens": 55733124.0,
      "step": 26860
    },
    {
      "entropy": 5.6440221786499025,
      "epoch": 2.436955732946299,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004405653278935927,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21672249734401702,
      "num_tokens": 55743438.0,
      "step": 26865
    },
    {
      "entropy": 5.510346174240112,
      "epoch": 2.4374092888243832,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044054341606341533,
      "loss": 4.8227,
      "mean_token_accuracy": 0.21453460305929184,
      "num_tokens": 55753602.0,
      "step": 26870
    },
    {
      "entropy": 5.505764961242676,
      "epoch": 2.4378628447024675,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004405215008097579,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22253549844026566,
      "num_tokens": 55763473.0,
      "step": 26875
    },
    {
      "entropy": 5.59206075668335,
      "epoch": 2.4383164005805513,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044049958213307364,
      "loss": 4.8864,
      "mean_token_accuracy": 0.21952882260084153,
      "num_tokens": 55772848.0,
      "step": 26880
    },
    {
      "entropy": 5.527996444702149,
      "epoch": 2.4387699564586356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044047766003381584,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22041210532188416,
      "num_tokens": 55783468.0,
      "step": 26885
    },
    {
      "entropy": 5.537233209609985,
      "epoch": 2.43922351233672,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044045573451243793,
      "loss": 4.9569,
      "mean_token_accuracy": 0.21993866711854934,
      "num_tokens": 55793619.0,
      "step": 26890
    },
    {
      "entropy": 5.569816637039184,
      "epoch": 2.439677068214804,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044043380556939326,
      "loss": 4.9353,
      "mean_token_accuracy": 0.20997764617204667,
      "num_tokens": 55803869.0,
      "step": 26895
    },
    {
      "entropy": 5.627328300476075,
      "epoch": 2.4401306240928884,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004404118732051353,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21249611228704451,
      "num_tokens": 55814754.0,
      "step": 26900
    },
    {
      "entropy": 5.576613998413086,
      "epoch": 2.4405841799709727,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004403899374201176,
      "loss": 4.888,
      "mean_token_accuracy": 0.21150109469890593,
      "num_tokens": 55824862.0,
      "step": 26905
    },
    {
      "entropy": 5.489608192443848,
      "epoch": 2.4410377358490565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044036799821479397,
      "loss": 4.8968,
      "mean_token_accuracy": 0.21324735581874849,
      "num_tokens": 55835506.0,
      "step": 26910
    },
    {
      "entropy": 5.601044845581055,
      "epoch": 2.4414912917271407,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044034605558961786,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20867536216974258,
      "num_tokens": 55846244.0,
      "step": 26915
    },
    {
      "entropy": 5.527788400650024,
      "epoch": 2.441944847605225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044032410954504325,
      "loss": 4.8862,
      "mean_token_accuracy": 0.222291462123394,
      "num_tokens": 55856192.0,
      "step": 26920
    },
    {
      "entropy": 5.570795631408691,
      "epoch": 2.4423984034833093,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044030216008152394,
      "loss": 4.9374,
      "mean_token_accuracy": 0.2087250292301178,
      "num_tokens": 55866509.0,
      "step": 26925
    },
    {
      "entropy": 5.595271015167237,
      "epoch": 2.442851959361393,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044028020719951376,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21611881256103516,
      "num_tokens": 55877321.0,
      "step": 26930
    },
    {
      "entropy": 5.596599817276001,
      "epoch": 2.4433055152394774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004402582508994669,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2107618883252144,
      "num_tokens": 55887933.0,
      "step": 26935
    },
    {
      "entropy": 5.519869232177735,
      "epoch": 2.4437590711175616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044023629118183726,
      "loss": 4.8762,
      "mean_token_accuracy": 0.21172595918178558,
      "num_tokens": 55897274.0,
      "step": 26940
    },
    {
      "entropy": 5.589644050598144,
      "epoch": 2.444212626995646,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044021432804707895,
      "loss": 5.0299,
      "mean_token_accuracy": 0.20810270458459854,
      "num_tokens": 55907393.0,
      "step": 26945
    },
    {
      "entropy": 5.546515941619873,
      "epoch": 2.44466618287373,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004401923614956463,
      "loss": 4.9148,
      "mean_token_accuracy": 0.20886614173650742,
      "num_tokens": 55918429.0,
      "step": 26950
    },
    {
      "entropy": 5.569817018508911,
      "epoch": 2.4451197387518144,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004401703915279935,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21486484855413437,
      "num_tokens": 55928733.0,
      "step": 26955
    },
    {
      "entropy": 5.615135669708252,
      "epoch": 2.4455732946298983,
      "grad_norm": 1.125,
      "learning_rate": 0.00044014841814457486,
      "loss": 4.88,
      "mean_token_accuracy": 0.2158233180642128,
      "num_tokens": 55939128.0,
      "step": 26960
    },
    {
      "entropy": 5.612682914733886,
      "epoch": 2.4460268505079825,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004401264413458449,
      "loss": 5.045,
      "mean_token_accuracy": 0.2082000896334648,
      "num_tokens": 55950237.0,
      "step": 26965
    },
    {
      "entropy": 5.577176237106324,
      "epoch": 2.446480406386067,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004401044611322581,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22138510644435883,
      "num_tokens": 55960436.0,
      "step": 26970
    },
    {
      "entropy": 5.550209188461304,
      "epoch": 2.446933962264151,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004400824775042689,
      "loss": 4.8352,
      "mean_token_accuracy": 0.21578322649002074,
      "num_tokens": 55970446.0,
      "step": 26975
    },
    {
      "entropy": 5.563942766189575,
      "epoch": 2.447387518142235,
      "grad_norm": 1.125,
      "learning_rate": 0.000440060490462332,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21210048198699952,
      "num_tokens": 55980706.0,
      "step": 26980
    },
    {
      "entropy": 5.5506431579589846,
      "epoch": 2.447841074020319,
      "grad_norm": 1.125,
      "learning_rate": 0.0004400385000069021,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2054441064596176,
      "num_tokens": 55990669.0,
      "step": 26985
    },
    {
      "entropy": 5.665261888504029,
      "epoch": 2.4482946298984034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044001650613843395,
      "loss": 5.0159,
      "mean_token_accuracy": 0.2040278732776642,
      "num_tokens": 56000709.0,
      "step": 26990
    },
    {
      "entropy": 5.625793933868408,
      "epoch": 2.4487481857764877,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004399945088573824,
      "loss": 4.9142,
      "mean_token_accuracy": 0.22189556807279587,
      "num_tokens": 56011435.0,
      "step": 26995
    },
    {
      "entropy": 5.586576795578003,
      "epoch": 2.449201741654572,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043997250816420246,
      "loss": 4.9081,
      "mean_token_accuracy": 0.21742316484451293,
      "num_tokens": 56021264.0,
      "step": 27000
    },
    {
      "epoch": 2.449201741654572,
      "eval_entropy": 5.310908531495349,
      "eval_loss": 5.009191989898682,
      "eval_mean_token_accuracy": 0.2188378827165964,
      "eval_num_tokens": 56021264.0,
      "eval_runtime": 20.6659,
      "eval_samples_per_second": 1711.035,
      "eval_steps_per_second": 213.879,
      "step": 27000
    },
    {
      "entropy": 5.570468521118164,
      "epoch": 2.449655297532656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004399505040593489,
      "loss": 4.9535,
      "mean_token_accuracy": 0.21505245566368103,
      "num_tokens": 56031463.0,
      "step": 27005
    },
    {
      "entropy": 5.518310689926148,
      "epoch": 2.45010885341074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043992849654327695,
      "loss": 4.8361,
      "mean_token_accuracy": 0.22150550931692123,
      "num_tokens": 56041195.0,
      "step": 27010
    },
    {
      "entropy": 5.566996955871582,
      "epoch": 2.4505624092888243,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004399064856164416,
      "loss": 4.9708,
      "mean_token_accuracy": 0.20919194519519807,
      "num_tokens": 56051840.0,
      "step": 27015
    },
    {
      "entropy": 5.593413543701172,
      "epoch": 2.4510159651669086,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043988447127929817,
      "loss": 4.9754,
      "mean_token_accuracy": 0.21148500442504883,
      "num_tokens": 56062535.0,
      "step": 27020
    },
    {
      "entropy": 5.578488445281982,
      "epoch": 2.451469521044993,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043986245353230173,
      "loss": 4.8841,
      "mean_token_accuracy": 0.22253292053937912,
      "num_tokens": 56072177.0,
      "step": 27025
    },
    {
      "entropy": 5.546461963653565,
      "epoch": 2.451923076923077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043984043237590785,
      "loss": 4.9689,
      "mean_token_accuracy": 0.2161324664950371,
      "num_tokens": 56081607.0,
      "step": 27030
    },
    {
      "entropy": 5.568353462219238,
      "epoch": 2.452376632801161,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043981840781057185,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20944924652576447,
      "num_tokens": 56093371.0,
      "step": 27035
    },
    {
      "entropy": 5.5949352264404295,
      "epoch": 2.452830188679245,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043979637983674913,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20938770920038224,
      "num_tokens": 56103973.0,
      "step": 27040
    },
    {
      "entropy": 5.565538311004639,
      "epoch": 2.4532837445573294,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004397743484548953,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20909886062145233,
      "num_tokens": 56113906.0,
      "step": 27045
    },
    {
      "entropy": 5.573693943023682,
      "epoch": 2.4537373004354137,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004397523136654659,
      "loss": 4.9105,
      "mean_token_accuracy": 0.210191810131073,
      "num_tokens": 56124172.0,
      "step": 27050
    },
    {
      "entropy": 5.5464723110198975,
      "epoch": 2.454190856313498,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043973027546891677,
      "loss": 4.8016,
      "mean_token_accuracy": 0.22593728750944136,
      "num_tokens": 56133088.0,
      "step": 27055
    },
    {
      "entropy": 5.549463558197021,
      "epoch": 2.4546444121915822,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004397082338657034,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21861016154289245,
      "num_tokens": 56143098.0,
      "step": 27060
    },
    {
      "entropy": 5.53235502243042,
      "epoch": 2.455097968069666,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043968618885628185,
      "loss": 4.8941,
      "mean_token_accuracy": 0.2166932627558708,
      "num_tokens": 56153889.0,
      "step": 27065
    },
    {
      "entropy": 5.540852499008179,
      "epoch": 2.4555515239477503,
      "grad_norm": 1.25,
      "learning_rate": 0.00043966414044110805,
      "loss": 4.8671,
      "mean_token_accuracy": 0.21665560603141784,
      "num_tokens": 56164164.0,
      "step": 27070
    },
    {
      "entropy": 5.6239914894104,
      "epoch": 2.4560050798258346,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043964208862063775,
      "loss": 4.9725,
      "mean_token_accuracy": 0.20838364064693451,
      "num_tokens": 56173172.0,
      "step": 27075
    },
    {
      "entropy": 5.590054845809936,
      "epoch": 2.456458635703919,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043962003339532714,
      "loss": 4.925,
      "mean_token_accuracy": 0.21700985580682755,
      "num_tokens": 56183127.0,
      "step": 27080
    },
    {
      "entropy": 5.617047834396362,
      "epoch": 2.4569121915820027,
      "grad_norm": 1.125,
      "learning_rate": 0.00043959797476563226,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20998311936855316,
      "num_tokens": 56192528.0,
      "step": 27085
    },
    {
      "entropy": 5.546414995193482,
      "epoch": 2.457365747460087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043957591273200943,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21502474844455718,
      "num_tokens": 56203239.0,
      "step": 27090
    },
    {
      "entropy": 5.59856219291687,
      "epoch": 2.4578193033381712,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004395538472949146,
      "loss": 4.9704,
      "mean_token_accuracy": 0.20955437868833543,
      "num_tokens": 56214068.0,
      "step": 27095
    },
    {
      "entropy": 5.61622805595398,
      "epoch": 2.4582728592162555,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004395317784548045,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20501068830490113,
      "num_tokens": 56224829.0,
      "step": 27100
    },
    {
      "entropy": 5.663437223434448,
      "epoch": 2.4587264150943398,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004395097062121351,
      "loss": 4.9764,
      "mean_token_accuracy": 0.20961254239082336,
      "num_tokens": 56235412.0,
      "step": 27105
    },
    {
      "entropy": 5.640355491638184,
      "epoch": 2.459179970972424,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043948763056736315,
      "loss": 4.9944,
      "mean_token_accuracy": 0.2067007914185524,
      "num_tokens": 56246785.0,
      "step": 27110
    },
    {
      "entropy": 5.5982475757598875,
      "epoch": 2.459633526850508,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000439465551520945,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20904777944087982,
      "num_tokens": 56256701.0,
      "step": 27115
    },
    {
      "entropy": 5.563984632492065,
      "epoch": 2.460087082728592,
      "grad_norm": 1.125,
      "learning_rate": 0.00043944346907333744,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21463871151208877,
      "num_tokens": 56267418.0,
      "step": 27120
    },
    {
      "entropy": 5.562194728851319,
      "epoch": 2.4605406386066764,
      "grad_norm": 1.015625,
      "learning_rate": 0.000439421383224997,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21297574192285537,
      "num_tokens": 56278232.0,
      "step": 27125
    },
    {
      "entropy": 5.635591316223144,
      "epoch": 2.4609941944847606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004393992939763804,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21307477504014968,
      "num_tokens": 56289866.0,
      "step": 27130
    },
    {
      "entropy": 5.593373823165893,
      "epoch": 2.4614477503628445,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043937720132794457,
      "loss": 4.9077,
      "mean_token_accuracy": 0.21929869055747986,
      "num_tokens": 56299595.0,
      "step": 27135
    },
    {
      "entropy": 5.610020589828491,
      "epoch": 2.4619013062409287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043935510528014625,
      "loss": 4.9751,
      "mean_token_accuracy": 0.21973224878311157,
      "num_tokens": 56310152.0,
      "step": 27140
    },
    {
      "entropy": 5.591468572616577,
      "epoch": 2.462354862119013,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004393330058334426,
      "loss": 5.0402,
      "mean_token_accuracy": 0.20449508130550384,
      "num_tokens": 56320288.0,
      "step": 27145
    },
    {
      "entropy": 5.622226905822754,
      "epoch": 2.4628084179970973,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043931090298829047,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20896437019109726,
      "num_tokens": 56330345.0,
      "step": 27150
    },
    {
      "entropy": 5.662629175186157,
      "epoch": 2.4632619738751815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043928879674514694,
      "loss": 5.0846,
      "mean_token_accuracy": 0.21637462973594665,
      "num_tokens": 56340996.0,
      "step": 27155
    },
    {
      "entropy": 5.495919227600098,
      "epoch": 2.463715529753266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004392666871044692,
      "loss": 4.8101,
      "mean_token_accuracy": 0.22434767037630082,
      "num_tokens": 56350326.0,
      "step": 27160
    },
    {
      "entropy": 5.587428092956543,
      "epoch": 2.4641690856313496,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043924457406671456,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2184249699115753,
      "num_tokens": 56360556.0,
      "step": 27165
    },
    {
      "entropy": 5.617318773269654,
      "epoch": 2.464622641509434,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004392224576323403,
      "loss": 4.9017,
      "mean_token_accuracy": 0.21992544382810592,
      "num_tokens": 56369980.0,
      "step": 27170
    },
    {
      "entropy": 5.587795972824097,
      "epoch": 2.465076197387518,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043920033780180367,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20803798735141754,
      "num_tokens": 56379974.0,
      "step": 27175
    },
    {
      "entropy": 5.587786102294922,
      "epoch": 2.4655297532656024,
      "grad_norm": 1.125,
      "learning_rate": 0.0004391782145755623,
      "loss": 4.8843,
      "mean_token_accuracy": 0.21582925468683242,
      "num_tokens": 56390542.0,
      "step": 27180
    },
    {
      "entropy": 5.689225053787231,
      "epoch": 2.4659833091436867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004391560879540736,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20784496814012526,
      "num_tokens": 56401755.0,
      "step": 27185
    },
    {
      "entropy": 5.564447259902954,
      "epoch": 2.4664368650217705,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004391339579377952,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2139232873916626,
      "num_tokens": 56412223.0,
      "step": 27190
    },
    {
      "entropy": 5.60780782699585,
      "epoch": 2.4668904208998548,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043911182452718476,
      "loss": 4.9794,
      "mean_token_accuracy": 0.21333203166723252,
      "num_tokens": 56422947.0,
      "step": 27195
    },
    {
      "entropy": 5.5757640361785885,
      "epoch": 2.467343976777939,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043908968772269984,
      "loss": 4.9293,
      "mean_token_accuracy": 0.2183648630976677,
      "num_tokens": 56433189.0,
      "step": 27200
    },
    {
      "entropy": 5.620698690414429,
      "epoch": 2.4677975326560233,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043906754752479836,
      "loss": 4.9579,
      "mean_token_accuracy": 0.21366307139396667,
      "num_tokens": 56444460.0,
      "step": 27205
    },
    {
      "entropy": 5.621269369125367,
      "epoch": 2.4682510885341076,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004390454039339383,
      "loss": 4.9845,
      "mean_token_accuracy": 0.2081169918179512,
      "num_tokens": 56454368.0,
      "step": 27210
    },
    {
      "entropy": 5.539461755752564,
      "epoch": 2.4687046444121914,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043902325695057743,
      "loss": 4.8793,
      "mean_token_accuracy": 0.21865448653697966,
      "num_tokens": 56464153.0,
      "step": 27215
    },
    {
      "entropy": 5.555964565277099,
      "epoch": 2.4691582002902757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043900110657517377,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21785708516836166,
      "num_tokens": 56473627.0,
      "step": 27220
    },
    {
      "entropy": 5.588455629348755,
      "epoch": 2.46961175616836,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004389789528081855,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2176984205842018,
      "num_tokens": 56483660.0,
      "step": 27225
    },
    {
      "entropy": 5.618128061294556,
      "epoch": 2.470065312046444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004389567956500707,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21988930851221083,
      "num_tokens": 56492653.0,
      "step": 27230
    },
    {
      "entropy": 5.671018075942993,
      "epoch": 2.4705188679245285,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004389346351012876,
      "loss": 5.0523,
      "mean_token_accuracy": 0.2028290942311287,
      "num_tokens": 56503289.0,
      "step": 27235
    },
    {
      "entropy": 5.635839033126831,
      "epoch": 2.4709724238026123,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043891247116229436,
      "loss": 4.9963,
      "mean_token_accuracy": 0.20613661110401155,
      "num_tokens": 56513843.0,
      "step": 27240
    },
    {
      "entropy": 5.625729131698608,
      "epoch": 2.4714259796806965,
      "grad_norm": 1.125,
      "learning_rate": 0.00043889030383354957,
      "loss": 4.97,
      "mean_token_accuracy": 0.20706405639648437,
      "num_tokens": 56524175.0,
      "step": 27245
    },
    {
      "entropy": 5.5597631454467775,
      "epoch": 2.471879535558781,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004388681331155114,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2226969212293625,
      "num_tokens": 56534976.0,
      "step": 27250
    },
    {
      "entropy": 5.607913303375244,
      "epoch": 2.472333091436865,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004388459590086386,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21343802511692048,
      "num_tokens": 56545967.0,
      "step": 27255
    },
    {
      "entropy": 5.615893077850342,
      "epoch": 2.4727866473149493,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043882378151338956,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20648104101419448,
      "num_tokens": 56555527.0,
      "step": 27260
    },
    {
      "entropy": 5.625734519958496,
      "epoch": 2.4732402031930336,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000438801600630223,
      "loss": 4.9561,
      "mean_token_accuracy": 0.2064664050936699,
      "num_tokens": 56565874.0,
      "step": 27265
    },
    {
      "entropy": 5.577193641662598,
      "epoch": 2.4736937590711174,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043877941635959755,
      "loss": 4.882,
      "mean_token_accuracy": 0.21218475699424744,
      "num_tokens": 56575452.0,
      "step": 27270
    },
    {
      "entropy": 5.497557878494263,
      "epoch": 2.4741473149492017,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043875722870197214,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2242387428879738,
      "num_tokens": 56586176.0,
      "step": 27275
    },
    {
      "entropy": 5.629019355773925,
      "epoch": 2.474600870827286,
      "grad_norm": 1.25,
      "learning_rate": 0.0004387350376578054,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21332286596298217,
      "num_tokens": 56595473.0,
      "step": 27280
    },
    {
      "entropy": 5.600218296051025,
      "epoch": 2.4750544267053702,
      "grad_norm": 1.125,
      "learning_rate": 0.0004387128432275563,
      "loss": 4.9419,
      "mean_token_accuracy": 0.21243224143981934,
      "num_tokens": 56606016.0,
      "step": 27285
    },
    {
      "entropy": 5.559001731872558,
      "epoch": 2.475507982583454,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000438690645411684,
      "loss": 4.9521,
      "mean_token_accuracy": 0.2098272144794464,
      "num_tokens": 56617674.0,
      "step": 27290
    },
    {
      "entropy": 5.595889949798584,
      "epoch": 2.4759615384615383,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043866844421064745,
      "loss": 4.918,
      "mean_token_accuracy": 0.21643660962581635,
      "num_tokens": 56628077.0,
      "step": 27295
    },
    {
      "entropy": 5.619894218444824,
      "epoch": 2.4764150943396226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043864623962490575,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2159027114510536,
      "num_tokens": 56638691.0,
      "step": 27300
    },
    {
      "entropy": 5.590474224090576,
      "epoch": 2.476868650217707,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004386240316549181,
      "loss": 4.9088,
      "mean_token_accuracy": 0.21426018923521042,
      "num_tokens": 56650338.0,
      "step": 27305
    },
    {
      "entropy": 5.562205076217651,
      "epoch": 2.477322206095791,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043860182030114374,
      "loss": 4.8651,
      "mean_token_accuracy": 0.2227303832769394,
      "num_tokens": 56660328.0,
      "step": 27310
    },
    {
      "entropy": 5.633257436752319,
      "epoch": 2.4777757619738754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043857960556404214,
      "loss": 5.008,
      "mean_token_accuracy": 0.21100738793611526,
      "num_tokens": 56670674.0,
      "step": 27315
    },
    {
      "entropy": 5.5805905818939205,
      "epoch": 2.478229317851959,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004385573874440725,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2148086979985237,
      "num_tokens": 56680656.0,
      "step": 27320
    },
    {
      "entropy": 5.547960376739502,
      "epoch": 2.4786828737300435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043853516594169446,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22031059563159944,
      "num_tokens": 56691039.0,
      "step": 27325
    },
    {
      "entropy": 5.58705849647522,
      "epoch": 2.4791364296081277,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004385129410573675,
      "loss": 4.9222,
      "mean_token_accuracy": 0.20640294700860978,
      "num_tokens": 56703101.0,
      "step": 27330
    },
    {
      "entropy": 5.619821214675904,
      "epoch": 2.479589985486212,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043849071279155124,
      "loss": 4.9872,
      "mean_token_accuracy": 0.218429596722126,
      "num_tokens": 56712351.0,
      "step": 27335
    },
    {
      "entropy": 5.572011566162109,
      "epoch": 2.480043541364296,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043846848114470537,
      "loss": 4.9407,
      "mean_token_accuracy": 0.22059279680252075,
      "num_tokens": 56722836.0,
      "step": 27340
    },
    {
      "entropy": 5.606530284881591,
      "epoch": 2.48049709724238,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043844624611728974,
      "loss": 4.9501,
      "mean_token_accuracy": 0.20923953354358674,
      "num_tokens": 56734032.0,
      "step": 27345
    },
    {
      "entropy": 5.61112380027771,
      "epoch": 2.4809506531204644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043842400770976406,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21352135837078096,
      "num_tokens": 56743536.0,
      "step": 27350
    },
    {
      "entropy": 5.558936548233032,
      "epoch": 2.4814042089985486,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004384017659225882,
      "loss": 4.9153,
      "mean_token_accuracy": 0.21404812783002852,
      "num_tokens": 56753475.0,
      "step": 27355
    },
    {
      "entropy": 5.512931776046753,
      "epoch": 2.481857764876633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043837952075622224,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2228367283940315,
      "num_tokens": 56763602.0,
      "step": 27360
    },
    {
      "entropy": 5.564649868011474,
      "epoch": 2.482311320754717,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004383572722111261,
      "loss": 4.8722,
      "mean_token_accuracy": 0.2213428407907486,
      "num_tokens": 56773859.0,
      "step": 27365
    },
    {
      "entropy": 5.554527521133423,
      "epoch": 2.482764876632801,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043833502028775995,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21437181979417802,
      "num_tokens": 56783810.0,
      "step": 27370
    },
    {
      "entropy": 5.593163251876831,
      "epoch": 2.4832184325108853,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043831276498658395,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20869463980197905,
      "num_tokens": 56794088.0,
      "step": 27375
    },
    {
      "entropy": 5.510219144821167,
      "epoch": 2.4836719883889695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004382905063080584,
      "loss": 4.8682,
      "mean_token_accuracy": 0.21661886125802993,
      "num_tokens": 56804623.0,
      "step": 27380
    },
    {
      "entropy": 5.548147821426392,
      "epoch": 2.484125544267054,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043826824425264357,
      "loss": 4.932,
      "mean_token_accuracy": 0.2082523375749588,
      "num_tokens": 56813842.0,
      "step": 27385
    },
    {
      "entropy": 5.619441175460816,
      "epoch": 2.484579100145138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004382459788207997,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20921426266431808,
      "num_tokens": 56823501.0,
      "step": 27390
    },
    {
      "entropy": 5.594589424133301,
      "epoch": 2.485032656023222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043822371001298754,
      "loss": 4.9554,
      "mean_token_accuracy": 0.21391973346471788,
      "num_tokens": 56834601.0,
      "step": 27395
    },
    {
      "entropy": 5.54300012588501,
      "epoch": 2.485486211901306,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004382014378296674,
      "loss": 4.9594,
      "mean_token_accuracy": 0.20829545557498932,
      "num_tokens": 56844701.0,
      "step": 27400
    },
    {
      "entropy": 5.532312822341919,
      "epoch": 2.4859397677793904,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004381791622713,
      "loss": 4.9275,
      "mean_token_accuracy": 0.21718046814203262,
      "num_tokens": 56854867.0,
      "step": 27405
    },
    {
      "entropy": 5.629356527328492,
      "epoch": 2.4863933236574747,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000438156883338346,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21532835215330123,
      "num_tokens": 56865688.0,
      "step": 27410
    },
    {
      "entropy": 5.6419556617736815,
      "epoch": 2.486846879535559,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043813460103126595,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2146413117647171,
      "num_tokens": 56876504.0,
      "step": 27415
    },
    {
      "entropy": 5.575489521026611,
      "epoch": 2.487300435413643,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043811231535052076,
      "loss": 4.898,
      "mean_token_accuracy": 0.21808527261018754,
      "num_tokens": 56886083.0,
      "step": 27420
    },
    {
      "entropy": 5.528043127059936,
      "epoch": 2.487753991291727,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004380900262965714,
      "loss": 4.9758,
      "mean_token_accuracy": 0.21830589771270753,
      "num_tokens": 56897079.0,
      "step": 27425
    },
    {
      "entropy": 5.58110933303833,
      "epoch": 2.4882075471698113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004380677338698787,
      "loss": 4.8889,
      "mean_token_accuracy": 0.21870796531438827,
      "num_tokens": 56906962.0,
      "step": 27430
    },
    {
      "entropy": 5.633279848098755,
      "epoch": 2.4886611030478956,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043804543807090374,
      "loss": 5.0077,
      "mean_token_accuracy": 0.20929981619119645,
      "num_tokens": 56918156.0,
      "step": 27435
    },
    {
      "entropy": 5.578549575805664,
      "epoch": 2.48911465892598,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043802313890010757,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2116858184337616,
      "num_tokens": 56928888.0,
      "step": 27440
    },
    {
      "entropy": 5.604401922225952,
      "epoch": 2.4895682148040637,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004380008363579513,
      "loss": 4.9251,
      "mean_token_accuracy": 0.21217681914567948,
      "num_tokens": 56939929.0,
      "step": 27445
    },
    {
      "entropy": 5.629007005691529,
      "epoch": 2.490021770682148,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043797853044489623,
      "loss": 4.8465,
      "mean_token_accuracy": 0.2207060232758522,
      "num_tokens": 56950992.0,
      "step": 27450
    },
    {
      "entropy": 5.669310283660889,
      "epoch": 2.490475326560232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004379562211614036,
      "loss": 5.0206,
      "mean_token_accuracy": 0.2106938138604164,
      "num_tokens": 56962118.0,
      "step": 27455
    },
    {
      "entropy": 5.563676357269287,
      "epoch": 2.4909288824383164,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004379339085079348,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21446316093206405,
      "num_tokens": 56972837.0,
      "step": 27460
    },
    {
      "entropy": 5.606867027282715,
      "epoch": 2.4913824383164007,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004379115924849512,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2118871331214905,
      "num_tokens": 56982794.0,
      "step": 27465
    },
    {
      "entropy": 5.67323784828186,
      "epoch": 2.491835994194485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004378892730929145,
      "loss": 5.0805,
      "mean_token_accuracy": 0.2094934642314911,
      "num_tokens": 56993979.0,
      "step": 27470
    },
    {
      "entropy": 5.622083377838135,
      "epoch": 2.492289550072569,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043786695033228604,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21009259968996047,
      "num_tokens": 57004505.0,
      "step": 27475
    },
    {
      "entropy": 5.632649374008179,
      "epoch": 2.492743105950653,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043784462420352747,
      "loss": 4.97,
      "mean_token_accuracy": 0.20422477275133133,
      "num_tokens": 57014121.0,
      "step": 27480
    },
    {
      "entropy": 5.660454797744751,
      "epoch": 2.4931966618287373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043782229470710067,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2153025597333908,
      "num_tokens": 57023748.0,
      "step": 27485
    },
    {
      "entropy": 5.545122814178467,
      "epoch": 2.4936502177068216,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004377999618434673,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21740242540836335,
      "num_tokens": 57034782.0,
      "step": 27490
    },
    {
      "entropy": 5.540149545669555,
      "epoch": 2.4941037735849054,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004377776256130892,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2168115645647049,
      "num_tokens": 57045136.0,
      "step": 27495
    },
    {
      "entropy": 5.610564136505127,
      "epoch": 2.4945573294629897,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043775528601642833,
      "loss": 5.0736,
      "mean_token_accuracy": 0.20102916061878204,
      "num_tokens": 57055893.0,
      "step": 27500
    },
    {
      "entropy": 5.597240543365478,
      "epoch": 2.495010885341074,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043773294305394665,
      "loss": 4.981,
      "mean_token_accuracy": 0.211080926656723,
      "num_tokens": 57067189.0,
      "step": 27505
    },
    {
      "entropy": 5.6404197216033936,
      "epoch": 2.4954644412191582,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043771059672610636,
      "loss": 4.9529,
      "mean_token_accuracy": 0.2158254489302635,
      "num_tokens": 57077374.0,
      "step": 27510
    },
    {
      "entropy": 5.595413112640381,
      "epoch": 2.4959179970972425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043768824703336933,
      "loss": 4.9032,
      "mean_token_accuracy": 0.21071065664291383,
      "num_tokens": 57088168.0,
      "step": 27515
    },
    {
      "entropy": 5.606458568572998,
      "epoch": 2.4963715529753268,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004376658939761979,
      "loss": 4.8896,
      "mean_token_accuracy": 0.2151413932442665,
      "num_tokens": 57098743.0,
      "step": 27520
    },
    {
      "entropy": 5.577711009979248,
      "epoch": 2.4968251088534106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004376435375550544,
      "loss": 5.0265,
      "mean_token_accuracy": 0.2076060429215431,
      "num_tokens": 57109854.0,
      "step": 27525
    },
    {
      "entropy": 5.554155397415161,
      "epoch": 2.497278664731495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043762117777040106,
      "loss": 4.8712,
      "mean_token_accuracy": 0.22206318527460098,
      "num_tokens": 57120296.0,
      "step": 27530
    },
    {
      "entropy": 5.548946571350098,
      "epoch": 2.497732220609579,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004375988146227003,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21424288153648377,
      "num_tokens": 57131295.0,
      "step": 27535
    },
    {
      "entropy": 5.577417039871216,
      "epoch": 2.4981857764876634,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004375764481124147,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21111445128917694,
      "num_tokens": 57141936.0,
      "step": 27540
    },
    {
      "entropy": 5.568736457824707,
      "epoch": 2.4986393323657476,
      "grad_norm": 1.125,
      "learning_rate": 0.00043755407824000664,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22450370639562606,
      "num_tokens": 57152510.0,
      "step": 27545
    },
    {
      "entropy": 5.604910373687744,
      "epoch": 2.4990928882438315,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043753170500593884,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21606001555919646,
      "num_tokens": 57162546.0,
      "step": 27550
    },
    {
      "entropy": 5.514680004119873,
      "epoch": 2.4995464441219157,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043750932841067397,
      "loss": 4.8462,
      "mean_token_accuracy": 0.21972191333770752,
      "num_tokens": 57174658.0,
      "step": 27555
    },
    {
      "entropy": 5.641879987716675,
      "epoch": 2.5,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043748694845467483,
      "loss": 5.0723,
      "mean_token_accuracy": 0.19958110600709916,
      "num_tokens": 57185157.0,
      "step": 27560
    },
    {
      "entropy": 5.584434175491333,
      "epoch": 2.5004535558780843,
      "grad_norm": 1.125,
      "learning_rate": 0.00043746456513840405,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21656010299921036,
      "num_tokens": 57195277.0,
      "step": 27565
    },
    {
      "entropy": 5.62018780708313,
      "epoch": 2.5009071117561685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004374421784623248,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21126658618450164,
      "num_tokens": 57205374.0,
      "step": 27570
    },
    {
      "entropy": 5.558691692352295,
      "epoch": 2.501360667634253,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043741978842689987,
      "loss": 4.9287,
      "mean_token_accuracy": 0.2144061177968979,
      "num_tokens": 57215700.0,
      "step": 27575
    },
    {
      "entropy": 5.600464963912964,
      "epoch": 2.5018142235123366,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004373973950325923,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21464745849370956,
      "num_tokens": 57225765.0,
      "step": 27580
    },
    {
      "entropy": 5.617864799499512,
      "epoch": 2.502267779390421,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043737499827986527,
      "loss": 4.9416,
      "mean_token_accuracy": 0.21197375357151033,
      "num_tokens": 57235913.0,
      "step": 27585
    },
    {
      "entropy": 5.5413017749786375,
      "epoch": 2.502721335268505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043735259816918184,
      "loss": 4.8642,
      "mean_token_accuracy": 0.21646612286567687,
      "num_tokens": 57246204.0,
      "step": 27590
    },
    {
      "entropy": 5.529222631454468,
      "epoch": 2.5031748911465894,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004373301947010053,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20813624560832977,
      "num_tokens": 57256167.0,
      "step": 27595
    },
    {
      "entropy": 5.540852451324463,
      "epoch": 2.5036284470246732,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043730778787579895,
      "loss": 4.893,
      "mean_token_accuracy": 0.21697479486465454,
      "num_tokens": 57266621.0,
      "step": 27600
    },
    {
      "entropy": 5.609730052947998,
      "epoch": 2.5040820029027575,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043728537769402627,
      "loss": 4.9262,
      "mean_token_accuracy": 0.213581383228302,
      "num_tokens": 57277282.0,
      "step": 27605
    },
    {
      "entropy": 5.599507427215576,
      "epoch": 2.5045355587808418,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004372629641561506,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2103717416524887,
      "num_tokens": 57287541.0,
      "step": 27610
    },
    {
      "entropy": 5.592021560668945,
      "epoch": 2.504989114658926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043724054726263547,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21137694418430328,
      "num_tokens": 57298158.0,
      "step": 27615
    },
    {
      "entropy": 5.625513219833374,
      "epoch": 2.5054426705370103,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004372181270139444,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20922743231058122,
      "num_tokens": 57308356.0,
      "step": 27620
    },
    {
      "entropy": 5.589460182189941,
      "epoch": 2.5058962264150946,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043719570341054127,
      "loss": 4.8697,
      "mean_token_accuracy": 0.22005687952041625,
      "num_tokens": 57317905.0,
      "step": 27625
    },
    {
      "entropy": 5.579386281967163,
      "epoch": 2.5063497822931784,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043717327645288956,
      "loss": 4.9731,
      "mean_token_accuracy": 0.21698225289583206,
      "num_tokens": 57328277.0,
      "step": 27630
    },
    {
      "entropy": 5.638761806488037,
      "epoch": 2.5068033381712627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043715084614145316,
      "loss": 5.0015,
      "mean_token_accuracy": 0.2112189382314682,
      "num_tokens": 57338975.0,
      "step": 27635
    },
    {
      "entropy": 5.586807107925415,
      "epoch": 2.507256894049347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000437128412476696,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21520341485738753,
      "num_tokens": 57349439.0,
      "step": 27640
    },
    {
      "entropy": 5.584844493865967,
      "epoch": 2.507710449927431,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004371059754590819,
      "loss": 4.9879,
      "mean_token_accuracy": 0.2099724069237709,
      "num_tokens": 57359414.0,
      "step": 27645
    },
    {
      "entropy": 5.644027757644653,
      "epoch": 2.508164005805515,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043708353508907494,
      "loss": 4.9941,
      "mean_token_accuracy": 0.2152398109436035,
      "num_tokens": 57369711.0,
      "step": 27650
    },
    {
      "entropy": 5.55789589881897,
      "epoch": 2.5086175616835993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004370610913671392,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22408596128225328,
      "num_tokens": 57380001.0,
      "step": 27655
    },
    {
      "entropy": 5.617283725738526,
      "epoch": 2.5090711175616836,
      "grad_norm": 1.25,
      "learning_rate": 0.00043703864429373875,
      "loss": 5.0598,
      "mean_token_accuracy": 0.1973607823252678,
      "num_tokens": 57390794.0,
      "step": 27660
    },
    {
      "entropy": 5.579990482330322,
      "epoch": 2.509524673439768,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043701619386933793,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20638088434934615,
      "num_tokens": 57401975.0,
      "step": 27665
    },
    {
      "entropy": 5.545427894592285,
      "epoch": 2.509978229317852,
      "grad_norm": 1.125,
      "learning_rate": 0.00043699374009440085,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22356573939323426,
      "num_tokens": 57411249.0,
      "step": 27670
    },
    {
      "entropy": 5.5337066650390625,
      "epoch": 2.5104317851959363,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004369712829693919,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2189682364463806,
      "num_tokens": 57422679.0,
      "step": 27675
    },
    {
      "entropy": 5.539751482009888,
      "epoch": 2.51088534107402,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004369488224947757,
      "loss": 4.935,
      "mean_token_accuracy": 0.21954985707998276,
      "num_tokens": 57433265.0,
      "step": 27680
    },
    {
      "entropy": 5.615825271606445,
      "epoch": 2.5113388969521044,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004369263586710164,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2133643940091133,
      "num_tokens": 57444095.0,
      "step": 27685
    },
    {
      "entropy": 5.599020051956177,
      "epoch": 2.5117924528301887,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004369038914985789,
      "loss": 4.9225,
      "mean_token_accuracy": 0.2126259282231331,
      "num_tokens": 57454059.0,
      "step": 27690
    },
    {
      "entropy": 5.581152868270874,
      "epoch": 2.512246008708273,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004368814209779276,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21774860769510268,
      "num_tokens": 57463940.0,
      "step": 27695
    },
    {
      "entropy": 5.526418876647949,
      "epoch": 2.512699564586357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043685894710952724,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21330235451459884,
      "num_tokens": 57474731.0,
      "step": 27700
    },
    {
      "entropy": 5.542140340805053,
      "epoch": 2.513153120464441,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004368364698938426,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2170833706855774,
      "num_tokens": 57485096.0,
      "step": 27705
    },
    {
      "entropy": 5.57192440032959,
      "epoch": 2.5136066763425253,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004368139893313385,
      "loss": 4.941,
      "mean_token_accuracy": 0.21175574958324433,
      "num_tokens": 57496371.0,
      "step": 27710
    },
    {
      "entropy": 5.624359273910523,
      "epoch": 2.5140602322206096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043679150542248,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2158511221408844,
      "num_tokens": 57506012.0,
      "step": 27715
    },
    {
      "entropy": 5.60762300491333,
      "epoch": 2.514513788098694,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043676901816773174,
      "loss": 4.97,
      "mean_token_accuracy": 0.21026700586080552,
      "num_tokens": 57516536.0,
      "step": 27720
    },
    {
      "entropy": 5.545338296890259,
      "epoch": 2.514967343976778,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004367465275675591,
      "loss": 4.859,
      "mean_token_accuracy": 0.21712665408849716,
      "num_tokens": 57527450.0,
      "step": 27725
    },
    {
      "entropy": 5.614373064041137,
      "epoch": 2.5154208998548624,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043672403362242704,
      "loss": 4.9882,
      "mean_token_accuracy": 0.21164269000291824,
      "num_tokens": 57538234.0,
      "step": 27730
    },
    {
      "entropy": 5.5832830429077145,
      "epoch": 2.515874455732946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043670153633280066,
      "loss": 4.911,
      "mean_token_accuracy": 0.21933975219726562,
      "num_tokens": 57547938.0,
      "step": 27735
    },
    {
      "entropy": 5.60312852859497,
      "epoch": 2.5163280116110305,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004366790356991454,
      "loss": 4.957,
      "mean_token_accuracy": 0.2140236094594002,
      "num_tokens": 57560553.0,
      "step": 27740
    },
    {
      "entropy": 5.5407579898834225,
      "epoch": 2.5167815674891147,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004366565317219265,
      "loss": 4.8992,
      "mean_token_accuracy": 0.21860406696796417,
      "num_tokens": 57570670.0,
      "step": 27745
    },
    {
      "entropy": 5.665750646591187,
      "epoch": 2.5172351233671986,
      "grad_norm": 1.0,
      "learning_rate": 0.00043663402440160924,
      "loss": 5.1212,
      "mean_token_accuracy": 0.20040595531463623,
      "num_tokens": 57582377.0,
      "step": 27750
    },
    {
      "entropy": 5.58102650642395,
      "epoch": 2.517688679245283,
      "grad_norm": 1.125,
      "learning_rate": 0.0004366115137386592,
      "loss": 4.8306,
      "mean_token_accuracy": 0.22278811037540436,
      "num_tokens": 57592764.0,
      "step": 27755
    },
    {
      "entropy": 5.569661664962768,
      "epoch": 2.518142235123367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004365889997335419,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20830441564321517,
      "num_tokens": 57603768.0,
      "step": 27760
    },
    {
      "entropy": 5.552487564086914,
      "epoch": 2.5185957910014514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043656648238672285,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21595267355442047,
      "num_tokens": 57614346.0,
      "step": 27765
    },
    {
      "entropy": 5.673894596099854,
      "epoch": 2.5190493468795356,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004365439616986678,
      "loss": 5.0397,
      "mean_token_accuracy": 0.20825873613357543,
      "num_tokens": 57625015.0,
      "step": 27770
    },
    {
      "entropy": 5.608899831771851,
      "epoch": 2.51950290275762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043652143766984247,
      "loss": 4.9988,
      "mean_token_accuracy": 0.20780922323465348,
      "num_tokens": 57635476.0,
      "step": 27775
    },
    {
      "entropy": 5.542952394485473,
      "epoch": 2.519956458635704,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043649891030071264,
      "loss": 4.8917,
      "mean_token_accuracy": 0.21576763242483138,
      "num_tokens": 57645512.0,
      "step": 27780
    },
    {
      "entropy": 5.569958066940307,
      "epoch": 2.520410014513788,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043647637959174425,
      "loss": 4.898,
      "mean_token_accuracy": 0.21391591429710388,
      "num_tokens": 57655766.0,
      "step": 27785
    },
    {
      "entropy": 5.5870044231414795,
      "epoch": 2.5208635703918723,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043645384554340315,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21954991668462753,
      "num_tokens": 57667325.0,
      "step": 27790
    },
    {
      "entropy": 5.545009422302246,
      "epoch": 2.5213171262699565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043643130815615536,
      "loss": 4.9286,
      "mean_token_accuracy": 0.20707884728908538,
      "num_tokens": 57677884.0,
      "step": 27795
    },
    {
      "entropy": 5.632546091079712,
      "epoch": 2.521770682148041,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043640876743046694,
      "loss": 4.9589,
      "mean_token_accuracy": 0.22201118171215056,
      "num_tokens": 57687765.0,
      "step": 27800
    },
    {
      "entropy": 5.569636011123658,
      "epoch": 2.5222242380261246,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004363862233668041,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21295838207006454,
      "num_tokens": 57697777.0,
      "step": 27805
    },
    {
      "entropy": 5.5323882579803465,
      "epoch": 2.522677793904209,
      "grad_norm": 1.0625,
      "learning_rate": 0.000436363675965633,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21258897781372071,
      "num_tokens": 57708765.0,
      "step": 27810
    },
    {
      "entropy": 5.626805973052979,
      "epoch": 2.523131349782293,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004363411252274201,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21650134921073913,
      "num_tokens": 57719658.0,
      "step": 27815
    },
    {
      "entropy": 5.5963387966156,
      "epoch": 2.5235849056603774,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043631857115263153,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22033073008060455,
      "num_tokens": 57730018.0,
      "step": 27820
    },
    {
      "entropy": 5.50908899307251,
      "epoch": 2.5240384615384617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004362960137417338,
      "loss": 4.865,
      "mean_token_accuracy": 0.2187896952033043,
      "num_tokens": 57741251.0,
      "step": 27825
    },
    {
      "entropy": 5.596281671524048,
      "epoch": 2.524492017416546,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004362734529951933,
      "loss": 4.9687,
      "mean_token_accuracy": 0.21493223160505295,
      "num_tokens": 57751020.0,
      "step": 27830
    },
    {
      "entropy": 5.573217535018921,
      "epoch": 2.5249455732946298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043625088891347685,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2189878150820732,
      "num_tokens": 57760989.0,
      "step": 27835
    },
    {
      "entropy": 5.633127069473266,
      "epoch": 2.525399129172714,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043622832149705086,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2171213984489441,
      "num_tokens": 57770882.0,
      "step": 27840
    },
    {
      "entropy": 5.6543049812316895,
      "epoch": 2.5258526850507983,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043620575074638206,
      "loss": 4.9974,
      "mean_token_accuracy": 0.2138979196548462,
      "num_tokens": 57781812.0,
      "step": 27845
    },
    {
      "entropy": 5.59008378982544,
      "epoch": 2.5263062409288826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043618317666193734,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22132632434368132,
      "num_tokens": 57792009.0,
      "step": 27850
    },
    {
      "entropy": 5.541980600357055,
      "epoch": 2.5267597968069664,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043616059924418346,
      "loss": 4.908,
      "mean_token_accuracy": 0.2208719477057457,
      "num_tokens": 57802109.0,
      "step": 27855
    },
    {
      "entropy": 5.57148928642273,
      "epoch": 2.5272133526850507,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004361380184935872,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21691748201847078,
      "num_tokens": 57812374.0,
      "step": 27860
    },
    {
      "entropy": 5.536352157592773,
      "epoch": 2.527666908563135,
      "grad_norm": 1.125,
      "learning_rate": 0.0004361154344106158,
      "loss": 4.8634,
      "mean_token_accuracy": 0.21948933154344558,
      "num_tokens": 57822936.0,
      "step": 27865
    },
    {
      "entropy": 5.669306802749634,
      "epoch": 2.528120464441219,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043609284699573604,
      "loss": 5.0946,
      "mean_token_accuracy": 0.20776402503252028,
      "num_tokens": 57832693.0,
      "step": 27870
    },
    {
      "entropy": 5.644558334350586,
      "epoch": 2.5285740203193035,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043607025624941523,
      "loss": 4.9716,
      "mean_token_accuracy": 0.21231866776943206,
      "num_tokens": 57843161.0,
      "step": 27875
    },
    {
      "entropy": 5.610720205307007,
      "epoch": 2.5290275761973877,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004360476621721204,
      "loss": 4.8953,
      "mean_token_accuracy": 0.22512662410736084,
      "num_tokens": 57853027.0,
      "step": 27880
    },
    {
      "entropy": 5.586286020278931,
      "epoch": 2.5294811320754715,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004360250647643189,
      "loss": 4.9348,
      "mean_token_accuracy": 0.22143570929765702,
      "num_tokens": 57862821.0,
      "step": 27885
    },
    {
      "entropy": 5.540043497085572,
      "epoch": 2.529934687953556,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043600246402647805,
      "loss": 4.8941,
      "mean_token_accuracy": 0.2220374420285225,
      "num_tokens": 57874065.0,
      "step": 27890
    },
    {
      "entropy": 5.566473865509034,
      "epoch": 2.53038824383164,
      "grad_norm": 1.125,
      "learning_rate": 0.0004359798599590652,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20926931351423264,
      "num_tokens": 57885190.0,
      "step": 27895
    },
    {
      "entropy": 5.5794103145599365,
      "epoch": 2.5308417997097243,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043595725256254786,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2170964553952217,
      "num_tokens": 57894234.0,
      "step": 27900
    },
    {
      "entropy": 5.614807987213135,
      "epoch": 2.531295355587808,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004359346418373935,
      "loss": 5.063,
      "mean_token_accuracy": 0.2021944284439087,
      "num_tokens": 57905401.0,
      "step": 27905
    },
    {
      "entropy": 5.544233179092407,
      "epoch": 2.5317489114658924,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043591202778406975,
      "loss": 4.9064,
      "mean_token_accuracy": 0.21363851577043533,
      "num_tokens": 57916097.0,
      "step": 27910
    },
    {
      "entropy": 5.510739517211914,
      "epoch": 2.5322024673439767,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004358894104030442,
      "loss": 4.8145,
      "mean_token_accuracy": 0.22497545927762985,
      "num_tokens": 57926871.0,
      "step": 27915
    },
    {
      "entropy": 5.534551811218262,
      "epoch": 2.532656023222061,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043586678969478464,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22367361187934875,
      "num_tokens": 57937344.0,
      "step": 27920
    },
    {
      "entropy": 5.617660713195801,
      "epoch": 2.5331095791001452,
      "grad_norm": 1.1875,
      "learning_rate": 0.000435844165659759,
      "loss": 4.9639,
      "mean_token_accuracy": 0.216617850959301,
      "num_tokens": 57947633.0,
      "step": 27925
    },
    {
      "entropy": 5.567021083831787,
      "epoch": 2.5335631349782295,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043582153829843493,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22023017555475236,
      "num_tokens": 57957948.0,
      "step": 27930
    },
    {
      "entropy": 5.596801614761352,
      "epoch": 2.5340166908563138,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004357989076112805,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21556100100278855,
      "num_tokens": 57967719.0,
      "step": 27935
    },
    {
      "entropy": 5.601662635803223,
      "epoch": 2.5344702467343976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004357762735987637,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21218422204256057,
      "num_tokens": 57978218.0,
      "step": 27940
    },
    {
      "entropy": 5.6088603973388675,
      "epoch": 2.534923802612482,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004357536362613525,
      "loss": 4.8648,
      "mean_token_accuracy": 0.220647431910038,
      "num_tokens": 57987860.0,
      "step": 27945
    },
    {
      "entropy": 5.56914529800415,
      "epoch": 2.535377358490566,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043573099559951516,
      "loss": 4.9285,
      "mean_token_accuracy": 0.21944055706262589,
      "num_tokens": 57997872.0,
      "step": 27950
    },
    {
      "entropy": 5.638042306900024,
      "epoch": 2.5358309143686504,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004357083516137199,
      "loss": 4.9991,
      "mean_token_accuracy": 0.20960499048233033,
      "num_tokens": 58007771.0,
      "step": 27955
    },
    {
      "entropy": 5.593334531784057,
      "epoch": 2.536284470246734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000435685704304435,
      "loss": 4.9307,
      "mean_token_accuracy": 0.21017040610313414,
      "num_tokens": 58018972.0,
      "step": 27960
    },
    {
      "entropy": 5.549252510070801,
      "epoch": 2.5367380261248185,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043566305367212884,
      "loss": 4.8182,
      "mean_token_accuracy": 0.21416020691394805,
      "num_tokens": 58028581.0,
      "step": 27965
    },
    {
      "entropy": 5.5674216747283936,
      "epoch": 2.5371915820029027,
      "grad_norm": 1.125,
      "learning_rate": 0.00043564039971726964,
      "loss": 4.9777,
      "mean_token_accuracy": 0.21665931344032288,
      "num_tokens": 58038877.0,
      "step": 27970
    },
    {
      "entropy": 5.575457763671875,
      "epoch": 2.537645137880987,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043561774244032614,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2171279102563858,
      "num_tokens": 58048777.0,
      "step": 27975
    },
    {
      "entropy": 5.625828075408935,
      "epoch": 2.5380986937590713,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043559508184176675,
      "loss": 4.9987,
      "mean_token_accuracy": 0.2120449960231781,
      "num_tokens": 58060270.0,
      "step": 27980
    },
    {
      "entropy": 5.568502759933471,
      "epoch": 2.5385522496371555,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043557241792206014,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21771903932094575,
      "num_tokens": 58070838.0,
      "step": 27985
    },
    {
      "entropy": 5.5928129196167,
      "epoch": 2.5390058055152394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043554975068167505,
      "loss": 5.0097,
      "mean_token_accuracy": 0.21062657237052917,
      "num_tokens": 58079933.0,
      "step": 27990
    },
    {
      "entropy": 5.541003656387329,
      "epoch": 2.5394593613933236,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004355270801210802,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21689495742321013,
      "num_tokens": 58090673.0,
      "step": 27995
    },
    {
      "entropy": 5.543783330917359,
      "epoch": 2.539912917271408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004355044062407444,
      "loss": 4.7834,
      "mean_token_accuracy": 0.22975331395864487,
      "num_tokens": 58100172.0,
      "step": 28000
    },
    {
      "entropy": 5.555199098587036,
      "epoch": 2.540366473149492,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004354817290411366,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21375659555196763,
      "num_tokens": 58108965.0,
      "step": 28005
    },
    {
      "entropy": 5.5565770149230955,
      "epoch": 2.540820029027576,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043545904852272573,
      "loss": 4.8235,
      "mean_token_accuracy": 0.22217636704444885,
      "num_tokens": 58120094.0,
      "step": 28010
    },
    {
      "entropy": 5.5980805397033695,
      "epoch": 2.5412735849056602,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043543636468598083,
      "loss": 4.9855,
      "mean_token_accuracy": 0.20712201595306395,
      "num_tokens": 58131525.0,
      "step": 28015
    },
    {
      "entropy": 5.61816987991333,
      "epoch": 2.5417271407837445,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043541367753137107,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21651554405689238,
      "num_tokens": 58141986.0,
      "step": 28020
    },
    {
      "entropy": 5.602068138122559,
      "epoch": 2.5421806966618288,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004353909870593656,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20951160788536072,
      "num_tokens": 58154122.0,
      "step": 28025
    },
    {
      "entropy": 5.578133010864258,
      "epoch": 2.542634252539913,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004353682932704336,
      "loss": 4.874,
      "mean_token_accuracy": 0.2249719262123108,
      "num_tokens": 58164148.0,
      "step": 28030
    },
    {
      "entropy": 5.5288416862487795,
      "epoch": 2.5430878084179973,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004353455961650444,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2150352865457535,
      "num_tokens": 58173836.0,
      "step": 28035
    },
    {
      "entropy": 5.552236127853393,
      "epoch": 2.543541364296081,
      "grad_norm": 1.125,
      "learning_rate": 0.0004353228957436675,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21269828528165818,
      "num_tokens": 58184598.0,
      "step": 28040
    },
    {
      "entropy": 5.551817512512207,
      "epoch": 2.5439949201741654,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043530019200677215,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21802981048822404,
      "num_tokens": 58194820.0,
      "step": 28045
    },
    {
      "entropy": 5.565774393081665,
      "epoch": 2.5444484760522497,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043527748495482806,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21097971498966217,
      "num_tokens": 58206589.0,
      "step": 28050
    },
    {
      "entropy": 5.617750787734986,
      "epoch": 2.544902031930334,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043525477458830473,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2160607174038887,
      "num_tokens": 58216750.0,
      "step": 28055
    },
    {
      "entropy": 5.548269557952881,
      "epoch": 2.5453555878084178,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004352320609076718,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2209152415394783,
      "num_tokens": 58228496.0,
      "step": 28060
    },
    {
      "entropy": 5.584446954727173,
      "epoch": 2.545809143686502,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043520934391339894,
      "loss": 4.8819,
      "mean_token_accuracy": 0.2128905862569809,
      "num_tokens": 58237844.0,
      "step": 28065
    },
    {
      "entropy": 5.508859348297119,
      "epoch": 2.5462626995645863,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004351866236059561,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21740521341562272,
      "num_tokens": 58248522.0,
      "step": 28070
    },
    {
      "entropy": 5.531892919540406,
      "epoch": 2.5467162554426706,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043516389998581305,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2148062378168106,
      "num_tokens": 58258492.0,
      "step": 28075
    },
    {
      "entropy": 5.530713367462158,
      "epoch": 2.547169811320755,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004351411730534398,
      "loss": 4.9086,
      "mean_token_accuracy": 0.20897070467472076,
      "num_tokens": 58268683.0,
      "step": 28080
    },
    {
      "entropy": 5.585899686813354,
      "epoch": 2.547623367198839,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004351184428093062,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20405336916446687,
      "num_tokens": 58279479.0,
      "step": 28085
    },
    {
      "entropy": 5.622782564163208,
      "epoch": 2.5480769230769234,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004350957092538825,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20888267159461976,
      "num_tokens": 58290751.0,
      "step": 28090
    },
    {
      "entropy": 5.630240869522095,
      "epoch": 2.548530478955007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004350729723876386,
      "loss": 5.016,
      "mean_token_accuracy": 0.2077210769057274,
      "num_tokens": 58302163.0,
      "step": 28095
    },
    {
      "entropy": 5.586409091949463,
      "epoch": 2.5489840348330914,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043505023221104497,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2068890079855919,
      "num_tokens": 58311488.0,
      "step": 28100
    },
    {
      "entropy": 5.551432752609253,
      "epoch": 2.5494375907111757,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043502748872457166,
      "loss": 4.9111,
      "mean_token_accuracy": 0.21354300826787947,
      "num_tokens": 58321599.0,
      "step": 28105
    },
    {
      "entropy": 5.565755414962768,
      "epoch": 2.5498911465892595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004350047419286892,
      "loss": 4.9043,
      "mean_token_accuracy": 0.21592913568019867,
      "num_tokens": 58330696.0,
      "step": 28110
    },
    {
      "entropy": 5.545326375961304,
      "epoch": 2.550344702467344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043498199182386775,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21693719923496246,
      "num_tokens": 58341793.0,
      "step": 28115
    },
    {
      "entropy": 5.607199239730835,
      "epoch": 2.550798258345428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043495923841057807,
      "loss": 5.0109,
      "mean_token_accuracy": 0.20901485979557038,
      "num_tokens": 58352260.0,
      "step": 28120
    },
    {
      "entropy": 5.676895236968994,
      "epoch": 2.5512518142235123,
      "grad_norm": 1.125,
      "learning_rate": 0.00043493648168929053,
      "loss": 5.0552,
      "mean_token_accuracy": 0.20567869544029235,
      "num_tokens": 58363052.0,
      "step": 28125
    },
    {
      "entropy": 5.665903997421265,
      "epoch": 2.5517053701015966,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004349137216604758,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21507498174905776,
      "num_tokens": 58374039.0,
      "step": 28130
    },
    {
      "entropy": 5.5490013599395756,
      "epoch": 2.552158925979681,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004348909583246045,
      "loss": 4.8045,
      "mean_token_accuracy": 0.22377109080553054,
      "num_tokens": 58383441.0,
      "step": 28135
    },
    {
      "entropy": 5.522780227661133,
      "epoch": 2.552612481857765,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004348681916821475,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21670966148376464,
      "num_tokens": 58394164.0,
      "step": 28140
    },
    {
      "entropy": 5.536119890213013,
      "epoch": 2.553066037735849,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004348454217335755,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21522781401872634,
      "num_tokens": 58404610.0,
      "step": 28145
    },
    {
      "entropy": 5.562320899963379,
      "epoch": 2.553519593613933,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043482264847935953,
      "loss": 4.8768,
      "mean_token_accuracy": 0.21156506538391112,
      "num_tokens": 58414512.0,
      "step": 28150
    },
    {
      "entropy": 5.6373857975006105,
      "epoch": 2.5539731494920175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043479987191997045,
      "loss": 5.0198,
      "mean_token_accuracy": 0.21469323486089706,
      "num_tokens": 58424669.0,
      "step": 28155
    },
    {
      "entropy": 5.511224222183228,
      "epoch": 2.5544267053701017,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004347770920558792,
      "loss": 4.8332,
      "mean_token_accuracy": 0.21764508932828902,
      "num_tokens": 58435348.0,
      "step": 28160
    },
    {
      "entropy": 5.608684349060058,
      "epoch": 2.5548802612481856,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043475430888755706,
      "loss": 4.9161,
      "mean_token_accuracy": 0.21417239904403687,
      "num_tokens": 58445534.0,
      "step": 28165
    },
    {
      "entropy": 5.591418504714966,
      "epoch": 2.55533381712627,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004347315224154751,
      "loss": 4.9592,
      "mean_token_accuracy": 0.21402875781059266,
      "num_tokens": 58456718.0,
      "step": 28170
    },
    {
      "entropy": 5.660646438598633,
      "epoch": 2.555787373004354,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004347087326401045,
      "loss": 4.9877,
      "mean_token_accuracy": 0.2094752609729767,
      "num_tokens": 58467176.0,
      "step": 28175
    },
    {
      "entropy": 5.582002782821656,
      "epoch": 2.5562409288824384,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043468593956191663,
      "loss": 4.8529,
      "mean_token_accuracy": 0.2182113915681839,
      "num_tokens": 58477642.0,
      "step": 28180
    },
    {
      "entropy": 5.547383594512939,
      "epoch": 2.5566944847605226,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004346631431813829,
      "loss": 4.929,
      "mean_token_accuracy": 0.2082124650478363,
      "num_tokens": 58489055.0,
      "step": 28185
    },
    {
      "entropy": 5.635054254531861,
      "epoch": 2.557148040638607,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004346403434989746,
      "loss": 4.9451,
      "mean_token_accuracy": 0.20965656787157058,
      "num_tokens": 58499062.0,
      "step": 28190
    },
    {
      "entropy": 5.605187797546387,
      "epoch": 2.5576015965166907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004346175405151633,
      "loss": 4.9504,
      "mean_token_accuracy": 0.20859958678483964,
      "num_tokens": 58509154.0,
      "step": 28195
    },
    {
      "entropy": 5.539422559738159,
      "epoch": 2.558055152394775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004345947342304207,
      "loss": 4.845,
      "mean_token_accuracy": 0.21875003278255462,
      "num_tokens": 58520056.0,
      "step": 28200
    },
    {
      "entropy": 5.5600813865661625,
      "epoch": 2.5585087082728593,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043457192464521827,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21735468059778212,
      "num_tokens": 58529934.0,
      "step": 28205
    },
    {
      "entropy": 5.522853803634644,
      "epoch": 2.5589622641509435,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004345491117600278,
      "loss": 4.8363,
      "mean_token_accuracy": 0.21823289692401887,
      "num_tokens": 58540013.0,
      "step": 28210
    },
    {
      "entropy": 5.6585124969482425,
      "epoch": 2.5594158200290273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043452629557532097,
      "loss": 4.9357,
      "mean_token_accuracy": 0.21542141437530518,
      "num_tokens": 58550282.0,
      "step": 28215
    },
    {
      "entropy": 5.650312662124634,
      "epoch": 2.5598693759071116,
      "grad_norm": 1.125,
      "learning_rate": 0.00043450347609156974,
      "loss": 4.964,
      "mean_token_accuracy": 0.20569372177124023,
      "num_tokens": 58560490.0,
      "step": 28220
    },
    {
      "entropy": 5.5479888916015625,
      "epoch": 2.560322931785196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043448065330924595,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21200328916311265,
      "num_tokens": 58570371.0,
      "step": 28225
    },
    {
      "entropy": 5.589256954193115,
      "epoch": 2.56077648766328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004344578272288216,
      "loss": 4.8843,
      "mean_token_accuracy": 0.21767188757658004,
      "num_tokens": 58580701.0,
      "step": 28230
    },
    {
      "entropy": 5.623743343353271,
      "epoch": 2.5612300435413644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043443499785076885,
      "loss": 4.9307,
      "mean_token_accuracy": 0.2170988515019417,
      "num_tokens": 58590903.0,
      "step": 28235
    },
    {
      "entropy": 5.612763690948486,
      "epoch": 2.5616835994194487,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004344121651755596,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21549489349126816,
      "num_tokens": 58600924.0,
      "step": 28240
    },
    {
      "entropy": 5.620186042785645,
      "epoch": 2.5621371552975325,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043438932920366615,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2145834818482399,
      "num_tokens": 58611369.0,
      "step": 28245
    },
    {
      "entropy": 5.6274457454681395,
      "epoch": 2.5625907111756168,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004343664899355608,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21357562839984895,
      "num_tokens": 58622725.0,
      "step": 28250
    },
    {
      "entropy": 5.580396556854248,
      "epoch": 2.563044267053701,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043434364737171587,
      "loss": 4.9171,
      "mean_token_accuracy": 0.2151700735092163,
      "num_tokens": 58634157.0,
      "step": 28255
    },
    {
      "entropy": 5.577631187438965,
      "epoch": 2.5634978229317853,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043432080151260366,
      "loss": 4.8506,
      "mean_token_accuracy": 0.21447116136550903,
      "num_tokens": 58643384.0,
      "step": 28260
    },
    {
      "entropy": 5.577665138244629,
      "epoch": 2.563951378809869,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043429795235869667,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21589313298463822,
      "num_tokens": 58653748.0,
      "step": 28265
    },
    {
      "entropy": 5.612186479568481,
      "epoch": 2.5644049346879534,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043427509991046746,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2085652694106102,
      "num_tokens": 58664619.0,
      "step": 28270
    },
    {
      "entropy": 5.558182144165039,
      "epoch": 2.5648584905660377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043425224416838857,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22760617434978486,
      "num_tokens": 58674506.0,
      "step": 28275
    },
    {
      "entropy": 5.610248851776123,
      "epoch": 2.565312046444122,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004342293851329327,
      "loss": 4.9803,
      "mean_token_accuracy": 0.20432157218456268,
      "num_tokens": 58686252.0,
      "step": 28280
    },
    {
      "entropy": 5.600450038909912,
      "epoch": 2.565765602322206,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043420652280457255,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2097615897655487,
      "num_tokens": 58696477.0,
      "step": 28285
    },
    {
      "entropy": 5.494926166534424,
      "epoch": 2.5662191582002905,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043418365718378096,
      "loss": 4.8055,
      "mean_token_accuracy": 0.2248273029923439,
      "num_tokens": 58706664.0,
      "step": 28290
    },
    {
      "entropy": 5.499495935440064,
      "epoch": 2.5666727140783747,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004341607882710307,
      "loss": 4.8971,
      "mean_token_accuracy": 0.2177201196551323,
      "num_tokens": 58716950.0,
      "step": 28295
    },
    {
      "entropy": 5.560776662826538,
      "epoch": 2.5671262699564585,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004341379160667949,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21756815910339355,
      "num_tokens": 58727723.0,
      "step": 28300
    },
    {
      "entropy": 5.570564651489258,
      "epoch": 2.567579825834543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004341150405715464,
      "loss": 5.005,
      "mean_token_accuracy": 0.2085478648543358,
      "num_tokens": 58737956.0,
      "step": 28305
    },
    {
      "entropy": 5.572387790679931,
      "epoch": 2.568033381712627,
      "grad_norm": 1.125,
      "learning_rate": 0.00043409216178575826,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2219129577279091,
      "num_tokens": 58748476.0,
      "step": 28310
    },
    {
      "entropy": 5.620368814468383,
      "epoch": 2.5684869375907113,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004340692797099037,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21920992732048034,
      "num_tokens": 58759604.0,
      "step": 28315
    },
    {
      "entropy": 5.601776790618897,
      "epoch": 2.568940493468795,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004340463943444559,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21959961205720901,
      "num_tokens": 58769781.0,
      "step": 28320
    },
    {
      "entropy": 5.58840970993042,
      "epoch": 2.5693940493468794,
      "grad_norm": 1.25,
      "learning_rate": 0.0004340235056898882,
      "loss": 4.9249,
      "mean_token_accuracy": 0.21749221235513688,
      "num_tokens": 58779384.0,
      "step": 28325
    },
    {
      "entropy": 5.6534069061279295,
      "epoch": 2.5698476052249637,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043400061374667374,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21296125203371047,
      "num_tokens": 58791953.0,
      "step": 28330
    },
    {
      "entropy": 5.545256519317627,
      "epoch": 2.570301161103048,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004339777185152862,
      "loss": 4.8945,
      "mean_token_accuracy": 0.21814947873353957,
      "num_tokens": 58802415.0,
      "step": 28335
    },
    {
      "entropy": 5.634825372695923,
      "epoch": 2.5707547169811322,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004339548199961989,
      "loss": 4.93,
      "mean_token_accuracy": 0.21505463123321533,
      "num_tokens": 58812582.0,
      "step": 28340
    },
    {
      "entropy": 5.59175853729248,
      "epoch": 2.5712082728592165,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043393191818988543,
      "loss": 4.9714,
      "mean_token_accuracy": 0.20902827382087708,
      "num_tokens": 58822987.0,
      "step": 28345
    },
    {
      "entropy": 5.553751277923584,
      "epoch": 2.5716618287373003,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004339090130968193,
      "loss": 4.8768,
      "mean_token_accuracy": 0.21788219213485718,
      "num_tokens": 58832651.0,
      "step": 28350
    },
    {
      "entropy": 5.586088085174561,
      "epoch": 2.5721153846153846,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004338861047174743,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2090076819062233,
      "num_tokens": 58842493.0,
      "step": 28355
    },
    {
      "entropy": 5.54956841468811,
      "epoch": 2.572568940493469,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004338631930523242,
      "loss": 4.8349,
      "mean_token_accuracy": 0.22456513494253158,
      "num_tokens": 58852428.0,
      "step": 28360
    },
    {
      "entropy": 5.521766948699951,
      "epoch": 2.573022496371553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004338402781018429,
      "loss": 4.8497,
      "mean_token_accuracy": 0.22391195446252823,
      "num_tokens": 58863234.0,
      "step": 28365
    },
    {
      "entropy": 5.546092367172241,
      "epoch": 2.573476052249637,
      "grad_norm": 1.203125,
      "learning_rate": 0.000433817359866504,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21715689301490784,
      "num_tokens": 58872632.0,
      "step": 28370
    },
    {
      "entropy": 5.672575998306274,
      "epoch": 2.573929608127721,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004337944383467817,
      "loss": 5.0997,
      "mean_token_accuracy": 0.2020232558250427,
      "num_tokens": 58882774.0,
      "step": 28375
    },
    {
      "entropy": 5.5224446773529055,
      "epoch": 2.5743831640058055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043377151354314996,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22404280304908752,
      "num_tokens": 58892380.0,
      "step": 28380
    },
    {
      "entropy": 5.600534152984619,
      "epoch": 2.5748367198838897,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004337485854560829,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20759045481681823,
      "num_tokens": 58902659.0,
      "step": 28385
    },
    {
      "entropy": 5.546986436843872,
      "epoch": 2.575290275761974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004337256540860545,
      "loss": 4.894,
      "mean_token_accuracy": 0.21621687859296798,
      "num_tokens": 58913979.0,
      "step": 28390
    },
    {
      "entropy": 5.619999170303345,
      "epoch": 2.5757438316400583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004337027194335393,
      "loss": 4.9747,
      "mean_token_accuracy": 0.21403544396162033,
      "num_tokens": 58924481.0,
      "step": 28395
    },
    {
      "entropy": 5.594887161254883,
      "epoch": 2.576197387518142,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043367978149901125,
      "loss": 4.9559,
      "mean_token_accuracy": 0.2066089317202568,
      "num_tokens": 58936287.0,
      "step": 28400
    },
    {
      "entropy": 5.6483711242675785,
      "epoch": 2.5766509433962264,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043365684028294493,
      "loss": 4.9416,
      "mean_token_accuracy": 0.21255273073911668,
      "num_tokens": 58946070.0,
      "step": 28405
    },
    {
      "entropy": 5.6377613067626955,
      "epoch": 2.5771044992743106,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043363389578581483,
      "loss": 5.0141,
      "mean_token_accuracy": 0.20137166529893874,
      "num_tokens": 58956083.0,
      "step": 28410
    },
    {
      "entropy": 5.606115674972534,
      "epoch": 2.577558055152395,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004336109480080953,
      "loss": 4.9259,
      "mean_token_accuracy": 0.20825573652982712,
      "num_tokens": 58967557.0,
      "step": 28415
    },
    {
      "entropy": 5.558809423446656,
      "epoch": 2.5780116110304787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004335879969502608,
      "loss": 4.8882,
      "mean_token_accuracy": 0.21217737793922425,
      "num_tokens": 58977509.0,
      "step": 28420
    },
    {
      "entropy": 5.49219241142273,
      "epoch": 2.578465166908563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043356504261278625,
      "loss": 4.773,
      "mean_token_accuracy": 0.23292024582624435,
      "num_tokens": 58987498.0,
      "step": 28425
    },
    {
      "entropy": 5.665225219726563,
      "epoch": 2.5789187227866472,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004335420849961462,
      "loss": 5.0418,
      "mean_token_accuracy": 0.20125290602445603,
      "num_tokens": 58998503.0,
      "step": 28430
    },
    {
      "entropy": 5.567272472381592,
      "epoch": 2.5793722786647315,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004335191241008154,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21268775016069413,
      "num_tokens": 59007384.0,
      "step": 28435
    },
    {
      "entropy": 5.6467200756073,
      "epoch": 2.5798258345428158,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004334961599272687,
      "loss": 5.0273,
      "mean_token_accuracy": 0.21258178502321243,
      "num_tokens": 59017739.0,
      "step": 28440
    },
    {
      "entropy": 5.622543811798096,
      "epoch": 2.5802793904209,
      "grad_norm": 1.25,
      "learning_rate": 0.0004334731924759811,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21672325879335402,
      "num_tokens": 59028439.0,
      "step": 28445
    },
    {
      "entropy": 5.567311716079712,
      "epoch": 2.5807329462989843,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004334502217474274,
      "loss": 4.9142,
      "mean_token_accuracy": 0.2108398988842964,
      "num_tokens": 59039250.0,
      "step": 28450
    },
    {
      "entropy": 5.650187587738037,
      "epoch": 2.581186502177068,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043342724774208277,
      "loss": 5.0031,
      "mean_token_accuracy": 0.2137565240263939,
      "num_tokens": 59049375.0,
      "step": 28455
    },
    {
      "entropy": 5.646983861923218,
      "epoch": 2.5816400580551524,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043340427046042237,
      "loss": 5.0323,
      "mean_token_accuracy": 0.2114769548177719,
      "num_tokens": 59059873.0,
      "step": 28460
    },
    {
      "entropy": 5.586476039886475,
      "epoch": 2.5820936139332367,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004333812899029212,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21517833471298217,
      "num_tokens": 59069288.0,
      "step": 28465
    },
    {
      "entropy": 5.575258493423462,
      "epoch": 2.5825471698113205,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043335830607005456,
      "loss": 4.9423,
      "mean_token_accuracy": 0.20982772707939149,
      "num_tokens": 59079339.0,
      "step": 28470
    },
    {
      "entropy": 5.597243881225586,
      "epoch": 2.5830007256894048,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004333353189622979,
      "loss": 4.9451,
      "mean_token_accuracy": 0.21003630012273788,
      "num_tokens": 59089883.0,
      "step": 28475
    },
    {
      "entropy": 5.646843814849854,
      "epoch": 2.583454281567489,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004333123285801264,
      "loss": 4.9829,
      "mean_token_accuracy": 0.21654506474733354,
      "num_tokens": 59100532.0,
      "step": 28480
    },
    {
      "entropy": 5.603966569900512,
      "epoch": 2.5839078374455733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004332893349240156,
      "loss": 4.9301,
      "mean_token_accuracy": 0.21421366333961486,
      "num_tokens": 59111295.0,
      "step": 28485
    },
    {
      "entropy": 5.610342264175415,
      "epoch": 2.5843613933236576,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004332663379944411,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2104869604110718,
      "num_tokens": 59120808.0,
      "step": 28490
    },
    {
      "entropy": 5.659372758865357,
      "epoch": 2.584814949201742,
      "grad_norm": 1.25,
      "learning_rate": 0.00043324333779187844,
      "loss": 5.004,
      "mean_token_accuracy": 0.21949607878923416,
      "num_tokens": 59131438.0,
      "step": 28495
    },
    {
      "entropy": 5.568179035186768,
      "epoch": 2.585268505079826,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043322033431680314,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21955933421850204,
      "num_tokens": 59141701.0,
      "step": 28500
    },
    {
      "entropy": 5.536624765396118,
      "epoch": 2.58572206095791,
      "grad_norm": 1.203125,
      "learning_rate": 0.000433197327569691,
      "loss": 4.8978,
      "mean_token_accuracy": 0.21859243512153625,
      "num_tokens": 59152184.0,
      "step": 28505
    },
    {
      "entropy": 5.55860276222229,
      "epoch": 2.586175616835994,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043317431755101785,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22313813120126724,
      "num_tokens": 59162177.0,
      "step": 28510
    },
    {
      "entropy": 5.66088900566101,
      "epoch": 2.5866291727140784,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004331513042612594,
      "loss": 5.0233,
      "mean_token_accuracy": 0.2112630531191826,
      "num_tokens": 59171964.0,
      "step": 28515
    },
    {
      "entropy": 5.587431955337524,
      "epoch": 2.5870827285921627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004331282877008918,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21516778618097304,
      "num_tokens": 59182224.0,
      "step": 28520
    },
    {
      "entropy": 5.594990587234497,
      "epoch": 2.5875362844702465,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004331052678703909,
      "loss": 4.922,
      "mean_token_accuracy": 0.21240060925483703,
      "num_tokens": 59191912.0,
      "step": 28525
    },
    {
      "entropy": 5.604145765304565,
      "epoch": 2.587989840348331,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004330822447702327,
      "loss": 4.8989,
      "mean_token_accuracy": 0.21341202557086944,
      "num_tokens": 59200984.0,
      "step": 28530
    },
    {
      "entropy": 5.6295576095581055,
      "epoch": 2.588443396226415,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043305921840089347,
      "loss": 4.9696,
      "mean_token_accuracy": 0.21466362923383714,
      "num_tokens": 59210672.0,
      "step": 28535
    },
    {
      "entropy": 5.612875938415527,
      "epoch": 2.5888969521044993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043303618876284935,
      "loss": 4.9622,
      "mean_token_accuracy": 0.21743858605623245,
      "num_tokens": 59220230.0,
      "step": 28540
    },
    {
      "entropy": 5.523725605010986,
      "epoch": 2.5893505079825836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043301315585657645,
      "loss": 4.8644,
      "mean_token_accuracy": 0.21540619283914567,
      "num_tokens": 59230615.0,
      "step": 28545
    },
    {
      "entropy": 5.5843648433685305,
      "epoch": 2.589804063860668,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004329901196825513,
      "loss": 4.9656,
      "mean_token_accuracy": 0.21518293917179107,
      "num_tokens": 59241717.0,
      "step": 28550
    },
    {
      "entropy": 5.647132205963135,
      "epoch": 2.5902576197387517,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043296708024125025,
      "loss": 4.9947,
      "mean_token_accuracy": 0.2105607733130455,
      "num_tokens": 59251791.0,
      "step": 28555
    },
    {
      "entropy": 5.555968236923218,
      "epoch": 2.590711175616836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043294403753314965,
      "loss": 4.8182,
      "mean_token_accuracy": 0.22235488146543503,
      "num_tokens": 59262434.0,
      "step": 28560
    },
    {
      "entropy": 5.564286708831787,
      "epoch": 2.59116473149492,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004329209915587261,
      "loss": 4.9898,
      "mean_token_accuracy": 0.21071625351905823,
      "num_tokens": 59272138.0,
      "step": 28565
    },
    {
      "entropy": 5.548455333709716,
      "epoch": 2.5916182873730045,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043289794231845624,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2155385732650757,
      "num_tokens": 59283470.0,
      "step": 28570
    },
    {
      "entropy": 5.547768402099609,
      "epoch": 2.5920718432510883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004328748898128168,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21359224021434783,
      "num_tokens": 59293670.0,
      "step": 28575
    },
    {
      "entropy": 5.607025098800659,
      "epoch": 2.5925253991291726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004328518340422842,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21235549300909043,
      "num_tokens": 59303728.0,
      "step": 28580
    },
    {
      "entropy": 5.57889757156372,
      "epoch": 2.592978955007257,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043282877500733555,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21655260771512985,
      "num_tokens": 59314271.0,
      "step": 28585
    },
    {
      "entropy": 5.5851446151733395,
      "epoch": 2.593432510885341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004328057127084476,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21399024426937102,
      "num_tokens": 59324581.0,
      "step": 28590
    },
    {
      "entropy": 5.596630477905274,
      "epoch": 2.5938860667634254,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043278264714609733,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21554073989391326,
      "num_tokens": 59333912.0,
      "step": 28595
    },
    {
      "entropy": 5.556984853744507,
      "epoch": 2.5943396226415096,
      "grad_norm": 1.125,
      "learning_rate": 0.00043275957832076173,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2158809334039688,
      "num_tokens": 59344245.0,
      "step": 28600
    },
    {
      "entropy": 5.583969497680664,
      "epoch": 2.5947931785195935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043273650623291775,
      "loss": 5.0088,
      "mean_token_accuracy": 0.21582089960575104,
      "num_tokens": 59353825.0,
      "step": 28605
    },
    {
      "entropy": 5.58726749420166,
      "epoch": 2.5952467343976777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004327134308830427,
      "loss": 4.9558,
      "mean_token_accuracy": 0.21076942086219788,
      "num_tokens": 59364901.0,
      "step": 28610
    },
    {
      "entropy": 5.564519739151001,
      "epoch": 2.595700290275762,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004326903522716137,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2169706791639328,
      "num_tokens": 59374772.0,
      "step": 28615
    },
    {
      "entropy": 5.528669786453247,
      "epoch": 2.5961538461538463,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000432667270399108,
      "loss": 4.8161,
      "mean_token_accuracy": 0.22522260695695878,
      "num_tokens": 59383640.0,
      "step": 28620
    },
    {
      "entropy": 5.594297933578491,
      "epoch": 2.59660740203193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043264418526600305,
      "loss": 4.8928,
      "mean_token_accuracy": 0.21794225573539733,
      "num_tokens": 59392928.0,
      "step": 28625
    },
    {
      "entropy": 5.631557893753052,
      "epoch": 2.5970609579100143,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004326210968727761,
      "loss": 4.9979,
      "mean_token_accuracy": 0.2088549554347992,
      "num_tokens": 59402969.0,
      "step": 28630
    },
    {
      "entropy": 5.6218935489654545,
      "epoch": 2.5975145137880986,
      "grad_norm": 1.125,
      "learning_rate": 0.0004325980052199047,
      "loss": 5.003,
      "mean_token_accuracy": 0.2038627967238426,
      "num_tokens": 59413439.0,
      "step": 28635
    },
    {
      "entropy": 5.629627132415772,
      "epoch": 2.597968069666183,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043257491030786644,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21281755566596985,
      "num_tokens": 59423403.0,
      "step": 28640
    },
    {
      "entropy": 5.610312032699585,
      "epoch": 2.598421625544267,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043255181213713883,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20955325216054915,
      "num_tokens": 59434519.0,
      "step": 28645
    },
    {
      "entropy": 5.609882116317749,
      "epoch": 2.5988751814223514,
      "grad_norm": 1.125,
      "learning_rate": 0.0004325287107081996,
      "loss": 4.948,
      "mean_token_accuracy": 0.20845523625612258,
      "num_tokens": 59445056.0,
      "step": 28650
    },
    {
      "entropy": 5.574291849136353,
      "epoch": 2.5993287373004357,
      "grad_norm": 1.25,
      "learning_rate": 0.0004325056060215265,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21606272011995314,
      "num_tokens": 59455037.0,
      "step": 28655
    },
    {
      "entropy": 5.568118667602539,
      "epoch": 2.5997822931785195,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004324824980775973,
      "loss": 4.9598,
      "mean_token_accuracy": 0.21121555864810942,
      "num_tokens": 59465093.0,
      "step": 28660
    },
    {
      "entropy": 5.576521301269532,
      "epoch": 2.6002358490566038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043245938687688995,
      "loss": 4.9992,
      "mean_token_accuracy": 0.2153582751750946,
      "num_tokens": 59475449.0,
      "step": 28665
    },
    {
      "entropy": 5.653523874282837,
      "epoch": 2.600689404934688,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043243627241988235,
      "loss": 4.9495,
      "mean_token_accuracy": 0.2126238077878952,
      "num_tokens": 59484990.0,
      "step": 28670
    },
    {
      "entropy": 5.643709564208985,
      "epoch": 2.6011429608127723,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043241315470705254,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2096727728843689,
      "num_tokens": 59495739.0,
      "step": 28675
    },
    {
      "entropy": 5.595979118347168,
      "epoch": 2.601596516690856,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043239003373887855,
      "loss": 4.933,
      "mean_token_accuracy": 0.21496637016534806,
      "num_tokens": 59506432.0,
      "step": 28680
    },
    {
      "entropy": 5.590441560745239,
      "epoch": 2.6020500725689404,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004323669095158385,
      "loss": 4.9388,
      "mean_token_accuracy": 0.211165452003479,
      "num_tokens": 59516585.0,
      "step": 28685
    },
    {
      "entropy": 5.57189245223999,
      "epoch": 2.6025036284470247,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004323437820384107,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2089263454079628,
      "num_tokens": 59526824.0,
      "step": 28690
    },
    {
      "entropy": 5.641707992553711,
      "epoch": 2.602957184325109,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004323206513070734,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21190085262060165,
      "num_tokens": 59537749.0,
      "step": 28695
    },
    {
      "entropy": 5.624490976333618,
      "epoch": 2.603410740203193,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000432297517322305,
      "loss": 4.9262,
      "mean_token_accuracy": 0.20382642596960068,
      "num_tokens": 59548310.0,
      "step": 28700
    },
    {
      "entropy": 5.636884260177612,
      "epoch": 2.6038642960812775,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004322743800845838,
      "loss": 4.9973,
      "mean_token_accuracy": 0.2139281451702118,
      "num_tokens": 59558191.0,
      "step": 28705
    },
    {
      "entropy": 5.602367115020752,
      "epoch": 2.6043178519593613,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043225123959438836,
      "loss": 5.0101,
      "mean_token_accuracy": 0.202468204498291,
      "num_tokens": 59568961.0,
      "step": 28710
    },
    {
      "entropy": 5.602570867538452,
      "epoch": 2.6047714078374455,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043222809585219716,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21022058576345443,
      "num_tokens": 59579137.0,
      "step": 28715
    },
    {
      "entropy": 5.450237035751343,
      "epoch": 2.60522496371553,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043220494885848896,
      "loss": 4.7484,
      "mean_token_accuracy": 0.22537932246923448,
      "num_tokens": 59589612.0,
      "step": 28720
    },
    {
      "entropy": 5.6324601650238035,
      "epoch": 2.605678519593614,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004321817986137423,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20888511687517167,
      "num_tokens": 59600332.0,
      "step": 28725
    },
    {
      "entropy": 5.5210652351379395,
      "epoch": 2.606132075471698,
      "grad_norm": 1.234375,
      "learning_rate": 0.000432158645118436,
      "loss": 4.8573,
      "mean_token_accuracy": 0.22621295005083084,
      "num_tokens": 59610491.0,
      "step": 28730
    },
    {
      "entropy": 5.707390689849854,
      "epoch": 2.606585631349782,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004321354883730488,
      "loss": 5.0966,
      "mean_token_accuracy": 0.20002554804086686,
      "num_tokens": 59620729.0,
      "step": 28735
    },
    {
      "entropy": 5.561369371414185,
      "epoch": 2.6070391872278664,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043211232837805973,
      "loss": 4.8887,
      "mean_token_accuracy": 0.21987232565879822,
      "num_tokens": 59631243.0,
      "step": 28740
    },
    {
      "entropy": 5.604061460494995,
      "epoch": 2.6074927431059507,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043208916513394767,
      "loss": 4.9173,
      "mean_token_accuracy": 0.21859145760536194,
      "num_tokens": 59642055.0,
      "step": 28745
    },
    {
      "entropy": 5.664089727401733,
      "epoch": 2.607946298984035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043206599864119164,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20990859270095824,
      "num_tokens": 59652144.0,
      "step": 28750
    },
    {
      "entropy": 5.606751394271851,
      "epoch": 2.6083998548621192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043204282890027074,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22807482182979583,
      "num_tokens": 59662367.0,
      "step": 28755
    },
    {
      "entropy": 5.64818639755249,
      "epoch": 2.608853410740203,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043201965591166413,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2180311530828476,
      "num_tokens": 59672729.0,
      "step": 28760
    },
    {
      "entropy": 5.649823617935181,
      "epoch": 2.6093069666182873,
      "grad_norm": 1.171875,
      "learning_rate": 0.000431996479675851,
      "loss": 5.0057,
      "mean_token_accuracy": 0.21764556765556337,
      "num_tokens": 59683282.0,
      "step": 28765
    },
    {
      "entropy": 5.633039426803589,
      "epoch": 2.6097605224963716,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043197330019331074,
      "loss": 5.0466,
      "mean_token_accuracy": 0.204258731007576,
      "num_tokens": 59694297.0,
      "step": 28770
    },
    {
      "entropy": 5.540487909317017,
      "epoch": 2.610214078374456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043195011746452244,
      "loss": 4.8653,
      "mean_token_accuracy": 0.217235167324543,
      "num_tokens": 59704149.0,
      "step": 28775
    },
    {
      "entropy": 5.627522087097168,
      "epoch": 2.6106676342525397,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043192693148996585,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21168753057718276,
      "num_tokens": 59714811.0,
      "step": 28780
    },
    {
      "entropy": 5.535090017318725,
      "epoch": 2.611121190130624,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004319037422701203,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2220741555094719,
      "num_tokens": 59724950.0,
      "step": 28785
    },
    {
      "entropy": 5.614667510986328,
      "epoch": 2.611574746008708,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004318805498054654,
      "loss": 5.0572,
      "mean_token_accuracy": 0.19790161401033401,
      "num_tokens": 59735905.0,
      "step": 28790
    },
    {
      "entropy": 5.678648233413696,
      "epoch": 2.6120283018867925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043185735409648077,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2086726114153862,
      "num_tokens": 59746743.0,
      "step": 28795
    },
    {
      "entropy": 5.645985221862793,
      "epoch": 2.6124818577648767,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043183415514364606,
      "loss": 4.9226,
      "mean_token_accuracy": 0.20773072242736818,
      "num_tokens": 59757205.0,
      "step": 28800
    },
    {
      "entropy": 5.563832569122314,
      "epoch": 2.612935413642961,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004318109529474411,
      "loss": 4.9307,
      "mean_token_accuracy": 0.21350997686386108,
      "num_tokens": 59767691.0,
      "step": 28805
    },
    {
      "entropy": 5.546957921981812,
      "epoch": 2.6133889695210453,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004317877475083456,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21252218782901763,
      "num_tokens": 59778618.0,
      "step": 28810
    },
    {
      "entropy": 5.612049341201782,
      "epoch": 2.613842525399129,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043176453882683955,
      "loss": 4.9772,
      "mean_token_accuracy": 0.2077659174799919,
      "num_tokens": 59789423.0,
      "step": 28815
    },
    {
      "entropy": 5.678122425079346,
      "epoch": 2.6142960812772134,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043174132690340287,
      "loss": 5.0231,
      "mean_token_accuracy": 0.20145198255777358,
      "num_tokens": 59800021.0,
      "step": 28820
    },
    {
      "entropy": 5.561448335647583,
      "epoch": 2.6147496371552976,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004317181117385156,
      "loss": 4.9133,
      "mean_token_accuracy": 0.20836929380893707,
      "num_tokens": 59810006.0,
      "step": 28825
    },
    {
      "entropy": 5.6187968254089355,
      "epoch": 2.6152031930333814,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043169489333265795,
      "loss": 4.97,
      "mean_token_accuracy": 0.21044576913118362,
      "num_tokens": 59819505.0,
      "step": 28830
    },
    {
      "entropy": 5.621313524246216,
      "epoch": 2.6156567489114657,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043167167168630977,
      "loss": 4.982,
      "mean_token_accuracy": 0.21647306084632872,
      "num_tokens": 59830439.0,
      "step": 28835
    },
    {
      "entropy": 5.603928470611573,
      "epoch": 2.61611030478955,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004316484467999516,
      "loss": 4.9398,
      "mean_token_accuracy": 0.21201158463954925,
      "num_tokens": 59841281.0,
      "step": 28840
    },
    {
      "entropy": 5.606333303451538,
      "epoch": 2.6165638606676342,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043162521867406376,
      "loss": 4.9801,
      "mean_token_accuracy": 0.20431240797042846,
      "num_tokens": 59853581.0,
      "step": 28845
    },
    {
      "entropy": 5.680047702789307,
      "epoch": 2.6170174165457185,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004316019873091263,
      "loss": 5.0169,
      "mean_token_accuracy": 0.21038201600313186,
      "num_tokens": 59864787.0,
      "step": 28850
    },
    {
      "entropy": 5.645145893096924,
      "epoch": 2.6174709724238028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004315787527056199,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21263731718063356,
      "num_tokens": 59875244.0,
      "step": 28855
    },
    {
      "entropy": 5.569312000274659,
      "epoch": 2.617924528301887,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004315555148640251,
      "loss": 4.9173,
      "mean_token_accuracy": 0.21303060948848723,
      "num_tokens": 59886535.0,
      "step": 28860
    },
    {
      "entropy": 5.576078128814697,
      "epoch": 2.618378084179971,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004315322737848222,
      "loss": 4.9217,
      "mean_token_accuracy": 0.21538650393486022,
      "num_tokens": 59896298.0,
      "step": 28865
    },
    {
      "entropy": 5.500714635848999,
      "epoch": 2.618831640058055,
      "grad_norm": 1.171875,
      "learning_rate": 0.000431509029468492,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21334036588668823,
      "num_tokens": 59907010.0,
      "step": 28870
    },
    {
      "entropy": 5.636637210845947,
      "epoch": 2.6192851959361394,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004314857819155153,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2183952048420906,
      "num_tokens": 59916247.0,
      "step": 28875
    },
    {
      "entropy": 5.537689352035523,
      "epoch": 2.6197387518142237,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043146253112637267,
      "loss": 4.8363,
      "mean_token_accuracy": 0.22253036201000215,
      "num_tokens": 59926680.0,
      "step": 28880
    },
    {
      "entropy": 5.517590570449829,
      "epoch": 2.6201923076923075,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043143927710154504,
      "loss": 4.86,
      "mean_token_accuracy": 0.22130801826715468,
      "num_tokens": 59936346.0,
      "step": 28885
    },
    {
      "entropy": 5.585208892822266,
      "epoch": 2.6206458635703918,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004314160198415133,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2234189748764038,
      "num_tokens": 59947007.0,
      "step": 28890
    },
    {
      "entropy": 5.630993127822876,
      "epoch": 2.621099419448476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043139275934675844,
      "loss": 4.9875,
      "mean_token_accuracy": 0.21387099623680114,
      "num_tokens": 59957123.0,
      "step": 28895
    },
    {
      "entropy": 5.602101898193359,
      "epoch": 2.6215529753265603,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004313694956177613,
      "loss": 4.9489,
      "mean_token_accuracy": 0.21782744973897933,
      "num_tokens": 59966566.0,
      "step": 28900
    },
    {
      "entropy": 5.610617542266846,
      "epoch": 2.6220065312046446,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043134622865500334,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21037882417440415,
      "num_tokens": 59976065.0,
      "step": 28905
    },
    {
      "entropy": 5.619112396240235,
      "epoch": 2.622460087082729,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004313229584589655,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21438599228858948,
      "num_tokens": 59986201.0,
      "step": 28910
    },
    {
      "entropy": 5.566736698150635,
      "epoch": 2.6229136429608126,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000431299685030129,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21046952307224273,
      "num_tokens": 59996574.0,
      "step": 28915
    },
    {
      "entropy": 5.608832836151123,
      "epoch": 2.623367198838897,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043127640836897516,
      "loss": 4.9444,
      "mean_token_accuracy": 0.218801411986351,
      "num_tokens": 60006082.0,
      "step": 28920
    },
    {
      "entropy": 5.612108373641968,
      "epoch": 2.623820754716981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004312531284759854,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2128320351243019,
      "num_tokens": 60015483.0,
      "step": 28925
    },
    {
      "entropy": 5.585036373138427,
      "epoch": 2.6242743105950654,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043122984535164116,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22925111949443816,
      "num_tokens": 60025984.0,
      "step": 28930
    },
    {
      "entropy": 5.623856115341186,
      "epoch": 2.6247278664731493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043120655899642386,
      "loss": 4.9516,
      "mean_token_accuracy": 0.20588652789592743,
      "num_tokens": 60036109.0,
      "step": 28935
    },
    {
      "entropy": 5.673203086853027,
      "epoch": 2.6251814223512335,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043118326941081513,
      "loss": 5.015,
      "mean_token_accuracy": 0.20424046218395234,
      "num_tokens": 60046546.0,
      "step": 28940
    },
    {
      "entropy": 5.6346159934997555,
      "epoch": 2.625634978229318,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004311599765952966,
      "loss": 5.0068,
      "mean_token_accuracy": 0.21607362776994704,
      "num_tokens": 60056768.0,
      "step": 28945
    },
    {
      "entropy": 5.637686634063721,
      "epoch": 2.626088534107402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043113668055034995,
      "loss": 5.0634,
      "mean_token_accuracy": 0.20704057812690735,
      "num_tokens": 60066962.0,
      "step": 28950
    },
    {
      "entropy": 5.6153168201446535,
      "epoch": 2.6265420899854863,
      "grad_norm": 1.25,
      "learning_rate": 0.0004311133812764569,
      "loss": 4.9419,
      "mean_token_accuracy": 0.21427322477102279,
      "num_tokens": 60076477.0,
      "step": 28955
    },
    {
      "entropy": 5.550201272964477,
      "epoch": 2.6269956458635706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043109007877409946,
      "loss": 4.8265,
      "mean_token_accuracy": 0.22151952981948853,
      "num_tokens": 60088060.0,
      "step": 28960
    },
    {
      "entropy": 5.597414588928222,
      "epoch": 2.6274492017416544,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043106677304375935,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21379705220460893,
      "num_tokens": 60098677.0,
      "step": 28965
    },
    {
      "entropy": 5.637823247909546,
      "epoch": 2.6279027576197387,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004310434640859186,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20918598175048828,
      "num_tokens": 60109054.0,
      "step": 28970
    },
    {
      "entropy": 5.587820386886596,
      "epoch": 2.628356313497823,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004310201519010592,
      "loss": 4.8596,
      "mean_token_accuracy": 0.21929230242967607,
      "num_tokens": 60119951.0,
      "step": 28975
    },
    {
      "entropy": 5.499077987670899,
      "epoch": 2.628809869375907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043099683648966335,
      "loss": 4.8327,
      "mean_token_accuracy": 0.21656721383333205,
      "num_tokens": 60130026.0,
      "step": 28980
    },
    {
      "entropy": 5.485948848724365,
      "epoch": 2.629263425253991,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004309735178522132,
      "loss": 4.8614,
      "mean_token_accuracy": 0.21893069446086882,
      "num_tokens": 60140748.0,
      "step": 28985
    },
    {
      "entropy": 5.570332002639771,
      "epoch": 2.6297169811320753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004309501959891909,
      "loss": 4.9161,
      "mean_token_accuracy": 0.21589876860380172,
      "num_tokens": 60151042.0,
      "step": 28990
    },
    {
      "entropy": 5.6394469261169435,
      "epoch": 2.6301705370101596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043092687090107883,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20734856873750687,
      "num_tokens": 60160758.0,
      "step": 28995
    },
    {
      "entropy": 5.531998872756958,
      "epoch": 2.630624092888244,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043090354258835934,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21061759293079377,
      "num_tokens": 60170249.0,
      "step": 29000
    },
    {
      "entropy": 5.531155824661255,
      "epoch": 2.631077648766328,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043088021105151483,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21220093667507173,
      "num_tokens": 60181216.0,
      "step": 29005
    },
    {
      "entropy": 5.546112728118897,
      "epoch": 2.6315312046444124,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004308568762910279,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21996376067399978,
      "num_tokens": 60190559.0,
      "step": 29010
    },
    {
      "entropy": 5.6920863628387455,
      "epoch": 2.6319847605224966,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043083353830738097,
      "loss": 5.031,
      "mean_token_accuracy": 0.203672294318676,
      "num_tokens": 60201268.0,
      "step": 29015
    },
    {
      "entropy": 5.5402655601501465,
      "epoch": 2.6324383164005805,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043081019710105683,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2213676854968071,
      "num_tokens": 60210907.0,
      "step": 29020
    },
    {
      "entropy": 5.521044445037842,
      "epoch": 2.6328918722786647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004307868526725381,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22310011386871337,
      "num_tokens": 60220298.0,
      "step": 29025
    },
    {
      "entropy": 5.514516162872314,
      "epoch": 2.633345428156749,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004307635050223075,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22228290885686874,
      "num_tokens": 60230811.0,
      "step": 29030
    },
    {
      "entropy": 5.612066030502319,
      "epoch": 2.6337989840348333,
      "grad_norm": 1.203125,
      "learning_rate": 0.000430740154150848,
      "loss": 4.8629,
      "mean_token_accuracy": 0.2183879643678665,
      "num_tokens": 60241211.0,
      "step": 29035
    },
    {
      "entropy": 5.522991895675659,
      "epoch": 2.634252539912917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004307168000586423,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22052494883537294,
      "num_tokens": 60251506.0,
      "step": 29040
    },
    {
      "entropy": 5.524680042266846,
      "epoch": 2.6347060957910013,
      "grad_norm": 1.5,
      "learning_rate": 0.0004306934427461736,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2325665459036827,
      "num_tokens": 60261416.0,
      "step": 29045
    },
    {
      "entropy": 5.5910240650177006,
      "epoch": 2.6351596516690856,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004306700822139248,
      "loss": 4.9386,
      "mean_token_accuracy": 0.211593659222126,
      "num_tokens": 60272315.0,
      "step": 29050
    },
    {
      "entropy": 5.6002648830413815,
      "epoch": 2.63561320754717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000430646718462379,
      "loss": 5.0014,
      "mean_token_accuracy": 0.21055780500173568,
      "num_tokens": 60282722.0,
      "step": 29055
    },
    {
      "entropy": 5.595898580551148,
      "epoch": 2.636066763425254,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004306233514920194,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21408590227365493,
      "num_tokens": 60292636.0,
      "step": 29060
    },
    {
      "entropy": 5.575620317459107,
      "epoch": 2.6365203193033384,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043059998130332933,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2143368050456047,
      "num_tokens": 60302979.0,
      "step": 29065
    },
    {
      "entropy": 5.599817514419556,
      "epoch": 2.6369738751814222,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043057660789679194,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21508259177207947,
      "num_tokens": 60313181.0,
      "step": 29070
    },
    {
      "entropy": 5.642970275878906,
      "epoch": 2.6374274310595065,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004305532312728906,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21591168642044067,
      "num_tokens": 60323401.0,
      "step": 29075
    },
    {
      "entropy": 5.640989685058594,
      "epoch": 2.6378809869375908,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004305298514321088,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20930237025022508,
      "num_tokens": 60333478.0,
      "step": 29080
    },
    {
      "entropy": 5.56065936088562,
      "epoch": 2.638334542815675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043050646837493014,
      "loss": 4.898,
      "mean_token_accuracy": 0.2156144365668297,
      "num_tokens": 60343980.0,
      "step": 29085
    },
    {
      "entropy": 5.585715627670288,
      "epoch": 2.638788098693759,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004304830821018381,
      "loss": 4.902,
      "mean_token_accuracy": 0.2183808535337448,
      "num_tokens": 60354262.0,
      "step": 29090
    },
    {
      "entropy": 5.652287578582763,
      "epoch": 2.639241654571843,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004304596926133162,
      "loss": 4.9969,
      "mean_token_accuracy": 0.21569766253232955,
      "num_tokens": 60364137.0,
      "step": 29095
    },
    {
      "entropy": 5.59980354309082,
      "epoch": 2.6396952104499274,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043043629990984823,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21163684576749803,
      "num_tokens": 60374528.0,
      "step": 29100
    },
    {
      "entropy": 5.580597925186157,
      "epoch": 2.6401487663280117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043041290399191797,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2219318628311157,
      "num_tokens": 60384880.0,
      "step": 29105
    },
    {
      "entropy": 5.564107131958008,
      "epoch": 2.640602322206096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303895048600093,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20896926820278167,
      "num_tokens": 60395543.0,
      "step": 29110
    },
    {
      "entropy": 5.609448099136353,
      "epoch": 2.64105587808418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004303661025146061,
      "loss": 4.9194,
      "mean_token_accuracy": 0.22017806619405747,
      "num_tokens": 60405432.0,
      "step": 29115
    },
    {
      "entropy": 5.663624382019043,
      "epoch": 2.641509433962264,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004303426969561922,
      "loss": 5.0049,
      "mean_token_accuracy": 0.20715552419424058,
      "num_tokens": 60416360.0,
      "step": 29120
    },
    {
      "entropy": 5.592259502410888,
      "epoch": 2.6419629898403483,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004303192881852518,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21244598925113678,
      "num_tokens": 60426371.0,
      "step": 29125
    },
    {
      "entropy": 5.588502359390259,
      "epoch": 2.6424165457184325,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043029587620226897,
      "loss": 4.9342,
      "mean_token_accuracy": 0.21446119397878646,
      "num_tokens": 60437505.0,
      "step": 29130
    },
    {
      "entropy": 5.599785852432251,
      "epoch": 2.642870101596517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004302724610077278,
      "loss": 4.9251,
      "mean_token_accuracy": 0.21085345447063447,
      "num_tokens": 60448170.0,
      "step": 29135
    },
    {
      "entropy": 5.661326694488525,
      "epoch": 2.6433236574746006,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004302490426021126,
      "loss": 5.022,
      "mean_token_accuracy": 0.20712724924087525,
      "num_tokens": 60458776.0,
      "step": 29140
    },
    {
      "entropy": 5.607163143157959,
      "epoch": 2.643777213352685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004302256209859076,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22149889469146727,
      "num_tokens": 60468279.0,
      "step": 29145
    },
    {
      "entropy": 5.566177225112915,
      "epoch": 2.644230769230769,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043020219615959727,
      "loss": 4.9869,
      "mean_token_accuracy": 0.20772133022546768,
      "num_tokens": 60478507.0,
      "step": 29150
    },
    {
      "entropy": 5.610563278198242,
      "epoch": 2.6446843251088534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004301787681236659,
      "loss": 5.0584,
      "mean_token_accuracy": 0.20073592513799668,
      "num_tokens": 60489016.0,
      "step": 29155
    },
    {
      "entropy": 5.686238098144531,
      "epoch": 2.6451378809869377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004301553368785981,
      "loss": 5.0042,
      "mean_token_accuracy": 0.20334473699331285,
      "num_tokens": 60500169.0,
      "step": 29160
    },
    {
      "entropy": 5.592184114456177,
      "epoch": 2.645591436865022,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004301319024248784,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2178480789065361,
      "num_tokens": 60511109.0,
      "step": 29165
    },
    {
      "entropy": 5.5679283618927,
      "epoch": 2.6460449927431062,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043010846476299145,
      "loss": 4.91,
      "mean_token_accuracy": 0.2227904751896858,
      "num_tokens": 60521735.0,
      "step": 29170
    },
    {
      "entropy": 5.6478653907775875,
      "epoch": 2.64649854862119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004300850238934219,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20595175325870513,
      "num_tokens": 60532071.0,
      "step": 29175
    },
    {
      "entropy": 5.646704149246216,
      "epoch": 2.6469521044992743,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043006157981665446,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2191327765583992,
      "num_tokens": 60541909.0,
      "step": 29180
    },
    {
      "entropy": 5.538444995880127,
      "epoch": 2.6474056603773586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043003813253317415,
      "loss": 4.8718,
      "mean_token_accuracy": 0.21934362202882768,
      "num_tokens": 60551361.0,
      "step": 29185
    },
    {
      "entropy": 5.552004623413086,
      "epoch": 2.6478592162554424,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043001468204346564,
      "loss": 4.9639,
      "mean_token_accuracy": 0.21291962563991546,
      "num_tokens": 60562112.0,
      "step": 29190
    },
    {
      "entropy": 5.623826313018799,
      "epoch": 2.6483127721335267,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042999122834801406,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20764190554618836,
      "num_tokens": 60572827.0,
      "step": 29195
    },
    {
      "entropy": 5.5257162094116214,
      "epoch": 2.648766328011611,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004299677714473044,
      "loss": 4.8801,
      "mean_token_accuracy": 0.21866969019174576,
      "num_tokens": 60582222.0,
      "step": 29200
    },
    {
      "entropy": 5.593174457550049,
      "epoch": 2.649219883889695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004299443113418217,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2189294531941414,
      "num_tokens": 60592943.0,
      "step": 29205
    },
    {
      "entropy": 5.625114297866821,
      "epoch": 2.6496734397677795,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042992084803205124,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21454366594552993,
      "num_tokens": 60603206.0,
      "step": 29210
    },
    {
      "entropy": 5.573830223083496,
      "epoch": 2.6501269956458637,
      "grad_norm": 1.203125,
      "learning_rate": 0.000429897381518478,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2208836033940315,
      "num_tokens": 60614776.0,
      "step": 29215
    },
    {
      "entropy": 5.5716935157775875,
      "epoch": 2.650580551523948,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042987391180158757,
      "loss": 4.9209,
      "mean_token_accuracy": 0.208964903652668,
      "num_tokens": 60625914.0,
      "step": 29220
    },
    {
      "entropy": 5.587363529205322,
      "epoch": 2.651034107402032,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042985043888186516,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21605506390333176,
      "num_tokens": 60636934.0,
      "step": 29225
    },
    {
      "entropy": 5.6113217830657955,
      "epoch": 2.651487663280116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004298269627597962,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20899191945791246,
      "num_tokens": 60647707.0,
      "step": 29230
    },
    {
      "entropy": 5.510380363464355,
      "epoch": 2.6519412191582004,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042980348343586624,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2141405686736107,
      "num_tokens": 60658691.0,
      "step": 29235
    },
    {
      "entropy": 5.601118040084839,
      "epoch": 2.6523947750362846,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042978000091056074,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21102332472801208,
      "num_tokens": 60668479.0,
      "step": 29240
    },
    {
      "entropy": 5.583303737640381,
      "epoch": 2.6528483309143684,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004297565151843653,
      "loss": 4.838,
      "mean_token_accuracy": 0.2179940462112427,
      "num_tokens": 60679158.0,
      "step": 29245
    },
    {
      "entropy": 5.646920108795166,
      "epoch": 2.6533018867924527,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004297330262577659,
      "loss": 5.0701,
      "mean_token_accuracy": 0.2052389770746231,
      "num_tokens": 60689363.0,
      "step": 29250
    },
    {
      "entropy": 5.564053058624268,
      "epoch": 2.653755442670537,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004297095341312479,
      "loss": 4.8326,
      "mean_token_accuracy": 0.22076878547668458,
      "num_tokens": 60699650.0,
      "step": 29255
    },
    {
      "entropy": 5.533982181549073,
      "epoch": 2.6542089985486212,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004296860388052974,
      "loss": 4.8973,
      "mean_token_accuracy": 0.21476273685693742,
      "num_tokens": 60709921.0,
      "step": 29260
    },
    {
      "entropy": 5.613673686981201,
      "epoch": 2.6546625544267055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042966254028040015,
      "loss": 4.883,
      "mean_token_accuracy": 0.22746307998895646,
      "num_tokens": 60720112.0,
      "step": 29265
    },
    {
      "entropy": 5.569327735900879,
      "epoch": 2.65511611030479,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042963903855704216,
      "loss": 4.854,
      "mean_token_accuracy": 0.20987479984760285,
      "num_tokens": 60730604.0,
      "step": 29270
    },
    {
      "entropy": 5.57329797744751,
      "epoch": 2.6555696661828736,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042961553363570946,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2159927800297737,
      "num_tokens": 60740896.0,
      "step": 29275
    },
    {
      "entropy": 5.626656913757325,
      "epoch": 2.656023222060958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004295920255168881,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21682820469141006,
      "num_tokens": 60751771.0,
      "step": 29280
    },
    {
      "entropy": 5.643973875045776,
      "epoch": 2.656476777939042,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042956851420106423,
      "loss": 4.9767,
      "mean_token_accuracy": 0.21327410638332367,
      "num_tokens": 60762802.0,
      "step": 29285
    },
    {
      "entropy": 5.534553098678589,
      "epoch": 2.6569303338171264,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004295449996887242,
      "loss": 4.8612,
      "mean_token_accuracy": 0.2147950768470764,
      "num_tokens": 60774102.0,
      "step": 29290
    },
    {
      "entropy": 5.554935693740845,
      "epoch": 2.65738388969521,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004295214819803541,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2114003375172615,
      "num_tokens": 60785383.0,
      "step": 29295
    },
    {
      "entropy": 5.618823003768921,
      "epoch": 2.6578374455732945,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042949796107644053,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21727352291345597,
      "num_tokens": 60795173.0,
      "step": 29300
    },
    {
      "entropy": 5.595835161209107,
      "epoch": 2.6582910014513788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042947443697746956,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2198912486433983,
      "num_tokens": 60805621.0,
      "step": 29305
    },
    {
      "entropy": 5.615415000915528,
      "epoch": 2.658744557329463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004294509096839279,
      "loss": 5.0421,
      "mean_token_accuracy": 0.2025065928697586,
      "num_tokens": 60816799.0,
      "step": 29310
    },
    {
      "entropy": 5.509220600128174,
      "epoch": 2.6591981132075473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042942737919630215,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2169080838561058,
      "num_tokens": 60826827.0,
      "step": 29315
    },
    {
      "entropy": 5.606493282318115,
      "epoch": 2.6596516690856316,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042940384551507876,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21344392150640487,
      "num_tokens": 60836567.0,
      "step": 29320
    },
    {
      "entropy": 5.6255066871643065,
      "epoch": 2.6601052249637154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004293803086407447,
      "loss": 4.9186,
      "mean_token_accuracy": 0.21058397144079208,
      "num_tokens": 60846723.0,
      "step": 29325
    },
    {
      "entropy": 5.656845569610596,
      "epoch": 2.6605587808417996,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004293567685737863,
      "loss": 5.0333,
      "mean_token_accuracy": 0.2048659861087799,
      "num_tokens": 60857434.0,
      "step": 29330
    },
    {
      "entropy": 5.635298252105713,
      "epoch": 2.661012336719884,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042933322531469074,
      "loss": 4.9821,
      "mean_token_accuracy": 0.21060677617788315,
      "num_tokens": 60868780.0,
      "step": 29335
    },
    {
      "entropy": 5.660285568237304,
      "epoch": 2.661465892597968,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042930967886394456,
      "loss": 4.988,
      "mean_token_accuracy": 0.2140889808535576,
      "num_tokens": 60879037.0,
      "step": 29340
    },
    {
      "entropy": 5.5364936828613285,
      "epoch": 2.661919448476052,
      "grad_norm": 1.03125,
      "learning_rate": 0.000429286129222035,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2132459357380867,
      "num_tokens": 60889543.0,
      "step": 29345
    },
    {
      "entropy": 5.554695510864258,
      "epoch": 2.6623730043541363,
      "grad_norm": 1.046875,
      "learning_rate": 0.000429262576389449,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2105930358171463,
      "num_tokens": 60900141.0,
      "step": 29350
    },
    {
      "entropy": 5.62404580116272,
      "epoch": 2.6628265602322205,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004292390203666736,
      "loss": 4.9737,
      "mean_token_accuracy": 0.2165334090590477,
      "num_tokens": 60910244.0,
      "step": 29355
    },
    {
      "entropy": 5.534553909301758,
      "epoch": 2.663280116110305,
      "grad_norm": 1.1875,
      "learning_rate": 0.000429215461154196,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22304699718952178,
      "num_tokens": 60919908.0,
      "step": 29360
    },
    {
      "entropy": 5.6069643020629885,
      "epoch": 2.663733671988389,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042919189875250324,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21423041969537734,
      "num_tokens": 60929096.0,
      "step": 29365
    },
    {
      "entropy": 5.618274927139282,
      "epoch": 2.6641872278664733,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042916833316208284,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21121857315301895,
      "num_tokens": 60939346.0,
      "step": 29370
    },
    {
      "entropy": 5.611999607086181,
      "epoch": 2.6646407837445576,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004291447643834219,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20974236279726027,
      "num_tokens": 60950364.0,
      "step": 29375
    },
    {
      "entropy": 5.561800813674926,
      "epoch": 2.6650943396226414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000429121192417008,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2221387267112732,
      "num_tokens": 60960905.0,
      "step": 29380
    },
    {
      "entropy": 5.613738965988159,
      "epoch": 2.6655478955007257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042909761726332864,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21911579966545106,
      "num_tokens": 60971281.0,
      "step": 29385
    },
    {
      "entropy": 5.579023218154907,
      "epoch": 2.66600145137881,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004290740389228712,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21611983478069305,
      "num_tokens": 60982522.0,
      "step": 29390
    },
    {
      "entropy": 5.575920820236206,
      "epoch": 2.6664550072568938,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042905045739612347,
      "loss": 4.9534,
      "mean_token_accuracy": 0.21015044897794724,
      "num_tokens": 60992738.0,
      "step": 29395
    },
    {
      "entropy": 5.674988174438477,
      "epoch": 2.666908563134978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042902687268357297,
      "loss": 5.0483,
      "mean_token_accuracy": 0.21066773682832718,
      "num_tokens": 61003665.0,
      "step": 29400
    },
    {
      "entropy": 5.723544454574585,
      "epoch": 2.6673621190130623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042900328478570745,
      "loss": 5.0558,
      "mean_token_accuracy": 0.2096869632601738,
      "num_tokens": 61014663.0,
      "step": 29405
    },
    {
      "entropy": 5.615775728225708,
      "epoch": 2.6678156748911466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004289796937030148,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2180102750658989,
      "num_tokens": 61024916.0,
      "step": 29410
    },
    {
      "entropy": 5.522702074050903,
      "epoch": 2.668269230769231,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042895609943598286,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21833383589982985,
      "num_tokens": 61035043.0,
      "step": 29415
    },
    {
      "entropy": 5.564690828323364,
      "epoch": 2.668722786647315,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004289325019850995,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21271272301673888,
      "num_tokens": 61046153.0,
      "step": 29420
    },
    {
      "entropy": 5.614724636077881,
      "epoch": 2.6691763425253994,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004289089013508528,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21095340549945832,
      "num_tokens": 61057009.0,
      "step": 29425
    },
    {
      "entropy": 5.654237365722656,
      "epoch": 2.669629898403483,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004288852975337308,
      "loss": 5.0517,
      "mean_token_accuracy": 0.20024534910917283,
      "num_tokens": 61068514.0,
      "step": 29430
    },
    {
      "entropy": 5.643145895004272,
      "epoch": 2.6700834542815675,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042886169053422165,
      "loss": 4.9547,
      "mean_token_accuracy": 0.21623586118221283,
      "num_tokens": 61078902.0,
      "step": 29435
    },
    {
      "entropy": 5.6215770721435545,
      "epoch": 2.6705370101596517,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004288380803528135,
      "loss": 4.9556,
      "mean_token_accuracy": 0.2185447558760643,
      "num_tokens": 61089216.0,
      "step": 29440
    },
    {
      "entropy": 5.548191499710083,
      "epoch": 2.670990566037736,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042881446698999465,
      "loss": 4.8563,
      "mean_token_accuracy": 0.21845725029706956,
      "num_tokens": 61098470.0,
      "step": 29445
    },
    {
      "entropy": 5.686565208435058,
      "epoch": 2.67144412191582,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004287908504462534,
      "loss": 5.065,
      "mean_token_accuracy": 0.21650877594947815,
      "num_tokens": 61108694.0,
      "step": 29450
    },
    {
      "entropy": 5.560374593734741,
      "epoch": 2.671897677793904,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042876723072207825,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21677561700344086,
      "num_tokens": 61119389.0,
      "step": 29455
    },
    {
      "entropy": 5.633248567581177,
      "epoch": 2.6723512336719883,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042874360781795757,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2091132178902626,
      "num_tokens": 61130491.0,
      "step": 29460
    },
    {
      "entropy": 5.604853534698487,
      "epoch": 2.6728047895500726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042871998173437987,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21624602973461152,
      "num_tokens": 61140404.0,
      "step": 29465
    },
    {
      "entropy": 5.654856014251709,
      "epoch": 2.673258345428157,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042869635247183373,
      "loss": 5.0122,
      "mean_token_accuracy": 0.2105361670255661,
      "num_tokens": 61150777.0,
      "step": 29470
    },
    {
      "entropy": 5.60025053024292,
      "epoch": 2.673711901306241,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042867272003080793,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21678118258714676,
      "num_tokens": 61160685.0,
      "step": 29475
    },
    {
      "entropy": 5.623936986923217,
      "epoch": 2.674165457184325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004286490844117911,
      "loss": 4.9528,
      "mean_token_accuracy": 0.21411225348711013,
      "num_tokens": 61170553.0,
      "step": 29480
    },
    {
      "entropy": 5.65347204208374,
      "epoch": 2.6746190130624092,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042862544561527203,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21210510432720184,
      "num_tokens": 61181396.0,
      "step": 29485
    },
    {
      "entropy": 5.601888847351074,
      "epoch": 2.6750725689404935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004286018036417396,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21526999771595,
      "num_tokens": 61192315.0,
      "step": 29490
    },
    {
      "entropy": 5.6078180313110355,
      "epoch": 2.6755261248185778,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042857815849168277,
      "loss": 4.9251,
      "mean_token_accuracy": 0.21740586012601854,
      "num_tokens": 61202719.0,
      "step": 29495
    },
    {
      "entropy": 5.521963262557984,
      "epoch": 2.6759796806966616,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004285545101655904,
      "loss": 4.8249,
      "mean_token_accuracy": 0.22316931635141374,
      "num_tokens": 61212703.0,
      "step": 29500
    },
    {
      "entropy": 5.482142686843872,
      "epoch": 2.676433236574746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004285308586639517,
      "loss": 4.8072,
      "mean_token_accuracy": 0.22762789130210875,
      "num_tokens": 61222406.0,
      "step": 29505
    },
    {
      "entropy": 5.568676710128784,
      "epoch": 2.67688679245283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042850720398725575,
      "loss": 5.0185,
      "mean_token_accuracy": 0.2032059445977211,
      "num_tokens": 61232758.0,
      "step": 29510
    },
    {
      "entropy": 5.55197582244873,
      "epoch": 2.6773403483309144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042848354613599164,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22180147767066954,
      "num_tokens": 61243715.0,
      "step": 29515
    },
    {
      "entropy": 5.638426828384399,
      "epoch": 2.6777939042089987,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042845988511064863,
      "loss": 4.928,
      "mean_token_accuracy": 0.21806416511535645,
      "num_tokens": 61254669.0,
      "step": 29520
    },
    {
      "entropy": 5.595401859283447,
      "epoch": 2.678247460087083,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042843622091171625,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21941980272531508,
      "num_tokens": 61264377.0,
      "step": 29525
    },
    {
      "entropy": 5.541515111923218,
      "epoch": 2.6787010159651667,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004284125535396836,
      "loss": 4.9277,
      "mean_token_accuracy": 0.21571475118398667,
      "num_tokens": 61275348.0,
      "step": 29530
    },
    {
      "entropy": 5.514275741577149,
      "epoch": 2.679154571843251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042838888299504024,
      "loss": 4.8108,
      "mean_token_accuracy": 0.2235444962978363,
      "num_tokens": 61285579.0,
      "step": 29535
    },
    {
      "entropy": 5.548592805862427,
      "epoch": 2.6796081277213353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042836520927827575,
      "loss": 4.8829,
      "mean_token_accuracy": 0.21949210315942763,
      "num_tokens": 61296037.0,
      "step": 29540
    },
    {
      "entropy": 5.594270086288452,
      "epoch": 2.6800616835994195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004283415323898797,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22319923490285873,
      "num_tokens": 61305882.0,
      "step": 29545
    },
    {
      "entropy": 5.553998422622681,
      "epoch": 2.6805152394775034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042831785233034156,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22105389535427095,
      "num_tokens": 61317117.0,
      "step": 29550
    },
    {
      "entropy": 5.600816679000855,
      "epoch": 2.6809687953555876,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042829416910015116,
      "loss": 4.9428,
      "mean_token_accuracy": 0.21059881448745726,
      "num_tokens": 61328190.0,
      "step": 29555
    },
    {
      "entropy": 5.572664546966553,
      "epoch": 2.681422351233672,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042827048269979834,
      "loss": 4.9347,
      "mean_token_accuracy": 0.2145954504609108,
      "num_tokens": 61337978.0,
      "step": 29560
    },
    {
      "entropy": 5.644948196411133,
      "epoch": 2.681875907111756,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004282467931297729,
      "loss": 4.9843,
      "mean_token_accuracy": 0.2076480895280838,
      "num_tokens": 61348523.0,
      "step": 29565
    },
    {
      "entropy": 5.561984539031982,
      "epoch": 2.6823294629898404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004282231003905647,
      "loss": 4.8244,
      "mean_token_accuracy": 0.22791755348443984,
      "num_tokens": 61359250.0,
      "step": 29570
    },
    {
      "entropy": 5.524264812469482,
      "epoch": 2.6827830188679247,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042819940448266376,
      "loss": 4.8421,
      "mean_token_accuracy": 0.21869919896125795,
      "num_tokens": 61369999.0,
      "step": 29575
    },
    {
      "entropy": 5.642303609848023,
      "epoch": 2.683236574746009,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042817570540656003,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20877804309129716,
      "num_tokens": 61380279.0,
      "step": 29580
    },
    {
      "entropy": 5.564070177078247,
      "epoch": 2.683690130624093,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004281520031627438,
      "loss": 4.9321,
      "mean_token_accuracy": 0.22326969504356384,
      "num_tokens": 61389794.0,
      "step": 29585
    },
    {
      "entropy": 5.6434056758880615,
      "epoch": 2.684143686502177,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042812829775170497,
      "loss": 5.0361,
      "mean_token_accuracy": 0.21006974577903748,
      "num_tokens": 61401136.0,
      "step": 29590
    },
    {
      "entropy": 5.588241958618164,
      "epoch": 2.6845972423802613,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042810458917393404,
      "loss": 4.9787,
      "mean_token_accuracy": 0.215427365899086,
      "num_tokens": 61412830.0,
      "step": 29595
    },
    {
      "entropy": 5.6094013214111325,
      "epoch": 2.6850507982583456,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004280808774299211,
      "loss": 4.8711,
      "mean_token_accuracy": 0.2218219369649887,
      "num_tokens": 61422168.0,
      "step": 29600
    },
    {
      "entropy": 5.541056680679321,
      "epoch": 2.6855043541364294,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042805716252015665,
      "loss": 4.9066,
      "mean_token_accuracy": 0.2150896817445755,
      "num_tokens": 61432471.0,
      "step": 29605
    },
    {
      "entropy": 5.564865398406982,
      "epoch": 2.6859579100145137,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042803344444513106,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21957983821630478,
      "num_tokens": 61443402.0,
      "step": 29610
    },
    {
      "entropy": 5.622091484069824,
      "epoch": 2.686411465892598,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004280097232053349,
      "loss": 4.915,
      "mean_token_accuracy": 0.2122504696249962,
      "num_tokens": 61454465.0,
      "step": 29615
    },
    {
      "entropy": 5.604333591461182,
      "epoch": 2.686865021770682,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004279859988012586,
      "loss": 4.9528,
      "mean_token_accuracy": 0.21692622750997542,
      "num_tokens": 61464671.0,
      "step": 29620
    },
    {
      "entropy": 5.5615818977355955,
      "epoch": 2.6873185776487665,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004279622712333929,
      "loss": 4.9085,
      "mean_token_accuracy": 0.22105324566364287,
      "num_tokens": 61474940.0,
      "step": 29625
    },
    {
      "entropy": 5.605070924758911,
      "epoch": 2.6877721335268507,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004279385405022284,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21035200208425522,
      "num_tokens": 61485240.0,
      "step": 29630
    },
    {
      "entropy": 5.56203556060791,
      "epoch": 2.6882256894049346,
      "grad_norm": 1.140625,
      "learning_rate": 0.000427914806608256,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22004778534173966,
      "num_tokens": 61495382.0,
      "step": 29635
    },
    {
      "entropy": 5.551201391220093,
      "epoch": 2.688679245283019,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004278910695519663,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22094402760267257,
      "num_tokens": 61505372.0,
      "step": 29640
    },
    {
      "entropy": 5.597899532318115,
      "epoch": 2.689132801161103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042786732933385046,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21836409866809844,
      "num_tokens": 61516039.0,
      "step": 29645
    },
    {
      "entropy": 5.593144273757934,
      "epoch": 2.6895863570391874,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004278435859543992,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21281750351190568,
      "num_tokens": 61525435.0,
      "step": 29650
    },
    {
      "entropy": 5.640108060836792,
      "epoch": 2.690039912917271,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004278198394141037,
      "loss": 4.9286,
      "mean_token_accuracy": 0.20959557741880416,
      "num_tokens": 61534714.0,
      "step": 29655
    },
    {
      "entropy": 5.679019641876221,
      "epoch": 2.6904934687953554,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042779608971345496,
      "loss": 5.0697,
      "mean_token_accuracy": 0.20783402323722838,
      "num_tokens": 61545841.0,
      "step": 29660
    },
    {
      "entropy": 5.5687628269195555,
      "epoch": 2.6909470246734397,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042777233685294407,
      "loss": 4.8382,
      "mean_token_accuracy": 0.227628718316555,
      "num_tokens": 61555355.0,
      "step": 29665
    },
    {
      "entropy": 5.580555009841919,
      "epoch": 2.691400580551524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042774858083306246,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21662163287401198,
      "num_tokens": 61565474.0,
      "step": 29670
    },
    {
      "entropy": 5.609775495529175,
      "epoch": 2.6918541364296082,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042772482165430117,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2260458156466484,
      "num_tokens": 61575631.0,
      "step": 29675
    },
    {
      "entropy": 5.568985223770142,
      "epoch": 2.6923076923076925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042770105931715167,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21264175325632095,
      "num_tokens": 61586398.0,
      "step": 29680
    },
    {
      "entropy": 5.581919479370117,
      "epoch": 2.6927612481857763,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004276772938221054,
      "loss": 4.8396,
      "mean_token_accuracy": 0.22089870870113373,
      "num_tokens": 61596630.0,
      "step": 29685
    },
    {
      "entropy": 5.603707885742187,
      "epoch": 2.6932148040638606,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004276535251696537,
      "loss": 4.9569,
      "mean_token_accuracy": 0.21582499295473098,
      "num_tokens": 61605965.0,
      "step": 29690
    },
    {
      "entropy": 5.613928699493409,
      "epoch": 2.693668359941945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004276297533602883,
      "loss": 4.993,
      "mean_token_accuracy": 0.21594161093235015,
      "num_tokens": 61616973.0,
      "step": 29695
    },
    {
      "entropy": 5.640426158905029,
      "epoch": 2.694121915820029,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042760597839450057,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20988346338272096,
      "num_tokens": 61627420.0,
      "step": 29700
    },
    {
      "entropy": 5.592198324203491,
      "epoch": 2.694575471698113,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004275822002727824,
      "loss": 4.9199,
      "mean_token_accuracy": 0.21896839737892151,
      "num_tokens": 61638315.0,
      "step": 29705
    },
    {
      "entropy": 5.547072124481201,
      "epoch": 2.6950290275761972,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042755841899562544,
      "loss": 4.9247,
      "mean_token_accuracy": 0.21824809312820434,
      "num_tokens": 61649294.0,
      "step": 29710
    },
    {
      "entropy": 5.625209617614746,
      "epoch": 2.6954825834542815,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042753463456352155,
      "loss": 4.9767,
      "mean_token_accuracy": 0.2093016967177391,
      "num_tokens": 61660312.0,
      "step": 29715
    },
    {
      "entropy": 5.577877998352051,
      "epoch": 2.6959361393323658,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004275108469769625,
      "loss": 4.9356,
      "mean_token_accuracy": 0.21825241595506667,
      "num_tokens": 61670539.0,
      "step": 29720
    },
    {
      "entropy": 5.631443214416504,
      "epoch": 2.69638969521045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042748705623644025,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2202342763543129,
      "num_tokens": 61679613.0,
      "step": 29725
    },
    {
      "entropy": 5.662536716461181,
      "epoch": 2.6968432510885343,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004274632623424468,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21422010213136672,
      "num_tokens": 61689714.0,
      "step": 29730
    },
    {
      "entropy": 5.665488910675049,
      "epoch": 2.6972968069666186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042743946529547435,
      "loss": 5.0085,
      "mean_token_accuracy": 0.21243639588356017,
      "num_tokens": 61700701.0,
      "step": 29735
    },
    {
      "entropy": 5.631139135360717,
      "epoch": 2.6977503628447024,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004274156650960148,
      "loss": 4.9756,
      "mean_token_accuracy": 0.21852733194828033,
      "num_tokens": 61711403.0,
      "step": 29740
    },
    {
      "entropy": 5.466707038879394,
      "epoch": 2.6982039187227866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004273918617445605,
      "loss": 4.7851,
      "mean_token_accuracy": 0.2232792854309082,
      "num_tokens": 61721787.0,
      "step": 29745
    },
    {
      "entropy": 5.652282047271728,
      "epoch": 2.698657474600871,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042736805524160363,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20965745449066162,
      "num_tokens": 61732702.0,
      "step": 29750
    },
    {
      "entropy": 5.577242136001587,
      "epoch": 2.6991110304789547,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042734424558763656,
      "loss": 4.9309,
      "mean_token_accuracy": 0.2205286592245102,
      "num_tokens": 61743477.0,
      "step": 29755
    },
    {
      "entropy": 5.646869659423828,
      "epoch": 2.699564586357039,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004273204327831516,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20892202407121657,
      "num_tokens": 61753620.0,
      "step": 29760
    },
    {
      "entropy": 5.637120628356934,
      "epoch": 2.7000181422351233,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004272966168286414,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22123782932758332,
      "num_tokens": 61763678.0,
      "step": 29765
    },
    {
      "entropy": 5.6065515041351315,
      "epoch": 2.7004716981132075,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042727279772459837,
      "loss": 4.9649,
      "mean_token_accuracy": 0.22268361002206802,
      "num_tokens": 61774419.0,
      "step": 29770
    },
    {
      "entropy": 5.638718748092652,
      "epoch": 2.700925253991292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000427248975471515,
      "loss": 4.9886,
      "mean_token_accuracy": 0.21097040474414824,
      "num_tokens": 61785546.0,
      "step": 29775
    },
    {
      "entropy": 5.622020864486695,
      "epoch": 2.701378809869376,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042722515006988397,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21545098721981049,
      "num_tokens": 61795846.0,
      "step": 29780
    },
    {
      "entropy": 5.592202425003052,
      "epoch": 2.7018323657474603,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004272013215201982,
      "loss": 4.964,
      "mean_token_accuracy": 0.21018354147672652,
      "num_tokens": 61806440.0,
      "step": 29785
    },
    {
      "entropy": 5.56593656539917,
      "epoch": 2.702285921625544,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042717748982295015,
      "loss": 4.9159,
      "mean_token_accuracy": 0.21833684891462327,
      "num_tokens": 61815371.0,
      "step": 29790
    },
    {
      "entropy": 5.647078561782837,
      "epoch": 2.7027394775036284,
      "grad_norm": 1.125,
      "learning_rate": 0.0004271536549786329,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21756405532360076,
      "num_tokens": 61825225.0,
      "step": 29795
    },
    {
      "entropy": 5.572627830505371,
      "epoch": 2.7031930333817127,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004271298169877393,
      "loss": 4.9573,
      "mean_token_accuracy": 0.2187994599342346,
      "num_tokens": 61836705.0,
      "step": 29800
    },
    {
      "entropy": 5.564535331726074,
      "epoch": 2.703646589259797,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004271059758507624,
      "loss": 4.902,
      "mean_token_accuracy": 0.22134292870759964,
      "num_tokens": 61847082.0,
      "step": 29805
    },
    {
      "entropy": 5.616335535049439,
      "epoch": 2.7041001451378808,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042708213156819505,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21321527361869813,
      "num_tokens": 61857130.0,
      "step": 29810
    },
    {
      "entropy": 5.670545768737793,
      "epoch": 2.704553701015965,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004270582841405305,
      "loss": 5.019,
      "mean_token_accuracy": 0.20896028429269792,
      "num_tokens": 61866796.0,
      "step": 29815
    },
    {
      "entropy": 5.6275512218475345,
      "epoch": 2.7050072568940493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004270344335682618,
      "loss": 4.9693,
      "mean_token_accuracy": 0.20980193912982942,
      "num_tokens": 61876356.0,
      "step": 29820
    },
    {
      "entropy": 5.583295822143555,
      "epoch": 2.7054608127721336,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042701057985188235,
      "loss": 4.9723,
      "mean_token_accuracy": 0.2077516570687294,
      "num_tokens": 61887374.0,
      "step": 29825
    },
    {
      "entropy": 5.654992198944091,
      "epoch": 2.705914368650218,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004269867229918854,
      "loss": 5.0036,
      "mean_token_accuracy": 0.2114068567752838,
      "num_tokens": 61897376.0,
      "step": 29830
    },
    {
      "entropy": 5.663776206970215,
      "epoch": 2.706367924528302,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042696286298876424,
      "loss": 5.0167,
      "mean_token_accuracy": 0.2103147938847542,
      "num_tokens": 61908212.0,
      "step": 29835
    },
    {
      "entropy": 5.620198488235474,
      "epoch": 2.706821480406386,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004269389998430124,
      "loss": 5.0451,
      "mean_token_accuracy": 0.19946016520261764,
      "num_tokens": 61920699.0,
      "step": 29840
    },
    {
      "entropy": 5.582410192489624,
      "epoch": 2.70727503628447,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004269151335551233,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2113411009311676,
      "num_tokens": 61931556.0,
      "step": 29845
    },
    {
      "entropy": 5.7171485900878904,
      "epoch": 2.7077285921625545,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004268912641255905,
      "loss": 5.0749,
      "mean_token_accuracy": 0.21553849875926973,
      "num_tokens": 61941483.0,
      "step": 29850
    },
    {
      "entropy": 5.64946403503418,
      "epoch": 2.7081821480406387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042686739155490763,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21388835608959197,
      "num_tokens": 61951262.0,
      "step": 29855
    },
    {
      "entropy": 5.658086347579956,
      "epoch": 2.7086357039187225,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004268435158435684,
      "loss": 4.9939,
      "mean_token_accuracy": 0.2089886859059334,
      "num_tokens": 61962423.0,
      "step": 29860
    },
    {
      "entropy": 5.6905848503112795,
      "epoch": 2.709089259796807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042681963699206667,
      "loss": 5.0591,
      "mean_token_accuracy": 0.20625297278165816,
      "num_tokens": 61973276.0,
      "step": 29865
    },
    {
      "entropy": 5.573126363754272,
      "epoch": 2.709542815674891,
      "grad_norm": 1.125,
      "learning_rate": 0.00042679575500089605,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22009770423173905,
      "num_tokens": 61984369.0,
      "step": 29870
    },
    {
      "entropy": 5.561296510696411,
      "epoch": 2.7099963715529753,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004267718698705506,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21264982372522354,
      "num_tokens": 61994224.0,
      "step": 29875
    },
    {
      "entropy": 5.580608129501343,
      "epoch": 2.7104499274310596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267479816015242,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21732757687568666,
      "num_tokens": 62004546.0,
      "step": 29880
    },
    {
      "entropy": 5.640032386779785,
      "epoch": 2.710903483309144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004267240901943109,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21724576652050018,
      "num_tokens": 62013924.0,
      "step": 29885
    },
    {
      "entropy": 5.58840913772583,
      "epoch": 2.7113570391872277,
      "grad_norm": 1.125,
      "learning_rate": 0.00042670019564940474,
      "loss": 4.927,
      "mean_token_accuracy": 0.21217675507068634,
      "num_tokens": 62026150.0,
      "step": 29890
    },
    {
      "entropy": 5.638037395477295,
      "epoch": 2.711810595065312,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004266762979672998,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21751690208911895,
      "num_tokens": 62036203.0,
      "step": 29895
    },
    {
      "entropy": 5.639289617538452,
      "epoch": 2.7122641509433962,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042665239714849044,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21079617738723755,
      "num_tokens": 62046961.0,
      "step": 29900
    },
    {
      "entropy": 5.556210947036743,
      "epoch": 2.7127177068214805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042662849319347073,
      "loss": 4.8904,
      "mean_token_accuracy": 0.2123594179749489,
      "num_tokens": 62057365.0,
      "step": 29905
    },
    {
      "entropy": 5.5851380825042725,
      "epoch": 2.7131712626995643,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042660458610273527,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21034496277570724,
      "num_tokens": 62067934.0,
      "step": 29910
    },
    {
      "entropy": 5.669028568267822,
      "epoch": 2.7136248185776486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004265806758767783,
      "loss": 4.9984,
      "mean_token_accuracy": 0.2056246057152748,
      "num_tokens": 62077203.0,
      "step": 29915
    },
    {
      "entropy": 5.628414964675903,
      "epoch": 2.714078374455733,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042655676251609435,
      "loss": 4.9303,
      "mean_token_accuracy": 0.21692422777414322,
      "num_tokens": 62087262.0,
      "step": 29920
    },
    {
      "entropy": 5.580858945846558,
      "epoch": 2.714531930333817,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042653284602117795,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21459153443574905,
      "num_tokens": 62097931.0,
      "step": 29925
    },
    {
      "entropy": 5.594641256332397,
      "epoch": 2.7149854862119014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004265089263925236,
      "loss": 5.0078,
      "mean_token_accuracy": 0.2076190382242203,
      "num_tokens": 62109478.0,
      "step": 29930
    },
    {
      "entropy": 5.676377820968628,
      "epoch": 2.7154390420899857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000426485003630626,
      "loss": 5.024,
      "mean_token_accuracy": 0.2074105679988861,
      "num_tokens": 62120735.0,
      "step": 29935
    },
    {
      "entropy": 5.643840599060058,
      "epoch": 2.71589259796807,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004264610777359801,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2116520255804062,
      "num_tokens": 62130950.0,
      "step": 29940
    },
    {
      "entropy": 5.649484014511108,
      "epoch": 2.7163461538461537,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004264371487090804,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20947919487953187,
      "num_tokens": 62141351.0,
      "step": 29945
    },
    {
      "entropy": 5.542219448089599,
      "epoch": 2.716799709724238,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042641321655042185,
      "loss": 4.8886,
      "mean_token_accuracy": 0.21609922051429747,
      "num_tokens": 62151388.0,
      "step": 29950
    },
    {
      "entropy": 5.669764995574951,
      "epoch": 2.7172532656023223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004263892812604995,
      "loss": 5.087,
      "mean_token_accuracy": 0.20519790947437286,
      "num_tokens": 62162809.0,
      "step": 29955
    },
    {
      "entropy": 5.601447343826294,
      "epoch": 2.7177068214804065,
      "grad_norm": 1.25,
      "learning_rate": 0.00042636534283980814,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21473787277936934,
      "num_tokens": 62172738.0,
      "step": 29960
    },
    {
      "entropy": 5.680136823654175,
      "epoch": 2.7181603773584904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000426341401288843,
      "loss": 5.0706,
      "mean_token_accuracy": 0.2054516166448593,
      "num_tokens": 62182508.0,
      "step": 29965
    },
    {
      "entropy": 5.657821750640869,
      "epoch": 2.7186139332365746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042631745660809905,
      "loss": 5.0386,
      "mean_token_accuracy": 0.20377095639705659,
      "num_tokens": 62192873.0,
      "step": 29970
    },
    {
      "entropy": 5.680954551696777,
      "epoch": 2.719067489114659,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042629350879807155,
      "loss": 5.053,
      "mean_token_accuracy": 0.20653095990419387,
      "num_tokens": 62204727.0,
      "step": 29975
    },
    {
      "entropy": 5.627367162704468,
      "epoch": 2.719521044992743,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042626955785925576,
      "loss": 4.9416,
      "mean_token_accuracy": 0.20963971614837645,
      "num_tokens": 62214268.0,
      "step": 29980
    },
    {
      "entropy": 5.655849647521973,
      "epoch": 2.7199746008708274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042624560379214693,
      "loss": 4.9922,
      "mean_token_accuracy": 0.21071461886167525,
      "num_tokens": 62224756.0,
      "step": 29985
    },
    {
      "entropy": 5.659814548492432,
      "epoch": 2.7204281567489117,
      "grad_norm": 1.25,
      "learning_rate": 0.0004262216465972405,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21357038319110871,
      "num_tokens": 62234065.0,
      "step": 29990
    },
    {
      "entropy": 5.636747360229492,
      "epoch": 2.7208817126269955,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004261976862750318,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21384130716323851,
      "num_tokens": 62243732.0,
      "step": 29995
    },
    {
      "entropy": 5.640603733062744,
      "epoch": 2.72133526850508,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004261737228260165,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2205175757408142,
      "num_tokens": 62253461.0,
      "step": 30000
    },
    {
      "epoch": 2.72133526850508,
      "eval_entropy": 5.327902805967029,
      "eval_loss": 4.993574619293213,
      "eval_mean_token_accuracy": 0.22100695479678084,
      "eval_num_tokens": 62253461.0,
      "eval_runtime": 20.622,
      "eval_samples_per_second": 1714.672,
      "eval_steps_per_second": 214.334,
      "step": 30000
    },
    {
      "entropy": 5.632916402816773,
      "epoch": 2.721788824383164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004261497562506901,
      "loss": 4.9856,
      "mean_token_accuracy": 0.21325458437204362,
      "num_tokens": 62263830.0,
      "step": 30005
    },
    {
      "entropy": 5.546909332275391,
      "epoch": 2.7222423802612483,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004261257865495483,
      "loss": 4.9025,
      "mean_token_accuracy": 0.2191924974322319,
      "num_tokens": 62274383.0,
      "step": 30010
    },
    {
      "entropy": 5.668188095092773,
      "epoch": 2.722695936139332,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004261018137230865,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19490485787391662,
      "num_tokens": 62286476.0,
      "step": 30015
    },
    {
      "entropy": 5.721460390090942,
      "epoch": 2.7231494920174164,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042607783777180085,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21309360712766648,
      "num_tokens": 62296192.0,
      "step": 30020
    },
    {
      "entropy": 5.606363344192505,
      "epoch": 2.7236030478955007,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042605385869618693,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22656063735485077,
      "num_tokens": 62305300.0,
      "step": 30025
    },
    {
      "entropy": 5.610333442687988,
      "epoch": 2.724056603773585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004260298764967408,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21354444324970245,
      "num_tokens": 62316802.0,
      "step": 30030
    },
    {
      "entropy": 5.508436489105224,
      "epoch": 2.724510159651669,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004260058911739583,
      "loss": 4.9424,
      "mean_token_accuracy": 0.2136654421687126,
      "num_tokens": 62326444.0,
      "step": 30035
    },
    {
      "entropy": 5.587986135482788,
      "epoch": 2.7249637155297535,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042598190272833547,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22007625699043273,
      "num_tokens": 62336872.0,
      "step": 30040
    },
    {
      "entropy": 5.524597644805908,
      "epoch": 2.7254172714078373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004259579111603685,
      "loss": 4.8361,
      "mean_token_accuracy": 0.21752707809209823,
      "num_tokens": 62346771.0,
      "step": 30045
    },
    {
      "entropy": 5.607333517074585,
      "epoch": 2.7258708272859216,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042593391647055334,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21466097235679626,
      "num_tokens": 62357503.0,
      "step": 30050
    },
    {
      "entropy": 5.54860200881958,
      "epoch": 2.726324383164006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004259099186593864,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2213379293680191,
      "num_tokens": 62367664.0,
      "step": 30055
    },
    {
      "entropy": 5.6246910572052,
      "epoch": 2.72677793904209,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004258859177273639,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2124890550971031,
      "num_tokens": 62376924.0,
      "step": 30060
    },
    {
      "entropy": 5.547257900238037,
      "epoch": 2.727231494920174,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004258619136749821,
      "loss": 4.8429,
      "mean_token_accuracy": 0.219496650993824,
      "num_tokens": 62387529.0,
      "step": 30065
    },
    {
      "entropy": 5.575240707397461,
      "epoch": 2.727685050798258,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004258379065027375,
      "loss": 4.9038,
      "mean_token_accuracy": 0.21730420738458633,
      "num_tokens": 62397833.0,
      "step": 30070
    },
    {
      "entropy": 5.673918008804321,
      "epoch": 2.7281386066763424,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004258138962111266,
      "loss": 4.9855,
      "mean_token_accuracy": 0.21244439482688904,
      "num_tokens": 62409022.0,
      "step": 30075
    },
    {
      "entropy": 5.627362155914307,
      "epoch": 2.7285921625544267,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004257898828006459,
      "loss": 4.9441,
      "mean_token_accuracy": 0.20647481828927994,
      "num_tokens": 62419105.0,
      "step": 30080
    },
    {
      "entropy": 5.642910432815552,
      "epoch": 2.729045718432511,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004257658662717919,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2226934775710106,
      "num_tokens": 62428371.0,
      "step": 30085
    },
    {
      "entropy": 5.61530704498291,
      "epoch": 2.7294992743105952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004257418466250615,
      "loss": 4.8957,
      "mean_token_accuracy": 0.21682464480400085,
      "num_tokens": 62439301.0,
      "step": 30090
    },
    {
      "entropy": 5.61889533996582,
      "epoch": 2.7299528301886795,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042571782386095124,
      "loss": 4.9006,
      "mean_token_accuracy": 0.21412883698940277,
      "num_tokens": 62449450.0,
      "step": 30095
    },
    {
      "entropy": 5.619385862350464,
      "epoch": 2.7304063860667633,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042569379797995793,
      "loss": 4.927,
      "mean_token_accuracy": 0.2092105895280838,
      "num_tokens": 62461051.0,
      "step": 30100
    },
    {
      "entropy": 5.5840870380401615,
      "epoch": 2.7308599419448476,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004256697689825785,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22211892753839493,
      "num_tokens": 62471755.0,
      "step": 30105
    },
    {
      "entropy": 5.558784008026123,
      "epoch": 2.731313497822932,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042564573686930987,
      "loss": 4.9025,
      "mean_token_accuracy": 0.21349219232797623,
      "num_tokens": 62482060.0,
      "step": 30110
    },
    {
      "entropy": 5.5939428329467775,
      "epoch": 2.7317670537010157,
      "grad_norm": 1.125,
      "learning_rate": 0.000425621701640649,
      "loss": 4.9578,
      "mean_token_accuracy": 0.21146677285432816,
      "num_tokens": 62493735.0,
      "step": 30115
    },
    {
      "entropy": 5.640731143951416,
      "epoch": 2.7322206095791,
      "grad_norm": 1.109375,
      "learning_rate": 0.000425597663297093,
      "loss": 4.9734,
      "mean_token_accuracy": 0.21488242149353026,
      "num_tokens": 62504582.0,
      "step": 30120
    },
    {
      "entropy": 5.536716079711914,
      "epoch": 2.7326741654571842,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004255736218391389,
      "loss": 4.9014,
      "mean_token_accuracy": 0.20783267319202423,
      "num_tokens": 62516665.0,
      "step": 30125
    },
    {
      "entropy": 5.618150234222412,
      "epoch": 2.7331277213352685,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042554957726728394,
      "loss": 4.9479,
      "mean_token_accuracy": 0.2149476155638695,
      "num_tokens": 62527745.0,
      "step": 30130
    },
    {
      "entropy": 5.617371129989624,
      "epoch": 2.7335812772133528,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004255255295820253,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21916296631097792,
      "num_tokens": 62538622.0,
      "step": 30135
    },
    {
      "entropy": 5.591380167007446,
      "epoch": 2.734034833091437,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042550147878386044,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21482049226760863,
      "num_tokens": 62548302.0,
      "step": 30140
    },
    {
      "entropy": 5.553401803970337,
      "epoch": 2.7344883889695213,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004254774248732867,
      "loss": 4.857,
      "mean_token_accuracy": 0.22325119376182556,
      "num_tokens": 62557150.0,
      "step": 30145
    },
    {
      "entropy": 5.7374327182769775,
      "epoch": 2.734941944847605,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042545336785080126,
      "loss": 5.1368,
      "mean_token_accuracy": 0.20294255018234253,
      "num_tokens": 62568291.0,
      "step": 30150
    },
    {
      "entropy": 5.6338053226470945,
      "epoch": 2.7353955007256894,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000425429307716902,
      "loss": 4.9289,
      "mean_token_accuracy": 0.21749451905488967,
      "num_tokens": 62579351.0,
      "step": 30155
    },
    {
      "entropy": 5.516463851928711,
      "epoch": 2.7358490566037736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042540524447208625,
      "loss": 4.8085,
      "mean_token_accuracy": 0.22037747055292128,
      "num_tokens": 62588614.0,
      "step": 30160
    },
    {
      "entropy": 5.62703948020935,
      "epoch": 2.736302612481858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042538117811685174,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21654695868492127,
      "num_tokens": 62598844.0,
      "step": 30165
    },
    {
      "entropy": 5.63707332611084,
      "epoch": 2.7367561683599417,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042535710865169616,
      "loss": 4.9567,
      "mean_token_accuracy": 0.20858405381441117,
      "num_tokens": 62608722.0,
      "step": 30170
    },
    {
      "entropy": 5.548538684844971,
      "epoch": 2.737209724238026,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004253330360771173,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21380723863840104,
      "num_tokens": 62618712.0,
      "step": 30175
    },
    {
      "entropy": 5.604543399810791,
      "epoch": 2.7376632801161103,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042530896039361284,
      "loss": 5.0565,
      "mean_token_accuracy": 0.20018149465322493,
      "num_tokens": 62629329.0,
      "step": 30180
    },
    {
      "entropy": 5.6446315288543705,
      "epoch": 2.7381168359941945,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042528488160168077,
      "loss": 4.986,
      "mean_token_accuracy": 0.20789050310850143,
      "num_tokens": 62639644.0,
      "step": 30185
    },
    {
      "entropy": 5.659021329879761,
      "epoch": 2.738570391872279,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042526079970181915,
      "loss": 5.008,
      "mean_token_accuracy": 0.20486198961734772,
      "num_tokens": 62650979.0,
      "step": 30190
    },
    {
      "entropy": 5.676254081726074,
      "epoch": 2.739023947750363,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004252367146945259,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20933170914649962,
      "num_tokens": 62661163.0,
      "step": 30195
    },
    {
      "entropy": 5.6342103481292725,
      "epoch": 2.739477503628447,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042521262658029905,
      "loss": 5.0285,
      "mean_token_accuracy": 0.20412251949310303,
      "num_tokens": 62672371.0,
      "step": 30200
    },
    {
      "entropy": 5.6026030540466305,
      "epoch": 2.739931059506531,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042518853535963673,
      "loss": 4.8968,
      "mean_token_accuracy": 0.21222202777862548,
      "num_tokens": 62682521.0,
      "step": 30205
    },
    {
      "entropy": 5.612781286239624,
      "epoch": 2.7403846153846154,
      "grad_norm": 1.484375,
      "learning_rate": 0.00042516444103303736,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23078083395957946,
      "num_tokens": 62692420.0,
      "step": 30210
    },
    {
      "entropy": 5.55757327079773,
      "epoch": 2.7408381712626997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042514034360099907,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21350128650665284,
      "num_tokens": 62702686.0,
      "step": 30215
    },
    {
      "entropy": 5.6044361114501955,
      "epoch": 2.7412917271407835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042511624306402015,
      "loss": 4.9528,
      "mean_token_accuracy": 0.21285638809204102,
      "num_tokens": 62713262.0,
      "step": 30220
    },
    {
      "entropy": 5.6452392578125,
      "epoch": 2.7417452830188678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004250921394225992,
      "loss": 4.9002,
      "mean_token_accuracy": 0.21436085402965546,
      "num_tokens": 62724460.0,
      "step": 30225
    },
    {
      "entropy": 5.6062828540802006,
      "epoch": 2.742198838896952,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004250680326772343,
      "loss": 4.9817,
      "mean_token_accuracy": 0.21282104700803756,
      "num_tokens": 62734190.0,
      "step": 30230
    },
    {
      "entropy": 5.617833518981934,
      "epoch": 2.7426523947750363,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004250439228284244,
      "loss": 4.9192,
      "mean_token_accuracy": 0.2149022102355957,
      "num_tokens": 62744323.0,
      "step": 30235
    },
    {
      "entropy": 5.622685670852661,
      "epoch": 2.7431059506531206,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042501980987666807,
      "loss": 4.9481,
      "mean_token_accuracy": 0.21487101167440414,
      "num_tokens": 62753868.0,
      "step": 30240
    },
    {
      "entropy": 5.627379846572876,
      "epoch": 2.743559506531205,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004249956938224636,
      "loss": 5.0314,
      "mean_token_accuracy": 0.21195608079433442,
      "num_tokens": 62763954.0,
      "step": 30245
    },
    {
      "entropy": 5.6044105052947994,
      "epoch": 2.7440130624092887,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042497157466631015,
      "loss": 4.918,
      "mean_token_accuracy": 0.22302944958209991,
      "num_tokens": 62774116.0,
      "step": 30250
    },
    {
      "entropy": 5.674649286270141,
      "epoch": 2.744466618287373,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042494745240870626,
      "loss": 4.99,
      "mean_token_accuracy": 0.2058614671230316,
      "num_tokens": 62784939.0,
      "step": 30255
    },
    {
      "entropy": 5.601968765258789,
      "epoch": 2.744920174165457,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042492332705015075,
      "loss": 4.9571,
      "mean_token_accuracy": 0.21411143690347673,
      "num_tokens": 62794988.0,
      "step": 30260
    },
    {
      "entropy": 5.625520038604736,
      "epoch": 2.7453737300435415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042489919859114273,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21706822514533997,
      "num_tokens": 62804984.0,
      "step": 30265
    },
    {
      "entropy": 5.643641519546509,
      "epoch": 2.7458272859216253,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042487506703218104,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20847680121660234,
      "num_tokens": 62816401.0,
      "step": 30270
    },
    {
      "entropy": 5.560655736923218,
      "epoch": 2.7462808417997095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004248509323737648,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22057805806398392,
      "num_tokens": 62826411.0,
      "step": 30275
    },
    {
      "entropy": 5.628651428222656,
      "epoch": 2.746734397677794,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004248267946163931,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20857589840888976,
      "num_tokens": 62837228.0,
      "step": 30280
    },
    {
      "entropy": 5.576494646072388,
      "epoch": 2.747187953555878,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042480265376056506,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22177619487047195,
      "num_tokens": 62847852.0,
      "step": 30285
    },
    {
      "entropy": 5.613132286071777,
      "epoch": 2.7476415094339623,
      "grad_norm": 1.125,
      "learning_rate": 0.00042477850980678,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20616811960935594,
      "num_tokens": 62859516.0,
      "step": 30290
    },
    {
      "entropy": 5.62861328125,
      "epoch": 2.7480950653120466,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004247543627555372,
      "loss": 4.9579,
      "mean_token_accuracy": 0.2111853152513504,
      "num_tokens": 62869757.0,
      "step": 30295
    },
    {
      "entropy": 5.613959884643554,
      "epoch": 2.748548621190131,
      "grad_norm": 1.296875,
      "learning_rate": 0.000424730212607336,
      "loss": 5.0148,
      "mean_token_accuracy": 0.2095622733235359,
      "num_tokens": 62879618.0,
      "step": 30300
    },
    {
      "entropy": 5.624938344955444,
      "epoch": 2.7490021770682147,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004247060593626758,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20830693542957307,
      "num_tokens": 62890491.0,
      "step": 30305
    },
    {
      "entropy": 5.697379207611084,
      "epoch": 2.749455732946299,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004246819030220562,
      "loss": 4.9849,
      "mean_token_accuracy": 0.22261755764484406,
      "num_tokens": 62899954.0,
      "step": 30310
    },
    {
      "entropy": 5.635549783706665,
      "epoch": 2.7499092888243832,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004246577435859767,
      "loss": 4.9793,
      "mean_token_accuracy": 0.2149518683552742,
      "num_tokens": 62911857.0,
      "step": 30315
    },
    {
      "entropy": 5.574631595611573,
      "epoch": 2.7503628447024675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004246335810549369,
      "loss": 4.9097,
      "mean_token_accuracy": 0.2194608524441719,
      "num_tokens": 62922023.0,
      "step": 30320
    },
    {
      "entropy": 5.647210216522216,
      "epoch": 2.7508164005805513,
      "grad_norm": 1.125,
      "learning_rate": 0.00042460941542943656,
      "loss": 4.9867,
      "mean_token_accuracy": 0.21846277862787247,
      "num_tokens": 62932845.0,
      "step": 30325
    },
    {
      "entropy": 5.632871198654175,
      "epoch": 2.7512699564586356,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004245852467099753,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2101821258664131,
      "num_tokens": 62942881.0,
      "step": 30330
    },
    {
      "entropy": 5.5770667552947994,
      "epoch": 2.75172351233672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004245610748970531,
      "loss": 4.8896,
      "mean_token_accuracy": 0.21483128517866135,
      "num_tokens": 62953957.0,
      "step": 30335
    },
    {
      "entropy": 5.554941320419312,
      "epoch": 2.752177068214804,
      "grad_norm": 1.25,
      "learning_rate": 0.00042453689999116974,
      "loss": 4.8983,
      "mean_token_accuracy": 0.21520988047122955,
      "num_tokens": 62964803.0,
      "step": 30340
    },
    {
      "entropy": 5.593675518035889,
      "epoch": 2.7526306240928884,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004245127219928252,
      "loss": 4.8991,
      "mean_token_accuracy": 0.21524002999067307,
      "num_tokens": 62975477.0,
      "step": 30345
    },
    {
      "entropy": 5.598050594329834,
      "epoch": 2.7530841799709727,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042448854090251944,
      "loss": 4.9608,
      "mean_token_accuracy": 0.2135425627231598,
      "num_tokens": 62985430.0,
      "step": 30350
    },
    {
      "entropy": 5.601573753356933,
      "epoch": 2.7535377358490565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042446435672075254,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2169000968337059,
      "num_tokens": 62997493.0,
      "step": 30355
    },
    {
      "entropy": 5.685353231430054,
      "epoch": 2.7539912917271407,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042444016944802467,
      "loss": 5.0242,
      "mean_token_accuracy": 0.2181790977716446,
      "num_tokens": 63007573.0,
      "step": 30360
    },
    {
      "entropy": 5.66036901473999,
      "epoch": 2.754444847605225,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042441597908483603,
      "loss": 4.9817,
      "mean_token_accuracy": 0.21657835245132445,
      "num_tokens": 63018372.0,
      "step": 30365
    },
    {
      "entropy": 5.59838342666626,
      "epoch": 2.7548984034833093,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004243917856316869,
      "loss": 4.9217,
      "mean_token_accuracy": 0.21629077792167664,
      "num_tokens": 63028632.0,
      "step": 30370
    },
    {
      "entropy": 5.6141571521759035,
      "epoch": 2.755351959361393,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042436758908907747,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2099437564611435,
      "num_tokens": 63038510.0,
      "step": 30375
    },
    {
      "entropy": 5.715930843353272,
      "epoch": 2.7558055152394774,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042434338945750833,
      "loss": 5.1083,
      "mean_token_accuracy": 0.19410379528999328,
      "num_tokens": 63049375.0,
      "step": 30380
    },
    {
      "entropy": 5.65672287940979,
      "epoch": 2.7562590711175616,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042431918673747974,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21619293242692947,
      "num_tokens": 63059422.0,
      "step": 30385
    },
    {
      "entropy": 5.6731163501739506,
      "epoch": 2.756712626995646,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042429498092949245,
      "loss": 5.0149,
      "mean_token_accuracy": 0.2059454470872879,
      "num_tokens": 63071101.0,
      "step": 30390
    },
    {
      "entropy": 5.546786546707153,
      "epoch": 2.75716618287373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004242707720340468,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22282183915376663,
      "num_tokens": 63081104.0,
      "step": 30395
    },
    {
      "entropy": 5.600458192825317,
      "epoch": 2.7576197387518144,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004242465600516436,
      "loss": 4.9681,
      "mean_token_accuracy": 0.2115418627858162,
      "num_tokens": 63091123.0,
      "step": 30400
    },
    {
      "entropy": 5.583108234405517,
      "epoch": 2.7580732946298983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042422234498278353,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21760235726833344,
      "num_tokens": 63101541.0,
      "step": 30405
    },
    {
      "entropy": 5.652334785461425,
      "epoch": 2.7585268505079825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042419812682796727,
      "loss": 5.0279,
      "mean_token_accuracy": 0.20402513295412064,
      "num_tokens": 63112012.0,
      "step": 30410
    },
    {
      "entropy": 5.6674047946929935,
      "epoch": 2.758980406386067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004241739055876957,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21688816398382188,
      "num_tokens": 63122021.0,
      "step": 30415
    },
    {
      "entropy": 5.68640022277832,
      "epoch": 2.759433962264151,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042414968126246985,
      "loss": 5.0485,
      "mean_token_accuracy": 0.2107762411236763,
      "num_tokens": 63132286.0,
      "step": 30420
    },
    {
      "entropy": 5.572261667251587,
      "epoch": 2.759887518142235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042412545385279055,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2180559977889061,
      "num_tokens": 63142159.0,
      "step": 30425
    },
    {
      "entropy": 5.56485390663147,
      "epoch": 2.760341074020319,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042410122335915883,
      "loss": 4.8849,
      "mean_token_accuracy": 0.217661452293396,
      "num_tokens": 63152225.0,
      "step": 30430
    },
    {
      "entropy": 5.591285991668701,
      "epoch": 2.7607946298984034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004240769897820758,
      "loss": 4.9359,
      "mean_token_accuracy": 0.2141361877322197,
      "num_tokens": 63162522.0,
      "step": 30435
    },
    {
      "entropy": 5.6189690113067625,
      "epoch": 2.7612481857764877,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042405275312204266,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2210882067680359,
      "num_tokens": 63172094.0,
      "step": 30440
    },
    {
      "entropy": 5.633256101608277,
      "epoch": 2.761701741654572,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042402851337956066,
      "loss": 4.9588,
      "mean_token_accuracy": 0.21410634964704514,
      "num_tokens": 63182122.0,
      "step": 30445
    },
    {
      "entropy": 5.582202386856079,
      "epoch": 2.762155297532656,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042400427055513094,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2122228890657425,
      "num_tokens": 63193174.0,
      "step": 30450
    },
    {
      "entropy": 5.609793853759766,
      "epoch": 2.7626088534107405,
      "grad_norm": 1.171875,
      "learning_rate": 0.000423980024649255,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22054361402988434,
      "num_tokens": 63203698.0,
      "step": 30455
    },
    {
      "entropy": 5.517379808425903,
      "epoch": 2.7630624092888243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004239557756624342,
      "loss": 4.8349,
      "mean_token_accuracy": 0.21590106785297394,
      "num_tokens": 63213343.0,
      "step": 30460
    },
    {
      "entropy": 5.652347421646118,
      "epoch": 2.7635159651669086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042393152359516994,
      "loss": 5.0091,
      "mean_token_accuracy": 0.21349549740552903,
      "num_tokens": 63223962.0,
      "step": 30465
    },
    {
      "entropy": 5.594416332244873,
      "epoch": 2.763969521044993,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042390726844796384,
      "loss": 4.904,
      "mean_token_accuracy": 0.22045750319957733,
      "num_tokens": 63234279.0,
      "step": 30470
    },
    {
      "entropy": 5.644147968292236,
      "epoch": 2.7644230769230766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004238830102213174,
      "loss": 5.0045,
      "mean_token_accuracy": 0.2150461569428444,
      "num_tokens": 63245567.0,
      "step": 30475
    },
    {
      "entropy": 5.6111259937286375,
      "epoch": 2.764876632801161,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004238587489157325,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2222388952970505,
      "num_tokens": 63256227.0,
      "step": 30480
    },
    {
      "entropy": 5.554749441146851,
      "epoch": 2.765330188679245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004238344845317107,
      "loss": 4.85,
      "mean_token_accuracy": 0.217688125371933,
      "num_tokens": 63266618.0,
      "step": 30485
    },
    {
      "entropy": 5.562039613723755,
      "epoch": 2.7657837445573294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004238102170697539,
      "loss": 4.8811,
      "mean_token_accuracy": 0.21864323318004608,
      "num_tokens": 63277130.0,
      "step": 30490
    },
    {
      "entropy": 5.536762666702271,
      "epoch": 2.7662373004354137,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004237859465303637,
      "loss": 4.8083,
      "mean_token_accuracy": 0.22327447384595872,
      "num_tokens": 63287351.0,
      "step": 30495
    },
    {
      "entropy": 5.531674861907959,
      "epoch": 2.766690856313498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042376167291404234,
      "loss": 4.8937,
      "mean_token_accuracy": 0.21843965947628022,
      "num_tokens": 63298573.0,
      "step": 30500
    },
    {
      "entropy": 5.598010063171387,
      "epoch": 2.7671444121915822,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042373739622129166,
      "loss": 4.9423,
      "mean_token_accuracy": 0.20653199702501296,
      "num_tokens": 63310740.0,
      "step": 30505
    },
    {
      "entropy": 5.5904233932495115,
      "epoch": 2.767597968069666,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042371311645261376,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21120081692934037,
      "num_tokens": 63321323.0,
      "step": 30510
    },
    {
      "entropy": 5.614881372451782,
      "epoch": 2.7680515239477503,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004236888336085107,
      "loss": 4.923,
      "mean_token_accuracy": 0.21754363179206848,
      "num_tokens": 63332322.0,
      "step": 30515
    },
    {
      "entropy": 5.57806167602539,
      "epoch": 2.7685050798258346,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004236645476894846,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22277876883745193,
      "num_tokens": 63342204.0,
      "step": 30520
    },
    {
      "entropy": 5.527768850326538,
      "epoch": 2.768958635703919,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004236402586960378,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21968706101179122,
      "num_tokens": 63352460.0,
      "step": 30525
    },
    {
      "entropy": 5.522051906585693,
      "epoch": 2.7694121915820027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004236159666286726,
      "loss": 4.8813,
      "mean_token_accuracy": 0.21861146092414857,
      "num_tokens": 63362474.0,
      "step": 30530
    },
    {
      "entropy": 5.640454578399658,
      "epoch": 2.769865747460087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042359167148789135,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22466543614864348,
      "num_tokens": 63372537.0,
      "step": 30535
    },
    {
      "entropy": 5.555916166305542,
      "epoch": 2.7703193033381712,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004235673732741964,
      "loss": 4.8276,
      "mean_token_accuracy": 0.2260143905878067,
      "num_tokens": 63382985.0,
      "step": 30540
    },
    {
      "entropy": 5.570925855636597,
      "epoch": 2.7707728592162555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042354307198809037,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21783771067857743,
      "num_tokens": 63393424.0,
      "step": 30545
    },
    {
      "entropy": 5.533125305175782,
      "epoch": 2.7712264150943398,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004235187676300757,
      "loss": 4.8299,
      "mean_token_accuracy": 0.22464577108621597,
      "num_tokens": 63403160.0,
      "step": 30550
    },
    {
      "entropy": 5.636246109008789,
      "epoch": 2.771679970972424,
      "grad_norm": 1.125,
      "learning_rate": 0.0004234944602006552,
      "loss": 4.9583,
      "mean_token_accuracy": 0.2132783517241478,
      "num_tokens": 63413001.0,
      "step": 30555
    },
    {
      "entropy": 5.625688600540161,
      "epoch": 2.772133526850508,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042347014970033124,
      "loss": 4.9076,
      "mean_token_accuracy": 0.2218951016664505,
      "num_tokens": 63422735.0,
      "step": 30560
    },
    {
      "entropy": 5.531408596038818,
      "epoch": 2.772587082728592,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004234458361296069,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21520378440618515,
      "num_tokens": 63432566.0,
      "step": 30565
    },
    {
      "entropy": 5.589576768875122,
      "epoch": 2.7730406386066764,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042342151948898476,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21632362753152848,
      "num_tokens": 63442412.0,
      "step": 30570
    },
    {
      "entropy": 5.59679684638977,
      "epoch": 2.7734941944847606,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042339719977896784,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21769312173128127,
      "num_tokens": 63453018.0,
      "step": 30575
    },
    {
      "entropy": 5.662491655349731,
      "epoch": 2.7739477503628445,
      "grad_norm": 1.1875,
      "learning_rate": 0.000423372877000059,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21131200045347215,
      "num_tokens": 63464286.0,
      "step": 30580
    },
    {
      "entropy": 5.55959038734436,
      "epoch": 2.7744013062409287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042334855115276117,
      "loss": 4.9431,
      "mean_token_accuracy": 0.213600817322731,
      "num_tokens": 63475138.0,
      "step": 30585
    },
    {
      "entropy": 5.623905515670776,
      "epoch": 2.774854862119013,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042332422223757756,
      "loss": 4.999,
      "mean_token_accuracy": 0.21151673048734665,
      "num_tokens": 63486104.0,
      "step": 30590
    },
    {
      "entropy": 5.556403493881225,
      "epoch": 2.7753084179970973,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042329989025501123,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2150347650051117,
      "num_tokens": 63496490.0,
      "step": 30595
    },
    {
      "entropy": 5.569782733917236,
      "epoch": 2.7757619738751815,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042327555520556536,
      "loss": 4.8906,
      "mean_token_accuracy": 0.21169639229774476,
      "num_tokens": 63506562.0,
      "step": 30600
    },
    {
      "entropy": 5.627491617202759,
      "epoch": 2.776215529753266,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042325121708974324,
      "loss": 4.9272,
      "mean_token_accuracy": 0.2239815905690193,
      "num_tokens": 63516641.0,
      "step": 30605
    },
    {
      "entropy": 5.622805070877075,
      "epoch": 2.7766690856313496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004232268759080482,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21638973504304887,
      "num_tokens": 63526864.0,
      "step": 30610
    },
    {
      "entropy": 5.586530923843384,
      "epoch": 2.777122641509434,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042320253166098354,
      "loss": 5.007,
      "mean_token_accuracy": 0.20974934697151185,
      "num_tokens": 63537160.0,
      "step": 30615
    },
    {
      "entropy": 5.639689874649048,
      "epoch": 2.777576197387518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042317818434905275,
      "loss": 4.8974,
      "mean_token_accuracy": 0.21598411500453948,
      "num_tokens": 63547485.0,
      "step": 30620
    },
    {
      "entropy": 5.61621241569519,
      "epoch": 2.7780297532656024,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004231538339727594,
      "loss": 4.8928,
      "mean_token_accuracy": 0.21545920968055726,
      "num_tokens": 63559369.0,
      "step": 30625
    },
    {
      "entropy": 5.693338298797608,
      "epoch": 2.7784833091436862,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042312948053260704,
      "loss": 5.092,
      "mean_token_accuracy": 0.20301660895347595,
      "num_tokens": 63570033.0,
      "step": 30630
    },
    {
      "entropy": 5.5835034370422365,
      "epoch": 2.7789368650217705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004231051240290992,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21866330057382583,
      "num_tokens": 63581274.0,
      "step": 30635
    },
    {
      "entropy": 5.638933849334717,
      "epoch": 2.7793904208998548,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042308076446273963,
      "loss": 4.9321,
      "mean_token_accuracy": 0.21271978169679642,
      "num_tokens": 63590903.0,
      "step": 30640
    },
    {
      "entropy": 5.56838641166687,
      "epoch": 2.779843976777939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042305640183403214,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20841511189937592,
      "num_tokens": 63600982.0,
      "step": 30645
    },
    {
      "entropy": 5.511862945556641,
      "epoch": 2.7802975326560233,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042303203614348057,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2130862757563591,
      "num_tokens": 63612411.0,
      "step": 30650
    },
    {
      "entropy": 5.567126846313476,
      "epoch": 2.7807510885341076,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004230076673915888,
      "loss": 4.8886,
      "mean_token_accuracy": 0.22152767181396485,
      "num_tokens": 63622128.0,
      "step": 30655
    },
    {
      "entropy": 5.615833806991577,
      "epoch": 2.781204644412192,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042298329557886074,
      "loss": 4.9554,
      "mean_token_accuracy": 0.21670996397733688,
      "num_tokens": 63632047.0,
      "step": 30660
    },
    {
      "entropy": 5.653778076171875,
      "epoch": 2.7816582002902757,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004229589207058003,
      "loss": 4.9648,
      "mean_token_accuracy": 0.21392653584480287,
      "num_tokens": 63642490.0,
      "step": 30665
    },
    {
      "entropy": 5.586833047866821,
      "epoch": 2.78211175616836,
      "grad_norm": 1.125,
      "learning_rate": 0.00042293454277291177,
      "loss": 4.9254,
      "mean_token_accuracy": 0.2138791263103485,
      "num_tokens": 63653206.0,
      "step": 30670
    },
    {
      "entropy": 5.529301452636719,
      "epoch": 2.782565312046444,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042291016178069916,
      "loss": 4.924,
      "mean_token_accuracy": 0.21557315289974213,
      "num_tokens": 63663427.0,
      "step": 30675
    },
    {
      "entropy": 5.534003829956054,
      "epoch": 2.7830188679245285,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004228857777296668,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21934759020805358,
      "num_tokens": 63674269.0,
      "step": 30680
    },
    {
      "entropy": 5.62264838218689,
      "epoch": 2.7834724238026123,
      "grad_norm": 1.125,
      "learning_rate": 0.0004228613906203188,
      "loss": 4.8938,
      "mean_token_accuracy": 0.21602804809808732,
      "num_tokens": 63683989.0,
      "step": 30685
    },
    {
      "entropy": 5.614567136764526,
      "epoch": 2.7839259796806965,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042283700045315955,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21105228066444398,
      "num_tokens": 63694540.0,
      "step": 30690
    },
    {
      "entropy": 5.589615201950073,
      "epoch": 2.784379535558781,
      "grad_norm": 1.125,
      "learning_rate": 0.00042281260722869355,
      "loss": 4.899,
      "mean_token_accuracy": 0.21366845369338988,
      "num_tokens": 63705149.0,
      "step": 30695
    },
    {
      "entropy": 5.5503425121307375,
      "epoch": 2.784833091436865,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042278821094742495,
      "loss": 4.881,
      "mean_token_accuracy": 0.2166728511452675,
      "num_tokens": 63714061.0,
      "step": 30700
    },
    {
      "entropy": 5.6025643825531,
      "epoch": 2.7852866473149493,
      "grad_norm": 1.125,
      "learning_rate": 0.00042276381160985874,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21572295576334,
      "num_tokens": 63724732.0,
      "step": 30705
    },
    {
      "entropy": 5.596027994155884,
      "epoch": 2.7857402031930336,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004227394092164991,
      "loss": 4.9727,
      "mean_token_accuracy": 0.21107783615589143,
      "num_tokens": 63734942.0,
      "step": 30710
    },
    {
      "entropy": 5.672230577468872,
      "epoch": 2.7861937590711174,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004227150037678509,
      "loss": 5.0425,
      "mean_token_accuracy": 0.21797527968883515,
      "num_tokens": 63745326.0,
      "step": 30715
    },
    {
      "entropy": 5.6225536346435545,
      "epoch": 2.7866473149492017,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004226905952644188,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2138058990240097,
      "num_tokens": 63755355.0,
      "step": 30720
    },
    {
      "entropy": 5.624581813812256,
      "epoch": 2.787100870827286,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042266618370670743,
      "loss": 4.9212,
      "mean_token_accuracy": 0.212918521463871,
      "num_tokens": 63765112.0,
      "step": 30725
    },
    {
      "entropy": 5.571261072158814,
      "epoch": 2.7875544267053702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042264176909522183,
      "loss": 4.8748,
      "mean_token_accuracy": 0.21993715316057205,
      "num_tokens": 63774680.0,
      "step": 30730
    },
    {
      "entropy": 5.621033334732056,
      "epoch": 2.788007982583454,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004226173514304669,
      "loss": 4.9844,
      "mean_token_accuracy": 0.21756456643342972,
      "num_tokens": 63784939.0,
      "step": 30735
    },
    {
      "entropy": 5.629879808425903,
      "epoch": 2.7884615384615383,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004225929307129474,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20408451408147812,
      "num_tokens": 63795836.0,
      "step": 30740
    },
    {
      "entropy": 5.574771404266357,
      "epoch": 2.7889150943396226,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042256850694316844,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2149980679154396,
      "num_tokens": 63805607.0,
      "step": 30745
    },
    {
      "entropy": 5.53808741569519,
      "epoch": 2.789368650217707,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004225440801216353,
      "loss": 4.888,
      "mean_token_accuracy": 0.21340310722589492,
      "num_tokens": 63815081.0,
      "step": 30750
    },
    {
      "entropy": 5.556040334701538,
      "epoch": 2.789822206095791,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004225196502488529,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2129345253109932,
      "num_tokens": 63825932.0,
      "step": 30755
    },
    {
      "entropy": 5.579881572723389,
      "epoch": 2.7902757619738754,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042249521732532657,
      "loss": 4.8605,
      "mean_token_accuracy": 0.21878161281347275,
      "num_tokens": 63835653.0,
      "step": 30760
    },
    {
      "entropy": 5.674952793121338,
      "epoch": 2.790729317851959,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004224707813515615,
      "loss": 5.0583,
      "mean_token_accuracy": 0.20467241555452348,
      "num_tokens": 63847657.0,
      "step": 30765
    },
    {
      "entropy": 5.666828441619873,
      "epoch": 2.7911828737300435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042244634232806324,
      "loss": 4.9602,
      "mean_token_accuracy": 0.2172686532139778,
      "num_tokens": 63858476.0,
      "step": 30770
    },
    {
      "entropy": 5.556519317626953,
      "epoch": 2.7916364296081277,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042242190025533694,
      "loss": 4.8279,
      "mean_token_accuracy": 0.22101556658744811,
      "num_tokens": 63867970.0,
      "step": 30775
    },
    {
      "entropy": 5.604297065734864,
      "epoch": 2.792089985486212,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004223974551338881,
      "loss": 4.9141,
      "mean_token_accuracy": 0.22202686071395875,
      "num_tokens": 63878294.0,
      "step": 30780
    },
    {
      "entropy": 5.545058012008667,
      "epoch": 2.792543541364296,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004223730069642224,
      "loss": 4.914,
      "mean_token_accuracy": 0.2130719855427742,
      "num_tokens": 63889315.0,
      "step": 30785
    },
    {
      "entropy": 5.520653200149536,
      "epoch": 2.79299709724238,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004223485557468454,
      "loss": 4.8923,
      "mean_token_accuracy": 0.21057825535535812,
      "num_tokens": 63899923.0,
      "step": 30790
    },
    {
      "entropy": 5.573180055618286,
      "epoch": 2.7934506531204644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042232410148226277,
      "loss": 4.8777,
      "mean_token_accuracy": 0.21802876740694047,
      "num_tokens": 63909813.0,
      "step": 30795
    },
    {
      "entropy": 5.607568979263306,
      "epoch": 2.7939042089985486,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042229964417098007,
      "loss": 4.9196,
      "mean_token_accuracy": 0.21219711899757385,
      "num_tokens": 63920931.0,
      "step": 30800
    },
    {
      "entropy": 5.55732946395874,
      "epoch": 2.794357764876633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042227518381350316,
      "loss": 4.868,
      "mean_token_accuracy": 0.22100864499807357,
      "num_tokens": 63930799.0,
      "step": 30805
    },
    {
      "entropy": 5.66492600440979,
      "epoch": 2.794811320754717,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042225072041033804,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2090522140264511,
      "num_tokens": 63941763.0,
      "step": 30810
    },
    {
      "entropy": 5.6471329689025875,
      "epoch": 2.7952648766328014,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042222625396199043,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20792181491851808,
      "num_tokens": 63952775.0,
      "step": 30815
    },
    {
      "entropy": 5.630888032913208,
      "epoch": 2.7957184325108853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004222017844689663,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21161658465862274,
      "num_tokens": 63964223.0,
      "step": 30820
    },
    {
      "entropy": 5.6678611755371096,
      "epoch": 2.7961719883889695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004221773119317719,
      "loss": 4.9954,
      "mean_token_accuracy": 0.21187895834445952,
      "num_tokens": 63976413.0,
      "step": 30825
    },
    {
      "entropy": 5.673243284225464,
      "epoch": 2.796625544267054,
      "grad_norm": 1.375,
      "learning_rate": 0.00042215283635091313,
      "loss": 4.96,
      "mean_token_accuracy": 0.21708431094884872,
      "num_tokens": 63986578.0,
      "step": 30830
    },
    {
      "entropy": 5.615045881271362,
      "epoch": 2.7970791001451376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004221283577268962,
      "loss": 4.9383,
      "mean_token_accuracy": 0.21961487084627151,
      "num_tokens": 63996856.0,
      "step": 30835
    },
    {
      "entropy": 5.556055927276612,
      "epoch": 2.797532656023222,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042210387606022724,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21939728856086732,
      "num_tokens": 64007700.0,
      "step": 30840
    },
    {
      "entropy": 5.660103797912598,
      "epoch": 2.797986211901306,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004220793913514128,
      "loss": 5.0529,
      "mean_token_accuracy": 0.20538849085569383,
      "num_tokens": 64018771.0,
      "step": 30845
    },
    {
      "entropy": 5.623592329025269,
      "epoch": 2.7984397677793904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004220549036009589,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2099561020731926,
      "num_tokens": 64029381.0,
      "step": 30850
    },
    {
      "entropy": 5.5213452816009525,
      "epoch": 2.7988933236574747,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042203041280937224,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21454301476478577,
      "num_tokens": 64039596.0,
      "step": 30855
    },
    {
      "entropy": 5.6389189720153805,
      "epoch": 2.799346879535559,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004220059189771592,
      "loss": 4.9549,
      "mean_token_accuracy": 0.22159379422664643,
      "num_tokens": 64049123.0,
      "step": 30860
    },
    {
      "entropy": 5.632424592971802,
      "epoch": 2.799800435413643,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042198142210482623,
      "loss": 4.9803,
      "mean_token_accuracy": 0.2122730866074562,
      "num_tokens": 64059426.0,
      "step": 30865
    },
    {
      "entropy": 5.575161647796631,
      "epoch": 2.800253991291727,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042195692219287997,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22846995741128923,
      "num_tokens": 64069227.0,
      "step": 30870
    },
    {
      "entropy": 5.630750513076782,
      "epoch": 2.8007075471698113,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004219324192418272,
      "loss": 4.9382,
      "mean_token_accuracy": 0.22248561233282088,
      "num_tokens": 64078482.0,
      "step": 30875
    },
    {
      "entropy": 5.5800941467285154,
      "epoch": 2.8011611030478956,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042190791325217447,
      "loss": 4.9028,
      "mean_token_accuracy": 0.216207854449749,
      "num_tokens": 64088664.0,
      "step": 30880
    },
    {
      "entropy": 5.6033862113952635,
      "epoch": 2.80161465892598,
      "grad_norm": 1.25,
      "learning_rate": 0.0004218834042244287,
      "loss": 4.8551,
      "mean_token_accuracy": 0.22169570922851561,
      "num_tokens": 64099316.0,
      "step": 30885
    },
    {
      "entropy": 5.567199945449829,
      "epoch": 2.8020682148040637,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042185889215909665,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21490092873573302,
      "num_tokens": 64109786.0,
      "step": 30890
    },
    {
      "entropy": 5.608392429351807,
      "epoch": 2.802521770682148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042183437705668535,
      "loss": 4.9045,
      "mean_token_accuracy": 0.22288060039281846,
      "num_tokens": 64120771.0,
      "step": 30895
    },
    {
      "entropy": 5.644888544082642,
      "epoch": 2.802975326560232,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042180985891770176,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21018804460763932,
      "num_tokens": 64130833.0,
      "step": 30900
    },
    {
      "entropy": 5.560075283050537,
      "epoch": 2.8034288824383164,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004217853377426528,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22247102856636047,
      "num_tokens": 64139672.0,
      "step": 30905
    },
    {
      "entropy": 5.7054849624633786,
      "epoch": 2.8038824383164007,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004217608135320456,
      "loss": 5.0529,
      "mean_token_accuracy": 0.21037883460521697,
      "num_tokens": 64150383.0,
      "step": 30910
    },
    {
      "entropy": 5.574859619140625,
      "epoch": 2.804335994194485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004217362862863875,
      "loss": 4.8606,
      "mean_token_accuracy": 0.22063290029764177,
      "num_tokens": 64160786.0,
      "step": 30915
    },
    {
      "entropy": 5.632054996490479,
      "epoch": 2.804789550072569,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042171175600618546,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21001256257295609,
      "num_tokens": 64172057.0,
      "step": 30920
    },
    {
      "entropy": 5.593688488006592,
      "epoch": 2.805243105950653,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042168722269194704,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21802769750356674,
      "num_tokens": 64182599.0,
      "step": 30925
    },
    {
      "entropy": 5.61622953414917,
      "epoch": 2.8056966618287373,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004216626863441794,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2088227942585945,
      "num_tokens": 64193356.0,
      "step": 30930
    },
    {
      "entropy": 5.676929092407226,
      "epoch": 2.8061502177068216,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042163814696339014,
      "loss": 5.0388,
      "mean_token_accuracy": 0.2096481665968895,
      "num_tokens": 64203625.0,
      "step": 30935
    },
    {
      "entropy": 5.618952989578247,
      "epoch": 2.8066037735849054,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004216136045500866,
      "loss": 5.0406,
      "mean_token_accuracy": 0.21401924192905425,
      "num_tokens": 64214742.0,
      "step": 30940
    },
    {
      "entropy": 5.628974580764771,
      "epoch": 2.8070573294629897,
      "grad_norm": 1.25,
      "learning_rate": 0.0004215890591047763,
      "loss": 4.9425,
      "mean_token_accuracy": 0.21468819975852965,
      "num_tokens": 64223811.0,
      "step": 30945
    },
    {
      "entropy": 5.607697010040283,
      "epoch": 2.807510885341074,
      "grad_norm": 1.125,
      "learning_rate": 0.00042156451062796686,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2281048372387886,
      "num_tokens": 64234854.0,
      "step": 30950
    },
    {
      "entropy": 5.597255277633667,
      "epoch": 2.8079644412191582,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042153995912016606,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21392620652914046,
      "num_tokens": 64245914.0,
      "step": 30955
    },
    {
      "entropy": 5.595199203491211,
      "epoch": 2.8084179970972425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004215154045818816,
      "loss": 5.0029,
      "mean_token_accuracy": 0.21000120639801026,
      "num_tokens": 64256217.0,
      "step": 30960
    },
    {
      "entropy": 5.631856536865234,
      "epoch": 2.8088715529753268,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004214908470136212,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20288413763046265,
      "num_tokens": 64267211.0,
      "step": 30965
    },
    {
      "entropy": 5.686559963226318,
      "epoch": 2.8093251088534106,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004214662864158927,
      "loss": 4.9735,
      "mean_token_accuracy": 0.21689907610416412,
      "num_tokens": 64277770.0,
      "step": 30970
    },
    {
      "entropy": 5.585747337341308,
      "epoch": 2.809778664731495,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042144172278920413,
      "loss": 4.8615,
      "mean_token_accuracy": 0.2203228995203972,
      "num_tokens": 64287973.0,
      "step": 30975
    },
    {
      "entropy": 5.566807556152344,
      "epoch": 2.810232220609579,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004214171561340634,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21033688634634018,
      "num_tokens": 64297898.0,
      "step": 30980
    },
    {
      "entropy": 5.590597438812256,
      "epoch": 2.8106857764876634,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042139258645097863,
      "loss": 4.8114,
      "mean_token_accuracy": 0.22951457649469376,
      "num_tokens": 64308114.0,
      "step": 30985
    },
    {
      "entropy": 5.567050170898438,
      "epoch": 2.811139332365747,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042136801374045775,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21565261632204055,
      "num_tokens": 64319586.0,
      "step": 30990
    },
    {
      "entropy": 5.6042194843292235,
      "epoch": 2.8115928882438315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004213434380030091,
      "loss": 4.9978,
      "mean_token_accuracy": 0.21070887446403502,
      "num_tokens": 64330612.0,
      "step": 30995
    },
    {
      "entropy": 5.6452186584472654,
      "epoch": 2.8120464441219157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004213188592391409,
      "loss": 5.015,
      "mean_token_accuracy": 0.21246811747550964,
      "num_tokens": 64341934.0,
      "step": 31000
    },
    {
      "entropy": 5.621395111083984,
      "epoch": 2.8125,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004212942774493614,
      "loss": 4.9173,
      "mean_token_accuracy": 0.21760687381029128,
      "num_tokens": 64352166.0,
      "step": 31005
    },
    {
      "entropy": 5.589327812194824,
      "epoch": 2.8129535558780843,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042126969263417885,
      "loss": 4.964,
      "mean_token_accuracy": 0.2079501509666443,
      "num_tokens": 64361995.0,
      "step": 31010
    },
    {
      "entropy": 5.610994625091553,
      "epoch": 2.8134071117561685,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042124510479410183,
      "loss": 4.8934,
      "mean_token_accuracy": 0.21078013777732849,
      "num_tokens": 64372336.0,
      "step": 31015
    },
    {
      "entropy": 5.679530954360962,
      "epoch": 2.813860667634253,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042122051392963885,
      "loss": 4.9445,
      "mean_token_accuracy": 0.22433409094810486,
      "num_tokens": 64382309.0,
      "step": 31020
    },
    {
      "entropy": 5.62575511932373,
      "epoch": 2.8143142235123366,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004211959200412984,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2136935696005821,
      "num_tokens": 64393205.0,
      "step": 31025
    },
    {
      "entropy": 5.560979747772217,
      "epoch": 2.814767779390421,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004211713231295889,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21102684140205383,
      "num_tokens": 64403068.0,
      "step": 31030
    },
    {
      "entropy": 5.627737188339234,
      "epoch": 2.815221335268505,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004211467231950194,
      "loss": 4.9375,
      "mean_token_accuracy": 0.21938664764165877,
      "num_tokens": 64414296.0,
      "step": 31035
    },
    {
      "entropy": 5.596076774597168,
      "epoch": 2.8156748911465894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042112212023809824,
      "loss": 4.9661,
      "mean_token_accuracy": 0.20868293195962906,
      "num_tokens": 64424748.0,
      "step": 31040
    },
    {
      "entropy": 5.631599569320679,
      "epoch": 2.8161284470246732,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004210975142593345,
      "loss": 4.8872,
      "mean_token_accuracy": 0.2139439269900322,
      "num_tokens": 64434723.0,
      "step": 31045
    },
    {
      "entropy": 5.686302137374878,
      "epoch": 2.8165820029027575,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004210729052592369,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20711847096681596,
      "num_tokens": 64445864.0,
      "step": 31050
    },
    {
      "entropy": 5.538408374786377,
      "epoch": 2.8170355587808418,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004210482932383144,
      "loss": 4.8721,
      "mean_token_accuracy": 0.21520453244447707,
      "num_tokens": 64456704.0,
      "step": 31055
    },
    {
      "entropy": 5.518161869049072,
      "epoch": 2.817489114658926,
      "grad_norm": 1.234375,
      "learning_rate": 0.000421023678197076,
      "loss": 4.9433,
      "mean_token_accuracy": 0.219936366379261,
      "num_tokens": 64466870.0,
      "step": 31060
    },
    {
      "entropy": 5.6191342830657955,
      "epoch": 2.8179426705370103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004209990601360307,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21801260113716125,
      "num_tokens": 64477069.0,
      "step": 31065
    },
    {
      "entropy": 5.670992422103882,
      "epoch": 2.8183962264150946,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042097443905568765,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20735147446393967,
      "num_tokens": 64487165.0,
      "step": 31070
    },
    {
      "entropy": 5.566104507446289,
      "epoch": 2.8188497822931784,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042094981495655596,
      "loss": 4.8265,
      "mean_token_accuracy": 0.2197411134839058,
      "num_tokens": 64497977.0,
      "step": 31075
    },
    {
      "entropy": 5.6225471019744875,
      "epoch": 2.8193033381712627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000420925187839145,
      "loss": 4.966,
      "mean_token_accuracy": 0.21211251765489578,
      "num_tokens": 64508767.0,
      "step": 31080
    },
    {
      "entropy": 5.55676589012146,
      "epoch": 2.819756894049347,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042090055770396383,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21581576019525528,
      "num_tokens": 64519040.0,
      "step": 31085
    },
    {
      "entropy": 5.546733093261719,
      "epoch": 2.820210449927431,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042087592455152206,
      "loss": 4.9006,
      "mean_token_accuracy": 0.22090413719415664,
      "num_tokens": 64529721.0,
      "step": 31090
    },
    {
      "entropy": 5.577895927429199,
      "epoch": 2.820664005805515,
      "grad_norm": 1.046875,
      "learning_rate": 0.000420851288382329,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20295707434415816,
      "num_tokens": 64541299.0,
      "step": 31095
    },
    {
      "entropy": 5.624560070037842,
      "epoch": 2.8211175616835993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042082664919689406,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21227426528930665,
      "num_tokens": 64550831.0,
      "step": 31100
    },
    {
      "entropy": 5.6072417259216305,
      "epoch": 2.8215711175616836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004208020069957269,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20949650108814238,
      "num_tokens": 64562076.0,
      "step": 31105
    },
    {
      "entropy": 5.525624847412109,
      "epoch": 2.822024673439768,
      "grad_norm": 1.109375,
      "learning_rate": 0.000420777361779337,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21800896972417833,
      "num_tokens": 64572955.0,
      "step": 31110
    },
    {
      "entropy": 5.642174816131591,
      "epoch": 2.822478229317852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042075271354823427,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2185727134346962,
      "num_tokens": 64584376.0,
      "step": 31115
    },
    {
      "entropy": 5.6680704116821286,
      "epoch": 2.8229317851959363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004207280623029281,
      "loss": 4.967,
      "mean_token_accuracy": 0.21231476217508316,
      "num_tokens": 64594522.0,
      "step": 31120
    },
    {
      "entropy": 5.633637762069702,
      "epoch": 2.82338534107402,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042070340804392853,
      "loss": 4.9713,
      "mean_token_accuracy": 0.22108199894428254,
      "num_tokens": 64604625.0,
      "step": 31125
    },
    {
      "entropy": 5.579609012603759,
      "epoch": 2.8238388969521044,
      "grad_norm": 1.25,
      "learning_rate": 0.0004206787507717454,
      "loss": 4.9559,
      "mean_token_accuracy": 0.21223720610141755,
      "num_tokens": 64614184.0,
      "step": 31130
    },
    {
      "entropy": 5.68731722831726,
      "epoch": 2.8242924528301887,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042065409048688856,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20688927471637725,
      "num_tokens": 64623019.0,
      "step": 31135
    },
    {
      "entropy": 5.616736316680909,
      "epoch": 2.824746008708273,
      "grad_norm": 1.21875,
      "learning_rate": 0.000420629427189868,
      "loss": 4.9311,
      "mean_token_accuracy": 0.212612546980381,
      "num_tokens": 64634017.0,
      "step": 31140
    },
    {
      "entropy": 5.579551267623901,
      "epoch": 2.825199564586357,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042060476088119376,
      "loss": 4.9207,
      "mean_token_accuracy": 0.22012752294540405,
      "num_tokens": 64645007.0,
      "step": 31145
    },
    {
      "entropy": 5.660323905944824,
      "epoch": 2.825653120464441,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000420580091561376,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20943098068237304,
      "num_tokens": 64655705.0,
      "step": 31150
    },
    {
      "entropy": 5.705381870269775,
      "epoch": 2.8261066763425253,
      "grad_norm": 1.125,
      "learning_rate": 0.00042055541923092485,
      "loss": 5.0283,
      "mean_token_accuracy": 0.21066718846559523,
      "num_tokens": 64666094.0,
      "step": 31155
    },
    {
      "entropy": 5.672518348693847,
      "epoch": 2.8265602322206096,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004205307438903505,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20768718868494035,
      "num_tokens": 64676107.0,
      "step": 31160
    },
    {
      "entropy": 5.5879474639892575,
      "epoch": 2.827013788098694,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004205060655401633,
      "loss": 4.8991,
      "mean_token_accuracy": 0.21730416864156724,
      "num_tokens": 64685846.0,
      "step": 31165
    },
    {
      "entropy": 5.548808813095093,
      "epoch": 2.827467343976778,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004204813841808736,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21948938071727753,
      "num_tokens": 64695959.0,
      "step": 31170
    },
    {
      "entropy": 5.586927032470703,
      "epoch": 2.8279208998548624,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004204566998129917,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2189114898443222,
      "num_tokens": 64705594.0,
      "step": 31175
    },
    {
      "entropy": 5.636942720413208,
      "epoch": 2.828374455732946,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042043201243702834,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21315570473670958,
      "num_tokens": 64715974.0,
      "step": 31180
    },
    {
      "entropy": 5.584293985366822,
      "epoch": 2.8288280116110305,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004204073220534938,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21684186160564423,
      "num_tokens": 64725846.0,
      "step": 31185
    },
    {
      "entropy": 5.567299652099609,
      "epoch": 2.8292815674891147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004203826286628988,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2201531320810318,
      "num_tokens": 64735789.0,
      "step": 31190
    },
    {
      "entropy": 5.547651290893555,
      "epoch": 2.8297351233671986,
      "grad_norm": 1.171875,
      "learning_rate": 0.000420357932265754,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22971422970294952,
      "num_tokens": 64745738.0,
      "step": 31195
    },
    {
      "entropy": 5.6054356575012205,
      "epoch": 2.830188679245283,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004203332328625701,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21167430728673936,
      "num_tokens": 64756523.0,
      "step": 31200
    },
    {
      "entropy": 5.6715209007263185,
      "epoch": 2.830642235123367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042030853045385796,
      "loss": 5.0277,
      "mean_token_accuracy": 0.20484247654676438,
      "num_tokens": 64767367.0,
      "step": 31205
    },
    {
      "entropy": 5.651865530014038,
      "epoch": 2.8310957910014514,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042028382504012837,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21473362445831298,
      "num_tokens": 64777729.0,
      "step": 31210
    },
    {
      "entropy": 5.709919738769531,
      "epoch": 2.8315493468795356,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042025911662189224,
      "loss": 4.9755,
      "mean_token_accuracy": 0.2129921168088913,
      "num_tokens": 64788223.0,
      "step": 31215
    },
    {
      "entropy": 5.599450159072876,
      "epoch": 2.83200290275762,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042023440519966057,
      "loss": 4.928,
      "mean_token_accuracy": 0.21461839973926544,
      "num_tokens": 64797367.0,
      "step": 31220
    },
    {
      "entropy": 5.61387152671814,
      "epoch": 2.832456458635704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042020969077394434,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2126474127173424,
      "num_tokens": 64807179.0,
      "step": 31225
    },
    {
      "entropy": 5.641021871566773,
      "epoch": 2.832910014513788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042018497334525475,
      "loss": 4.862,
      "mean_token_accuracy": 0.20722534358501435,
      "num_tokens": 64817276.0,
      "step": 31230
    },
    {
      "entropy": 5.596399116516113,
      "epoch": 2.8333635703918723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004201602529141029,
      "loss": 4.945,
      "mean_token_accuracy": 0.21266809850931168,
      "num_tokens": 64827509.0,
      "step": 31235
    },
    {
      "entropy": 5.5908284187316895,
      "epoch": 2.8338171262699565,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042013552948100003,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2150601014494896,
      "num_tokens": 64836578.0,
      "step": 31240
    },
    {
      "entropy": 5.547596216201782,
      "epoch": 2.834270682148041,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004201108030464574,
      "loss": 4.8814,
      "mean_token_accuracy": 0.2227599397301674,
      "num_tokens": 64846518.0,
      "step": 31245
    },
    {
      "entropy": 5.682690858840942,
      "epoch": 2.8347242380261246,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042008607361098646,
      "loss": 4.9209,
      "mean_token_accuracy": 0.2164894625544548,
      "num_tokens": 64856304.0,
      "step": 31250
    },
    {
      "entropy": 5.5925287246704105,
      "epoch": 2.835177793904209,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004200613411750985,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2263615161180496,
      "num_tokens": 64867381.0,
      "step": 31255
    },
    {
      "entropy": 5.602374029159546,
      "epoch": 2.835631349782293,
      "grad_norm": 1.25,
      "learning_rate": 0.000420036605739305,
      "loss": 4.9574,
      "mean_token_accuracy": 0.2157713457942009,
      "num_tokens": 64877306.0,
      "step": 31260
    },
    {
      "entropy": 5.516693449020385,
      "epoch": 2.8360849056603774,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004200118673041176,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2164098784327507,
      "num_tokens": 64886932.0,
      "step": 31265
    },
    {
      "entropy": 5.62765588760376,
      "epoch": 2.8365384615384617,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041998712587004776,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21025226265192032,
      "num_tokens": 64897387.0,
      "step": 31270
    },
    {
      "entropy": 5.675957298278808,
      "epoch": 2.836992017416546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041996238143760726,
      "loss": 4.9654,
      "mean_token_accuracy": 0.21875430792570114,
      "num_tokens": 64907251.0,
      "step": 31275
    },
    {
      "entropy": 5.583954572677612,
      "epoch": 2.8374455732946298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041993763400730776,
      "loss": 4.913,
      "mean_token_accuracy": 0.21687745451927185,
      "num_tokens": 64917300.0,
      "step": 31280
    },
    {
      "entropy": 5.536740446090699,
      "epoch": 2.837899129172714,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004199128835796611,
      "loss": 4.8714,
      "mean_token_accuracy": 0.22533662021160125,
      "num_tokens": 64927133.0,
      "step": 31285
    },
    {
      "entropy": 5.578914594650269,
      "epoch": 2.8383526850507983,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000419888130155179,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21498076468706132,
      "num_tokens": 64937789.0,
      "step": 31290
    },
    {
      "entropy": 5.670873975753784,
      "epoch": 2.8388062409288826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004198633737343735,
      "loss": 4.952,
      "mean_token_accuracy": 0.21623307466506958,
      "num_tokens": 64948127.0,
      "step": 31295
    },
    {
      "entropy": 5.650634765625,
      "epoch": 2.8392597968069664,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004198386143177565,
      "loss": 4.9908,
      "mean_token_accuracy": 0.21361104398965836,
      "num_tokens": 64959632.0,
      "step": 31300
    },
    {
      "entropy": 5.507171535491944,
      "epoch": 2.8397133526850507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041981385190584003,
      "loss": 4.7715,
      "mean_token_accuracy": 0.2209300711750984,
      "num_tokens": 64970411.0,
      "step": 31305
    },
    {
      "entropy": 5.551486968994141,
      "epoch": 2.840166908563135,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004197890864991362,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21426456719636916,
      "num_tokens": 64980748.0,
      "step": 31310
    },
    {
      "entropy": 5.561504077911377,
      "epoch": 2.840620464441219,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004197643180981572,
      "loss": 4.8534,
      "mean_token_accuracy": 0.21824093163013458,
      "num_tokens": 64990818.0,
      "step": 31315
    },
    {
      "entropy": 5.600249576568603,
      "epoch": 2.8410740203193035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004197395467034152,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21965514421463012,
      "num_tokens": 65001993.0,
      "step": 31320
    },
    {
      "entropy": 5.558138227462768,
      "epoch": 2.8415275761973877,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004197147723154225,
      "loss": 4.886,
      "mean_token_accuracy": 0.22362725287675858,
      "num_tokens": 65013007.0,
      "step": 31325
    },
    {
      "entropy": 5.664604043960571,
      "epoch": 2.8419811320754715,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004196899949346914,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2142934888601303,
      "num_tokens": 65023595.0,
      "step": 31330
    },
    {
      "entropy": 5.600962066650391,
      "epoch": 2.842434687953556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004196652145617343,
      "loss": 4.8959,
      "mean_token_accuracy": 0.21897254139184952,
      "num_tokens": 65033631.0,
      "step": 31335
    },
    {
      "entropy": 5.672798156738281,
      "epoch": 2.84288824383164,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004196404311970638,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2126334637403488,
      "num_tokens": 65044059.0,
      "step": 31340
    },
    {
      "entropy": 5.59052209854126,
      "epoch": 2.8433417997097243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004196156448411922,
      "loss": 4.9177,
      "mean_token_accuracy": 0.21347975581884385,
      "num_tokens": 65053931.0,
      "step": 31345
    },
    {
      "entropy": 5.591944217681885,
      "epoch": 2.843795355587808,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041959085549463234,
      "loss": 4.9213,
      "mean_token_accuracy": 0.21769054532051085,
      "num_tokens": 65064011.0,
      "step": 31350
    },
    {
      "entropy": 5.654287385940552,
      "epoch": 2.8442489114658924,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004195660631578966,
      "loss": 4.9885,
      "mean_token_accuracy": 0.21189521700143815,
      "num_tokens": 65074417.0,
      "step": 31355
    },
    {
      "entropy": 5.596863317489624,
      "epoch": 2.8447024673439767,
      "grad_norm": 1.1875,
      "learning_rate": 0.000419541267831498,
      "loss": 4.8969,
      "mean_token_accuracy": 0.22106014043092728,
      "num_tokens": 65084593.0,
      "step": 31360
    },
    {
      "entropy": 5.57124547958374,
      "epoch": 2.845156023222061,
      "grad_norm": 1.171875,
      "learning_rate": 0.000419516469515949,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22277182787656785,
      "num_tokens": 65095047.0,
      "step": 31365
    },
    {
      "entropy": 5.631705808639526,
      "epoch": 2.8456095791001452,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041949166821176264,
      "loss": 5.0032,
      "mean_token_accuracy": 0.2164821818470955,
      "num_tokens": 65104434.0,
      "step": 31370
    },
    {
      "entropy": 5.678917980194091,
      "epoch": 2.8460631349782295,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041946686391945183,
      "loss": 4.9769,
      "mean_token_accuracy": 0.21663412749767302,
      "num_tokens": 65114039.0,
      "step": 31375
    },
    {
      "entropy": 5.71366171836853,
      "epoch": 2.8465166908563138,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004194420566395294,
      "loss": 5.0491,
      "mean_token_accuracy": 0.20536773651838303,
      "num_tokens": 65125387.0,
      "step": 31380
    },
    {
      "entropy": 5.637423467636109,
      "epoch": 2.8469702467343976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004194172463725084,
      "loss": 4.9679,
      "mean_token_accuracy": 0.2191828489303589,
      "num_tokens": 65136351.0,
      "step": 31385
    },
    {
      "entropy": 5.585445976257324,
      "epoch": 2.847423802612482,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041939243311890195,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21333178132772446,
      "num_tokens": 65147343.0,
      "step": 31390
    },
    {
      "entropy": 5.546621465682984,
      "epoch": 2.847877358490566,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041936761687922324,
      "loss": 4.8875,
      "mean_token_accuracy": 0.222720368206501,
      "num_tokens": 65157097.0,
      "step": 31395
    },
    {
      "entropy": 5.605692672729492,
      "epoch": 2.8483309143686504,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041934279765398535,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21638421565294266,
      "num_tokens": 65167546.0,
      "step": 31400
    },
    {
      "entropy": 5.668427658081055,
      "epoch": 2.848784470246734,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004193179754437017,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20696936845779418,
      "num_tokens": 65178497.0,
      "step": 31405
    },
    {
      "entropy": 5.586981344223022,
      "epoch": 2.8492380261248185,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004192931502488855,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22228772193193436,
      "num_tokens": 65189530.0,
      "step": 31410
    },
    {
      "entropy": 5.60063591003418,
      "epoch": 2.8496915820029027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041926832207005025,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21276939660310745,
      "num_tokens": 65199836.0,
      "step": 31415
    },
    {
      "entropy": 5.507954788208008,
      "epoch": 2.850145137880987,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004192434909077092,
      "loss": 4.7683,
      "mean_token_accuracy": 0.22527859956026078,
      "num_tokens": 65210208.0,
      "step": 31420
    },
    {
      "entropy": 5.583266258239746,
      "epoch": 2.8505986937590713,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041921865676237605,
      "loss": 4.9268,
      "mean_token_accuracy": 0.21439330875873566,
      "num_tokens": 65220954.0,
      "step": 31425
    },
    {
      "entropy": 5.644137525558472,
      "epoch": 2.8510522496371555,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041919381963456437,
      "loss": 4.9392,
      "mean_token_accuracy": 0.21211770176887512,
      "num_tokens": 65230870.0,
      "step": 31430
    },
    {
      "entropy": 5.635395431518555,
      "epoch": 2.8515058055152394,
      "grad_norm": 1.25,
      "learning_rate": 0.0004191689795247877,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22319147288799285,
      "num_tokens": 65241913.0,
      "step": 31435
    },
    {
      "entropy": 5.688692855834961,
      "epoch": 2.8519593613933236,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004191441364335598,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21712172031402588,
      "num_tokens": 65252513.0,
      "step": 31440
    },
    {
      "entropy": 5.554673337936402,
      "epoch": 2.852412917271408,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041911929036139457,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21490330696105958,
      "num_tokens": 65262695.0,
      "step": 31445
    },
    {
      "entropy": 5.639480209350586,
      "epoch": 2.852866473149492,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041909444130880553,
      "loss": 4.9946,
      "mean_token_accuracy": 0.2159951239824295,
      "num_tokens": 65274131.0,
      "step": 31450
    },
    {
      "entropy": 5.599239158630371,
      "epoch": 2.853320029027576,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041906958927630673,
      "loss": 4.9229,
      "mean_token_accuracy": 0.2150556668639183,
      "num_tokens": 65285080.0,
      "step": 31455
    },
    {
      "entropy": 5.735603189468383,
      "epoch": 2.8537735849056602,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004190447342644122,
      "loss": 5.1101,
      "mean_token_accuracy": 0.2141365796327591,
      "num_tokens": 65296512.0,
      "step": 31460
    },
    {
      "entropy": 5.5775209903717045,
      "epoch": 2.8542271407837445,
      "grad_norm": 1.25,
      "learning_rate": 0.0004190198762736358,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2224300131201744,
      "num_tokens": 65306054.0,
      "step": 31465
    },
    {
      "entropy": 5.547127199172974,
      "epoch": 2.8546806966618288,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004189950153044917,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21349292397499084,
      "num_tokens": 65315046.0,
      "step": 31470
    },
    {
      "entropy": 5.581344699859619,
      "epoch": 2.855134252539913,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004189701513574939,
      "loss": 4.8884,
      "mean_token_accuracy": 0.21237158179283142,
      "num_tokens": 65325482.0,
      "step": 31475
    },
    {
      "entropy": 5.564690732955933,
      "epoch": 2.8555878084179973,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004189452844331567,
      "loss": 4.8583,
      "mean_token_accuracy": 0.21760592311620713,
      "num_tokens": 65335518.0,
      "step": 31480
    },
    {
      "entropy": 5.601111650466919,
      "epoch": 2.856041364296081,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041892041453199443,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2173466458916664,
      "num_tokens": 65346079.0,
      "step": 31485
    },
    {
      "entropy": 5.560744714736939,
      "epoch": 2.8564949201741654,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004188955416545212,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2206578239798546,
      "num_tokens": 65356719.0,
      "step": 31490
    },
    {
      "entropy": 5.647084045410156,
      "epoch": 2.8569484760522497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004188706658012516,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2081117421388626,
      "num_tokens": 65367877.0,
      "step": 31495
    },
    {
      "entropy": 5.680764293670654,
      "epoch": 2.857402031930334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041884578697269994,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20874454826116562,
      "num_tokens": 65379624.0,
      "step": 31500
    },
    {
      "entropy": 5.629879760742187,
      "epoch": 2.8578555878084178,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004188209051693807,
      "loss": 4.9922,
      "mean_token_accuracy": 0.21204937994480133,
      "num_tokens": 65390417.0,
      "step": 31505
    },
    {
      "entropy": 5.501463937759399,
      "epoch": 2.858309143686502,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041879602039180856,
      "loss": 4.7681,
      "mean_token_accuracy": 0.2275952711701393,
      "num_tokens": 65401681.0,
      "step": 31510
    },
    {
      "entropy": 5.556026792526245,
      "epoch": 2.8587626995645863,
      "grad_norm": 1.25,
      "learning_rate": 0.00041877113264049805,
      "loss": 4.8873,
      "mean_token_accuracy": 0.21970682591199875,
      "num_tokens": 65412520.0,
      "step": 31515
    },
    {
      "entropy": 5.617603731155396,
      "epoch": 2.8592162554426706,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041874624191596383,
      "loss": 4.9313,
      "mean_token_accuracy": 0.21787261962890625,
      "num_tokens": 65423201.0,
      "step": 31520
    },
    {
      "entropy": 5.57971715927124,
      "epoch": 2.859669811320755,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004187213482187207,
      "loss": 4.9476,
      "mean_token_accuracy": 0.21197264790534973,
      "num_tokens": 65432778.0,
      "step": 31525
    },
    {
      "entropy": 5.6699371337890625,
      "epoch": 2.860123367198839,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004186964515492835,
      "loss": 4.947,
      "mean_token_accuracy": 0.21519640684127808,
      "num_tokens": 65442323.0,
      "step": 31530
    },
    {
      "entropy": 5.60116810798645,
      "epoch": 2.8605769230769234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004186715519081671,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21230805218219756,
      "num_tokens": 65451980.0,
      "step": 31535
    },
    {
      "entropy": 5.5135551452636715,
      "epoch": 2.861030478955007,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004186466492958864,
      "loss": 4.8813,
      "mean_token_accuracy": 0.21719976216554643,
      "num_tokens": 65463359.0,
      "step": 31540
    },
    {
      "entropy": 5.607174587249756,
      "epoch": 2.8614840348330914,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004186217437129562,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21512628197669983,
      "num_tokens": 65474155.0,
      "step": 31545
    },
    {
      "entropy": 5.631038093566895,
      "epoch": 2.8619375907111757,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041859683515989196,
      "loss": 4.8724,
      "mean_token_accuracy": 0.21913151890039445,
      "num_tokens": 65484884.0,
      "step": 31550
    },
    {
      "entropy": 5.630547952651978,
      "epoch": 2.8623911465892595,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041857192363720845,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21555111408233643,
      "num_tokens": 65494380.0,
      "step": 31555
    },
    {
      "entropy": 5.585837507247925,
      "epoch": 2.862844702467344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004185470091454211,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21548750698566438,
      "num_tokens": 65505735.0,
      "step": 31560
    },
    {
      "entropy": 5.61636209487915,
      "epoch": 2.863298258345428,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004185220916850449,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2144456148147583,
      "num_tokens": 65515956.0,
      "step": 31565
    },
    {
      "entropy": 5.596684074401855,
      "epoch": 2.8637518142235123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041849717125659537,
      "loss": 4.8554,
      "mean_token_accuracy": 0.218240687251091,
      "num_tokens": 65526369.0,
      "step": 31570
    },
    {
      "entropy": 5.603471994400024,
      "epoch": 2.8642053701015966,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004184722478605877,
      "loss": 4.8929,
      "mean_token_accuracy": 0.22364125698804854,
      "num_tokens": 65536032.0,
      "step": 31575
    },
    {
      "entropy": 5.567645168304443,
      "epoch": 2.864658925979681,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004184473214975374,
      "loss": 4.9614,
      "mean_token_accuracy": 0.20941196829080583,
      "num_tokens": 65546121.0,
      "step": 31580
    },
    {
      "entropy": 5.606190538406372,
      "epoch": 2.865112481857765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041842239216795994,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21521912068128585,
      "num_tokens": 65555915.0,
      "step": 31585
    },
    {
      "entropy": 5.608984661102295,
      "epoch": 2.865566037735849,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041839745987237086,
      "loss": 4.9359,
      "mean_token_accuracy": 0.20711799263954161,
      "num_tokens": 65566267.0,
      "step": 31590
    },
    {
      "entropy": 5.596514320373535,
      "epoch": 2.866019593613933,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004183725246112857,
      "loss": 4.914,
      "mean_token_accuracy": 0.22463058084249496,
      "num_tokens": 65577126.0,
      "step": 31595
    },
    {
      "entropy": 5.6551513671875,
      "epoch": 2.8664731494920175,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041834758638522034,
      "loss": 5.0326,
      "mean_token_accuracy": 0.20194878727197646,
      "num_tokens": 65587158.0,
      "step": 31600
    },
    {
      "entropy": 5.6248420715332035,
      "epoch": 2.8669267053701017,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041832264519469026,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2107012987136841,
      "num_tokens": 65597720.0,
      "step": 31605
    },
    {
      "entropy": 5.689760780334472,
      "epoch": 2.8673802612481856,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041829770104021143,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21968888491392136,
      "num_tokens": 65607409.0,
      "step": 31610
    },
    {
      "entropy": 5.618808603286743,
      "epoch": 2.86783381712627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041827275392229956,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22368093729019164,
      "num_tokens": 65616785.0,
      "step": 31615
    },
    {
      "entropy": 5.604651355743409,
      "epoch": 2.868287373004354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004182478038414706,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20952381938695908,
      "num_tokens": 65625794.0,
      "step": 31620
    },
    {
      "entropy": 5.526165246963501,
      "epoch": 2.8687409288824384,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041822285079824056,
      "loss": 4.8121,
      "mean_token_accuracy": 0.2223937079310417,
      "num_tokens": 65635789.0,
      "step": 31625
    },
    {
      "entropy": 5.644211196899414,
      "epoch": 2.8691944847605226,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004181978947931256,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20847029089927674,
      "num_tokens": 65646266.0,
      "step": 31630
    },
    {
      "entropy": 5.627982759475708,
      "epoch": 2.869648040638607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004181729358266415,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21238478124141694,
      "num_tokens": 65656538.0,
      "step": 31635
    },
    {
      "entropy": 5.680902433395386,
      "epoch": 2.8701015965166907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041814797389930474,
      "loss": 4.9415,
      "mean_token_accuracy": 0.2185475692152977,
      "num_tokens": 65666926.0,
      "step": 31640
    },
    {
      "entropy": 5.621043920516968,
      "epoch": 2.870555152394775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004181230090116313,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2103895977139473,
      "num_tokens": 65677219.0,
      "step": 31645
    },
    {
      "entropy": 5.589084482192993,
      "epoch": 2.8710087082728593,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041809804116413763,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21239991039037703,
      "num_tokens": 65688281.0,
      "step": 31650
    },
    {
      "entropy": 5.735374021530151,
      "epoch": 2.8714622641509435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041807307035734,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20635735392570495,
      "num_tokens": 65698815.0,
      "step": 31655
    },
    {
      "entropy": 5.630136966705322,
      "epoch": 2.8719158200290273,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004180480965917548,
      "loss": 4.9498,
      "mean_token_accuracy": 0.21364252865314484,
      "num_tokens": 65709481.0,
      "step": 31660
    },
    {
      "entropy": 5.66035270690918,
      "epoch": 2.8723693759071116,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041802311986789854,
      "loss": 4.9308,
      "mean_token_accuracy": 0.20607939064502717,
      "num_tokens": 65720561.0,
      "step": 31665
    },
    {
      "entropy": 5.653276872634888,
      "epoch": 2.872822931785196,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041799814018628765,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21045362800359727,
      "num_tokens": 65730384.0,
      "step": 31670
    },
    {
      "entropy": 5.612490224838257,
      "epoch": 2.87327648766328,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041797315754743883,
      "loss": 4.936,
      "mean_token_accuracy": 0.21761937886476518,
      "num_tokens": 65740711.0,
      "step": 31675
    },
    {
      "entropy": 5.704002571105957,
      "epoch": 2.8737300435413644,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041794817195186867,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20669930875301362,
      "num_tokens": 65751033.0,
      "step": 31680
    },
    {
      "entropy": 5.688816356658935,
      "epoch": 2.8741835994194487,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004179231834000939,
      "loss": 5.013,
      "mean_token_accuracy": 0.20879815518856049,
      "num_tokens": 65761500.0,
      "step": 31685
    },
    {
      "entropy": 5.560088014602661,
      "epoch": 2.8746371552975325,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041789819189263117,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22089462727308273,
      "num_tokens": 65771095.0,
      "step": 31690
    },
    {
      "entropy": 5.578502655029297,
      "epoch": 2.8750907111756168,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041787319742999756,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21886488795280457,
      "num_tokens": 65782058.0,
      "step": 31695
    },
    {
      "entropy": 5.593817472457886,
      "epoch": 2.875544267053701,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004178482000127098,
      "loss": 4.9596,
      "mean_token_accuracy": 0.2222698673605919,
      "num_tokens": 65791710.0,
      "step": 31700
    },
    {
      "entropy": 5.639909410476685,
      "epoch": 2.8759978229317853,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004178231996412848,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20786140561103822,
      "num_tokens": 65802180.0,
      "step": 31705
    },
    {
      "entropy": 5.5874964714050295,
      "epoch": 2.876451378809869,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004177981963162396,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22670624703168868,
      "num_tokens": 65812998.0,
      "step": 31710
    },
    {
      "entropy": 5.600241851806641,
      "epoch": 2.8769049346879534,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004177731900380914,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21553351879119872,
      "num_tokens": 65824210.0,
      "step": 31715
    },
    {
      "entropy": 5.5797172546386715,
      "epoch": 2.8773584905660377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004177481808073572,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22095848470926285,
      "num_tokens": 65834322.0,
      "step": 31720
    },
    {
      "entropy": 5.594752216339112,
      "epoch": 2.877812046444122,
      "grad_norm": 1.125,
      "learning_rate": 0.00041772316862455426,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2244741439819336,
      "num_tokens": 65845347.0,
      "step": 31725
    },
    {
      "entropy": 5.640364313125611,
      "epoch": 2.878265602322206,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041769815349019986,
      "loss": 4.9498,
      "mean_token_accuracy": 0.20914436727762223,
      "num_tokens": 65855616.0,
      "step": 31730
    },
    {
      "entropy": 5.650536251068115,
      "epoch": 2.8787191582002905,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00041767313540481125,
      "loss": 4.9208,
      "mean_token_accuracy": 0.2161816403269768,
      "num_tokens": 65865542.0,
      "step": 31735
    },
    {
      "entropy": 5.592294454574585,
      "epoch": 2.8791727140783747,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004176481143689058,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21320167630910875,
      "num_tokens": 65874478.0,
      "step": 31740
    },
    {
      "entropy": 5.651920795440674,
      "epoch": 2.8796262699564585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004176230903830011,
      "loss": 4.9551,
      "mean_token_accuracy": 0.2180263802409172,
      "num_tokens": 65884609.0,
      "step": 31745
    },
    {
      "entropy": 5.642696285247803,
      "epoch": 2.880079825834543,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004175980634476145,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21794142872095107,
      "num_tokens": 65894486.0,
      "step": 31750
    },
    {
      "entropy": 5.576766109466552,
      "epoch": 2.880533381712627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004175730335632635,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2173640325665474,
      "num_tokens": 65904643.0,
      "step": 31755
    },
    {
      "entropy": 5.522447299957276,
      "epoch": 2.8809869375907113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041754800073046593,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21958731561899186,
      "num_tokens": 65914107.0,
      "step": 31760
    },
    {
      "entropy": 5.579465913772583,
      "epoch": 2.881440493468795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004175229649497394,
      "loss": 4.8592,
      "mean_token_accuracy": 0.2164247527718544,
      "num_tokens": 65925112.0,
      "step": 31765
    },
    {
      "entropy": 5.684026718139648,
      "epoch": 2.8818940493468794,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004174979262216016,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21679816395044327,
      "num_tokens": 65934624.0,
      "step": 31770
    },
    {
      "entropy": 5.6303047180175785,
      "epoch": 2.8823476052249637,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004174728845465704,
      "loss": 4.9094,
      "mean_token_accuracy": 0.21469669044017792,
      "num_tokens": 65945951.0,
      "step": 31775
    },
    {
      "entropy": 5.533226585388183,
      "epoch": 2.882801161103048,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004174478399251636,
      "loss": 4.8376,
      "mean_token_accuracy": 0.22649696320295334,
      "num_tokens": 65956871.0,
      "step": 31780
    },
    {
      "entropy": 5.694487047195435,
      "epoch": 2.8832547169811322,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004174227923578992,
      "loss": 5.0485,
      "mean_token_accuracy": 0.2127063751220703,
      "num_tokens": 65967814.0,
      "step": 31785
    },
    {
      "entropy": 5.6321759700775145,
      "epoch": 2.8837082728592165,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004173977418452952,
      "loss": 4.945,
      "mean_token_accuracy": 0.20828794687986374,
      "num_tokens": 65978118.0,
      "step": 31790
    },
    {
      "entropy": 5.570476722717285,
      "epoch": 2.8841618287373003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041737268838786955,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2200121432542801,
      "num_tokens": 65988240.0,
      "step": 31795
    },
    {
      "entropy": 5.608204889297485,
      "epoch": 2.8846153846153846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004173476319861404,
      "loss": 4.876,
      "mean_token_accuracy": 0.21797575950622558,
      "num_tokens": 65998126.0,
      "step": 31800
    },
    {
      "entropy": 5.619529390335083,
      "epoch": 2.885068940493469,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004173225726406261,
      "loss": 4.9346,
      "mean_token_accuracy": 0.22170098423957824,
      "num_tokens": 66009118.0,
      "step": 31805
    },
    {
      "entropy": 5.641887760162353,
      "epoch": 2.885522496371553,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041729751035184457,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2094158411026001,
      "num_tokens": 66019340.0,
      "step": 31810
    },
    {
      "entropy": 5.611714315414429,
      "epoch": 2.885976052249637,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041727244512031425,
      "loss": 4.9003,
      "mean_token_accuracy": 0.2167496055364609,
      "num_tokens": 66029784.0,
      "step": 31815
    },
    {
      "entropy": 5.6632606983184814,
      "epoch": 2.886429608127721,
      "grad_norm": 1.25,
      "learning_rate": 0.0004172473769465536,
      "loss": 5.0469,
      "mean_token_accuracy": 0.2013155400753021,
      "num_tokens": 66039987.0,
      "step": 31820
    },
    {
      "entropy": 5.6277772903442385,
      "epoch": 2.8868831640058055,
      "grad_norm": 1.25,
      "learning_rate": 0.00041722230583108097,
      "loss": 4.965,
      "mean_token_accuracy": 0.2157362937927246,
      "num_tokens": 66049150.0,
      "step": 31825
    },
    {
      "entropy": 5.64697790145874,
      "epoch": 2.8873367198838897,
      "grad_norm": 1.125,
      "learning_rate": 0.00041719723177441467,
      "loss": 4.9671,
      "mean_token_accuracy": 0.21378494501113893,
      "num_tokens": 66060719.0,
      "step": 31830
    },
    {
      "entropy": 5.682665681838989,
      "epoch": 2.887790275761974,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004171721547770735,
      "loss": 4.9665,
      "mean_token_accuracy": 0.21503500044345855,
      "num_tokens": 66071126.0,
      "step": 31835
    },
    {
      "entropy": 5.635784912109375,
      "epoch": 2.8882438316400583,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004171470748395759,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22183056324720382,
      "num_tokens": 66080928.0,
      "step": 31840
    },
    {
      "entropy": 5.652846622467041,
      "epoch": 2.888697387518142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004171219919624406,
      "loss": 4.916,
      "mean_token_accuracy": 0.21347257643938064,
      "num_tokens": 66090900.0,
      "step": 31845
    },
    {
      "entropy": 5.608264684677124,
      "epoch": 2.8891509433962264,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041709690614618624,
      "loss": 4.9693,
      "mean_token_accuracy": 0.2088766023516655,
      "num_tokens": 66100720.0,
      "step": 31850
    },
    {
      "entropy": 5.587069129943847,
      "epoch": 2.8896044992743106,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041707181739133163,
      "loss": 4.9584,
      "mean_token_accuracy": 0.21069999933242797,
      "num_tokens": 66110780.0,
      "step": 31855
    },
    {
      "entropy": 5.633911228179931,
      "epoch": 2.890058055152395,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004170467256983957,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22332895696163177,
      "num_tokens": 66120976.0,
      "step": 31860
    },
    {
      "entropy": 5.641289472579956,
      "epoch": 2.8905116110304787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004170216310678973,
      "loss": 4.9198,
      "mean_token_accuracy": 0.22116368561983107,
      "num_tokens": 66130221.0,
      "step": 31865
    },
    {
      "entropy": 5.573739147186279,
      "epoch": 2.890965166908563,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041699653350035527,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2218844324350357,
      "num_tokens": 66140060.0,
      "step": 31870
    },
    {
      "entropy": 5.578892850875855,
      "epoch": 2.8914187227866472,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041697143299628886,
      "loss": 4.9294,
      "mean_token_accuracy": 0.2219777524471283,
      "num_tokens": 66151640.0,
      "step": 31875
    },
    {
      "entropy": 5.565376663208008,
      "epoch": 2.8918722786647315,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041694632955621697,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22564877718687057,
      "num_tokens": 66162923.0,
      "step": 31880
    },
    {
      "entropy": 5.629379320144653,
      "epoch": 2.8923258345428158,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041692122318065885,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2127934455871582,
      "num_tokens": 66173108.0,
      "step": 31885
    },
    {
      "entropy": 5.608965492248535,
      "epoch": 2.8927793904209,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041689611387013363,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22055374085903168,
      "num_tokens": 66185280.0,
      "step": 31890
    },
    {
      "entropy": 5.562539577484131,
      "epoch": 2.8932329462989843,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004168710016251606,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2128838062286377,
      "num_tokens": 66196020.0,
      "step": 31895
    },
    {
      "entropy": 5.586726188659668,
      "epoch": 2.893686502177068,
      "grad_norm": 1.125,
      "learning_rate": 0.0004168458864462592,
      "loss": 4.848,
      "mean_token_accuracy": 0.21899206191301346,
      "num_tokens": 66205753.0,
      "step": 31900
    },
    {
      "entropy": 5.64402961730957,
      "epoch": 2.8941400580551524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041682076833394864,
      "loss": 5.0152,
      "mean_token_accuracy": 0.21260420382022857,
      "num_tokens": 66218415.0,
      "step": 31905
    },
    {
      "entropy": 5.634943580627441,
      "epoch": 2.8945936139332367,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041679564728874845,
      "loss": 5.0084,
      "mean_token_accuracy": 0.21050702184438705,
      "num_tokens": 66228090.0,
      "step": 31910
    },
    {
      "entropy": 5.563312005996704,
      "epoch": 2.8950471698113205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004167705233111782,
      "loss": 4.865,
      "mean_token_accuracy": 0.22121773660182953,
      "num_tokens": 66237587.0,
      "step": 31915
    },
    {
      "entropy": 5.60017762184143,
      "epoch": 2.8955007256894048,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041674539640175733,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2185591533780098,
      "num_tokens": 66248162.0,
      "step": 31920
    },
    {
      "entropy": 5.585758066177368,
      "epoch": 2.895954281567489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004167202665610055,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22525439858436586,
      "num_tokens": 66258370.0,
      "step": 31925
    },
    {
      "entropy": 5.600293588638306,
      "epoch": 2.8964078374455733,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041669513378944254,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21419663578271866,
      "num_tokens": 66269267.0,
      "step": 31930
    },
    {
      "entropy": 5.6684362411499025,
      "epoch": 2.8968613933236576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041666999808758815,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21446811705827712,
      "num_tokens": 66280077.0,
      "step": 31935
    },
    {
      "entropy": 5.641676902770996,
      "epoch": 2.897314949201742,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000416644859455962,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22321052700281144,
      "num_tokens": 66291678.0,
      "step": 31940
    },
    {
      "entropy": 5.592736625671387,
      "epoch": 2.897768505079826,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004166197178950841,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21754908114671706,
      "num_tokens": 66302046.0,
      "step": 31945
    },
    {
      "entropy": 5.584465217590332,
      "epoch": 2.89822206095791,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041659457340547436,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21626962572336197,
      "num_tokens": 66311888.0,
      "step": 31950
    },
    {
      "entropy": 5.574025344848633,
      "epoch": 2.898675616835994,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041656942598765275,
      "loss": 4.8829,
      "mean_token_accuracy": 0.2203891918063164,
      "num_tokens": 66321507.0,
      "step": 31955
    },
    {
      "entropy": 5.544738721847534,
      "epoch": 2.8991291727140784,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041654427564213924,
      "loss": 4.8648,
      "mean_token_accuracy": 0.22370833903551102,
      "num_tokens": 66331458.0,
      "step": 31960
    },
    {
      "entropy": 5.6023890495300295,
      "epoch": 2.8995827285921627,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041651912236945416,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22036738842725753,
      "num_tokens": 66340875.0,
      "step": 31965
    },
    {
      "entropy": 5.642870759963989,
      "epoch": 2.9000362844702465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041649396617011753,
      "loss": 4.9819,
      "mean_token_accuracy": 0.20874171257019042,
      "num_tokens": 66350784.0,
      "step": 31970
    },
    {
      "entropy": 5.6391205310821535,
      "epoch": 2.900489840348331,
      "grad_norm": 1.125,
      "learning_rate": 0.0004164688070446496,
      "loss": 4.9899,
      "mean_token_accuracy": 0.21138345450162888,
      "num_tokens": 66361591.0,
      "step": 31975
    },
    {
      "entropy": 5.576123046875,
      "epoch": 2.900943396226415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004164436449935707,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22128966003656386,
      "num_tokens": 66371540.0,
      "step": 31980
    },
    {
      "entropy": 5.618195819854736,
      "epoch": 2.9013969521044993,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041641848001740126,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21928317099809647,
      "num_tokens": 66382053.0,
      "step": 31985
    },
    {
      "entropy": 5.593534517288208,
      "epoch": 2.9018505079825836,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004163933121166615,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21862017661333083,
      "num_tokens": 66392563.0,
      "step": 31990
    },
    {
      "entropy": 5.583109045028687,
      "epoch": 2.902304063860668,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000416368141291872,
      "loss": 4.8749,
      "mean_token_accuracy": 0.2170826643705368,
      "num_tokens": 66403582.0,
      "step": 31995
    },
    {
      "entropy": 5.525620412826538,
      "epoch": 2.9027576197387517,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041634296754355346,
      "loss": 4.8235,
      "mean_token_accuracy": 0.22272140830755233,
      "num_tokens": 66413713.0,
      "step": 32000
    },
    {
      "entropy": 5.582666110992432,
      "epoch": 2.903211175616836,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004163177908722262,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2152288094162941,
      "num_tokens": 66424802.0,
      "step": 32005
    },
    {
      "entropy": 5.600276708602905,
      "epoch": 2.90366473149492,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041629261127841105,
      "loss": 4.8683,
      "mean_token_accuracy": 0.22269952297210693,
      "num_tokens": 66434716.0,
      "step": 32010
    },
    {
      "entropy": 5.5774510383605955,
      "epoch": 2.9041182873730045,
      "grad_norm": 1.25,
      "learning_rate": 0.00041626742876262876,
      "loss": 4.9347,
      "mean_token_accuracy": 0.2132781520485878,
      "num_tokens": 66445725.0,
      "step": 32015
    },
    {
      "entropy": 5.614554500579834,
      "epoch": 2.9045718432510883,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041624224332539997,
      "loss": 4.9759,
      "mean_token_accuracy": 0.20886536538600922,
      "num_tokens": 66457492.0,
      "step": 32020
    },
    {
      "entropy": 5.6291656494140625,
      "epoch": 2.9050253991291726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004162170549672455,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21748114973306656,
      "num_tokens": 66467490.0,
      "step": 32025
    },
    {
      "entropy": 5.626606416702271,
      "epoch": 2.905478955007257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004161918636886865,
      "loss": 5.0097,
      "mean_token_accuracy": 0.21209583431482315,
      "num_tokens": 66478160.0,
      "step": 32030
    },
    {
      "entropy": 5.669289064407349,
      "epoch": 2.905932510885341,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004161666694902437,
      "loss": 4.9169,
      "mean_token_accuracy": 0.21435584425926207,
      "num_tokens": 66488181.0,
      "step": 32035
    },
    {
      "entropy": 5.653158140182495,
      "epoch": 2.9063860667634254,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004161414723724383,
      "loss": 4.9697,
      "mean_token_accuracy": 0.22207406908273697,
      "num_tokens": 66498507.0,
      "step": 32040
    },
    {
      "entropy": 5.5459442138671875,
      "epoch": 2.9068396226415096,
      "grad_norm": 1.125,
      "learning_rate": 0.0004161162723357912,
      "loss": 4.8466,
      "mean_token_accuracy": 0.22229903638362886,
      "num_tokens": 66507752.0,
      "step": 32045
    },
    {
      "entropy": 5.5540379047393795,
      "epoch": 2.9072931785195935,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004160910693808236,
      "loss": 4.9289,
      "mean_token_accuracy": 0.21004885584115982,
      "num_tokens": 66519360.0,
      "step": 32050
    },
    {
      "entropy": 5.659834241867065,
      "epoch": 2.9077467343976777,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041606586350805686,
      "loss": 4.9897,
      "mean_token_accuracy": 0.2120181292295456,
      "num_tokens": 66529790.0,
      "step": 32055
    },
    {
      "entropy": 5.6144193649292,
      "epoch": 2.908200290275762,
      "grad_norm": 1.15625,
      "learning_rate": 0.000416040654718012,
      "loss": 4.8676,
      "mean_token_accuracy": 0.22535613626241685,
      "num_tokens": 66539617.0,
      "step": 32060
    },
    {
      "entropy": 5.585898780822754,
      "epoch": 2.9086538461538463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041601544301121045,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2196207672357559,
      "num_tokens": 66549861.0,
      "step": 32065
    },
    {
      "entropy": 5.672889518737793,
      "epoch": 2.90910740203193,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004159902283881737,
      "loss": 5.0364,
      "mean_token_accuracy": 0.21636117845773697,
      "num_tokens": 66560388.0,
      "step": 32070
    },
    {
      "entropy": 5.690124416351319,
      "epoch": 2.9095609579100143,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000415965010849423,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21009019166231155,
      "num_tokens": 66570795.0,
      "step": 32075
    },
    {
      "entropy": 5.645208120346069,
      "epoch": 2.9100145137880986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041593979039548,
      "loss": 4.9451,
      "mean_token_accuracy": 0.20885093361139298,
      "num_tokens": 66580268.0,
      "step": 32080
    },
    {
      "entropy": 5.612254428863525,
      "epoch": 2.910468069666183,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004159145670268662,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21114560663700105,
      "num_tokens": 66591192.0,
      "step": 32085
    },
    {
      "entropy": 5.623955774307251,
      "epoch": 2.910921625544267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004158893407441033,
      "loss": 4.9968,
      "mean_token_accuracy": 0.21587304323911666,
      "num_tokens": 66600768.0,
      "step": 32090
    },
    {
      "entropy": 5.716529178619385,
      "epoch": 2.9113751814223514,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041586411154771285,
      "loss": 4.9719,
      "mean_token_accuracy": 0.21363837122917176,
      "num_tokens": 66612207.0,
      "step": 32095
    },
    {
      "entropy": 5.57863187789917,
      "epoch": 2.9118287373004357,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041583887943821667,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22028097063302993,
      "num_tokens": 66622175.0,
      "step": 32100
    },
    {
      "entropy": 5.588039255142212,
      "epoch": 2.9122822931785195,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004158136444161367,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2128696247935295,
      "num_tokens": 66633936.0,
      "step": 32105
    },
    {
      "entropy": 5.650552034378052,
      "epoch": 2.9127358490566038,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041578840648199455,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20419620871543884,
      "num_tokens": 66643502.0,
      "step": 32110
    },
    {
      "entropy": 5.614045572280884,
      "epoch": 2.913189404934688,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004157631656363124,
      "loss": 4.9208,
      "mean_token_accuracy": 0.21725158840417863,
      "num_tokens": 66654890.0,
      "step": 32115
    },
    {
      "entropy": 5.652681827545166,
      "epoch": 2.9136429608127723,
      "grad_norm": 1.09375,
      "learning_rate": 0.000415737921879612,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2010793849825859,
      "num_tokens": 66667037.0,
      "step": 32120
    },
    {
      "entropy": 5.603060245513916,
      "epoch": 2.914096516690856,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041571267521241554,
      "loss": 4.8915,
      "mean_token_accuracy": 0.213776995241642,
      "num_tokens": 66677291.0,
      "step": 32125
    },
    {
      "entropy": 5.5318738460540775,
      "epoch": 2.9145500725689404,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004156874256352451,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2148596927523613,
      "num_tokens": 66687421.0,
      "step": 32130
    },
    {
      "entropy": 5.674749851226807,
      "epoch": 2.9150036284470247,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004156621731486228,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2059122547507286,
      "num_tokens": 66697190.0,
      "step": 32135
    },
    {
      "entropy": 5.644549131393433,
      "epoch": 2.915457184325109,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000415636917753071,
      "loss": 5.0034,
      "mean_token_accuracy": 0.2133711650967598,
      "num_tokens": 66707255.0,
      "step": 32140
    },
    {
      "entropy": 5.560888004302979,
      "epoch": 2.915910740203193,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004156116594491119,
      "loss": 4.84,
      "mean_token_accuracy": 0.22024239003658294,
      "num_tokens": 66716451.0,
      "step": 32145
    },
    {
      "entropy": 5.570637559890747,
      "epoch": 2.9163642960812775,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004155863982372678,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21990731954574586,
      "num_tokens": 66727157.0,
      "step": 32150
    },
    {
      "entropy": 5.630880689620971,
      "epoch": 2.9168178519593613,
      "grad_norm": 1.125,
      "learning_rate": 0.00041556113411806117,
      "loss": 4.9247,
      "mean_token_accuracy": 0.2164222851395607,
      "num_tokens": 66737523.0,
      "step": 32155
    },
    {
      "entropy": 5.629849147796631,
      "epoch": 2.9172714078374455,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041553586709201453,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21538535058498381,
      "num_tokens": 66748183.0,
      "step": 32160
    },
    {
      "entropy": 5.650327491760254,
      "epoch": 2.91772496371553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004155105971596503,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22188195288181306,
      "num_tokens": 66758794.0,
      "step": 32165
    },
    {
      "entropy": 5.681767988204956,
      "epoch": 2.918178519593614,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041548532432149105,
      "loss": 5.0156,
      "mean_token_accuracy": 0.21754937320947648,
      "num_tokens": 66769916.0,
      "step": 32170
    },
    {
      "entropy": 5.639793348312378,
      "epoch": 2.918632075471698,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004154600485780595,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22241226732730865,
      "num_tokens": 66779658.0,
      "step": 32175
    },
    {
      "entropy": 5.579885864257813,
      "epoch": 2.919085631349782,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041543476992987843,
      "loss": 4.8697,
      "mean_token_accuracy": 0.22080554962158203,
      "num_tokens": 66788909.0,
      "step": 32180
    },
    {
      "entropy": 5.685889005661011,
      "epoch": 2.9195391872278664,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041540948837747056,
      "loss": 4.971,
      "mean_token_accuracy": 0.21447289139032363,
      "num_tokens": 66799519.0,
      "step": 32185
    },
    {
      "entropy": 5.581276988983154,
      "epoch": 2.9199927431059507,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004153842039213586,
      "loss": 4.8906,
      "mean_token_accuracy": 0.21626971960067748,
      "num_tokens": 66808915.0,
      "step": 32190
    },
    {
      "entropy": 5.6084143161773685,
      "epoch": 2.920446298984035,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004153589165620656,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20788655430078506,
      "num_tokens": 66818819.0,
      "step": 32195
    },
    {
      "entropy": 5.491002702713013,
      "epoch": 2.9208998548621192,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041533362630011435,
      "loss": 4.8135,
      "mean_token_accuracy": 0.22761358469724655,
      "num_tokens": 66829274.0,
      "step": 32200
    },
    {
      "entropy": 5.675210857391358,
      "epoch": 2.921353410740203,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000415308333136028,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21928514093160628,
      "num_tokens": 66839659.0,
      "step": 32205
    },
    {
      "entropy": 5.624256610870361,
      "epoch": 2.9218069666182873,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00041528303707032957,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21585683524608612,
      "num_tokens": 66850415.0,
      "step": 32210
    },
    {
      "entropy": 5.600710535049439,
      "epoch": 2.9222605224963716,
      "grad_norm": 1.25,
      "learning_rate": 0.0004152577381035422,
      "loss": 4.8661,
      "mean_token_accuracy": 0.22674308568239213,
      "num_tokens": 66860663.0,
      "step": 32215
    },
    {
      "entropy": 5.656810331344604,
      "epoch": 2.922714078374456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041523243623618907,
      "loss": 4.9208,
      "mean_token_accuracy": 0.21767379641532897,
      "num_tokens": 66871473.0,
      "step": 32220
    },
    {
      "entropy": 5.5699444770812985,
      "epoch": 2.9231676342525397,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041520713146879333,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22040941566228867,
      "num_tokens": 66881609.0,
      "step": 32225
    },
    {
      "entropy": 5.642283487319946,
      "epoch": 2.923621190130624,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041518182380187853,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21329072564840318,
      "num_tokens": 66891780.0,
      "step": 32230
    },
    {
      "entropy": 5.662907314300537,
      "epoch": 2.924074746008708,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041515651323596774,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20695139467716217,
      "num_tokens": 66902251.0,
      "step": 32235
    },
    {
      "entropy": 5.639466190338135,
      "epoch": 2.9245283018867925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041513119977158464,
      "loss": 4.957,
      "mean_token_accuracy": 0.21719592362642287,
      "num_tokens": 66912538.0,
      "step": 32240
    },
    {
      "entropy": 5.620833015441894,
      "epoch": 2.9249818577648767,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004151058834092526,
      "loss": 4.9288,
      "mean_token_accuracy": 0.21754180341959,
      "num_tokens": 66922648.0,
      "step": 32245
    },
    {
      "entropy": 5.588517904281616,
      "epoch": 2.925435413642961,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041508056414949515,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21842678189277648,
      "num_tokens": 66933115.0,
      "step": 32250
    },
    {
      "entropy": 5.671231269836426,
      "epoch": 2.9258889695210453,
      "grad_norm": 1.125,
      "learning_rate": 0.000415055241992836,
      "loss": 5.069,
      "mean_token_accuracy": 0.20341045707464217,
      "num_tokens": 66943727.0,
      "step": 32255
    },
    {
      "entropy": 5.648931837081909,
      "epoch": 2.926342525399129,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004150299169397987,
      "loss": 4.9009,
      "mean_token_accuracy": 0.21552292704582215,
      "num_tokens": 66953488.0,
      "step": 32260
    },
    {
      "entropy": 5.7079362869262695,
      "epoch": 2.9267960812772134,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004150045889909071,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20397693812847137,
      "num_tokens": 66964586.0,
      "step": 32265
    },
    {
      "entropy": 5.609491395950317,
      "epoch": 2.9272496371552976,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004149792581466848,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21083665490150452,
      "num_tokens": 66975032.0,
      "step": 32270
    },
    {
      "entropy": 5.591258382797241,
      "epoch": 2.9277031930333814,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041495392440765586,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21726235300302504,
      "num_tokens": 66984910.0,
      "step": 32275
    },
    {
      "entropy": 5.591516351699829,
      "epoch": 2.9281567489114657,
      "grad_norm": 1.25,
      "learning_rate": 0.00041492858777434407,
      "loss": 4.7341,
      "mean_token_accuracy": 0.2323855698108673,
      "num_tokens": 66995238.0,
      "step": 32280
    },
    {
      "entropy": 5.601317834854126,
      "epoch": 2.92861030478955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004149032482472734,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2219078943133354,
      "num_tokens": 67005596.0,
      "step": 32285
    },
    {
      "entropy": 5.7226598262786865,
      "epoch": 2.9290638606676342,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041487790582696803,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19873730093240738,
      "num_tokens": 67017312.0,
      "step": 32290
    },
    {
      "entropy": 5.627328205108642,
      "epoch": 2.9295174165457185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004148525605139518,
      "loss": 4.875,
      "mean_token_accuracy": 0.22271230667829514,
      "num_tokens": 67026692.0,
      "step": 32295
    },
    {
      "entropy": 5.554681205749512,
      "epoch": 2.9299709724238028,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041482721230874897,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21969154924154283,
      "num_tokens": 67036801.0,
      "step": 32300
    },
    {
      "entropy": 5.577628087997437,
      "epoch": 2.930424528301887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041480186121188383,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22599717527627944,
      "num_tokens": 67047262.0,
      "step": 32305
    },
    {
      "entropy": 5.611681652069092,
      "epoch": 2.930878084179971,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004147765072238806,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21240981817245483,
      "num_tokens": 67058019.0,
      "step": 32310
    },
    {
      "entropy": 5.664731884002686,
      "epoch": 2.931331640058055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004147511503452635,
      "loss": 5.0093,
      "mean_token_accuracy": 0.21111210733652114,
      "num_tokens": 67068844.0,
      "step": 32315
    },
    {
      "entropy": 5.6392110824584964,
      "epoch": 2.9317851959361394,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041472579057655703,
      "loss": 4.8601,
      "mean_token_accuracy": 0.2270249679684639,
      "num_tokens": 67077400.0,
      "step": 32320
    },
    {
      "entropy": 5.633264112472534,
      "epoch": 2.9322387518142237,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004147004279182856,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20062269568443297,
      "num_tokens": 67088352.0,
      "step": 32325
    },
    {
      "entropy": 5.69848895072937,
      "epoch": 2.9326923076923075,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004146750623709737,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21285239309072496,
      "num_tokens": 67097879.0,
      "step": 32330
    },
    {
      "entropy": 5.630833959579467,
      "epoch": 2.9331458635703918,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041464969393514595,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22104258835315704,
      "num_tokens": 67108355.0,
      "step": 32335
    },
    {
      "entropy": 5.625765562057495,
      "epoch": 2.933599419448476,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004146243226113269,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2136353299021721,
      "num_tokens": 67118348.0,
      "step": 32340
    },
    {
      "entropy": 5.527565765380859,
      "epoch": 2.9340529753265603,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041459894840004135,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22264847457408904,
      "num_tokens": 67128944.0,
      "step": 32345
    },
    {
      "entropy": 5.657931852340698,
      "epoch": 2.9345065312046446,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041457357130181396,
      "loss": 4.9715,
      "mean_token_accuracy": 0.2092956617474556,
      "num_tokens": 67139365.0,
      "step": 32350
    },
    {
      "entropy": 5.620879411697388,
      "epoch": 2.934960087082729,
      "grad_norm": 1.125,
      "learning_rate": 0.00041454819131716945,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21241387724876404,
      "num_tokens": 67150406.0,
      "step": 32355
    },
    {
      "entropy": 5.684853267669678,
      "epoch": 2.9354136429608126,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041452280844663295,
      "loss": 5.0893,
      "mean_token_accuracy": 0.20633440166711808,
      "num_tokens": 67161631.0,
      "step": 32360
    },
    {
      "entropy": 5.596579360961914,
      "epoch": 2.935867198838897,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004144974226907291,
      "loss": 4.947,
      "mean_token_accuracy": 0.21707209050655366,
      "num_tokens": 67171735.0,
      "step": 32365
    },
    {
      "entropy": 5.629430627822876,
      "epoch": 2.936320754716981,
      "grad_norm": 1.1875,
      "learning_rate": 0.000414472034049983,
      "loss": 4.9615,
      "mean_token_accuracy": 0.21705561578273774,
      "num_tokens": 67181616.0,
      "step": 32370
    },
    {
      "entropy": 5.640116548538208,
      "epoch": 2.9367743105950654,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004144466425249197,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2170005351305008,
      "num_tokens": 67192519.0,
      "step": 32375
    },
    {
      "entropy": 5.654494857788086,
      "epoch": 2.9372278664731493,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004144212481160643,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21257491260766984,
      "num_tokens": 67202805.0,
      "step": 32380
    },
    {
      "entropy": 5.615114545822143,
      "epoch": 2.9376814223512335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004143958508239419,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2178141340613365,
      "num_tokens": 67212778.0,
      "step": 32385
    },
    {
      "entropy": 5.589913082122803,
      "epoch": 2.938134978229318,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004143704506490779,
      "loss": 4.9972,
      "mean_token_accuracy": 0.21148374676704407,
      "num_tokens": 67222588.0,
      "step": 32390
    },
    {
      "entropy": 5.685904264450073,
      "epoch": 2.938588534107402,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004143450475919975,
      "loss": 4.9455,
      "mean_token_accuracy": 0.21193605810403823,
      "num_tokens": 67233022.0,
      "step": 32395
    },
    {
      "entropy": 5.603614234924317,
      "epoch": 2.9390420899854863,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004143196416532259,
      "loss": 4.9442,
      "mean_token_accuracy": 0.2156660333275795,
      "num_tokens": 67242631.0,
      "step": 32400
    },
    {
      "entropy": 5.559706783294677,
      "epoch": 2.9394956458635706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004142942328332887,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22285099625587462,
      "num_tokens": 67252376.0,
      "step": 32405
    },
    {
      "entropy": 5.632914686203003,
      "epoch": 2.9399492017416544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004142688211327111,
      "loss": 4.8714,
      "mean_token_accuracy": 0.21618175655603408,
      "num_tokens": 67262209.0,
      "step": 32410
    },
    {
      "entropy": 5.679757499694825,
      "epoch": 2.9404027576197387,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004142434065520189,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2097629651427269,
      "num_tokens": 67272893.0,
      "step": 32415
    },
    {
      "entropy": 5.590439653396606,
      "epoch": 2.940856313497823,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004142179890917376,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21640788316726683,
      "num_tokens": 67284017.0,
      "step": 32420
    },
    {
      "entropy": 5.569850015640259,
      "epoch": 2.941309869375907,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041419256875239275,
      "loss": 4.8994,
      "mean_token_accuracy": 0.22171291708946228,
      "num_tokens": 67293674.0,
      "step": 32425
    },
    {
      "entropy": 5.57862958908081,
      "epoch": 2.941763425253991,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004141671455345101,
      "loss": 4.9419,
      "mean_token_accuracy": 0.21450210213661194,
      "num_tokens": 67303759.0,
      "step": 32430
    },
    {
      "entropy": 5.676352024078369,
      "epoch": 2.9422169811320753,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041414171943861543,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2144734889268875,
      "num_tokens": 67314962.0,
      "step": 32435
    },
    {
      "entropy": 5.679864740371704,
      "epoch": 2.9426705370101596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004141162904652346,
      "loss": 5.0533,
      "mean_token_accuracy": 0.20631122291088105,
      "num_tokens": 67325210.0,
      "step": 32440
    },
    {
      "entropy": 5.6194719791412355,
      "epoch": 2.943124092888244,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004140908586148934,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21291278451681137,
      "num_tokens": 67335110.0,
      "step": 32445
    },
    {
      "entropy": 5.638901090621948,
      "epoch": 2.943577648766328,
      "grad_norm": 1.125,
      "learning_rate": 0.0004140654238881178,
      "loss": 5.0435,
      "mean_token_accuracy": 0.2125014528632164,
      "num_tokens": 67345872.0,
      "step": 32450
    },
    {
      "entropy": 5.637533044815063,
      "epoch": 2.9440312046444124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041403998628543375,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21328679472208023,
      "num_tokens": 67356532.0,
      "step": 32455
    },
    {
      "entropy": 5.622058010101318,
      "epoch": 2.9444847605224966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041401454580736744,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2226173236966133,
      "num_tokens": 67366603.0,
      "step": 32460
    },
    {
      "entropy": 5.708301973342896,
      "epoch": 2.9449383164005805,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004139891024544448,
      "loss": 5.047,
      "mean_token_accuracy": 0.21689256876707078,
      "num_tokens": 67377487.0,
      "step": 32465
    },
    {
      "entropy": 5.688707780838013,
      "epoch": 2.9453918722786647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004139636562271921,
      "loss": 5.0289,
      "mean_token_accuracy": 0.21270985156297684,
      "num_tokens": 67387704.0,
      "step": 32470
    },
    {
      "entropy": 5.662493801116943,
      "epoch": 2.945845428156749,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041393820712613567,
      "loss": 4.9919,
      "mean_token_accuracy": 0.2091204598546028,
      "num_tokens": 67398409.0,
      "step": 32475
    },
    {
      "entropy": 5.63400411605835,
      "epoch": 2.9462989840348333,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041391275515180166,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21090206652879714,
      "num_tokens": 67408905.0,
      "step": 32480
    },
    {
      "entropy": 5.625855731964111,
      "epoch": 2.946752539912917,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004138873003047165,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22361928671598436,
      "num_tokens": 67418466.0,
      "step": 32485
    },
    {
      "entropy": 5.542302846908569,
      "epoch": 2.9472060957910013,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004138618425854066,
      "loss": 4.931,
      "mean_token_accuracy": 0.2166370004415512,
      "num_tokens": 67428597.0,
      "step": 32490
    },
    {
      "entropy": 5.594061994552613,
      "epoch": 2.9476596516690856,
      "grad_norm": 1.25,
      "learning_rate": 0.00041383638199439824,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21714912801980973,
      "num_tokens": 67439158.0,
      "step": 32495
    },
    {
      "entropy": 5.665826368331909,
      "epoch": 2.94811320754717,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041381091853221835,
      "loss": 4.9617,
      "mean_token_accuracy": 0.21099960058927536,
      "num_tokens": 67449540.0,
      "step": 32500
    },
    {
      "entropy": 5.611739921569824,
      "epoch": 2.948566763425254,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004137854521993932,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21903390288352967,
      "num_tokens": 67459385.0,
      "step": 32505
    },
    {
      "entropy": 5.643102741241455,
      "epoch": 2.9490203193033384,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041375998299644946,
      "loss": 4.9932,
      "mean_token_accuracy": 0.21498721092939377,
      "num_tokens": 67469597.0,
      "step": 32510
    },
    {
      "entropy": 5.61246542930603,
      "epoch": 2.9494738751814222,
      "grad_norm": 1.25,
      "learning_rate": 0.000413734510923914,
      "loss": 4.8602,
      "mean_token_accuracy": 0.21646612882614136,
      "num_tokens": 67479271.0,
      "step": 32515
    },
    {
      "entropy": 5.577236032485962,
      "epoch": 2.9499274310595065,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004137090359823135,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2210137739777565,
      "num_tokens": 67488997.0,
      "step": 32520
    },
    {
      "entropy": 5.594775915145874,
      "epoch": 2.9503809869375908,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041368355817217474,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21761196702718735,
      "num_tokens": 67499507.0,
      "step": 32525
    },
    {
      "entropy": 5.623895502090454,
      "epoch": 2.950834542815675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041365807749402467,
      "loss": 4.8751,
      "mean_token_accuracy": 0.22030483037233353,
      "num_tokens": 67509916.0,
      "step": 32530
    },
    {
      "entropy": 5.613157272338867,
      "epoch": 2.951288098693759,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004136325939483902,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2098008692264557,
      "num_tokens": 67519759.0,
      "step": 32535
    },
    {
      "entropy": 5.614022159576416,
      "epoch": 2.951741654571843,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004136071075357984,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21799196153879166,
      "num_tokens": 67529261.0,
      "step": 32540
    },
    {
      "entropy": 5.619625425338745,
      "epoch": 2.9521952104499274,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004135816182567763,
      "loss": 4.9213,
      "mean_token_accuracy": 0.21682696789503098,
      "num_tokens": 67539084.0,
      "step": 32545
    },
    {
      "entropy": 5.662848567962646,
      "epoch": 2.9526487663280117,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000413556126111851,
      "loss": 4.9778,
      "mean_token_accuracy": 0.2152765229344368,
      "num_tokens": 67548551.0,
      "step": 32550
    },
    {
      "entropy": 5.618746423721314,
      "epoch": 2.953102322206096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004135306311015497,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21302825957536697,
      "num_tokens": 67558364.0,
      "step": 32555
    },
    {
      "entropy": 5.660374736785888,
      "epoch": 2.95355587808418,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004135051332263996,
      "loss": 4.9221,
      "mean_token_accuracy": 0.2231972724199295,
      "num_tokens": 67569257.0,
      "step": 32560
    },
    {
      "entropy": 5.613333082199096,
      "epoch": 2.954009433962264,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041347963248692814,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20575937181711196,
      "num_tokens": 67580135.0,
      "step": 32565
    },
    {
      "entropy": 5.599939680099487,
      "epoch": 2.9544629898403483,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004134541288836626,
      "loss": 4.9353,
      "mean_token_accuracy": 0.21489370316267015,
      "num_tokens": 67589705.0,
      "step": 32570
    },
    {
      "entropy": 5.644237804412842,
      "epoch": 2.9549165457184325,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041342862241713034,
      "loss": 4.89,
      "mean_token_accuracy": 0.2166994646191597,
      "num_tokens": 67599648.0,
      "step": 32575
    },
    {
      "entropy": 5.662343168258667,
      "epoch": 2.955370101596517,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041340311308785886,
      "loss": 4.8951,
      "mean_token_accuracy": 0.21581293493509293,
      "num_tokens": 67609624.0,
      "step": 32580
    },
    {
      "entropy": 5.5733386993408205,
      "epoch": 2.9558236574746006,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041337760089637577,
      "loss": 4.9252,
      "mean_token_accuracy": 0.2156374931335449,
      "num_tokens": 67619596.0,
      "step": 32585
    },
    {
      "entropy": 5.599661159515381,
      "epoch": 2.956277213352685,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041335208584320865,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2136589229106903,
      "num_tokens": 67629667.0,
      "step": 32590
    },
    {
      "entropy": 5.674688863754272,
      "epoch": 2.956730769230769,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004133265679288851,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2157885566353798,
      "num_tokens": 67640048.0,
      "step": 32595
    },
    {
      "entropy": 5.688122081756592,
      "epoch": 2.9571843251088534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004133010471539329,
      "loss": 5.0098,
      "mean_token_accuracy": 0.21059911847114562,
      "num_tokens": 67650731.0,
      "step": 32600
    },
    {
      "entropy": 5.597215557098389,
      "epoch": 2.9576378809869377,
      "grad_norm": 1.125,
      "learning_rate": 0.00041327552351887983,
      "loss": 4.9273,
      "mean_token_accuracy": 0.2153029352426529,
      "num_tokens": 67661696.0,
      "step": 32605
    },
    {
      "entropy": 5.62748908996582,
      "epoch": 2.958091436865022,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041324999702425366,
      "loss": 4.9533,
      "mean_token_accuracy": 0.2140968382358551,
      "num_tokens": 67671758.0,
      "step": 32610
    },
    {
      "entropy": 5.585301160812378,
      "epoch": 2.9585449927431062,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004132244676705823,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21776215583086014,
      "num_tokens": 67682488.0,
      "step": 32615
    },
    {
      "entropy": 5.675340986251831,
      "epoch": 2.95899854862119,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004131989354583938,
      "loss": 4.9496,
      "mean_token_accuracy": 0.21965046972036362,
      "num_tokens": 67692945.0,
      "step": 32620
    },
    {
      "entropy": 5.641361427307129,
      "epoch": 2.9594521044992743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000413173400388216,
      "loss": 4.8613,
      "mean_token_accuracy": 0.21642902493476868,
      "num_tokens": 67702924.0,
      "step": 32625
    },
    {
      "entropy": 5.6127790927886965,
      "epoch": 2.9599056603773586,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004131478624605771,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21458469033241273,
      "num_tokens": 67714284.0,
      "step": 32630
    },
    {
      "entropy": 5.637450075149536,
      "epoch": 2.9603592162554424,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004131223216760052,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20886660665273665,
      "num_tokens": 67725902.0,
      "step": 32635
    },
    {
      "entropy": 5.65799617767334,
      "epoch": 2.9608127721335267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004130967780350286,
      "loss": 4.921,
      "mean_token_accuracy": 0.21364203691482545,
      "num_tokens": 67736268.0,
      "step": 32640
    },
    {
      "entropy": 5.710126209259033,
      "epoch": 2.961266328011611,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004130712315381753,
      "loss": 4.962,
      "mean_token_accuracy": 0.22410699874162673,
      "num_tokens": 67745709.0,
      "step": 32645
    },
    {
      "entropy": 5.6640753746032715,
      "epoch": 2.961719883889695,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041304568218597385,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21398856341838837,
      "num_tokens": 67756092.0,
      "step": 32650
    },
    {
      "entropy": 5.646423768997193,
      "epoch": 2.9621734397677795,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041302012997895246,
      "loss": 4.9561,
      "mean_token_accuracy": 0.21489520072937013,
      "num_tokens": 67766224.0,
      "step": 32655
    },
    {
      "entropy": 5.70157380104065,
      "epoch": 2.9626269956458637,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004129945749176395,
      "loss": 5.0903,
      "mean_token_accuracy": 0.21241241246461867,
      "num_tokens": 67776787.0,
      "step": 32660
    },
    {
      "entropy": 5.62306957244873,
      "epoch": 2.963080551523948,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041296901700256375,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21550153195858002,
      "num_tokens": 67787924.0,
      "step": 32665
    },
    {
      "entropy": 5.6585205554962155,
      "epoch": 2.963534107402032,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004129434562342535,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2023211821913719,
      "num_tokens": 67798544.0,
      "step": 32670
    },
    {
      "entropy": 5.589351844787598,
      "epoch": 2.963987663280116,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041291789261323734,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21614918112754822,
      "num_tokens": 67809528.0,
      "step": 32675
    },
    {
      "entropy": 5.516956758499146,
      "epoch": 2.9644412191582004,
      "grad_norm": 1.125,
      "learning_rate": 0.00041289232614004416,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23332008570432664,
      "num_tokens": 67820263.0,
      "step": 32680
    },
    {
      "entropy": 5.7073241710662845,
      "epoch": 2.9648947750362846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004128667568152024,
      "loss": 5.0145,
      "mean_token_accuracy": 0.20994754433631896,
      "num_tokens": 67830687.0,
      "step": 32685
    },
    {
      "entropy": 5.742996454238892,
      "epoch": 2.9653483309143684,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041284118463924107,
      "loss": 5.0012,
      "mean_token_accuracy": 0.20630993843078613,
      "num_tokens": 67841714.0,
      "step": 32690
    },
    {
      "entropy": 5.59776291847229,
      "epoch": 2.9658018867924527,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004128156096126889,
      "loss": 4.9308,
      "mean_token_accuracy": 0.2160676062107086,
      "num_tokens": 67851952.0,
      "step": 32695
    },
    {
      "entropy": 5.5455316543579105,
      "epoch": 2.966255442670537,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041279003173607477,
      "loss": 4.8196,
      "mean_token_accuracy": 0.21863433271646499,
      "num_tokens": 67861958.0,
      "step": 32700
    },
    {
      "entropy": 5.569171380996704,
      "epoch": 2.9667089985486212,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004127644510099278,
      "loss": 4.8125,
      "mean_token_accuracy": 0.22394121438264847,
      "num_tokens": 67871654.0,
      "step": 32705
    },
    {
      "entropy": 5.570980644226074,
      "epoch": 2.9671625544267055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004127388674347767,
      "loss": 4.9057,
      "mean_token_accuracy": 0.22241017520427703,
      "num_tokens": 67881678.0,
      "step": 32710
    },
    {
      "entropy": 5.5784557342529295,
      "epoch": 2.96761611030479,
      "grad_norm": 1.125,
      "learning_rate": 0.00041271328101115075,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21070809960365294,
      "num_tokens": 67892663.0,
      "step": 32715
    },
    {
      "entropy": 5.742793178558349,
      "epoch": 2.9680696661828736,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041268769173957915,
      "loss": 5.0946,
      "mean_token_accuracy": 0.20620316565036773,
      "num_tokens": 67903308.0,
      "step": 32720
    },
    {
      "entropy": 5.677955484390258,
      "epoch": 2.968523222060958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041266209962059097,
      "loss": 4.913,
      "mean_token_accuracy": 0.21964753568172454,
      "num_tokens": 67913092.0,
      "step": 32725
    },
    {
      "entropy": 5.618656778335572,
      "epoch": 2.968976777939042,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004126365046547154,
      "loss": 4.9118,
      "mean_token_accuracy": 0.22545998841524123,
      "num_tokens": 67923708.0,
      "step": 32730
    },
    {
      "entropy": 5.556598711013794,
      "epoch": 2.9694303338171264,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004126109068424819,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2105742424726486,
      "num_tokens": 67934164.0,
      "step": 32735
    },
    {
      "entropy": 5.669560670852661,
      "epoch": 2.96988388969521,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004125853061844198,
      "loss": 5.0164,
      "mean_token_accuracy": 0.20944148451089858,
      "num_tokens": 67943162.0,
      "step": 32740
    },
    {
      "entropy": 5.681047248840332,
      "epoch": 2.9703374455732945,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004125597026810585,
      "loss": 5.014,
      "mean_token_accuracy": 0.21345773637294768,
      "num_tokens": 67953943.0,
      "step": 32745
    },
    {
      "entropy": 5.580320644378662,
      "epoch": 2.9707910014513788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041253409633292747,
      "loss": 4.907,
      "mean_token_accuracy": 0.21638807654380798,
      "num_tokens": 67964603.0,
      "step": 32750
    },
    {
      "entropy": 5.653650856018066,
      "epoch": 2.971244557329463,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041250848714055626,
      "loss": 4.9801,
      "mean_token_accuracy": 0.21489916741847992,
      "num_tokens": 67974541.0,
      "step": 32755
    },
    {
      "entropy": 5.652291536331177,
      "epoch": 2.9716981132075473,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004124828751044746,
      "loss": 4.9515,
      "mean_token_accuracy": 0.20700111091136933,
      "num_tokens": 67984968.0,
      "step": 32760
    },
    {
      "entropy": 5.621393728256225,
      "epoch": 2.9721516690856316,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004124572602252119,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2190841406583786,
      "num_tokens": 67995567.0,
      "step": 32765
    },
    {
      "entropy": 5.674803781509399,
      "epoch": 2.9726052249637154,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004124316425032981,
      "loss": 5.0553,
      "mean_token_accuracy": 0.20576776564121246,
      "num_tokens": 68008530.0,
      "step": 32770
    },
    {
      "entropy": 5.708988618850708,
      "epoch": 2.9730587808417996,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041240602193926294,
      "loss": 5.014,
      "mean_token_accuracy": 0.21019861698150635,
      "num_tokens": 68018393.0,
      "step": 32775
    },
    {
      "entropy": 5.693976020812988,
      "epoch": 2.973512336719884,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004123803985336362,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21103900372982026,
      "num_tokens": 68027677.0,
      "step": 32780
    },
    {
      "entropy": 5.616236925125122,
      "epoch": 2.973965892597968,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041235477228694786,
      "loss": 4.8312,
      "mean_token_accuracy": 0.22508990317583083,
      "num_tokens": 68037157.0,
      "step": 32785
    },
    {
      "entropy": 5.622634601593018,
      "epoch": 2.974419448476052,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041232914319972775,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22021604776382447,
      "num_tokens": 68047793.0,
      "step": 32790
    },
    {
      "entropy": 5.627812910079956,
      "epoch": 2.9748730043541363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041230351127250603,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21828629225492477,
      "num_tokens": 68058155.0,
      "step": 32795
    },
    {
      "entropy": 5.668702793121338,
      "epoch": 2.9753265602322205,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041227787650581266,
      "loss": 4.8697,
      "mean_token_accuracy": 0.22094268053770066,
      "num_tokens": 68067440.0,
      "step": 32800
    },
    {
      "entropy": 5.63725643157959,
      "epoch": 2.975780116110305,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041225223890017785,
      "loss": 4.9663,
      "mean_token_accuracy": 0.21871229112148285,
      "num_tokens": 68077815.0,
      "step": 32805
    },
    {
      "entropy": 5.656004333496094,
      "epoch": 2.976233671988389,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041222659845613173,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2228584349155426,
      "num_tokens": 68087692.0,
      "step": 32810
    },
    {
      "entropy": 5.583959674835205,
      "epoch": 2.9766872278664733,
      "grad_norm": 1.25,
      "learning_rate": 0.00041220095517420464,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21812089830636977,
      "num_tokens": 68097774.0,
      "step": 32815
    },
    {
      "entropy": 5.606702470779419,
      "epoch": 2.9771407837445576,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004121753090549268,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2166294664144516,
      "num_tokens": 68107992.0,
      "step": 32820
    },
    {
      "entropy": 5.588299131393432,
      "epoch": 2.9775943396226414,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041214966009882857,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22164697498083114,
      "num_tokens": 68117103.0,
      "step": 32825
    },
    {
      "entropy": 5.689782953262329,
      "epoch": 2.9780478955007257,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004121240083064405,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21295422017574311,
      "num_tokens": 68127951.0,
      "step": 32830
    },
    {
      "entropy": 5.694617986679077,
      "epoch": 2.97850145137881,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041209835367829293,
      "loss": 5.0053,
      "mean_token_accuracy": 0.21347694098949432,
      "num_tokens": 68138229.0,
      "step": 32835
    },
    {
      "entropy": 5.555681800842285,
      "epoch": 2.9789550072568938,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041207269621491654,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21759978234767913,
      "num_tokens": 68148019.0,
      "step": 32840
    },
    {
      "entropy": 5.567492580413818,
      "epoch": 2.979408563134978,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004120470359168418,
      "loss": 4.9352,
      "mean_token_accuracy": 0.20564557015895843,
      "num_tokens": 68158704.0,
      "step": 32845
    },
    {
      "entropy": 5.6813959121704105,
      "epoch": 2.9798621190130623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041202137278459946,
      "loss": 4.946,
      "mean_token_accuracy": 0.2196308732032776,
      "num_tokens": 68169105.0,
      "step": 32850
    },
    {
      "entropy": 5.6532128810882565,
      "epoch": 2.9803156748911466,
      "grad_norm": 1.125,
      "learning_rate": 0.0004119957068187202,
      "loss": 4.8721,
      "mean_token_accuracy": 0.21854404211044312,
      "num_tokens": 68179778.0,
      "step": 32855
    },
    {
      "entropy": 5.623121643066407,
      "epoch": 2.980769230769231,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004119700380197348,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21520705074071883,
      "num_tokens": 68189405.0,
      "step": 32860
    },
    {
      "entropy": 5.511805057525635,
      "epoch": 2.981222786647315,
      "grad_norm": 1.25,
      "learning_rate": 0.00041194436638817417,
      "loss": 4.8594,
      "mean_token_accuracy": 0.2265425905585289,
      "num_tokens": 68199351.0,
      "step": 32865
    },
    {
      "entropy": 5.663418102264404,
      "epoch": 2.9816763425253994,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041191869192456907,
      "loss": 5.0217,
      "mean_token_accuracy": 0.21237728297710418,
      "num_tokens": 68209904.0,
      "step": 32870
    },
    {
      "entropy": 5.685230398178101,
      "epoch": 2.982129898403483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004118930146294506,
      "loss": 5.0164,
      "mean_token_accuracy": 0.2111487627029419,
      "num_tokens": 68221754.0,
      "step": 32875
    },
    {
      "entropy": 5.657987499237061,
      "epoch": 2.9825834542815675,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004118673345033496,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2101563557982445,
      "num_tokens": 68232328.0,
      "step": 32880
    },
    {
      "entropy": 5.56987247467041,
      "epoch": 2.9830370101596517,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004118416515467974,
      "loss": 4.8615,
      "mean_token_accuracy": 0.2152936726808548,
      "num_tokens": 68242812.0,
      "step": 32885
    },
    {
      "entropy": 5.6841230392456055,
      "epoch": 2.983490566037736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004118159657603248,
      "loss": 5.0067,
      "mean_token_accuracy": 0.20781323462724685,
      "num_tokens": 68253888.0,
      "step": 32890
    },
    {
      "entropy": 5.678138589859008,
      "epoch": 2.98394412191582,
      "grad_norm": 1.25,
      "learning_rate": 0.0004117902771444633,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21844129264354706,
      "num_tokens": 68264877.0,
      "step": 32895
    },
    {
      "entropy": 5.731426858901978,
      "epoch": 2.984397677793904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000411764585699744,
      "loss": 5.0304,
      "mean_token_accuracy": 0.204776369035244,
      "num_tokens": 68275863.0,
      "step": 32900
    },
    {
      "entropy": 5.671324920654297,
      "epoch": 2.9848512336719883,
      "grad_norm": 1.125,
      "learning_rate": 0.00041173889142669806,
      "loss": 4.952,
      "mean_token_accuracy": 0.21256603598594664,
      "num_tokens": 68286768.0,
      "step": 32905
    },
    {
      "entropy": 5.598630094528199,
      "epoch": 2.9853047895500726,
      "grad_norm": 1.125,
      "learning_rate": 0.0004117131943258572,
      "loss": 4.889,
      "mean_token_accuracy": 0.22051786929368972,
      "num_tokens": 68296990.0,
      "step": 32910
    },
    {
      "entropy": 5.540438270568847,
      "epoch": 2.985758345428157,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041168749439775255,
      "loss": 4.8733,
      "mean_token_accuracy": 0.21655682027339934,
      "num_tokens": 68307646.0,
      "step": 32915
    },
    {
      "entropy": 5.670006847381591,
      "epoch": 2.986211901306241,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004116617916429156,
      "loss": 5.0244,
      "mean_token_accuracy": 0.2048196941614151,
      "num_tokens": 68319586.0,
      "step": 32920
    },
    {
      "entropy": 5.736089468002319,
      "epoch": 2.986665457184325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004116360860618781,
      "loss": 4.9894,
      "mean_token_accuracy": 0.2112579584121704,
      "num_tokens": 68329692.0,
      "step": 32925
    },
    {
      "entropy": 5.69001579284668,
      "epoch": 2.9871190130624092,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004116103776551715,
      "loss": 4.9471,
      "mean_token_accuracy": 0.2142190381884575,
      "num_tokens": 68339589.0,
      "step": 32930
    },
    {
      "entropy": 5.647755146026611,
      "epoch": 2.9875725689404935,
      "grad_norm": 1.125,
      "learning_rate": 0.00041158466642332743,
      "loss": 5.012,
      "mean_token_accuracy": 0.20411770343780516,
      "num_tokens": 68349485.0,
      "step": 32935
    },
    {
      "entropy": 5.647630310058593,
      "epoch": 2.9880261248185778,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004115589523668777,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22176818698644638,
      "num_tokens": 68359695.0,
      "step": 32940
    },
    {
      "entropy": 5.584457778930664,
      "epoch": 2.9884796806966616,
      "grad_norm": 1.125,
      "learning_rate": 0.000411533235486354,
      "loss": 4.8175,
      "mean_token_accuracy": 0.2219352290034294,
      "num_tokens": 68369591.0,
      "step": 32945
    },
    {
      "entropy": 5.638293361663818,
      "epoch": 2.988933236574746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004115075157822883,
      "loss": 4.9651,
      "mean_token_accuracy": 0.21667128652334214,
      "num_tokens": 68380382.0,
      "step": 32950
    },
    {
      "entropy": 5.5126505374908445,
      "epoch": 2.98938679245283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004114817932552122,
      "loss": 4.8299,
      "mean_token_accuracy": 0.22157813012599945,
      "num_tokens": 68389865.0,
      "step": 32955
    },
    {
      "entropy": 5.587721443176269,
      "epoch": 2.9898403483309144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041145606790565797,
      "loss": 4.848,
      "mean_token_accuracy": 0.22686622738838197,
      "num_tokens": 68399267.0,
      "step": 32960
    },
    {
      "entropy": 5.6585770606994625,
      "epoch": 2.9902939042089987,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004114303397341575,
      "loss": 4.9677,
      "mean_token_accuracy": 0.21847444921731948,
      "num_tokens": 68409480.0,
      "step": 32965
    },
    {
      "entropy": 5.759487581253052,
      "epoch": 2.990747460087083,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004114046087412427,
      "loss": 5.0621,
      "mean_token_accuracy": 0.21091356724500657,
      "num_tokens": 68420019.0,
      "step": 32970
    },
    {
      "entropy": 5.5844038963317875,
      "epoch": 2.9912010159651667,
      "grad_norm": 1.25,
      "learning_rate": 0.000411378874927446,
      "loss": 4.8916,
      "mean_token_accuracy": 0.22049353122711182,
      "num_tokens": 68429967.0,
      "step": 32975
    },
    {
      "entropy": 5.568133401870727,
      "epoch": 2.991654571843251,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004113531382932993,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2174044206738472,
      "num_tokens": 68441639.0,
      "step": 32980
    },
    {
      "entropy": 5.680133819580078,
      "epoch": 2.9921081277213353,
      "grad_norm": 1.09375,
      "learning_rate": 0.000411327398839335,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20901804119348527,
      "num_tokens": 68452781.0,
      "step": 32985
    },
    {
      "entropy": 5.646609878540039,
      "epoch": 2.9925616835994195,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004113016565660854,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2187865272164345,
      "num_tokens": 68463353.0,
      "step": 32990
    },
    {
      "entropy": 5.55716233253479,
      "epoch": 2.9930152394775034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041127591147408274,
      "loss": 4.907,
      "mean_token_accuracy": 0.20799251645803452,
      "num_tokens": 68473507.0,
      "step": 32995
    },
    {
      "entropy": 5.619699001312256,
      "epoch": 2.9934687953555876,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041125016356385945,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21613216400146484,
      "num_tokens": 68483737.0,
      "step": 33000
    },
    {
      "epoch": 2.9934687953555876,
      "eval_entropy": 5.388790743167584,
      "eval_loss": 4.98230504989624,
      "eval_mean_token_accuracy": 0.22280678409257088,
      "eval_num_tokens": 68483737.0,
      "eval_runtime": 20.6588,
      "eval_samples_per_second": 1711.617,
      "eval_steps_per_second": 213.952,
      "step": 33000
    },
    {
      "entropy": 5.655807256698608,
      "epoch": 2.993922351233672,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041122441283594817,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2172561466693878,
      "num_tokens": 68494013.0,
      "step": 33005
    },
    {
      "entropy": 5.604366636276245,
      "epoch": 2.994375907111756,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004111986592908813,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22452555745840072,
      "num_tokens": 68504361.0,
      "step": 33010
    },
    {
      "entropy": 5.587920141220093,
      "epoch": 2.9948294629898404,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041117290292919134,
      "loss": 4.8609,
      "mean_token_accuracy": 0.221909761428833,
      "num_tokens": 68514825.0,
      "step": 33015
    },
    {
      "entropy": 5.6046061515808105,
      "epoch": 2.9952830188679247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004111471437514111,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21499638706445695,
      "num_tokens": 68524858.0,
      "step": 33020
    },
    {
      "entropy": 5.599625158309936,
      "epoch": 2.995736574746009,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004111213817580733,
      "loss": 4.8385,
      "mean_token_accuracy": 0.2205299213528633,
      "num_tokens": 68534481.0,
      "step": 33025
    },
    {
      "entropy": 5.585771274566651,
      "epoch": 2.996190130624093,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004110956169497106,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21296273022890091,
      "num_tokens": 68545436.0,
      "step": 33030
    },
    {
      "entropy": 5.583476209640503,
      "epoch": 2.996643686502177,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004110698493268558,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22294741868972778,
      "num_tokens": 68555434.0,
      "step": 33035
    },
    {
      "entropy": 5.642462682723999,
      "epoch": 2.9970972423802613,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004110440788900419,
      "loss": 5.0442,
      "mean_token_accuracy": 0.21189566403627397,
      "num_tokens": 68566295.0,
      "step": 33040
    },
    {
      "entropy": 5.577358722686768,
      "epoch": 2.9975507982583456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041101830563980174,
      "loss": 4.8539,
      "mean_token_accuracy": 0.2244438037276268,
      "num_tokens": 68575927.0,
      "step": 33045
    },
    {
      "entropy": 5.668047857284546,
      "epoch": 2.9980043541364294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041099252957666843,
      "loss": 4.9653,
      "mean_token_accuracy": 0.2177191987633705,
      "num_tokens": 68586254.0,
      "step": 33050
    },
    {
      "entropy": 5.565159559249878,
      "epoch": 2.9984579100145137,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041096675070117496,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21760228872299195,
      "num_tokens": 68596881.0,
      "step": 33055
    },
    {
      "entropy": 5.556361150741577,
      "epoch": 2.998911465892598,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041094096901385434,
      "loss": 4.8287,
      "mean_token_accuracy": 0.22269561588764192,
      "num_tokens": 68606875.0,
      "step": 33060
    },
    {
      "entropy": 5.67962532043457,
      "epoch": 2.999365021770682,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004109151845152399,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21584342271089554,
      "num_tokens": 68617011.0,
      "step": 33065
    },
    {
      "entropy": 5.678834629058838,
      "epoch": 2.9998185776487665,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004108893972058649,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21373047977685927,
      "num_tokens": 68627764.0,
      "step": 33070
    },
    {
      "entropy": 5.660772180557251,
      "epoch": 3.0002721335268503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041086360708626236,
      "loss": 4.9232,
      "mean_token_accuracy": 0.2192634165287018,
      "num_tokens": 68637272.0,
      "step": 33075
    },
    {
      "entropy": 5.565616655349731,
      "epoch": 3.0007256894049346,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041083781415696585,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21277209371328354,
      "num_tokens": 68648058.0,
      "step": 33080
    },
    {
      "entropy": 5.5635254859924315,
      "epoch": 3.001179245283019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004108120184185088,
      "loss": 4.805,
      "mean_token_accuracy": 0.2236357167363167,
      "num_tokens": 68658174.0,
      "step": 33085
    },
    {
      "entropy": 5.637840366363525,
      "epoch": 3.001632801161103,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041078621987142457,
      "loss": 4.8467,
      "mean_token_accuracy": 0.22610540241003035,
      "num_tokens": 68669281.0,
      "step": 33090
    },
    {
      "entropy": 5.590885257720947,
      "epoch": 3.0020863570391874,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004107604185162467,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21254901438951493,
      "num_tokens": 68679879.0,
      "step": 33095
    },
    {
      "entropy": 5.525386905670166,
      "epoch": 3.0025399129172716,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041073461435350877,
      "loss": 4.8058,
      "mean_token_accuracy": 0.2221756562590599,
      "num_tokens": 68689852.0,
      "step": 33100
    },
    {
      "entropy": 5.553326749801636,
      "epoch": 3.0029934687953554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041070880738374447,
      "loss": 4.879,
      "mean_token_accuracy": 0.21886124163866044,
      "num_tokens": 68700149.0,
      "step": 33105
    },
    {
      "entropy": 5.597825431823731,
      "epoch": 3.0034470246734397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041068299760748734,
      "loss": 4.8719,
      "mean_token_accuracy": 0.22189686000347136,
      "num_tokens": 68710837.0,
      "step": 33110
    },
    {
      "entropy": 5.614534330368042,
      "epoch": 3.003900580551524,
      "grad_norm": 1.125,
      "learning_rate": 0.00041065718502527134,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2179928243160248,
      "num_tokens": 68721451.0,
      "step": 33115
    },
    {
      "entropy": 5.578591299057007,
      "epoch": 3.0043541364296082,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041063136963763013,
      "loss": 4.8614,
      "mean_token_accuracy": 0.22229890376329423,
      "num_tokens": 68732911.0,
      "step": 33120
    },
    {
      "entropy": 5.595605134963989,
      "epoch": 3.0048076923076925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004106055514450977,
      "loss": 4.8231,
      "mean_token_accuracy": 0.22759455889463426,
      "num_tokens": 68742650.0,
      "step": 33125
    },
    {
      "entropy": 5.5638525009155275,
      "epoch": 3.0052612481857763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004105797304482079,
      "loss": 4.8328,
      "mean_token_accuracy": 0.22094301879405975,
      "num_tokens": 68753660.0,
      "step": 33130
    },
    {
      "entropy": 5.568468713760376,
      "epoch": 3.0057148040638606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004105539066474947,
      "loss": 4.762,
      "mean_token_accuracy": 0.22654962837696074,
      "num_tokens": 68763868.0,
      "step": 33135
    },
    {
      "entropy": 5.59303650856018,
      "epoch": 3.006168359941945,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041052808004349214,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22007373422384263,
      "num_tokens": 68773911.0,
      "step": 33140
    },
    {
      "entropy": 5.642910814285278,
      "epoch": 3.006621915820029,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004105022506367344,
      "loss": 4.9678,
      "mean_token_accuracy": 0.21345291584730147,
      "num_tokens": 68786423.0,
      "step": 33145
    },
    {
      "entropy": 5.635404300689697,
      "epoch": 3.0070754716981134,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004104764184277555,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21514613479375838,
      "num_tokens": 68797329.0,
      "step": 33150
    },
    {
      "entropy": 5.666480016708374,
      "epoch": 3.0075290275761972,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004104505834170898,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21784710884094238,
      "num_tokens": 68808447.0,
      "step": 33155
    },
    {
      "entropy": 5.636412954330444,
      "epoch": 3.0079825834542815,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041042474560527153,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2221631035208702,
      "num_tokens": 68819381.0,
      "step": 33160
    },
    {
      "entropy": 5.545153617858887,
      "epoch": 3.0084361393323658,
      "grad_norm": 1.1875,
      "learning_rate": 0.000410398904992835,
      "loss": 4.8246,
      "mean_token_accuracy": 0.22535058856010437,
      "num_tokens": 68829319.0,
      "step": 33165
    },
    {
      "entropy": 5.599809694290161,
      "epoch": 3.00888969521045,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041037306158031464,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22253966927528382,
      "num_tokens": 68838640.0,
      "step": 33170
    },
    {
      "entropy": 5.646754407882691,
      "epoch": 3.0093432510885343,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004103472153682448,
      "loss": 4.8637,
      "mean_token_accuracy": 0.21811421811580659,
      "num_tokens": 68847960.0,
      "step": 33175
    },
    {
      "entropy": 5.618949508666992,
      "epoch": 3.009796806966618,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041032136635716016,
      "loss": 4.8798,
      "mean_token_accuracy": 0.21817819625139237,
      "num_tokens": 68858851.0,
      "step": 33180
    },
    {
      "entropy": 5.5537933826446535,
      "epoch": 3.0102503628447024,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004102955145475951,
      "loss": 4.8448,
      "mean_token_accuracy": 0.2198275536298752,
      "num_tokens": 68869006.0,
      "step": 33185
    },
    {
      "entropy": 5.689778518676758,
      "epoch": 3.0107039187227866,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041026965994008435,
      "loss": 4.9725,
      "mean_token_accuracy": 0.21148451864719392,
      "num_tokens": 68879606.0,
      "step": 33190
    },
    {
      "entropy": 5.561501836776733,
      "epoch": 3.011157474600871,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041024380253516253,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23179419040679933,
      "num_tokens": 68889862.0,
      "step": 33195
    },
    {
      "entropy": 5.719213008880615,
      "epoch": 3.011611030478955,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041021794233336444,
      "loss": 4.9491,
      "mean_token_accuracy": 0.21639616787433624,
      "num_tokens": 68900512.0,
      "step": 33200
    },
    {
      "entropy": 5.622331953048706,
      "epoch": 3.012064586357039,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004101920793352249,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21384023427963256,
      "num_tokens": 68910475.0,
      "step": 33205
    },
    {
      "entropy": 5.673051500320435,
      "epoch": 3.0125181422351233,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004101662135412786,
      "loss": 4.9297,
      "mean_token_accuracy": 0.21494946479797364,
      "num_tokens": 68920583.0,
      "step": 33210
    },
    {
      "entropy": 5.543615484237671,
      "epoch": 3.0129716981132075,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004101403449520606,
      "loss": 4.8092,
      "mean_token_accuracy": 0.2217329666018486,
      "num_tokens": 68931016.0,
      "step": 33215
    },
    {
      "entropy": 5.577709531784057,
      "epoch": 3.013425253991292,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041011447356810584,
      "loss": 4.8417,
      "mean_token_accuracy": 0.21917263716459273,
      "num_tokens": 68941489.0,
      "step": 33220
    },
    {
      "entropy": 5.656309318542481,
      "epoch": 3.013878809869376,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004100885993899493,
      "loss": 4.9288,
      "mean_token_accuracy": 0.21367194205522538,
      "num_tokens": 68952322.0,
      "step": 33225
    },
    {
      "entropy": 5.638419771194458,
      "epoch": 3.01433236574746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004100627224181261,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21681924611330033,
      "num_tokens": 68962570.0,
      "step": 33230
    },
    {
      "entropy": 5.649652481079102,
      "epoch": 3.014785921625544,
      "grad_norm": 1.125,
      "learning_rate": 0.00041003684265317143,
      "loss": 4.8534,
      "mean_token_accuracy": 0.22429651468992234,
      "num_tokens": 68973733.0,
      "step": 33235
    },
    {
      "entropy": 5.605364513397217,
      "epoch": 3.0152394775036284,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041001096009562045,
      "loss": 4.8794,
      "mean_token_accuracy": 0.22386542409658433,
      "num_tokens": 68983526.0,
      "step": 33240
    },
    {
      "entropy": 5.704187202453613,
      "epoch": 3.0156930333817127,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004099850747460084,
      "loss": 4.9613,
      "mean_token_accuracy": 0.22057290226221085,
      "num_tokens": 68993968.0,
      "step": 33245
    },
    {
      "entropy": 5.618980503082275,
      "epoch": 3.016146589259797,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040995918660487055,
      "loss": 4.8377,
      "mean_token_accuracy": 0.23325806260108947,
      "num_tokens": 69003599.0,
      "step": 33250
    },
    {
      "entropy": 5.577439785003662,
      "epoch": 3.0166001451378808,
      "grad_norm": 1.25,
      "learning_rate": 0.0004099332956727424,
      "loss": 4.8673,
      "mean_token_accuracy": 0.21647975891828536,
      "num_tokens": 69013686.0,
      "step": 33255
    },
    {
      "entropy": 5.620430660247803,
      "epoch": 3.017053701015965,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004099074019501593,
      "loss": 4.8957,
      "mean_token_accuracy": 0.21302349269390106,
      "num_tokens": 69023803.0,
      "step": 33260
    },
    {
      "entropy": 5.591452693939209,
      "epoch": 3.0175072568940493,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004098815054376567,
      "loss": 4.7861,
      "mean_token_accuracy": 0.22871512919664383,
      "num_tokens": 69033885.0,
      "step": 33265
    },
    {
      "entropy": 5.589730310440063,
      "epoch": 3.0179608127721336,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004098556061357702,
      "loss": 4.8407,
      "mean_token_accuracy": 0.22262195646762847,
      "num_tokens": 69044471.0,
      "step": 33270
    },
    {
      "entropy": 5.543532085418701,
      "epoch": 3.018414368650218,
      "grad_norm": 1.125,
      "learning_rate": 0.00040982970404503544,
      "loss": 4.7505,
      "mean_token_accuracy": 0.22955181151628495,
      "num_tokens": 69055109.0,
      "step": 33275
    },
    {
      "entropy": 5.545413494110107,
      "epoch": 3.018867924528302,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040980379916598803,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22021372020244598,
      "num_tokens": 69064692.0,
      "step": 33280
    },
    {
      "entropy": 5.613268947601318,
      "epoch": 3.019321480406386,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004097778914991637,
      "loss": 4.8574,
      "mean_token_accuracy": 0.22543514519929886,
      "num_tokens": 69075331.0,
      "step": 33285
    },
    {
      "entropy": 5.618462419509887,
      "epoch": 3.01977503628447,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040975198104509827,
      "loss": 4.8593,
      "mean_token_accuracy": 0.22036763429641723,
      "num_tokens": 69085017.0,
      "step": 33290
    },
    {
      "entropy": 5.548881959915161,
      "epoch": 3.0202285921625545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004097260678043275,
      "loss": 4.8678,
      "mean_token_accuracy": 0.21486279517412185,
      "num_tokens": 69096128.0,
      "step": 33295
    },
    {
      "entropy": 5.6427265167236325,
      "epoch": 3.0206821480406387,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040970015177738725,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21123552620410918,
      "num_tokens": 69106479.0,
      "step": 33300
    },
    {
      "entropy": 5.710194063186646,
      "epoch": 3.021135703918723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004096742329648136,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21697250306606292,
      "num_tokens": 69116346.0,
      "step": 33305
    },
    {
      "entropy": 5.629936933517456,
      "epoch": 3.021589259796807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004096483113671425,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22117064744234086,
      "num_tokens": 69126489.0,
      "step": 33310
    },
    {
      "entropy": 5.57441349029541,
      "epoch": 3.022042815674891,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040962238698490994,
      "loss": 4.8652,
      "mean_token_accuracy": 0.2204676628112793,
      "num_tokens": 69136618.0,
      "step": 33315
    },
    {
      "entropy": 5.649873208999634,
      "epoch": 3.0224963715529753,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004095964598186522,
      "loss": 4.92,
      "mean_token_accuracy": 0.21439631134271622,
      "num_tokens": 69146242.0,
      "step": 33320
    },
    {
      "entropy": 5.623918104171753,
      "epoch": 3.0229499274310596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004095705298689053,
      "loss": 4.8537,
      "mean_token_accuracy": 0.21990257501602173,
      "num_tokens": 69157269.0,
      "step": 33325
    },
    {
      "entropy": 5.7010791301727295,
      "epoch": 3.023403483309144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040954459713620563,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21316420137882233,
      "num_tokens": 69169260.0,
      "step": 33330
    },
    {
      "entropy": 5.716715335845947,
      "epoch": 3.0238570391872277,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004095186616210893,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2040004700422287,
      "num_tokens": 69179815.0,
      "step": 33335
    },
    {
      "entropy": 5.6676500797271725,
      "epoch": 3.024310595065312,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004094927233240928,
      "loss": 4.972,
      "mean_token_accuracy": 0.2176794484257698,
      "num_tokens": 69189959.0,
      "step": 33340
    },
    {
      "entropy": 5.628020620346069,
      "epoch": 3.0247641509433962,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040946678224575264,
      "loss": 4.8723,
      "mean_token_accuracy": 0.21981181353330612,
      "num_tokens": 69199485.0,
      "step": 33345
    },
    {
      "entropy": 5.574540615081787,
      "epoch": 3.0252177068214805,
      "grad_norm": 1.3125,
      "learning_rate": 0.000409440838386605,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21889398992061615,
      "num_tokens": 69209180.0,
      "step": 33350
    },
    {
      "entropy": 5.635682249069214,
      "epoch": 3.0256712626995648,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004094148917471866,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21798319220542908,
      "num_tokens": 69220648.0,
      "step": 33355
    },
    {
      "entropy": 5.553546714782715,
      "epoch": 3.0261248185776486,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000409388942328034,
      "loss": 4.7629,
      "mean_token_accuracy": 0.2267109766602516,
      "num_tokens": 69230685.0,
      "step": 33360
    },
    {
      "entropy": 5.553228282928467,
      "epoch": 3.026578374455733,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004093629901296838,
      "loss": 4.9227,
      "mean_token_accuracy": 0.21376700401306153,
      "num_tokens": 69240407.0,
      "step": 33365
    },
    {
      "entropy": 5.718560123443604,
      "epoch": 3.027031930333817,
      "grad_norm": 1.25,
      "learning_rate": 0.0004093370351526727,
      "loss": 4.9529,
      "mean_token_accuracy": 0.21289327889680862,
      "num_tokens": 69250624.0,
      "step": 33370
    },
    {
      "entropy": 5.717520999908447,
      "epoch": 3.0274854862119014,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004093110773975376,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20824849009513854,
      "num_tokens": 69261131.0,
      "step": 33375
    },
    {
      "entropy": 5.592414999008179,
      "epoch": 3.0279390420899857,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004092851168648152,
      "loss": 4.9164,
      "mean_token_accuracy": 0.20945546329021453,
      "num_tokens": 69270668.0,
      "step": 33380
    },
    {
      "entropy": 5.582691621780396,
      "epoch": 3.0283925979680695,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004092591535550422,
      "loss": 4.8308,
      "mean_token_accuracy": 0.22511018961668014,
      "num_tokens": 69280528.0,
      "step": 33385
    },
    {
      "entropy": 5.57068829536438,
      "epoch": 3.0288461538461537,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004092331874687558,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22823573052883148,
      "num_tokens": 69290376.0,
      "step": 33390
    },
    {
      "entropy": 5.638965034484864,
      "epoch": 3.029299709724238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040920721860649285,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22777412682771683,
      "num_tokens": 69301766.0,
      "step": 33395
    },
    {
      "entropy": 5.620216751098633,
      "epoch": 3.0297532656023223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004091812469687904,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21777494698762895,
      "num_tokens": 69312668.0,
      "step": 33400
    },
    {
      "entropy": 5.6081806182861325,
      "epoch": 3.0302068214804065,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004091552725561856,
      "loss": 4.9184,
      "mean_token_accuracy": 0.20867687910795213,
      "num_tokens": 69323195.0,
      "step": 33405
    },
    {
      "entropy": 5.602424287796021,
      "epoch": 3.0306603773584904,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004091292953692156,
      "loss": 4.8711,
      "mean_token_accuracy": 0.2210137277841568,
      "num_tokens": 69333812.0,
      "step": 33410
    },
    {
      "entropy": 5.621130561828613,
      "epoch": 3.0311139332365746,
      "grad_norm": 1.125,
      "learning_rate": 0.00040910331540841755,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22838663458824157,
      "num_tokens": 69344378.0,
      "step": 33415
    },
    {
      "entropy": 5.587288951873779,
      "epoch": 3.031567489114659,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040907733267432877,
      "loss": 4.8781,
      "mean_token_accuracy": 0.22115831971168518,
      "num_tokens": 69354592.0,
      "step": 33420
    },
    {
      "entropy": 5.628437328338623,
      "epoch": 3.032021044992743,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040905134716748656,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21539197415113448,
      "num_tokens": 69365011.0,
      "step": 33425
    },
    {
      "entropy": 5.571248960494995,
      "epoch": 3.0324746008708274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004090253588884284,
      "loss": 4.8867,
      "mean_token_accuracy": 0.21312977820634843,
      "num_tokens": 69377010.0,
      "step": 33430
    },
    {
      "entropy": 5.630991649627686,
      "epoch": 3.0329281567489113,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040899936783769157,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2216227889060974,
      "num_tokens": 69387263.0,
      "step": 33435
    },
    {
      "entropy": 5.638337564468384,
      "epoch": 3.0333817126269955,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040897337401581374,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22455921322107314,
      "num_tokens": 69399188.0,
      "step": 33440
    },
    {
      "entropy": 5.621446037292481,
      "epoch": 3.03383526850508,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040894737742333244,
      "loss": 4.8899,
      "mean_token_accuracy": 0.21498363167047502,
      "num_tokens": 69408271.0,
      "step": 33445
    },
    {
      "entropy": 5.679814243316651,
      "epoch": 3.034288824383164,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004089213780607851,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20780993551015853,
      "num_tokens": 69418424.0,
      "step": 33450
    },
    {
      "entropy": 5.655559444427491,
      "epoch": 3.0347423802612483,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004088953759287096,
      "loss": 4.8907,
      "mean_token_accuracy": 0.2269052356481552,
      "num_tokens": 69427802.0,
      "step": 33455
    },
    {
      "entropy": 5.616131734848023,
      "epoch": 3.0351959361393326,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004088693710276436,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21780091226100923,
      "num_tokens": 69437581.0,
      "step": 33460
    },
    {
      "entropy": 5.6072611808776855,
      "epoch": 3.0356494920174164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040884336335812494,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22064584642648696,
      "num_tokens": 69447711.0,
      "step": 33465
    },
    {
      "entropy": 5.674621725082398,
      "epoch": 3.0361030478955007,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004088173529206914,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2255007028579712,
      "num_tokens": 69457692.0,
      "step": 33470
    },
    {
      "entropy": 5.617612743377686,
      "epoch": 3.036556603773585,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004087913397158809,
      "loss": 4.8312,
      "mean_token_accuracy": 0.22081757634878157,
      "num_tokens": 69468034.0,
      "step": 33475
    },
    {
      "entropy": 5.568518733978271,
      "epoch": 3.037010159651669,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040876532374423135,
      "loss": 4.8508,
      "mean_token_accuracy": 0.22349444180727004,
      "num_tokens": 69477719.0,
      "step": 33480
    },
    {
      "entropy": 5.603288698196411,
      "epoch": 3.0374637155297535,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004087393050062809,
      "loss": 4.8818,
      "mean_token_accuracy": 0.21587074398994446,
      "num_tokens": 69487261.0,
      "step": 33485
    },
    {
      "entropy": 5.6113794326782225,
      "epoch": 3.0379172714078373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040871328350256745,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2233819842338562,
      "num_tokens": 69496658.0,
      "step": 33490
    },
    {
      "entropy": 5.713962650299072,
      "epoch": 3.0383708272859216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040868725923362927,
      "loss": 5.0183,
      "mean_token_accuracy": 0.2100777268409729,
      "num_tokens": 69507197.0,
      "step": 33495
    },
    {
      "entropy": 5.610123252868652,
      "epoch": 3.038824383164006,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040866123220000455,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21952326893806456,
      "num_tokens": 69516905.0,
      "step": 33500
    },
    {
      "entropy": 5.629584217071534,
      "epoch": 3.03927793904209,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040863520240223144,
      "loss": 4.8326,
      "mean_token_accuracy": 0.22321782410144805,
      "num_tokens": 69526869.0,
      "step": 33505
    },
    {
      "entropy": 5.662524604797364,
      "epoch": 3.0397314949201744,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004086091698408483,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21661264449357986,
      "num_tokens": 69537058.0,
      "step": 33510
    },
    {
      "entropy": 5.664655303955078,
      "epoch": 3.040185050798258,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040858313451639343,
      "loss": 4.9798,
      "mean_token_accuracy": 0.20732590556144714,
      "num_tokens": 69547617.0,
      "step": 33515
    },
    {
      "entropy": 5.64209566116333,
      "epoch": 3.0406386066763424,
      "grad_norm": 1.125,
      "learning_rate": 0.0004085570964294053,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21068992018699645,
      "num_tokens": 69558733.0,
      "step": 33520
    },
    {
      "entropy": 5.717536878585816,
      "epoch": 3.0410921625544267,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004085310555804225,
      "loss": 4.9868,
      "mean_token_accuracy": 0.21742648929357528,
      "num_tokens": 69570011.0,
      "step": 33525
    },
    {
      "entropy": 5.719327163696289,
      "epoch": 3.041545718432511,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040850501196998336,
      "loss": 5.0076,
      "mean_token_accuracy": 0.21230290234088897,
      "num_tokens": 69580894.0,
      "step": 33530
    },
    {
      "entropy": 5.6406091213226315,
      "epoch": 3.0419992743105952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040847896559862654,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21515671014785767,
      "num_tokens": 69591438.0,
      "step": 33535
    },
    {
      "entropy": 5.613695383071899,
      "epoch": 3.042452830188679,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004084529164668907,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21364197880029678,
      "num_tokens": 69601745.0,
      "step": 33540
    },
    {
      "entropy": 5.658150720596313,
      "epoch": 3.0429063860667633,
      "grad_norm": 1.25,
      "learning_rate": 0.00040842686457531453,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21762502193450928,
      "num_tokens": 69611872.0,
      "step": 33545
    },
    {
      "entropy": 5.639253711700439,
      "epoch": 3.0433599419448476,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004084008099244368,
      "loss": 4.868,
      "mean_token_accuracy": 0.21429318338632583,
      "num_tokens": 69621673.0,
      "step": 33550
    },
    {
      "entropy": 5.562106704711914,
      "epoch": 3.043813497822932,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040837475251479633,
      "loss": 4.814,
      "mean_token_accuracy": 0.2233399495482445,
      "num_tokens": 69632090.0,
      "step": 33555
    },
    {
      "entropy": 5.561004781723023,
      "epoch": 3.044267053701016,
      "grad_norm": 1.15625,
      "learning_rate": 0.000408348692346932,
      "loss": 4.8554,
      "mean_token_accuracy": 0.22518489360809327,
      "num_tokens": 69642920.0,
      "step": 33560
    },
    {
      "entropy": 5.629204845428466,
      "epoch": 3.0447206095791,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040832262942138277,
      "loss": 4.9635,
      "mean_token_accuracy": 0.21594253033399582,
      "num_tokens": 69653936.0,
      "step": 33565
    },
    {
      "entropy": 5.595352268218994,
      "epoch": 3.0451741654571842,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004082965637386875,
      "loss": 4.8272,
      "mean_token_accuracy": 0.22017356157302856,
      "num_tokens": 69663362.0,
      "step": 33570
    },
    {
      "entropy": 5.6082963943481445,
      "epoch": 3.0456277213352685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004082704952993854,
      "loss": 4.8409,
      "mean_token_accuracy": 0.2237624317407608,
      "num_tokens": 69673178.0,
      "step": 33575
    },
    {
      "entropy": 5.668405055999756,
      "epoch": 3.0460812772133528,
      "grad_norm": 1.125,
      "learning_rate": 0.0004082444241040155,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21675197184085845,
      "num_tokens": 69683627.0,
      "step": 33580
    },
    {
      "entropy": 5.593782234191894,
      "epoch": 3.046534833091437,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004082183501531169,
      "loss": 4.907,
      "mean_token_accuracy": 0.2181442439556122,
      "num_tokens": 69694010.0,
      "step": 33585
    },
    {
      "entropy": 5.603965950012207,
      "epoch": 3.046988388969521,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040819227344722893,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21518379151821138,
      "num_tokens": 69704888.0,
      "step": 33590
    },
    {
      "entropy": 5.58859658241272,
      "epoch": 3.047441944847605,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040816619398689074,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22560120820999147,
      "num_tokens": 69714718.0,
      "step": 33595
    },
    {
      "entropy": 5.608825016021728,
      "epoch": 3.0478955007256894,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040814011177264174,
      "loss": 4.8621,
      "mean_token_accuracy": 0.21948602199554443,
      "num_tokens": 69726076.0,
      "step": 33600
    },
    {
      "entropy": 5.568218040466308,
      "epoch": 3.0483490566037736,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040811402680502137,
      "loss": 4.841,
      "mean_token_accuracy": 0.2172714203596115,
      "num_tokens": 69736507.0,
      "step": 33605
    },
    {
      "entropy": 5.5609088897705075,
      "epoch": 3.048802612481858,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040808793908456885,
      "loss": 4.852,
      "mean_token_accuracy": 0.21985357850790024,
      "num_tokens": 69746061.0,
      "step": 33610
    },
    {
      "entropy": 5.62335467338562,
      "epoch": 3.0492561683599417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004080618486118241,
      "loss": 4.9231,
      "mean_token_accuracy": 0.21722546815872193,
      "num_tokens": 69756493.0,
      "step": 33615
    },
    {
      "entropy": 5.6466676712036135,
      "epoch": 3.049709724238026,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004080357553873261,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2215881437063217,
      "num_tokens": 69766937.0,
      "step": 33620
    },
    {
      "entropy": 5.594765472412109,
      "epoch": 3.0501632801161103,
      "grad_norm": 1.1875,
      "learning_rate": 0.000408009659411615,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22119349837303162,
      "num_tokens": 69776872.0,
      "step": 33625
    },
    {
      "entropy": 5.598317098617554,
      "epoch": 3.0506168359941945,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004079835606852301,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2186738595366478,
      "num_tokens": 69788145.0,
      "step": 33630
    },
    {
      "entropy": 5.614964771270752,
      "epoch": 3.051070391872279,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004079574592087114,
      "loss": 4.8809,
      "mean_token_accuracy": 0.21911578327417375,
      "num_tokens": 69797353.0,
      "step": 33635
    },
    {
      "entropy": 5.58411922454834,
      "epoch": 3.051523947750363,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040793135498259854,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22167650908231734,
      "num_tokens": 69806915.0,
      "step": 33640
    },
    {
      "entropy": 5.590922689437866,
      "epoch": 3.051977503628447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040790524800743134,
      "loss": 4.8762,
      "mean_token_accuracy": 0.21981202960014343,
      "num_tokens": 69817695.0,
      "step": 33645
    },
    {
      "entropy": 5.553254222869873,
      "epoch": 3.052431059506531,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004078791382837498,
      "loss": 4.8028,
      "mean_token_accuracy": 0.2348508670926094,
      "num_tokens": 69827451.0,
      "step": 33650
    },
    {
      "entropy": 5.654099321365356,
      "epoch": 3.0528846153846154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040785302581209374,
      "loss": 4.8927,
      "mean_token_accuracy": 0.22072887271642685,
      "num_tokens": 69837897.0,
      "step": 33655
    },
    {
      "entropy": 5.615628433227539,
      "epoch": 3.0533381712626997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040782691059300326,
      "loss": 4.8573,
      "mean_token_accuracy": 0.22448501437902452,
      "num_tokens": 69847378.0,
      "step": 33660
    },
    {
      "entropy": 5.579566860198975,
      "epoch": 3.053791727140784,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004078007926270185,
      "loss": 4.9243,
      "mean_token_accuracy": 0.2194240555167198,
      "num_tokens": 69857522.0,
      "step": 33665
    },
    {
      "entropy": 5.597421550750733,
      "epoch": 3.0542452830188678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040777467191467946,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2233879014849663,
      "num_tokens": 69868384.0,
      "step": 33670
    },
    {
      "entropy": 5.6122270107269285,
      "epoch": 3.054698838896952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004077485484565264,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21967986822128296,
      "num_tokens": 69878761.0,
      "step": 33675
    },
    {
      "entropy": 5.662349510192871,
      "epoch": 3.0551523947750363,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004077224222530994,
      "loss": 4.8609,
      "mean_token_accuracy": 0.21498922854661942,
      "num_tokens": 69888366.0,
      "step": 33680
    },
    {
      "entropy": 5.623615264892578,
      "epoch": 3.0556059506531206,
      "grad_norm": 1.09375,
      "learning_rate": 0.000407696293304939,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21239088475704193,
      "num_tokens": 69898489.0,
      "step": 33685
    },
    {
      "entropy": 5.576485586166382,
      "epoch": 3.056059506531205,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004076701616125854,
      "loss": 4.8471,
      "mean_token_accuracy": 0.21782163083553313,
      "num_tokens": 69908425.0,
      "step": 33690
    },
    {
      "entropy": 5.661604166030884,
      "epoch": 3.0565130624092887,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004076440271765791,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21908647418022156,
      "num_tokens": 69918488.0,
      "step": 33695
    },
    {
      "entropy": 5.531662702560425,
      "epoch": 3.056966618287373,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004076178899974605,
      "loss": 4.7933,
      "mean_token_accuracy": 0.23034947216510773,
      "num_tokens": 69928831.0,
      "step": 33700
    },
    {
      "entropy": 5.609535217285156,
      "epoch": 3.057420174165457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040759175007577017,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21517693400382995,
      "num_tokens": 69938974.0,
      "step": 33705
    },
    {
      "entropy": 5.577897357940674,
      "epoch": 3.0578737300435415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004075656074120486,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2197770282626152,
      "num_tokens": 69948804.0,
      "step": 33710
    },
    {
      "entropy": 5.640744161605835,
      "epoch": 3.0583272859216257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004075394620068365,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22303461283445358,
      "num_tokens": 69958945.0,
      "step": 33715
    },
    {
      "entropy": 5.598249816894532,
      "epoch": 3.0587808417997095,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040751331386067456,
      "loss": 4.8476,
      "mean_token_accuracy": 0.22164223492145538,
      "num_tokens": 69968826.0,
      "step": 33720
    },
    {
      "entropy": 5.641958808898925,
      "epoch": 3.059234397677794,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004074871629741035,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21732646077871323,
      "num_tokens": 69978961.0,
      "step": 33725
    },
    {
      "entropy": 5.559868907928466,
      "epoch": 3.059687953555878,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004074610093476641,
      "loss": 4.9092,
      "mean_token_accuracy": 0.218568617105484,
      "num_tokens": 69989848.0,
      "step": 33730
    },
    {
      "entropy": 5.595981121063232,
      "epoch": 3.0601415094339623,
      "grad_norm": 1.25,
      "learning_rate": 0.00040743485298189744,
      "loss": 4.8499,
      "mean_token_accuracy": 0.2210436075925827,
      "num_tokens": 69999607.0,
      "step": 33735
    },
    {
      "entropy": 5.605841207504272,
      "epoch": 3.0605950653120466,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040740869387734415,
      "loss": 4.8988,
      "mean_token_accuracy": 0.21666444838047028,
      "num_tokens": 70010446.0,
      "step": 33740
    },
    {
      "entropy": 5.571853971481323,
      "epoch": 3.0610486211901304,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004073825320345453,
      "loss": 4.8009,
      "mean_token_accuracy": 0.22282686531543733,
      "num_tokens": 70021148.0,
      "step": 33745
    },
    {
      "entropy": 5.654863500595093,
      "epoch": 3.0615021770682147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040735636745404196,
      "loss": 4.96,
      "mean_token_accuracy": 0.2122356802225113,
      "num_tokens": 70030765.0,
      "step": 33750
    },
    {
      "entropy": 5.662096452713013,
      "epoch": 3.061955732946299,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040733020013637526,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21841022223234177,
      "num_tokens": 70041717.0,
      "step": 33755
    },
    {
      "entropy": 5.68151068687439,
      "epoch": 3.0624092888243832,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040730403008208633,
      "loss": 4.9477,
      "mean_token_accuracy": 0.22137918174266816,
      "num_tokens": 70052178.0,
      "step": 33760
    },
    {
      "entropy": 5.606624031066895,
      "epoch": 3.0628628447024675,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004072778572917163,
      "loss": 4.9038,
      "mean_token_accuracy": 0.21718788743019105,
      "num_tokens": 70062819.0,
      "step": 33765
    },
    {
      "entropy": 5.551634454727173,
      "epoch": 3.0633164005805513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004072516817658065,
      "loss": 4.9068,
      "mean_token_accuracy": 0.21947057396173478,
      "num_tokens": 70073966.0,
      "step": 33770
    },
    {
      "entropy": 5.625006103515625,
      "epoch": 3.0637699564586356,
      "grad_norm": 1.125,
      "learning_rate": 0.00040722550350489825,
      "loss": 4.8915,
      "mean_token_accuracy": 0.21627938151359558,
      "num_tokens": 70083884.0,
      "step": 33775
    },
    {
      "entropy": 5.674850797653198,
      "epoch": 3.06422351233672,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004071993225095327,
      "loss": 4.8703,
      "mean_token_accuracy": 0.21777383536100386,
      "num_tokens": 70094910.0,
      "step": 33780
    },
    {
      "entropy": 5.567288732528686,
      "epoch": 3.064677068214804,
      "grad_norm": 1.375,
      "learning_rate": 0.0004071731387802517,
      "loss": 4.7206,
      "mean_token_accuracy": 0.23919747024774551,
      "num_tokens": 70104480.0,
      "step": 33785
    },
    {
      "entropy": 5.570855808258057,
      "epoch": 3.0651306240928884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004071469523175964,
      "loss": 4.8594,
      "mean_token_accuracy": 0.21891711354255677,
      "num_tokens": 70114003.0,
      "step": 33790
    },
    {
      "entropy": 5.630770921707153,
      "epoch": 3.065584179970972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040712076312210844,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21823698580265044,
      "num_tokens": 70123604.0,
      "step": 33795
    },
    {
      "entropy": 5.617650127410888,
      "epoch": 3.0660377358490565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040709457119432945,
      "loss": 4.926,
      "mean_token_accuracy": 0.21489761173725128,
      "num_tokens": 70133895.0,
      "step": 33800
    },
    {
      "entropy": 5.6311156272888185,
      "epoch": 3.0664912917271407,
      "grad_norm": 1.125,
      "learning_rate": 0.00040706837653480113,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22623943090438842,
      "num_tokens": 70145063.0,
      "step": 33805
    },
    {
      "entropy": 5.623827266693115,
      "epoch": 3.066944847605225,
      "grad_norm": 1.296875,
      "learning_rate": 0.000407042179144065,
      "loss": 4.958,
      "mean_token_accuracy": 0.21199622601270676,
      "num_tokens": 70155456.0,
      "step": 33810
    },
    {
      "entropy": 5.523899555206299,
      "epoch": 3.0673984034833093,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040701597902266306,
      "loss": 4.8075,
      "mean_token_accuracy": 0.2228490799665451,
      "num_tokens": 70165427.0,
      "step": 33815
    },
    {
      "entropy": 5.6857884407043455,
      "epoch": 3.067851959361393,
      "grad_norm": 1.1875,
      "learning_rate": 0.000406989776171137,
      "loss": 4.941,
      "mean_token_accuracy": 0.21535684168338776,
      "num_tokens": 70175161.0,
      "step": 33820
    },
    {
      "entropy": 5.612390184402466,
      "epoch": 3.0683055152394774,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040696357059002877,
      "loss": 4.8945,
      "mean_token_accuracy": 0.223714242875576,
      "num_tokens": 70185026.0,
      "step": 33825
    },
    {
      "entropy": 5.668047618865967,
      "epoch": 3.0687590711175616,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004069373622798802,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22745070308446885,
      "num_tokens": 70195645.0,
      "step": 33830
    },
    {
      "entropy": 5.64789924621582,
      "epoch": 3.069212626995646,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004069111512412334,
      "loss": 4.9661,
      "mean_token_accuracy": 0.21733848601579667,
      "num_tokens": 70206591.0,
      "step": 33835
    },
    {
      "entropy": 5.6326210498809814,
      "epoch": 3.06966618287373,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040688493747463036,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22249459475278854,
      "num_tokens": 70216945.0,
      "step": 33840
    },
    {
      "entropy": 5.555872583389283,
      "epoch": 3.0701197387518144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004068587209806132,
      "loss": 4.7804,
      "mean_token_accuracy": 0.2328471839427948,
      "num_tokens": 70226983.0,
      "step": 33845
    },
    {
      "entropy": 5.599160385131836,
      "epoch": 3.0705732946298983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040683250175972414,
      "loss": 4.8636,
      "mean_token_accuracy": 0.21896474659442902,
      "num_tokens": 70237730.0,
      "step": 33850
    },
    {
      "entropy": 5.651721715927124,
      "epoch": 3.0710268505079825,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004068062798125053,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21341179311275482,
      "num_tokens": 70248499.0,
      "step": 33855
    },
    {
      "entropy": 5.599911260604858,
      "epoch": 3.071480406386067,
      "grad_norm": 1.203125,
      "learning_rate": 0.000406780055139499,
      "loss": 4.8763,
      "mean_token_accuracy": 0.21700489968061448,
      "num_tokens": 70259294.0,
      "step": 33860
    },
    {
      "entropy": 5.651598072052002,
      "epoch": 3.071933962264151,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004067538277412476,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21679080724716188,
      "num_tokens": 70269400.0,
      "step": 33865
    },
    {
      "entropy": 5.652189302444458,
      "epoch": 3.0723875181422353,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004067275976182935,
      "loss": 4.9574,
      "mean_token_accuracy": 0.21994108855724334,
      "num_tokens": 70279380.0,
      "step": 33870
    },
    {
      "entropy": 5.640937185287475,
      "epoch": 3.072841074020319,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040670136477117906,
      "loss": 4.8866,
      "mean_token_accuracy": 0.2217855617403984,
      "num_tokens": 70289919.0,
      "step": 33875
    },
    {
      "entropy": 5.699266958236694,
      "epoch": 3.0732946298984034,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040667512920044684,
      "loss": 4.994,
      "mean_token_accuracy": 0.21905118376016616,
      "num_tokens": 70300136.0,
      "step": 33880
    },
    {
      "entropy": 5.597797870635986,
      "epoch": 3.0737481857764877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040664889090663944,
      "loss": 4.796,
      "mean_token_accuracy": 0.22733922600746154,
      "num_tokens": 70310072.0,
      "step": 33885
    },
    {
      "entropy": 5.6391500473022464,
      "epoch": 3.074201741654572,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040662264989029933,
      "loss": 4.879,
      "mean_token_accuracy": 0.22217219769954683,
      "num_tokens": 70320185.0,
      "step": 33890
    },
    {
      "entropy": 5.617372179031372,
      "epoch": 3.074655297532656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004065964061519694,
      "loss": 4.8779,
      "mean_token_accuracy": 0.2218205600976944,
      "num_tokens": 70330211.0,
      "step": 33895
    },
    {
      "entropy": 5.638485527038574,
      "epoch": 3.07510885341074,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004065701596921922,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21446337848901748,
      "num_tokens": 70341044.0,
      "step": 33900
    },
    {
      "entropy": 5.650462293624878,
      "epoch": 3.0755624092888243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004065439105115106,
      "loss": 4.9337,
      "mean_token_accuracy": 0.22086102813482283,
      "num_tokens": 70351320.0,
      "step": 33905
    },
    {
      "entropy": 5.602971076965332,
      "epoch": 3.0760159651669086,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040651765861046734,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22849898487329484,
      "num_tokens": 70361214.0,
      "step": 33910
    },
    {
      "entropy": 5.642775344848633,
      "epoch": 3.076469521044993,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040649140398960537,
      "loss": 4.921,
      "mean_token_accuracy": 0.22190209478139877,
      "num_tokens": 70371083.0,
      "step": 33915
    },
    {
      "entropy": 5.638441133499145,
      "epoch": 3.076923076923077,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040646514664946775,
      "loss": 5.0183,
      "mean_token_accuracy": 0.21401370763778688,
      "num_tokens": 70381987.0,
      "step": 33920
    },
    {
      "entropy": 5.58901572227478,
      "epoch": 3.077376632801161,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004064388865905974,
      "loss": 4.8425,
      "mean_token_accuracy": 0.2244996890425682,
      "num_tokens": 70393462.0,
      "step": 33925
    },
    {
      "entropy": 5.579142665863037,
      "epoch": 3.077830188679245,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004064126238135373,
      "loss": 4.7486,
      "mean_token_accuracy": 0.22805921882390975,
      "num_tokens": 70403664.0,
      "step": 33930
    },
    {
      "entropy": 5.600571727752685,
      "epoch": 3.0782837445573294,
      "grad_norm": 1.125,
      "learning_rate": 0.0004063863583188308,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21349358707666397,
      "num_tokens": 70414588.0,
      "step": 33935
    },
    {
      "entropy": 5.571170854568481,
      "epoch": 3.0787373004354137,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004063600901070207,
      "loss": 4.941,
      "mean_token_accuracy": 0.21524872928857802,
      "num_tokens": 70425572.0,
      "step": 33940
    },
    {
      "entropy": 5.604797840118408,
      "epoch": 3.079190856313498,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040633381917865057,
      "loss": 4.8442,
      "mean_token_accuracy": 0.21919606029987335,
      "num_tokens": 70436140.0,
      "step": 33945
    },
    {
      "entropy": 5.580093669891357,
      "epoch": 3.079644412191582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004063075455342636,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22368747293949126,
      "num_tokens": 70447117.0,
      "step": 33950
    },
    {
      "entropy": 5.623998498916626,
      "epoch": 3.080097968069666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004062812691744032,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2238440290093422,
      "num_tokens": 70457756.0,
      "step": 33955
    },
    {
      "entropy": 5.63613920211792,
      "epoch": 3.0805515239477503,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004062549900996126,
      "loss": 4.9433,
      "mean_token_accuracy": 0.21454330384731293,
      "num_tokens": 70468671.0,
      "step": 33960
    },
    {
      "entropy": 5.634016609191894,
      "epoch": 3.0810050798258346,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004062287083104354,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22401514202356337,
      "num_tokens": 70478212.0,
      "step": 33965
    },
    {
      "entropy": 5.585205030441284,
      "epoch": 3.081458635703919,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040620242380741504,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2190607324242592,
      "num_tokens": 70489062.0,
      "step": 33970
    },
    {
      "entropy": 5.625957870483399,
      "epoch": 3.0819121915820027,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004061761365910951,
      "loss": 4.9443,
      "mean_token_accuracy": 0.22176443785429,
      "num_tokens": 70499732.0,
      "step": 33975
    },
    {
      "entropy": 5.6337282180786135,
      "epoch": 3.082365747460087,
      "grad_norm": 1.25,
      "learning_rate": 0.00040614984666201927,
      "loss": 4.8948,
      "mean_token_accuracy": 0.21288188844919204,
      "num_tokens": 70509777.0,
      "step": 33980
    },
    {
      "entropy": 5.63773512840271,
      "epoch": 3.0828193033381712,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004061235540207312,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21612088531255721,
      "num_tokens": 70520674.0,
      "step": 33985
    },
    {
      "entropy": 5.631599187850952,
      "epoch": 3.0832728592162555,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004060972586677746,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22063713520765305,
      "num_tokens": 70531115.0,
      "step": 33990
    },
    {
      "entropy": 5.584438419342041,
      "epoch": 3.0837264150943398,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004060709606036933,
      "loss": 4.8683,
      "mean_token_accuracy": 0.22172345072031022,
      "num_tokens": 70540726.0,
      "step": 33995
    },
    {
      "entropy": 5.623329210281372,
      "epoch": 3.084179970972424,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004060446598290311,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2237810105085373,
      "num_tokens": 70551227.0,
      "step": 34000
    },
    {
      "entropy": 5.564810132980346,
      "epoch": 3.084633526850508,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000406018356344332,
      "loss": 4.875,
      "mean_token_accuracy": 0.21612243503332138,
      "num_tokens": 70560815.0,
      "step": 34005
    },
    {
      "entropy": 5.666827774047851,
      "epoch": 3.085087082728592,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004059920501501398,
      "loss": 4.9493,
      "mean_token_accuracy": 0.2166752338409424,
      "num_tokens": 70570247.0,
      "step": 34010
    },
    {
      "entropy": 5.6672851085662845,
      "epoch": 3.0855406386066764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004059657412469987,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21635723114013672,
      "num_tokens": 70581497.0,
      "step": 34015
    },
    {
      "entropy": 5.608234643936157,
      "epoch": 3.0859941944847606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004059394296354527,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22011708319187165,
      "num_tokens": 70593133.0,
      "step": 34020
    },
    {
      "entropy": 5.609160947799682,
      "epoch": 3.086447750362845,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004059131153160459,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22407814413309096,
      "num_tokens": 70603953.0,
      "step": 34025
    },
    {
      "entropy": 5.560619926452636,
      "epoch": 3.0869013062409287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004058867982893225,
      "loss": 4.8481,
      "mean_token_accuracy": 0.22599049806594848,
      "num_tokens": 70615292.0,
      "step": 34030
    },
    {
      "entropy": 5.663981866836548,
      "epoch": 3.087354862119013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004058604785558268,
      "loss": 5.0145,
      "mean_token_accuracy": 0.2074570193886757,
      "num_tokens": 70626681.0,
      "step": 34035
    },
    {
      "entropy": 5.662851715087891,
      "epoch": 3.0878084179970973,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040583415611610305,
      "loss": 4.935,
      "mean_token_accuracy": 0.2140041247010231,
      "num_tokens": 70637339.0,
      "step": 34040
    },
    {
      "entropy": 5.704662322998047,
      "epoch": 3.0882619738751815,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004058078309706956,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22234832793474196,
      "num_tokens": 70647980.0,
      "step": 34045
    },
    {
      "entropy": 5.586560487747192,
      "epoch": 3.088715529753266,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004057815031201488,
      "loss": 4.7292,
      "mean_token_accuracy": 0.23724641501903534,
      "num_tokens": 70657042.0,
      "step": 34050
    },
    {
      "entropy": 5.601404905319214,
      "epoch": 3.0891690856313496,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004057551725650073,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21587942391633988,
      "num_tokens": 70668542.0,
      "step": 34055
    },
    {
      "entropy": 5.637851047515869,
      "epoch": 3.089622641509434,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040572883930581544,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21750420480966567,
      "num_tokens": 70679332.0,
      "step": 34060
    },
    {
      "entropy": 5.7466113567352295,
      "epoch": 3.090076197387518,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004057025033431179,
      "loss": 5.0266,
      "mean_token_accuracy": 0.21392195224761962,
      "num_tokens": 70689897.0,
      "step": 34065
    },
    {
      "entropy": 5.632059240341187,
      "epoch": 3.0905297532656024,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040567616467745933,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22339710146188735,
      "num_tokens": 70699965.0,
      "step": 34070
    },
    {
      "entropy": 5.65618200302124,
      "epoch": 3.0909833091436867,
      "grad_norm": 1.25,
      "learning_rate": 0.0004056498233093843,
      "loss": 4.9552,
      "mean_token_accuracy": 0.22152122110128403,
      "num_tokens": 70710583.0,
      "step": 34075
    },
    {
      "entropy": 5.557923030853272,
      "epoch": 3.0914368650217705,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040562347923943766,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23478762060403824,
      "num_tokens": 70721800.0,
      "step": 34080
    },
    {
      "entropy": 5.592049312591553,
      "epoch": 3.0918904208998548,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004055971324681642,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21415081173181533,
      "num_tokens": 70731527.0,
      "step": 34085
    },
    {
      "entropy": 5.653995752334595,
      "epoch": 3.092343976777939,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004055707829961087,
      "loss": 4.9013,
      "mean_token_accuracy": 0.22178112268447875,
      "num_tokens": 70740821.0,
      "step": 34090
    },
    {
      "entropy": 5.630556106567383,
      "epoch": 3.0927975326560233,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040554443082381627,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22050863802433013,
      "num_tokens": 70750506.0,
      "step": 34095
    },
    {
      "entropy": 5.607876491546631,
      "epoch": 3.0932510885341076,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004055180759518315,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22196728885173797,
      "num_tokens": 70762009.0,
      "step": 34100
    },
    {
      "entropy": 5.560287809371948,
      "epoch": 3.0937046444121914,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004054917183806999,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2206227406859398,
      "num_tokens": 70772030.0,
      "step": 34105
    },
    {
      "entropy": 5.648917388916016,
      "epoch": 3.0941582002902757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040546535811096626,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21337200850248336,
      "num_tokens": 70782539.0,
      "step": 34110
    },
    {
      "entropy": 5.789413785934448,
      "epoch": 3.09461175616836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040543899514317564,
      "loss": 5.0427,
      "mean_token_accuracy": 0.20537374168634415,
      "num_tokens": 70793433.0,
      "step": 34115
    },
    {
      "entropy": 5.650471019744873,
      "epoch": 3.095065312046444,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004054126294778735,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21279655694961547,
      "num_tokens": 70802919.0,
      "step": 34120
    },
    {
      "entropy": 5.571983909606933,
      "epoch": 3.0955188679245285,
      "grad_norm": 1.125,
      "learning_rate": 0.00040538626111560487,
      "loss": 4.8412,
      "mean_token_accuracy": 0.2223627358675003,
      "num_tokens": 70813827.0,
      "step": 34125
    },
    {
      "entropy": 5.682165908813476,
      "epoch": 3.0959724238026123,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040535989005691515,
      "loss": 4.9449,
      "mean_token_accuracy": 0.2172440230846405,
      "num_tokens": 70824117.0,
      "step": 34130
    },
    {
      "entropy": 5.556173610687256,
      "epoch": 3.0964259796806965,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040533351630234985,
      "loss": 4.8249,
      "mean_token_accuracy": 0.22764098197221755,
      "num_tokens": 70833387.0,
      "step": 34135
    },
    {
      "entropy": 5.600375604629517,
      "epoch": 3.096879535558781,
      "grad_norm": 1.203125,
      "learning_rate": 0.000405307139852454,
      "loss": 4.8591,
      "mean_token_accuracy": 0.22222230583429337,
      "num_tokens": 70844363.0,
      "step": 34140
    },
    {
      "entropy": 5.64898681640625,
      "epoch": 3.097333091436865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040528076070777335,
      "loss": 4.9166,
      "mean_token_accuracy": 0.21835437566041946,
      "num_tokens": 70854759.0,
      "step": 34145
    },
    {
      "entropy": 5.539999198913574,
      "epoch": 3.0977866473149493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004052543788688534,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2176368311047554,
      "num_tokens": 70865682.0,
      "step": 34150
    },
    {
      "entropy": 5.705310201644897,
      "epoch": 3.098240203193033,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004052279943362396,
      "loss": 4.9397,
      "mean_token_accuracy": 0.2154989242553711,
      "num_tokens": 70875449.0,
      "step": 34155
    },
    {
      "entropy": 5.652715063095092,
      "epoch": 3.0986937590711174,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004052016071104779,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2179457113146782,
      "num_tokens": 70885885.0,
      "step": 34160
    },
    {
      "entropy": 5.599583864212036,
      "epoch": 3.0991473149492017,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004051752171921136,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22145263701677323,
      "num_tokens": 70896320.0,
      "step": 34165
    },
    {
      "entropy": 5.568074798583984,
      "epoch": 3.099600870827286,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040514882458169267,
      "loss": 4.8132,
      "mean_token_accuracy": 0.2256337061524391,
      "num_tokens": 70907074.0,
      "step": 34170
    },
    {
      "entropy": 5.605743265151977,
      "epoch": 3.1000544267053702,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004051224292797609,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2163526952266693,
      "num_tokens": 70916567.0,
      "step": 34175
    },
    {
      "entropy": 5.629853200912476,
      "epoch": 3.100507982583454,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040509603128686413,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22729743719100953,
      "num_tokens": 70926163.0,
      "step": 34180
    },
    {
      "entropy": 5.6725012302398685,
      "epoch": 3.1009615384615383,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004050696306035483,
      "loss": 4.8829,
      "mean_token_accuracy": 0.21403191089630128,
      "num_tokens": 70937358.0,
      "step": 34185
    },
    {
      "entropy": 5.5880731582641605,
      "epoch": 3.1014150943396226,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040504322723035937,
      "loss": 4.8772,
      "mean_token_accuracy": 0.2194350019097328,
      "num_tokens": 70949305.0,
      "step": 34190
    },
    {
      "entropy": 5.5348611831665036,
      "epoch": 3.101868650217707,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040501682116784326,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2258891612291336,
      "num_tokens": 70959146.0,
      "step": 34195
    },
    {
      "entropy": 5.546608352661133,
      "epoch": 3.102322206095791,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040499041241654626,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21654050797224045,
      "num_tokens": 70968828.0,
      "step": 34200
    },
    {
      "entropy": 5.587368154525757,
      "epoch": 3.1027757619738754,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004049640009770144,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22193701118230819,
      "num_tokens": 70979733.0,
      "step": 34205
    },
    {
      "entropy": 5.672473382949829,
      "epoch": 3.103229317851959,
      "grad_norm": 1.25,
      "learning_rate": 0.0004049375868497939,
      "loss": 4.9225,
      "mean_token_accuracy": 0.2205413907766342,
      "num_tokens": 70989726.0,
      "step": 34210
    },
    {
      "entropy": 5.656386852264404,
      "epoch": 3.1036828737300435,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040491117003543087,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2169051706790924,
      "num_tokens": 71000715.0,
      "step": 34215
    },
    {
      "entropy": 5.563777923583984,
      "epoch": 3.1041364296081277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040488475053447186,
      "loss": 4.7298,
      "mean_token_accuracy": 0.23253817111253738,
      "num_tokens": 71011453.0,
      "step": 34220
    },
    {
      "entropy": 5.620495557785034,
      "epoch": 3.104589985486212,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040485832834746306,
      "loss": 4.9256,
      "mean_token_accuracy": 0.21509634107351303,
      "num_tokens": 71021477.0,
      "step": 34225
    },
    {
      "entropy": 5.56983323097229,
      "epoch": 3.1050435413642963,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040483190347495097,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21912508606910705,
      "num_tokens": 71031706.0,
      "step": 34230
    },
    {
      "entropy": 5.647357511520386,
      "epoch": 3.10549709724238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000404805475917482,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22162728905677795,
      "num_tokens": 71042882.0,
      "step": 34235
    },
    {
      "entropy": 5.629679536819458,
      "epoch": 3.1059506531204644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040477904567560275,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22511965334415435,
      "num_tokens": 71052804.0,
      "step": 34240
    },
    {
      "entropy": 5.650507593154908,
      "epoch": 3.1064042089985486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004047526127498597,
      "loss": 4.9868,
      "mean_token_accuracy": 0.20449872314929962,
      "num_tokens": 71064349.0,
      "step": 34245
    },
    {
      "entropy": 5.590665483474732,
      "epoch": 3.106857764876633,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040472617714079964,
      "loss": 4.9086,
      "mean_token_accuracy": 0.218218632042408,
      "num_tokens": 71075754.0,
      "step": 34250
    },
    {
      "entropy": 5.642666578292847,
      "epoch": 3.107311320754717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004046997388489691,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2216945096850395,
      "num_tokens": 71085338.0,
      "step": 34255
    },
    {
      "entropy": 5.558362007141113,
      "epoch": 3.107764876632801,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004046732978749149,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2154309943318367,
      "num_tokens": 71097089.0,
      "step": 34260
    },
    {
      "entropy": 5.532902479171753,
      "epoch": 3.1082184325108853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000404646854219184,
      "loss": 4.7966,
      "mean_token_accuracy": 0.2322097212076187,
      "num_tokens": 71106990.0,
      "step": 34265
    },
    {
      "entropy": 5.5698882102966305,
      "epoch": 3.1086719883889695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000404620407882323,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22159672081470488,
      "num_tokens": 71117634.0,
      "step": 34270
    },
    {
      "entropy": 5.54518461227417,
      "epoch": 3.109125544267054,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040459395886487885,
      "loss": 4.8049,
      "mean_token_accuracy": 0.2202684372663498,
      "num_tokens": 71127717.0,
      "step": 34275
    },
    {
      "entropy": 5.603554058074951,
      "epoch": 3.109579100145138,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004045675071673988,
      "loss": 4.8174,
      "mean_token_accuracy": 0.22675131261348724,
      "num_tokens": 71137172.0,
      "step": 34280
    },
    {
      "entropy": 5.651407766342163,
      "epoch": 3.110032656023222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004045410527904296,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2083875447511673,
      "num_tokens": 71146843.0,
      "step": 34285
    },
    {
      "entropy": 5.554104900360107,
      "epoch": 3.110486211901306,
      "grad_norm": 1.25,
      "learning_rate": 0.0004045145957345184,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23007616251707078,
      "num_tokens": 71156619.0,
      "step": 34290
    },
    {
      "entropy": 5.618606901168823,
      "epoch": 3.1109397677793904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004044881360002123,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22491240054368972,
      "num_tokens": 71167234.0,
      "step": 34295
    },
    {
      "entropy": 5.615027379989624,
      "epoch": 3.1113933236574747,
      "grad_norm": 1.125,
      "learning_rate": 0.0004044616735880586,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22390307039022445,
      "num_tokens": 71178049.0,
      "step": 34300
    },
    {
      "entropy": 5.585771560668945,
      "epoch": 3.111846879535559,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040443520849860453,
      "loss": 4.81,
      "mean_token_accuracy": 0.22132199108600617,
      "num_tokens": 71188568.0,
      "step": 34305
    },
    {
      "entropy": 5.641154861450195,
      "epoch": 3.1123004354136428,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040440874073239733,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21335671097040176,
      "num_tokens": 71199004.0,
      "step": 34310
    },
    {
      "entropy": 5.570372152328491,
      "epoch": 3.112753991291727,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040438227028998436,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22319533824920654,
      "num_tokens": 71209239.0,
      "step": 34315
    },
    {
      "entropy": 5.619079065322876,
      "epoch": 3.1132075471698113,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040435579717191313,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21719261556863784,
      "num_tokens": 71219469.0,
      "step": 34320
    },
    {
      "entropy": 5.609042549133301,
      "epoch": 3.1136611030478956,
      "grad_norm": 1.25,
      "learning_rate": 0.00040432932137873107,
      "loss": 4.9335,
      "mean_token_accuracy": 0.2149026244878769,
      "num_tokens": 71230670.0,
      "step": 34325
    },
    {
      "entropy": 5.679176712036133,
      "epoch": 3.11411465892598,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004043028429109857,
      "loss": 4.9925,
      "mean_token_accuracy": 0.2172279730439186,
      "num_tokens": 71241512.0,
      "step": 34330
    },
    {
      "entropy": 5.627719497680664,
      "epoch": 3.1145682148040637,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004042763617692245,
      "loss": 4.8422,
      "mean_token_accuracy": 0.22052338123321533,
      "num_tokens": 71250869.0,
      "step": 34335
    },
    {
      "entropy": 5.603387022018433,
      "epoch": 3.115021770682148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040424987795399523,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2186043456196785,
      "num_tokens": 71260577.0,
      "step": 34340
    },
    {
      "entropy": 5.623979616165161,
      "epoch": 3.115475326560232,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004042233914658455,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2218081012368202,
      "num_tokens": 71270185.0,
      "step": 34345
    },
    {
      "entropy": 5.539167213439941,
      "epoch": 3.1159288824383164,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040419690230532324,
      "loss": 4.8463,
      "mean_token_accuracy": 0.22802520543336868,
      "num_tokens": 71279427.0,
      "step": 34350
    },
    {
      "entropy": 5.584182500839233,
      "epoch": 3.1163824383164007,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040417041047297606,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22426612675189972,
      "num_tokens": 71290857.0,
      "step": 34355
    },
    {
      "entropy": 5.637346792221069,
      "epoch": 3.116835994194485,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004041439159693518,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2221457377076149,
      "num_tokens": 71300971.0,
      "step": 34360
    },
    {
      "entropy": 5.659772205352783,
      "epoch": 3.117289550072569,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004041174187949985,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21609605699777604,
      "num_tokens": 71311017.0,
      "step": 34365
    },
    {
      "entropy": 5.649938869476318,
      "epoch": 3.117743105950653,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040409091895046413,
      "loss": 4.9825,
      "mean_token_accuracy": 0.21190449893474578,
      "num_tokens": 71321546.0,
      "step": 34370
    },
    {
      "entropy": 5.594947814941406,
      "epoch": 3.1181966618287373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040406441643629656,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21915542632341384,
      "num_tokens": 71332345.0,
      "step": 34375
    },
    {
      "entropy": 5.6774999618530275,
      "epoch": 3.1186502177068216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040403791125304405,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21980152428150176,
      "num_tokens": 71342670.0,
      "step": 34380
    },
    {
      "entropy": 5.61887640953064,
      "epoch": 3.119103773584906,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004040114034012546,
      "loss": 4.8196,
      "mean_token_accuracy": 0.2188290163874626,
      "num_tokens": 71353440.0,
      "step": 34385
    },
    {
      "entropy": 5.626083898544311,
      "epoch": 3.1195573294629897,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040398489288147646,
      "loss": 4.9558,
      "mean_token_accuracy": 0.21995634585618973,
      "num_tokens": 71363199.0,
      "step": 34390
    },
    {
      "entropy": 5.61260757446289,
      "epoch": 3.120010885341074,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040395837969425786,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21791732609272002,
      "num_tokens": 71372925.0,
      "step": 34395
    },
    {
      "entropy": 5.650451803207398,
      "epoch": 3.1204644412191582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004039318638401471,
      "loss": 4.9111,
      "mean_token_accuracy": 0.21911511868238448,
      "num_tokens": 71383350.0,
      "step": 34400
    },
    {
      "entropy": 5.613069200515747,
      "epoch": 3.1209179970972425,
      "grad_norm": 1.25,
      "learning_rate": 0.0004039053453196926,
      "loss": 4.8233,
      "mean_token_accuracy": 0.23012498766183853,
      "num_tokens": 71394006.0,
      "step": 34405
    },
    {
      "entropy": 5.558915090560913,
      "epoch": 3.1213715529753268,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004038788241334427,
      "loss": 4.8659,
      "mean_token_accuracy": 0.21994834393262863,
      "num_tokens": 71403945.0,
      "step": 34410
    },
    {
      "entropy": 5.622850179672241,
      "epoch": 3.1218251088534106,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004038523002819458,
      "loss": 4.9112,
      "mean_token_accuracy": 0.21319989562034608,
      "num_tokens": 71414238.0,
      "step": 34415
    },
    {
      "entropy": 5.595477724075318,
      "epoch": 3.122278664731495,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040382577376575055,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21439329236745835,
      "num_tokens": 71423586.0,
      "step": 34420
    },
    {
      "entropy": 5.607227325439453,
      "epoch": 3.122732220609579,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004037992445854054,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22491985261440278,
      "num_tokens": 71434261.0,
      "step": 34425
    },
    {
      "entropy": 5.608119153976441,
      "epoch": 3.1231857764876634,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040377271274145916,
      "loss": 4.8831,
      "mean_token_accuracy": 0.22074744552373887,
      "num_tokens": 71443952.0,
      "step": 34430
    },
    {
      "entropy": 5.66825647354126,
      "epoch": 3.1236393323657476,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004037461782344603,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22169191539287567,
      "num_tokens": 71454051.0,
      "step": 34435
    },
    {
      "entropy": 5.597895765304566,
      "epoch": 3.1240928882438315,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040371964106495777,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2259342685341835,
      "num_tokens": 71464579.0,
      "step": 34440
    },
    {
      "entropy": 5.598602676391602,
      "epoch": 3.1245464441219157,
      "grad_norm": 1.25,
      "learning_rate": 0.00040369310123350015,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22197497338056565,
      "num_tokens": 71474631.0,
      "step": 34445
    },
    {
      "entropy": 5.687271547317505,
      "epoch": 3.125,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040366655874063645,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22455458641052245,
      "num_tokens": 71485266.0,
      "step": 34450
    },
    {
      "entropy": 5.5737861633300785,
      "epoch": 3.1254535558780843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004036400135869156,
      "loss": 4.7584,
      "mean_token_accuracy": 0.23174856156110762,
      "num_tokens": 71495248.0,
      "step": 34455
    },
    {
      "entropy": 5.585037755966186,
      "epoch": 3.1259071117561685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040361346577288645,
      "loss": 4.8553,
      "mean_token_accuracy": 0.22350795120000838,
      "num_tokens": 71505600.0,
      "step": 34460
    },
    {
      "entropy": 5.6106870651245115,
      "epoch": 3.1263606676342524,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040358691529909796,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22294523566961288,
      "num_tokens": 71516634.0,
      "step": 34465
    },
    {
      "entropy": 5.68028244972229,
      "epoch": 3.1268142235123366,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040356036216609943,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2163221687078476,
      "num_tokens": 71526745.0,
      "step": 34470
    },
    {
      "entropy": 5.701898574829102,
      "epoch": 3.127267779390421,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040353380637443974,
      "loss": 5.0099,
      "mean_token_accuracy": 0.20294490605592727,
      "num_tokens": 71537015.0,
      "step": 34475
    },
    {
      "entropy": 5.60182147026062,
      "epoch": 3.127721335268505,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040350724792466826,
      "loss": 4.7702,
      "mean_token_accuracy": 0.2262040987610817,
      "num_tokens": 71546660.0,
      "step": 34480
    },
    {
      "entropy": 5.578854274749756,
      "epoch": 3.1281748911465894,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004034806868173341,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22003268897533418,
      "num_tokens": 71557309.0,
      "step": 34485
    },
    {
      "entropy": 5.552659893035889,
      "epoch": 3.1286284470246732,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004034541230529866,
      "loss": 4.7871,
      "mean_token_accuracy": 0.22338122576475145,
      "num_tokens": 71567651.0,
      "step": 34490
    },
    {
      "entropy": 5.569345378875733,
      "epoch": 3.1290820029027575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040342755663217507,
      "loss": 4.8246,
      "mean_token_accuracy": 0.2244902566075325,
      "num_tokens": 71577538.0,
      "step": 34495
    },
    {
      "entropy": 5.5653339385986325,
      "epoch": 3.1295355587808418,
      "grad_norm": 1.15625,
      "learning_rate": 0.000403400987555449,
      "loss": 4.8409,
      "mean_token_accuracy": 0.2266332596540451,
      "num_tokens": 71588299.0,
      "step": 34500
    },
    {
      "entropy": 5.623280048370361,
      "epoch": 3.129989114658926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004033744158233577,
      "loss": 4.937,
      "mean_token_accuracy": 0.21450992077589034,
      "num_tokens": 71598989.0,
      "step": 34505
    },
    {
      "entropy": 5.644419574737549,
      "epoch": 3.1304426705370103,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040334784143645087,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21343223452568055,
      "num_tokens": 71608596.0,
      "step": 34510
    },
    {
      "entropy": 5.553900718688965,
      "epoch": 3.1308962264150946,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004033212643952779,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2250541105866432,
      "num_tokens": 71619182.0,
      "step": 34515
    },
    {
      "entropy": 5.612643527984619,
      "epoch": 3.1313497822931784,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040329468470038857,
      "loss": 4.9061,
      "mean_token_accuracy": 0.21922181397676468,
      "num_tokens": 71629526.0,
      "step": 34520
    },
    {
      "entropy": 5.663424348831176,
      "epoch": 3.1318033381712627,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004032681023523324,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21724731475114822,
      "num_tokens": 71640146.0,
      "step": 34525
    },
    {
      "entropy": 5.621704196929931,
      "epoch": 3.132256894049347,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004032415173516592,
      "loss": 4.8004,
      "mean_token_accuracy": 0.22204351276159287,
      "num_tokens": 71649652.0,
      "step": 34530
    },
    {
      "entropy": 5.61730318069458,
      "epoch": 3.132710449927431,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004032149296989187,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22444941401481627,
      "num_tokens": 71659264.0,
      "step": 34535
    },
    {
      "entropy": 5.61833643913269,
      "epoch": 3.133164005805515,
      "grad_norm": 1.25,
      "learning_rate": 0.00040318833939466083,
      "loss": 4.8903,
      "mean_token_accuracy": 0.2178942382335663,
      "num_tokens": 71669175.0,
      "step": 34540
    },
    {
      "entropy": 5.572068166732788,
      "epoch": 3.1336175616835993,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004031617464394354,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22315966784954072,
      "num_tokens": 71679537.0,
      "step": 34545
    },
    {
      "entropy": 5.625502586364746,
      "epoch": 3.1340711175616836,
      "grad_norm": 1.25,
      "learning_rate": 0.0004031351508337925,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2195585012435913,
      "num_tokens": 71690024.0,
      "step": 34550
    },
    {
      "entropy": 5.686118745803833,
      "epoch": 3.134524673439768,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004031085525782819,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21952289789915086,
      "num_tokens": 71700019.0,
      "step": 34555
    },
    {
      "entropy": 5.644762468338013,
      "epoch": 3.134978229317852,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004030819516734539,
      "loss": 4.9301,
      "mean_token_accuracy": 0.20568177253007888,
      "num_tokens": 71710570.0,
      "step": 34560
    },
    {
      "entropy": 5.609179210662842,
      "epoch": 3.1354317851959363,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040305534811985844,
      "loss": 4.9252,
      "mean_token_accuracy": 0.22606145441532136,
      "num_tokens": 71720619.0,
      "step": 34565
    },
    {
      "entropy": 5.5872642993927,
      "epoch": 3.13588534107402,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004030287419180457,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22151790261268617,
      "num_tokens": 71730201.0,
      "step": 34570
    },
    {
      "entropy": 5.643096542358398,
      "epoch": 3.1363388969521044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040300213306856606,
      "loss": 4.8998,
      "mean_token_accuracy": 0.22392575144767762,
      "num_tokens": 71741368.0,
      "step": 34575
    },
    {
      "entropy": 5.585963392257691,
      "epoch": 3.1367924528301887,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004029755215719696,
      "loss": 4.8117,
      "mean_token_accuracy": 0.22652245461940765,
      "num_tokens": 71751361.0,
      "step": 34580
    },
    {
      "entropy": 5.626019906997681,
      "epoch": 3.137246008708273,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040294890742880677,
      "loss": 4.8623,
      "mean_token_accuracy": 0.2234893262386322,
      "num_tokens": 71761164.0,
      "step": 34585
    },
    {
      "entropy": 5.591586017608643,
      "epoch": 3.1376995645863572,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040292229063962787,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22718488723039626,
      "num_tokens": 71770726.0,
      "step": 34590
    },
    {
      "entropy": 5.636309623718262,
      "epoch": 3.138153120464441,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004028956712049834,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22407835274934768,
      "num_tokens": 71781124.0,
      "step": 34595
    },
    {
      "entropy": 5.6022295475006105,
      "epoch": 3.1386066763425253,
      "grad_norm": 1.15625,
      "learning_rate": 0.000402869049125424,
      "loss": 4.8514,
      "mean_token_accuracy": 0.21854931563138963,
      "num_tokens": 71791218.0,
      "step": 34600
    },
    {
      "entropy": 5.610520124435425,
      "epoch": 3.1390602322206096,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004028424244015,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22201108783483506,
      "num_tokens": 71801013.0,
      "step": 34605
    },
    {
      "entropy": 5.637994575500488,
      "epoch": 3.139513788098694,
      "grad_norm": 1.171875,
      "learning_rate": 0.000402815797033762,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2094426929950714,
      "num_tokens": 71811742.0,
      "step": 34610
    },
    {
      "entropy": 5.560222387313843,
      "epoch": 3.139967343976778,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004027891670227609,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2140335962176323,
      "num_tokens": 71822265.0,
      "step": 34615
    },
    {
      "entropy": 5.627309799194336,
      "epoch": 3.140420899854862,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004027625343690472,
      "loss": 4.8382,
      "mean_token_accuracy": 0.219781094789505,
      "num_tokens": 71832090.0,
      "step": 34620
    },
    {
      "entropy": 5.603528928756714,
      "epoch": 3.140874455732946,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004027358990731717,
      "loss": 4.8648,
      "mean_token_accuracy": 0.22248990088701248,
      "num_tokens": 71841381.0,
      "step": 34625
    },
    {
      "entropy": 5.588434362411499,
      "epoch": 3.1413280116110305,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040270926113568524,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21590789556503295,
      "num_tokens": 71851383.0,
      "step": 34630
    },
    {
      "entropy": 5.592743062973023,
      "epoch": 3.1417815674891147,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040268262055713873,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22258922159671785,
      "num_tokens": 71862588.0,
      "step": 34635
    },
    {
      "entropy": 5.606044101715088,
      "epoch": 3.142235123367199,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004026559773380831,
      "loss": 4.848,
      "mean_token_accuracy": 0.22057845443487167,
      "num_tokens": 71873975.0,
      "step": 34640
    },
    {
      "entropy": 5.683474063873291,
      "epoch": 3.142688679245283,
      "grad_norm": 1.375,
      "learning_rate": 0.0004026293314790694,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2113919660449028,
      "num_tokens": 71884084.0,
      "step": 34645
    },
    {
      "entropy": 5.585837554931641,
      "epoch": 3.143142235123367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004026026829806485,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21480174064636232,
      "num_tokens": 71894643.0,
      "step": 34650
    },
    {
      "entropy": 5.5260179996490475,
      "epoch": 3.1435957910014514,
      "grad_norm": 1.25,
      "learning_rate": 0.00040257603184337167,
      "loss": 4.8161,
      "mean_token_accuracy": 0.22148592174053192,
      "num_tokens": 71905245.0,
      "step": 34655
    },
    {
      "entropy": 5.635767316818237,
      "epoch": 3.1440493468795356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040254937806778995,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2123051702976227,
      "num_tokens": 71916473.0,
      "step": 34660
    },
    {
      "entropy": 5.624259424209595,
      "epoch": 3.14450290275762,
      "grad_norm": 1.125,
      "learning_rate": 0.00040252272165445456,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22096159309148788,
      "num_tokens": 71927465.0,
      "step": 34665
    },
    {
      "entropy": 5.640947771072388,
      "epoch": 3.1449564586357037,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004024960626039168,
      "loss": 4.8938,
      "mean_token_accuracy": 0.22290985137224198,
      "num_tokens": 71937380.0,
      "step": 34670
    },
    {
      "entropy": 5.600855922698974,
      "epoch": 3.145410014513788,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040246940091672806,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2250762850046158,
      "num_tokens": 71948088.0,
      "step": 34675
    },
    {
      "entropy": 5.62983775138855,
      "epoch": 3.1458635703918723,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004024427365934395,
      "loss": 4.838,
      "mean_token_accuracy": 0.22383915185928344,
      "num_tokens": 71958284.0,
      "step": 34680
    },
    {
      "entropy": 5.612975645065307,
      "epoch": 3.1463171262699565,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040241606963460283,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21289132982492448,
      "num_tokens": 71968138.0,
      "step": 34685
    },
    {
      "entropy": 5.664534997940064,
      "epoch": 3.146770682148041,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040238940004076924,
      "loss": 4.9593,
      "mean_token_accuracy": 0.21742864400148393,
      "num_tokens": 71978275.0,
      "step": 34690
    },
    {
      "entropy": 5.586671018600464,
      "epoch": 3.1472242380261246,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040236272781249047,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2201538622379303,
      "num_tokens": 71988875.0,
      "step": 34695
    },
    {
      "entropy": 5.638438892364502,
      "epoch": 3.147677793904209,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040233605295031795,
      "loss": 4.8601,
      "mean_token_accuracy": 0.2229277342557907,
      "num_tokens": 71999560.0,
      "step": 34700
    },
    {
      "entropy": 5.638119268417358,
      "epoch": 3.148131349782293,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004023093754548034,
      "loss": 4.889,
      "mean_token_accuracy": 0.22454825639724732,
      "num_tokens": 72009970.0,
      "step": 34705
    },
    {
      "entropy": 5.624862957000732,
      "epoch": 3.1485849056603774,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004022826953264986,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22717690467834473,
      "num_tokens": 72020196.0,
      "step": 34710
    },
    {
      "entropy": 5.586294746398925,
      "epoch": 3.1490384615384617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040225601256595525,
      "loss": 4.8148,
      "mean_token_accuracy": 0.22161322236061096,
      "num_tokens": 72031456.0,
      "step": 34715
    },
    {
      "entropy": 5.622655630111694,
      "epoch": 3.149492017416546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000402229327173725,
      "loss": 4.9443,
      "mean_token_accuracy": 0.2180787667632103,
      "num_tokens": 72042260.0,
      "step": 34720
    },
    {
      "entropy": 5.698824262619018,
      "epoch": 3.1499455732946298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040220263915035997,
      "loss": 5.0102,
      "mean_token_accuracy": 0.20166546255350112,
      "num_tokens": 72053463.0,
      "step": 34725
    },
    {
      "entropy": 5.749280071258545,
      "epoch": 3.150399129172714,
      "grad_norm": 1.25,
      "learning_rate": 0.00040217594849641195,
      "loss": 5.001,
      "mean_token_accuracy": 0.20983909219503402,
      "num_tokens": 72064508.0,
      "step": 34730
    },
    {
      "entropy": 5.576893901824951,
      "epoch": 3.1508526850507983,
      "grad_norm": 1.25,
      "learning_rate": 0.00040214925521243287,
      "loss": 4.8943,
      "mean_token_accuracy": 0.21324467211961745,
      "num_tokens": 72074758.0,
      "step": 34735
    },
    {
      "entropy": 5.597273683547973,
      "epoch": 3.1513062409288826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040212255929897474,
      "loss": 4.8601,
      "mean_token_accuracy": 0.22307022362947465,
      "num_tokens": 72085548.0,
      "step": 34740
    },
    {
      "entropy": 5.614319133758545,
      "epoch": 3.1517597968069664,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004020958607565897,
      "loss": 4.9774,
      "mean_token_accuracy": 0.2147376850247383,
      "num_tokens": 72096773.0,
      "step": 34745
    },
    {
      "entropy": 5.632926559448242,
      "epoch": 3.1522133526850507,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004020691595858299,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21896272301673889,
      "num_tokens": 72106832.0,
      "step": 34750
    },
    {
      "entropy": 5.633471155166626,
      "epoch": 3.152666908563135,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040204245578724754,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21699773967266084,
      "num_tokens": 72117758.0,
      "step": 34755
    },
    {
      "entropy": 5.638801765441895,
      "epoch": 3.153120464441219,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040201574936139475,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2162176862359047,
      "num_tokens": 72127522.0,
      "step": 34760
    },
    {
      "entropy": 5.565290021896362,
      "epoch": 3.1535740203193035,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004019890403088239,
      "loss": 4.8964,
      "mean_token_accuracy": 0.22387027591466904,
      "num_tokens": 72138310.0,
      "step": 34765
    },
    {
      "entropy": 5.589483213424683,
      "epoch": 3.1540275761973877,
      "grad_norm": 1.25,
      "learning_rate": 0.00040196232863008735,
      "loss": 4.9189,
      "mean_token_accuracy": 0.22281512171030043,
      "num_tokens": 72148958.0,
      "step": 34770
    },
    {
      "entropy": 5.669883108139038,
      "epoch": 3.1544811320754715,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040193561432573754,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22320373654365538,
      "num_tokens": 72158594.0,
      "step": 34775
    },
    {
      "entropy": 5.688465309143067,
      "epoch": 3.154934687953556,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004019088973963268,
      "loss": 4.9651,
      "mean_token_accuracy": 0.2170100286602974,
      "num_tokens": 72168552.0,
      "step": 34780
    },
    {
      "entropy": 5.604195976257325,
      "epoch": 3.15538824383164,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040188217784240785,
      "loss": 4.8433,
      "mean_token_accuracy": 0.222317111492157,
      "num_tokens": 72178584.0,
      "step": 34785
    },
    {
      "entropy": 5.595452451705933,
      "epoch": 3.1558417997097243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000401855455664533,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22452037483453752,
      "num_tokens": 72189144.0,
      "step": 34790
    },
    {
      "entropy": 5.6572870254516605,
      "epoch": 3.1562953555878086,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004018287308632551,
      "loss": 4.9045,
      "mean_token_accuracy": 0.22017716318368913,
      "num_tokens": 72198326.0,
      "step": 34795
    },
    {
      "entropy": 5.655667352676391,
      "epoch": 3.1567489114658924,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004018020034391266,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21748037934303283,
      "num_tokens": 72207916.0,
      "step": 34800
    },
    {
      "entropy": 5.657133388519287,
      "epoch": 3.1572024673439767,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004017752733927004,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22074204981327056,
      "num_tokens": 72219111.0,
      "step": 34805
    },
    {
      "entropy": 5.62120566368103,
      "epoch": 3.157656023222061,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004017485407245293,
      "loss": 4.91,
      "mean_token_accuracy": 0.22101642936468124,
      "num_tokens": 72230081.0,
      "step": 34810
    },
    {
      "entropy": 5.5638477325439455,
      "epoch": 3.1581095791001452,
      "grad_norm": 1.296875,
      "learning_rate": 0.000401721805435166,
      "loss": 4.8412,
      "mean_token_accuracy": 0.22440837621688842,
      "num_tokens": 72240693.0,
      "step": 34815
    },
    {
      "entropy": 5.526685953140259,
      "epoch": 3.1585631349782295,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040169506752516356,
      "loss": 4.8116,
      "mean_token_accuracy": 0.2326302409172058,
      "num_tokens": 72250925.0,
      "step": 34820
    },
    {
      "entropy": 5.547344970703125,
      "epoch": 3.1590166908563133,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004016683269950748,
      "loss": 4.8229,
      "mean_token_accuracy": 0.2293567731976509,
      "num_tokens": 72261027.0,
      "step": 34825
    },
    {
      "entropy": 5.63401837348938,
      "epoch": 3.1594702467343976,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040164158384545267,
      "loss": 4.7801,
      "mean_token_accuracy": 0.2347550079226494,
      "num_tokens": 72269430.0,
      "step": 34830
    },
    {
      "entropy": 5.541888332366943,
      "epoch": 3.159923802612482,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004016148380768504,
      "loss": 4.8173,
      "mean_token_accuracy": 0.22614311128854753,
      "num_tokens": 72280207.0,
      "step": 34835
    },
    {
      "entropy": 5.629393196105957,
      "epoch": 3.160377358490566,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040158808968982103,
      "loss": 4.9876,
      "mean_token_accuracy": 0.22297317683696746,
      "num_tokens": 72290093.0,
      "step": 34840
    },
    {
      "entropy": 5.591180944442749,
      "epoch": 3.1608309143686504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040156133868491766,
      "loss": 4.8083,
      "mean_token_accuracy": 0.22583741843700408,
      "num_tokens": 72300203.0,
      "step": 34845
    },
    {
      "entropy": 5.624714422225952,
      "epoch": 3.161284470246734,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040153458506269365,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21573832333087922,
      "num_tokens": 72311778.0,
      "step": 34850
    },
    {
      "entropy": 5.624976778030396,
      "epoch": 3.1617380261248185,
      "grad_norm": 1.296875,
      "learning_rate": 0.000401507828823702,
      "loss": 4.862,
      "mean_token_accuracy": 0.22003922164440154,
      "num_tokens": 72322142.0,
      "step": 34855
    },
    {
      "entropy": 5.652061033248901,
      "epoch": 3.1621915820029027,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040148106996849636,
      "loss": 4.939,
      "mean_token_accuracy": 0.21862932592630385,
      "num_tokens": 72332312.0,
      "step": 34860
    },
    {
      "entropy": 5.522206974029541,
      "epoch": 3.162645137880987,
      "grad_norm": 1.25,
      "learning_rate": 0.00040145430849762993,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2223895251750946,
      "num_tokens": 72342157.0,
      "step": 34865
    },
    {
      "entropy": 5.633262014389038,
      "epoch": 3.1630986937590713,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040142754441165615,
      "loss": 4.9032,
      "mean_token_accuracy": 0.21996026039123534,
      "num_tokens": 72353417.0,
      "step": 34870
    },
    {
      "entropy": 5.666251230239868,
      "epoch": 3.1635522496371555,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004014007777111285,
      "loss": 4.8583,
      "mean_token_accuracy": 0.22641475796699523,
      "num_tokens": 72363744.0,
      "step": 34875
    },
    {
      "entropy": 5.633161830902099,
      "epoch": 3.1640058055152394,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040137400839660066,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21772806495428085,
      "num_tokens": 72375241.0,
      "step": 34880
    },
    {
      "entropy": 5.6322822093963625,
      "epoch": 3.1644593613933236,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040134723646862607,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21977696418762208,
      "num_tokens": 72385475.0,
      "step": 34885
    },
    {
      "entropy": 5.613815355300903,
      "epoch": 3.164912917271408,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040132046192775835,
      "loss": 4.9637,
      "mean_token_accuracy": 0.21981662660837173,
      "num_tokens": 72396417.0,
      "step": 34890
    },
    {
      "entropy": 5.6317102909088135,
      "epoch": 3.165366473149492,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004012936847745514,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22171031087636947,
      "num_tokens": 72406169.0,
      "step": 34895
    },
    {
      "entropy": 5.636847257614136,
      "epoch": 3.165820029027576,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004012669050095589,
      "loss": 4.9063,
      "mean_token_accuracy": 0.21598950028419495,
      "num_tokens": 72416522.0,
      "step": 34900
    },
    {
      "entropy": 5.657953691482544,
      "epoch": 3.1662735849056602,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004012401226333345,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2198248729109764,
      "num_tokens": 72426000.0,
      "step": 34905
    },
    {
      "entropy": 5.639194393157959,
      "epoch": 3.1667271407837445,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004012133376464322,
      "loss": 4.9869,
      "mean_token_accuracy": 0.2110868737101555,
      "num_tokens": 72435952.0,
      "step": 34910
    },
    {
      "entropy": 5.650312232971191,
      "epoch": 3.1671806966618288,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040118655004940597,
      "loss": 4.9467,
      "mean_token_accuracy": 0.2171285942196846,
      "num_tokens": 72449957.0,
      "step": 34915
    },
    {
      "entropy": 5.622080850601196,
      "epoch": 3.167634252539913,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004011597598428097,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2260361298918724,
      "num_tokens": 72459956.0,
      "step": 34920
    },
    {
      "entropy": 5.588940715789795,
      "epoch": 3.1680878084179973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004011329670271975,
      "loss": 4.8308,
      "mean_token_accuracy": 0.2241721048951149,
      "num_tokens": 72471107.0,
      "step": 34925
    },
    {
      "entropy": 5.5678815841674805,
      "epoch": 3.168541364296081,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040110617160312334,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22342223376035691,
      "num_tokens": 72481428.0,
      "step": 34930
    },
    {
      "entropy": 5.6213951587677,
      "epoch": 3.1689949201741654,
      "grad_norm": 1.25,
      "learning_rate": 0.0004010793735711415,
      "loss": 4.928,
      "mean_token_accuracy": 0.21212025880813598,
      "num_tokens": 72491133.0,
      "step": 34935
    },
    {
      "entropy": 5.549967670440674,
      "epoch": 3.1694484760522497,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040105257293180605,
      "loss": 4.8305,
      "mean_token_accuracy": 0.22898245602846146,
      "num_tokens": 72502047.0,
      "step": 34940
    },
    {
      "entropy": 5.6256640434265135,
      "epoch": 3.169902031930334,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004010257696856712,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21532421112060546,
      "num_tokens": 72513120.0,
      "step": 34945
    },
    {
      "entropy": 5.688839244842529,
      "epoch": 3.170355587808418,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040099896383329135,
      "loss": 5.0015,
      "mean_token_accuracy": 0.22066273391246796,
      "num_tokens": 72523118.0,
      "step": 34950
    },
    {
      "entropy": 5.685703706741333,
      "epoch": 3.170809143686502,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040097215537522086,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21737915724515916,
      "num_tokens": 72533697.0,
      "step": 34955
    },
    {
      "entropy": 5.54258279800415,
      "epoch": 3.1712626995645863,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040094534431201406,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23132589757442473,
      "num_tokens": 72543661.0,
      "step": 34960
    },
    {
      "entropy": 5.628453111648559,
      "epoch": 3.1717162554426706,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004009185306442255,
      "loss": 4.9677,
      "mean_token_accuracy": 0.2180483266711235,
      "num_tokens": 72554551.0,
      "step": 34965
    },
    {
      "entropy": 5.649918031692505,
      "epoch": 3.172169811320755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040089171437240954,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21000791490077972,
      "num_tokens": 72565134.0,
      "step": 34970
    },
    {
      "entropy": 5.615237665176392,
      "epoch": 3.172623367198839,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040086489549712097,
      "loss": 4.896,
      "mean_token_accuracy": 0.2244971662759781,
      "num_tokens": 72574981.0,
      "step": 34975
    },
    {
      "entropy": 5.63043999671936,
      "epoch": 3.173076923076923,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004008380740189142,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2216053009033203,
      "num_tokens": 72585073.0,
      "step": 34980
    },
    {
      "entropy": 5.680956888198852,
      "epoch": 3.173530478955007,
      "grad_norm": 1.203125,
      "learning_rate": 0.000400811249938344,
      "loss": 4.9045,
      "mean_token_accuracy": 0.2270319476723671,
      "num_tokens": 72595719.0,
      "step": 34985
    },
    {
      "entropy": 5.691639041900634,
      "epoch": 3.1739840348330914,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004007844232559652,
      "loss": 4.8925,
      "mean_token_accuracy": 0.21655526012182236,
      "num_tokens": 72605276.0,
      "step": 34990
    },
    {
      "entropy": 5.660261392593384,
      "epoch": 3.1744375907111757,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004007575939723324,
      "loss": 4.9393,
      "mean_token_accuracy": 0.21795331686735153,
      "num_tokens": 72614964.0,
      "step": 34995
    },
    {
      "entropy": 5.660168313980103,
      "epoch": 3.17489114658926,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004007307620880004,
      "loss": 4.9311,
      "mean_token_accuracy": 0.2194843575358391,
      "num_tokens": 72625461.0,
      "step": 35000
    },
    {
      "entropy": 5.631026840209961,
      "epoch": 3.175344702467344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040070392760352424,
      "loss": 4.9287,
      "mean_token_accuracy": 0.22143623232841492,
      "num_tokens": 72636192.0,
      "step": 35005
    },
    {
      "entropy": 5.669824457168579,
      "epoch": 3.175798258345428,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040067709051945895,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2148689240217209,
      "num_tokens": 72648432.0,
      "step": 35010
    },
    {
      "entropy": 5.632280921936035,
      "epoch": 3.1762518142235123,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004006502508363593,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2156403437256813,
      "num_tokens": 72658678.0,
      "step": 35015
    },
    {
      "entropy": 5.712354135513306,
      "epoch": 3.1767053701015966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040062340855478046,
      "loss": 5.0534,
      "mean_token_accuracy": 0.2102057456970215,
      "num_tokens": 72669191.0,
      "step": 35020
    },
    {
      "entropy": 5.638961601257324,
      "epoch": 3.177158925979681,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040059656367527754,
      "loss": 4.8681,
      "mean_token_accuracy": 0.2249057933688164,
      "num_tokens": 72679327.0,
      "step": 35025
    },
    {
      "entropy": 5.633044338226318,
      "epoch": 3.1776124818577647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004005697161984056,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21578340381383895,
      "num_tokens": 72689673.0,
      "step": 35030
    },
    {
      "entropy": 5.614214658737183,
      "epoch": 3.178066037735849,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040054286612472006,
      "loss": 4.9667,
      "mean_token_accuracy": 0.20773992985486983,
      "num_tokens": 72700606.0,
      "step": 35035
    },
    {
      "entropy": 5.715998983383178,
      "epoch": 3.178519593613933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040051601345477593,
      "loss": 4.8959,
      "mean_token_accuracy": 0.2197418689727783,
      "num_tokens": 72711195.0,
      "step": 35040
    },
    {
      "entropy": 5.6628399848937985,
      "epoch": 3.1789731494920175,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040048915818912875,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2142451986670494,
      "num_tokens": 72721160.0,
      "step": 35045
    },
    {
      "entropy": 5.606712341308594,
      "epoch": 3.1794267053701017,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040046230032833374,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21797827780246734,
      "num_tokens": 72730867.0,
      "step": 35050
    },
    {
      "entropy": 5.594057321548462,
      "epoch": 3.1798802612481856,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004004354398729464,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21410192549228668,
      "num_tokens": 72740453.0,
      "step": 35055
    },
    {
      "entropy": 5.670143795013428,
      "epoch": 3.18033381712627,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004004085768235222,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2144264131784439,
      "num_tokens": 72751395.0,
      "step": 35060
    },
    {
      "entropy": 5.630158948898315,
      "epoch": 3.180787373004354,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040038171118061673,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22281871140003204,
      "num_tokens": 72761300.0,
      "step": 35065
    },
    {
      "entropy": 5.594234037399292,
      "epoch": 3.1812409288824384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004003548429447855,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22644988000392913,
      "num_tokens": 72772011.0,
      "step": 35070
    },
    {
      "entropy": 5.634019613265991,
      "epoch": 3.1816944847605226,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004003279721165841,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21830470860004425,
      "num_tokens": 72782521.0,
      "step": 35075
    },
    {
      "entropy": 5.5902174949646,
      "epoch": 3.182148040638607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004003010986965683,
      "loss": 4.8103,
      "mean_token_accuracy": 0.22426186203956605,
      "num_tokens": 72793125.0,
      "step": 35080
    },
    {
      "entropy": 5.6961350440979,
      "epoch": 3.1826015965166907,
      "grad_norm": 1.234375,
      "learning_rate": 0.000400274222685294,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2188430055975914,
      "num_tokens": 72804541.0,
      "step": 35085
    },
    {
      "entropy": 5.7512688636779785,
      "epoch": 3.183055152394775,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040024734408331664,
      "loss": 4.9949,
      "mean_token_accuracy": 0.22102211117744447,
      "num_tokens": 72814447.0,
      "step": 35090
    },
    {
      "entropy": 5.653251791000367,
      "epoch": 3.1835087082728593,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040022046289119246,
      "loss": 4.9335,
      "mean_token_accuracy": 0.22518065571784973,
      "num_tokens": 72823545.0,
      "step": 35095
    },
    {
      "entropy": 5.610453510284424,
      "epoch": 3.1839622641509435,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040019357910947714,
      "loss": 4.7855,
      "mean_token_accuracy": 0.22406467497348787,
      "num_tokens": 72833803.0,
      "step": 35100
    },
    {
      "entropy": 5.680115413665772,
      "epoch": 3.1844158200290273,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040016669273872666,
      "loss": 4.9328,
      "mean_token_accuracy": 0.2164497271180153,
      "num_tokens": 72843383.0,
      "step": 35105
    },
    {
      "entropy": 5.7054911136627195,
      "epoch": 3.1848693759071116,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040013980377949716,
      "loss": 4.9762,
      "mean_token_accuracy": 0.21454016566276551,
      "num_tokens": 72853285.0,
      "step": 35110
    },
    {
      "entropy": 5.637636518478393,
      "epoch": 3.185322931785196,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040011291223234456,
      "loss": 4.9538,
      "mean_token_accuracy": 0.21610147058963775,
      "num_tokens": 72863556.0,
      "step": 35115
    },
    {
      "entropy": 5.640329504013062,
      "epoch": 3.18577648766328,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040008601809782507,
      "loss": 4.8336,
      "mean_token_accuracy": 0.2251451939344406,
      "num_tokens": 72873296.0,
      "step": 35120
    },
    {
      "entropy": 5.668989276885986,
      "epoch": 3.1862300435413644,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004000591213764948,
      "loss": 4.9825,
      "mean_token_accuracy": 0.20847248286008835,
      "num_tokens": 72884738.0,
      "step": 35125
    },
    {
      "entropy": 5.691875886917114,
      "epoch": 3.1866835994194487,
      "grad_norm": 1.453125,
      "learning_rate": 0.00040003222206891017,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20698761194944382,
      "num_tokens": 72894573.0,
      "step": 35130
    },
    {
      "entropy": 5.567454862594604,
      "epoch": 3.1871371552975325,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004000053201756272,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22272568494081496,
      "num_tokens": 72904976.0,
      "step": 35135
    },
    {
      "entropy": 5.612062549591064,
      "epoch": 3.1875907111756168,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003999784156972024,
      "loss": 4.9129,
      "mean_token_accuracy": 0.21314848512411116,
      "num_tokens": 72915152.0,
      "step": 35140
    },
    {
      "entropy": 5.6872917175292965,
      "epoch": 3.188044267053701,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003999515086341921,
      "loss": 4.9969,
      "mean_token_accuracy": 0.21276690661907197,
      "num_tokens": 72925450.0,
      "step": 35145
    },
    {
      "entropy": 5.622509717941284,
      "epoch": 3.1884978229317853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003999245989871527,
      "loss": 4.8371,
      "mean_token_accuracy": 0.22074327170848845,
      "num_tokens": 72935990.0,
      "step": 35150
    },
    {
      "entropy": 5.610948801040649,
      "epoch": 3.1889513788098696,
      "grad_norm": 1.25,
      "learning_rate": 0.00039989768675664084,
      "loss": 4.7773,
      "mean_token_accuracy": 0.22453771084547042,
      "num_tokens": 72945662.0,
      "step": 35155
    },
    {
      "entropy": 5.562100028991699,
      "epoch": 3.1894049346879534,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039987077194321305,
      "loss": 4.7521,
      "mean_token_accuracy": 0.2287294700741768,
      "num_tokens": 72955313.0,
      "step": 35160
    },
    {
      "entropy": 5.677704954147339,
      "epoch": 3.1898584905660377,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039984385454742576,
      "loss": 5.0146,
      "mean_token_accuracy": 0.20997219979763032,
      "num_tokens": 72965347.0,
      "step": 35165
    },
    {
      "entropy": 5.59677267074585,
      "epoch": 3.190312046444122,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039981693456983574,
      "loss": 4.8249,
      "mean_token_accuracy": 0.22215677350759505,
      "num_tokens": 72976794.0,
      "step": 35170
    },
    {
      "entropy": 5.597756052017212,
      "epoch": 3.190765602322206,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003997900120109998,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2243738904595375,
      "num_tokens": 72986837.0,
      "step": 35175
    },
    {
      "entropy": 5.600722742080689,
      "epoch": 3.1912191582002905,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003997630868714745,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21610475033521653,
      "num_tokens": 72996819.0,
      "step": 35180
    },
    {
      "entropy": 5.581478214263916,
      "epoch": 3.1916727140783743,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039973615915181687,
      "loss": 4.7794,
      "mean_token_accuracy": 0.22965004295110703,
      "num_tokens": 73005963.0,
      "step": 35185
    },
    {
      "entropy": 5.610789489746094,
      "epoch": 3.1921262699564585,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039970922885258363,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2169812560081482,
      "num_tokens": 73015912.0,
      "step": 35190
    },
    {
      "entropy": 5.562548494338989,
      "epoch": 3.192579825834543,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003996822959743318,
      "loss": 4.8046,
      "mean_token_accuracy": 0.22984550297260284,
      "num_tokens": 73026528.0,
      "step": 35195
    },
    {
      "entropy": 5.629562854766846,
      "epoch": 3.193033381712627,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003996553605176183,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2149041324853897,
      "num_tokens": 73036255.0,
      "step": 35200
    },
    {
      "entropy": 5.655352020263672,
      "epoch": 3.1934869375907113,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003996284224830002,
      "loss": 5.0011,
      "mean_token_accuracy": 0.21483247429132463,
      "num_tokens": 73046686.0,
      "step": 35205
    },
    {
      "entropy": 5.6188966751098635,
      "epoch": 3.193940493468795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039960148187103453,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2261397659778595,
      "num_tokens": 73056070.0,
      "step": 35210
    },
    {
      "entropy": 5.70836615562439,
      "epoch": 3.1943940493468794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039957453868227845,
      "loss": 4.9427,
      "mean_token_accuracy": 0.22494465857744217,
      "num_tokens": 73066288.0,
      "step": 35215
    },
    {
      "entropy": 5.64421329498291,
      "epoch": 3.1948476052249637,
      "grad_norm": 1.25,
      "learning_rate": 0.00039954759291728915,
      "loss": 4.9006,
      "mean_token_accuracy": 0.21916361898183823,
      "num_tokens": 73076297.0,
      "step": 35220
    },
    {
      "entropy": 5.6523832321167,
      "epoch": 3.195301161103048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039952064457662395,
      "loss": 4.9785,
      "mean_token_accuracy": 0.21674781888723374,
      "num_tokens": 73086249.0,
      "step": 35225
    },
    {
      "entropy": 5.581540584564209,
      "epoch": 3.1957547169811322,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039949369366084,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22464638501405715,
      "num_tokens": 73096358.0,
      "step": 35230
    },
    {
      "entropy": 5.576098346710205,
      "epoch": 3.196208272859216,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039946674017049484,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22194795459508895,
      "num_tokens": 73106891.0,
      "step": 35235
    },
    {
      "entropy": 5.5393165111541744,
      "epoch": 3.1966618287373003,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039943978410614565,
      "loss": 4.8461,
      "mean_token_accuracy": 0.22040769308805466,
      "num_tokens": 73117106.0,
      "step": 35240
    },
    {
      "entropy": 5.629960393905639,
      "epoch": 3.1971153846153846,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039941282546835017,
      "loss": 4.9235,
      "mean_token_accuracy": 0.22212623953819274,
      "num_tokens": 73127112.0,
      "step": 35245
    },
    {
      "entropy": 5.632509374618531,
      "epoch": 3.197568940493469,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039938586425766564,
      "loss": 4.8958,
      "mean_token_accuracy": 0.22116705030202866,
      "num_tokens": 73137751.0,
      "step": 35250
    },
    {
      "entropy": 5.659935188293457,
      "epoch": 3.198022496371553,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003993589004746498,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22556164264678955,
      "num_tokens": 73149303.0,
      "step": 35255
    },
    {
      "entropy": 5.602596807479858,
      "epoch": 3.198476052249637,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039933193411986024,
      "loss": 4.8818,
      "mean_token_accuracy": 0.2224455252289772,
      "num_tokens": 73160025.0,
      "step": 35260
    },
    {
      "entropy": 5.588270854949951,
      "epoch": 3.198929608127721,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039930496519385454,
      "loss": 4.869,
      "mean_token_accuracy": 0.22529269903898239,
      "num_tokens": 73170721.0,
      "step": 35265
    },
    {
      "entropy": 5.643705558776856,
      "epoch": 3.1993831640058055,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039927799369719047,
      "loss": 4.9377,
      "mean_token_accuracy": 0.2158379837870598,
      "num_tokens": 73180902.0,
      "step": 35270
    },
    {
      "entropy": 5.537416172027588,
      "epoch": 3.1998367198838897,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003992510196304258,
      "loss": 4.7598,
      "mean_token_accuracy": 0.22765354961156845,
      "num_tokens": 73191259.0,
      "step": 35275
    },
    {
      "entropy": 5.614968538284302,
      "epoch": 3.200290275761974,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039922404299411846,
      "loss": 4.8887,
      "mean_token_accuracy": 0.22845399230718613,
      "num_tokens": 73201654.0,
      "step": 35280
    },
    {
      "entropy": 5.624163722991943,
      "epoch": 3.2007438316400583,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003991970637888262,
      "loss": 4.9304,
      "mean_token_accuracy": 0.2075518861413002,
      "num_tokens": 73212658.0,
      "step": 35285
    },
    {
      "entropy": 5.67769227027893,
      "epoch": 3.201197387518142,
      "grad_norm": 1.296875,
      "learning_rate": 0.000399170082015107,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2146735519170761,
      "num_tokens": 73222883.0,
      "step": 35290
    },
    {
      "entropy": 5.599152755737305,
      "epoch": 3.2016509433962264,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000399143097673519,
      "loss": 4.8653,
      "mean_token_accuracy": 0.22059713453054428,
      "num_tokens": 73233731.0,
      "step": 35295
    },
    {
      "entropy": 5.600179052352905,
      "epoch": 3.2021044992743106,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039911611076461995,
      "loss": 4.8098,
      "mean_token_accuracy": 0.23347573727369308,
      "num_tokens": 73243794.0,
      "step": 35300
    },
    {
      "entropy": 5.625681829452515,
      "epoch": 3.202558055152395,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039908912128896817,
      "loss": 4.8797,
      "mean_token_accuracy": 0.21602460891008377,
      "num_tokens": 73254708.0,
      "step": 35305
    },
    {
      "entropy": 5.56280665397644,
      "epoch": 3.203011611030479,
      "grad_norm": 1.25,
      "learning_rate": 0.0003990621292471217,
      "loss": 4.8724,
      "mean_token_accuracy": 0.2140805959701538,
      "num_tokens": 73265311.0,
      "step": 35310
    },
    {
      "entropy": 5.72505669593811,
      "epoch": 3.203465166908563,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039903513463963874,
      "loss": 5.0589,
      "mean_token_accuracy": 0.21287549585103988,
      "num_tokens": 73275886.0,
      "step": 35315
    },
    {
      "entropy": 5.665661716461182,
      "epoch": 3.2039187227866472,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003990081374670777,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2260631799697876,
      "num_tokens": 73285903.0,
      "step": 35320
    },
    {
      "entropy": 5.600179815292359,
      "epoch": 3.2043722786647315,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039898113772999665,
      "loss": 4.881,
      "mean_token_accuracy": 0.21989596933126448,
      "num_tokens": 73296571.0,
      "step": 35325
    },
    {
      "entropy": 5.630857992172241,
      "epoch": 3.2048258345428158,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039895413542895427,
      "loss": 4.9393,
      "mean_token_accuracy": 0.21024704426527024,
      "num_tokens": 73307105.0,
      "step": 35330
    },
    {
      "entropy": 5.6180963039398195,
      "epoch": 3.2052793904209,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003989271305645086,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22188646495342254,
      "num_tokens": 73316664.0,
      "step": 35335
    },
    {
      "entropy": 5.6249833583831785,
      "epoch": 3.205732946298984,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039890012313721835,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22378720790147782,
      "num_tokens": 73327438.0,
      "step": 35340
    },
    {
      "entropy": 5.629421186447144,
      "epoch": 3.206186502177068,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000398873113147642,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22626577019691468,
      "num_tokens": 73337635.0,
      "step": 35345
    },
    {
      "entropy": 5.589259052276612,
      "epoch": 3.2066400580551524,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039884610059633807,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22069051265716552,
      "num_tokens": 73347032.0,
      "step": 35350
    },
    {
      "entropy": 5.700584125518799,
      "epoch": 3.2070936139332367,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003988190854838652,
      "loss": 4.9656,
      "mean_token_accuracy": 0.21592655628919602,
      "num_tokens": 73357780.0,
      "step": 35355
    },
    {
      "entropy": 5.67203049659729,
      "epoch": 3.207547169811321,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039879206781078214,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22255441993474961,
      "num_tokens": 73367831.0,
      "step": 35360
    },
    {
      "entropy": 5.642022132873535,
      "epoch": 3.2080007256894048,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003987650475776476,
      "loss": 4.8832,
      "mean_token_accuracy": 0.2203782916069031,
      "num_tokens": 73378588.0,
      "step": 35365
    },
    {
      "entropy": 5.592491340637207,
      "epoch": 3.208454281567489,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039873802478502027,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22168018966913222,
      "num_tokens": 73388414.0,
      "step": 35370
    },
    {
      "entropy": 5.606138134002686,
      "epoch": 3.2089078374455733,
      "grad_norm": 1.25,
      "learning_rate": 0.0003987109994334591,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22078590840101242,
      "num_tokens": 73398891.0,
      "step": 35375
    },
    {
      "entropy": 5.6265462875366214,
      "epoch": 3.2093613933236576,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003986839715235229,
      "loss": 4.938,
      "mean_token_accuracy": 0.21807720959186555,
      "num_tokens": 73409605.0,
      "step": 35380
    },
    {
      "entropy": 5.722667026519775,
      "epoch": 3.209814949201742,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039865694105577063,
      "loss": 4.9622,
      "mean_token_accuracy": 0.21125759780406952,
      "num_tokens": 73419103.0,
      "step": 35385
    },
    {
      "entropy": 5.590012073516846,
      "epoch": 3.2102685050798256,
      "grad_norm": 1.25,
      "learning_rate": 0.0003986299080307614,
      "loss": 4.8757,
      "mean_token_accuracy": 0.22253487408161163,
      "num_tokens": 73429134.0,
      "step": 35390
    },
    {
      "entropy": 5.630838298797608,
      "epoch": 3.21072206095791,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003986028724490541,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20802294313907624,
      "num_tokens": 73440382.0,
      "step": 35395
    },
    {
      "entropy": 5.631938171386719,
      "epoch": 3.211175616835994,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003985758343112079,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22155732810497283,
      "num_tokens": 73451300.0,
      "step": 35400
    },
    {
      "entropy": 5.6054424285888675,
      "epoch": 3.2116291727140784,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039854879361778193,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2249677747488022,
      "num_tokens": 73461339.0,
      "step": 35405
    },
    {
      "entropy": 5.643403100967407,
      "epoch": 3.2120827285921627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039852175036933545,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21941620260477065,
      "num_tokens": 73470961.0,
      "step": 35410
    },
    {
      "entropy": 5.7341409683227536,
      "epoch": 3.2125362844702465,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039849470456642777,
      "loss": 4.9799,
      "mean_token_accuracy": 0.21070910543203353,
      "num_tokens": 73481492.0,
      "step": 35415
    },
    {
      "entropy": 5.604830980300903,
      "epoch": 3.212989840348331,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039846765620961806,
      "loss": 4.8537,
      "mean_token_accuracy": 0.21573233157396315,
      "num_tokens": 73492351.0,
      "step": 35420
    },
    {
      "entropy": 5.639754962921143,
      "epoch": 3.213443396226415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003984406052994658,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21697274297475816,
      "num_tokens": 73503265.0,
      "step": 35425
    },
    {
      "entropy": 5.643771028518676,
      "epoch": 3.2138969521044993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003984135518365303,
      "loss": 4.9252,
      "mean_token_accuracy": 0.2145065635442734,
      "num_tokens": 73513804.0,
      "step": 35430
    },
    {
      "entropy": 5.677305603027344,
      "epoch": 3.2143505079825836,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039838649582137113,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21593131273984909,
      "num_tokens": 73523453.0,
      "step": 35435
    },
    {
      "entropy": 5.720432758331299,
      "epoch": 3.214804063860668,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039835943725454777,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20846251547336578,
      "num_tokens": 73533931.0,
      "step": 35440
    },
    {
      "entropy": 5.694847488403321,
      "epoch": 3.2152576197387517,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039833237613661984,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2214327409863472,
      "num_tokens": 73544562.0,
      "step": 35445
    },
    {
      "entropy": 5.640496635437012,
      "epoch": 3.215711175616836,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003983053124681469,
      "loss": 4.8945,
      "mean_token_accuracy": 0.22305275797843932,
      "num_tokens": 73555149.0,
      "step": 35450
    },
    {
      "entropy": 5.564590978622436,
      "epoch": 3.21616473149492,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039827824624968873,
      "loss": 4.823,
      "mean_token_accuracy": 0.2190001577138901,
      "num_tokens": 73565131.0,
      "step": 35455
    },
    {
      "entropy": 5.592260074615479,
      "epoch": 3.2166182873730045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039825117748180493,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22566003054380418,
      "num_tokens": 73574185.0,
      "step": 35460
    },
    {
      "entropy": 5.678895950317383,
      "epoch": 3.2170718432510883,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039822410616505545,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2118372231721878,
      "num_tokens": 73585733.0,
      "step": 35465
    },
    {
      "entropy": 5.716606712341308,
      "epoch": 3.2175253991291726,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039819703230000006,
      "loss": 4.9926,
      "mean_token_accuracy": 0.2108761712908745,
      "num_tokens": 73597077.0,
      "step": 35470
    },
    {
      "entropy": 5.56656494140625,
      "epoch": 3.217978955007257,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003981699558871986,
      "loss": 4.9504,
      "mean_token_accuracy": 0.2193862110376358,
      "num_tokens": 73607885.0,
      "step": 35475
    },
    {
      "entropy": 5.622937774658203,
      "epoch": 3.218432510885341,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000398142876927211,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21663307696580886,
      "num_tokens": 73618769.0,
      "step": 35480
    },
    {
      "entropy": 5.715946435928345,
      "epoch": 3.2188860667634254,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039811579542059744,
      "loss": 5.0061,
      "mean_token_accuracy": 0.21169230937957764,
      "num_tokens": 73629345.0,
      "step": 35485
    },
    {
      "entropy": 5.671459341049195,
      "epoch": 3.2193396226415096,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039808871136791786,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2142948254942894,
      "num_tokens": 73639698.0,
      "step": 35490
    },
    {
      "entropy": 5.5969727516174315,
      "epoch": 3.2197931785195935,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039806162476973227,
      "loss": 4.9002,
      "mean_token_accuracy": 0.2204391047358513,
      "num_tokens": 73650954.0,
      "step": 35495
    },
    {
      "entropy": 5.670774650573731,
      "epoch": 3.2202467343976777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000398034535626601,
      "loss": 4.9822,
      "mean_token_accuracy": 0.2123030126094818,
      "num_tokens": 73661195.0,
      "step": 35500
    },
    {
      "entropy": 5.7414649486541744,
      "epoch": 3.220700290275762,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039800744393908414,
      "loss": 5.0235,
      "mean_token_accuracy": 0.21257992535829545,
      "num_tokens": 73670512.0,
      "step": 35505
    },
    {
      "entropy": 5.658827781677246,
      "epoch": 3.2211538461538463,
      "grad_norm": 1.1875,
      "learning_rate": 0.000397980349707742,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21366484761238097,
      "num_tokens": 73681927.0,
      "step": 35510
    },
    {
      "entropy": 5.701186227798462,
      "epoch": 3.2216074020319305,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039795325293313493,
      "loss": 4.9455,
      "mean_token_accuracy": 0.21942930221557616,
      "num_tokens": 73692025.0,
      "step": 35515
    },
    {
      "entropy": 5.650922250747681,
      "epoch": 3.2220609579100143,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039792615361582324,
      "loss": 4.899,
      "mean_token_accuracy": 0.21642192751169204,
      "num_tokens": 73702278.0,
      "step": 35520
    },
    {
      "entropy": 5.7090349197387695,
      "epoch": 3.2225145137880986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039789905175636726,
      "loss": 5.0006,
      "mean_token_accuracy": 0.21112528294324875,
      "num_tokens": 73713511.0,
      "step": 35525
    },
    {
      "entropy": 5.57156548500061,
      "epoch": 3.222968069666183,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039787194735532765,
      "loss": 4.8076,
      "mean_token_accuracy": 0.22535914927721024,
      "num_tokens": 73723683.0,
      "step": 35530
    },
    {
      "entropy": 5.639390373229981,
      "epoch": 3.223421625544267,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003978448404132649,
      "loss": 4.8811,
      "mean_token_accuracy": 0.22362838089466094,
      "num_tokens": 73734191.0,
      "step": 35535
    },
    {
      "entropy": 5.680836391448975,
      "epoch": 3.2238751814223514,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039781773093073956,
      "loss": 4.922,
      "mean_token_accuracy": 0.21638190895318984,
      "num_tokens": 73743747.0,
      "step": 35540
    },
    {
      "entropy": 5.598128414154052,
      "epoch": 3.2243287373004352,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003977906189083122,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22042784988880157,
      "num_tokens": 73754466.0,
      "step": 35545
    },
    {
      "entropy": 5.641989135742188,
      "epoch": 3.2247822931785195,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003977635043465435,
      "loss": 4.8767,
      "mean_token_accuracy": 0.21699256002902984,
      "num_tokens": 73765948.0,
      "step": 35550
    },
    {
      "entropy": 5.5435771465301515,
      "epoch": 3.2252358490566038,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003977363872459943,
      "loss": 4.8485,
      "mean_token_accuracy": 0.2227562829852104,
      "num_tokens": 73776003.0,
      "step": 35555
    },
    {
      "entropy": 5.708752870559692,
      "epoch": 3.225689404934688,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003977092676072254,
      "loss": 5.0667,
      "mean_token_accuracy": 0.21447840631008147,
      "num_tokens": 73787821.0,
      "step": 35560
    },
    {
      "entropy": 5.658638525009155,
      "epoch": 3.2261429608127723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039768214543079746,
      "loss": 4.9121,
      "mean_token_accuracy": 0.21536696404218675,
      "num_tokens": 73799207.0,
      "step": 35565
    },
    {
      "entropy": 5.641917896270752,
      "epoch": 3.226596516690856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003976550207172716,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2239322379231453,
      "num_tokens": 73809987.0,
      "step": 35570
    },
    {
      "entropy": 5.56656551361084,
      "epoch": 3.2270500725689404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039762789346720854,
      "loss": 4.8754,
      "mean_token_accuracy": 0.2103703185915947,
      "num_tokens": 73819075.0,
      "step": 35575
    },
    {
      "entropy": 5.632872867584228,
      "epoch": 3.2275036284470247,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003976007636811694,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22548563927412033,
      "num_tokens": 73829831.0,
      "step": 35580
    },
    {
      "entropy": 5.677143001556397,
      "epoch": 3.227957184325109,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003975736313597153,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22132644355297088,
      "num_tokens": 73839826.0,
      "step": 35585
    },
    {
      "entropy": 5.621141004562378,
      "epoch": 3.228410740203193,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003975464965034072,
      "loss": 4.8521,
      "mean_token_accuracy": 0.22440076023340225,
      "num_tokens": 73850006.0,
      "step": 35590
    },
    {
      "entropy": 5.635663843154907,
      "epoch": 3.228864296081277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003975193591128064,
      "loss": 4.901,
      "mean_token_accuracy": 0.21671372950077056,
      "num_tokens": 73860309.0,
      "step": 35595
    },
    {
      "entropy": 5.656742811203003,
      "epoch": 3.2293178519593613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039749221918847406,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20660825818777084,
      "num_tokens": 73871613.0,
      "step": 35600
    },
    {
      "entropy": 5.556329965591431,
      "epoch": 3.2297714078374455,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003974650767309713,
      "loss": 4.8159,
      "mean_token_accuracy": 0.22392415553331374,
      "num_tokens": 73882303.0,
      "step": 35605
    },
    {
      "entropy": 5.690486145019531,
      "epoch": 3.23022496371553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039743793174085965,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2083287999033928,
      "num_tokens": 73893548.0,
      "step": 35610
    },
    {
      "entropy": 5.670873308181763,
      "epoch": 3.230678519593614,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039741078421870024,
      "loss": 4.8977,
      "mean_token_accuracy": 0.21701249331235886,
      "num_tokens": 73903934.0,
      "step": 35615
    },
    {
      "entropy": 5.681013298034668,
      "epoch": 3.231132075471698,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039738363416505474,
      "loss": 5.0092,
      "mean_token_accuracy": 0.21038107126951217,
      "num_tokens": 73913813.0,
      "step": 35620
    },
    {
      "entropy": 5.662494802474976,
      "epoch": 3.231585631349782,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039735648158048444,
      "loss": 4.9407,
      "mean_token_accuracy": 0.2096915066242218,
      "num_tokens": 73924339.0,
      "step": 35625
    },
    {
      "entropy": 5.634679746627808,
      "epoch": 3.2320391872278664,
      "grad_norm": 1.25,
      "learning_rate": 0.0003973293264655509,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2128859207034111,
      "num_tokens": 73935318.0,
      "step": 35630
    },
    {
      "entropy": 5.623164558410645,
      "epoch": 3.2324927431059507,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003973021688208158,
      "loss": 4.867,
      "mean_token_accuracy": 0.22606242448091507,
      "num_tokens": 73945547.0,
      "step": 35635
    },
    {
      "entropy": 5.6686169624328615,
      "epoch": 3.232946298984035,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003972750086468406,
      "loss": 4.9059,
      "mean_token_accuracy": 0.2174498438835144,
      "num_tokens": 73956195.0,
      "step": 35640
    },
    {
      "entropy": 5.649539661407471,
      "epoch": 3.2333998548621192,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039724784594418714,
      "loss": 4.8446,
      "mean_token_accuracy": 0.22769755423069,
      "num_tokens": 73966843.0,
      "step": 35645
    },
    {
      "entropy": 5.61407151222229,
      "epoch": 3.233853410740203,
      "grad_norm": 1.140625,
      "learning_rate": 0.000397220680713417,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21844768673181533,
      "num_tokens": 73978516.0,
      "step": 35650
    },
    {
      "entropy": 5.5622883319854735,
      "epoch": 3.2343069666182873,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039719351295509203,
      "loss": 4.8353,
      "mean_token_accuracy": 0.22558247148990632,
      "num_tokens": 73988376.0,
      "step": 35655
    },
    {
      "entropy": 5.599919557571411,
      "epoch": 3.2347605224963716,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039716634266977413,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2243556335568428,
      "num_tokens": 73998232.0,
      "step": 35660
    },
    {
      "entropy": 5.609225940704346,
      "epoch": 3.235214078374456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003971391698580252,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2171994373202324,
      "num_tokens": 74008726.0,
      "step": 35665
    },
    {
      "entropy": 5.644869995117188,
      "epoch": 3.23566763425254,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039711199452040696,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22001542896032333,
      "num_tokens": 74018746.0,
      "step": 35670
    },
    {
      "entropy": 5.640611839294434,
      "epoch": 3.236121190130624,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039708481665748157,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22133959829807281,
      "num_tokens": 74028615.0,
      "step": 35675
    },
    {
      "entropy": 5.649672031402588,
      "epoch": 3.236574746008708,
      "grad_norm": 1.125,
      "learning_rate": 0.0003970576362698111,
      "loss": 4.9401,
      "mean_token_accuracy": 0.2181209221482277,
      "num_tokens": 74039219.0,
      "step": 35680
    },
    {
      "entropy": 5.633511400222778,
      "epoch": 3.2370283018867925,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039703045335795766,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22398090362548828,
      "num_tokens": 74049385.0,
      "step": 35685
    },
    {
      "entropy": 5.711008930206299,
      "epoch": 3.2374818577648767,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003970032679224833,
      "loss": 5.0416,
      "mean_token_accuracy": 0.20649878978729247,
      "num_tokens": 74059437.0,
      "step": 35690
    },
    {
      "entropy": 5.702699613571167,
      "epoch": 3.237935413642961,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039697607996395023,
      "loss": 4.9695,
      "mean_token_accuracy": 0.209164160490036,
      "num_tokens": 74068834.0,
      "step": 35695
    },
    {
      "entropy": 5.689772510528565,
      "epoch": 3.238388969521045,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003969488894829208,
      "loss": 4.9421,
      "mean_token_accuracy": 0.213767808675766,
      "num_tokens": 74079757.0,
      "step": 35700
    },
    {
      "entropy": 5.675665712356567,
      "epoch": 3.238842525399129,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003969216964799573,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2203963503241539,
      "num_tokens": 74091092.0,
      "step": 35705
    },
    {
      "entropy": 5.587155389785766,
      "epoch": 3.2392960812772134,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003968945009556219,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22699668556451796,
      "num_tokens": 74102272.0,
      "step": 35710
    },
    {
      "entropy": 5.5951841354370115,
      "epoch": 3.2397496371552976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003968673029104773,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22325626015663147,
      "num_tokens": 74113101.0,
      "step": 35715
    },
    {
      "entropy": 5.689749670028687,
      "epoch": 3.240203193033382,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003968401023450858,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22083511352539062,
      "num_tokens": 74123180.0,
      "step": 35720
    },
    {
      "entropy": 5.618878889083862,
      "epoch": 3.2406567489114657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039681289926000985,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2248257130384445,
      "num_tokens": 74133675.0,
      "step": 35725
    },
    {
      "entropy": 5.590866994857788,
      "epoch": 3.24111030478955,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003967856936558122,
      "loss": 4.8172,
      "mean_token_accuracy": 0.2213743433356285,
      "num_tokens": 74143472.0,
      "step": 35730
    },
    {
      "entropy": 5.635987854003906,
      "epoch": 3.2415638606676342,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039675848553305526,
      "loss": 4.9032,
      "mean_token_accuracy": 0.21613086760044098,
      "num_tokens": 74154682.0,
      "step": 35735
    },
    {
      "entropy": 5.642532873153686,
      "epoch": 3.2420174165457185,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003967312748923019,
      "loss": 4.9155,
      "mean_token_accuracy": 0.2224576100707054,
      "num_tokens": 74164745.0,
      "step": 35740
    },
    {
      "entropy": 5.614141464233398,
      "epoch": 3.2424709724238028,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039670406173411473,
      "loss": 4.8122,
      "mean_token_accuracy": 0.22684427052736283,
      "num_tokens": 74175388.0,
      "step": 35745
    },
    {
      "entropy": 5.624486970901489,
      "epoch": 3.2429245283018866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003966768460590566,
      "loss": 4.8641,
      "mean_token_accuracy": 0.21987419575452805,
      "num_tokens": 74185929.0,
      "step": 35750
    },
    {
      "entropy": 5.641559648513794,
      "epoch": 3.243378084179971,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039664962786769013,
      "loss": 4.8714,
      "mean_token_accuracy": 0.22296553552150727,
      "num_tokens": 74196165.0,
      "step": 35755
    },
    {
      "entropy": 5.632019186019898,
      "epoch": 3.243831640058055,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039662240716057854,
      "loss": 4.8721,
      "mean_token_accuracy": 0.22270023077726364,
      "num_tokens": 74206843.0,
      "step": 35760
    },
    {
      "entropy": 5.656398010253906,
      "epoch": 3.2442851959361394,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003965951839382845,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2233958825469017,
      "num_tokens": 74216854.0,
      "step": 35765
    },
    {
      "entropy": 5.6067023277282715,
      "epoch": 3.2447387518142237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000396567958201371,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22269334644079208,
      "num_tokens": 74226158.0,
      "step": 35770
    },
    {
      "entropy": 5.601041603088379,
      "epoch": 3.2451923076923075,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003965407299504013,
      "loss": 4.9146,
      "mean_token_accuracy": 0.222706201672554,
      "num_tokens": 74236860.0,
      "step": 35775
    },
    {
      "entropy": 5.638400554656982,
      "epoch": 3.2456458635703918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003965134991859382,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22387139350175858,
      "num_tokens": 74247706.0,
      "step": 35780
    },
    {
      "entropy": 5.687844610214233,
      "epoch": 3.246099419448476,
      "grad_norm": 1.203125,
      "learning_rate": 0.000396486265908545,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2200131133198738,
      "num_tokens": 74259479.0,
      "step": 35785
    },
    {
      "entropy": 5.616486120223999,
      "epoch": 3.2465529753265603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003964590301187849,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21905033588409423,
      "num_tokens": 74270371.0,
      "step": 35790
    },
    {
      "entropy": 5.623467445373535,
      "epoch": 3.2470065312046446,
      "grad_norm": 1.375,
      "learning_rate": 0.0003964317918172211,
      "loss": 4.8514,
      "mean_token_accuracy": 0.23185165226459503,
      "num_tokens": 74279637.0,
      "step": 35795
    },
    {
      "entropy": 5.66420350074768,
      "epoch": 3.247460087082729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039640455100441684,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21282645016908647,
      "num_tokens": 74289889.0,
      "step": 35800
    },
    {
      "entropy": 5.673099660873413,
      "epoch": 3.2479136429608126,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039637730768093564,
      "loss": 4.923,
      "mean_token_accuracy": 0.2277629092335701,
      "num_tokens": 74300949.0,
      "step": 35805
    },
    {
      "entropy": 5.648768043518066,
      "epoch": 3.248367198838897,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003963500618473407,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22065519392490388,
      "num_tokens": 74312077.0,
      "step": 35810
    },
    {
      "entropy": 5.602660465240478,
      "epoch": 3.248820754716981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003963228135041956,
      "loss": 4.9658,
      "mean_token_accuracy": 0.21205090433359147,
      "num_tokens": 74322610.0,
      "step": 35815
    },
    {
      "entropy": 5.5728880882263185,
      "epoch": 3.2492743105950654,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039629556265206373,
      "loss": 4.8184,
      "mean_token_accuracy": 0.21757723838090898,
      "num_tokens": 74331970.0,
      "step": 35820
    },
    {
      "entropy": 5.619731903076172,
      "epoch": 3.2497278664731493,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003962683092915088,
      "loss": 4.8552,
      "mean_token_accuracy": 0.2270527794957161,
      "num_tokens": 74342812.0,
      "step": 35825
    },
    {
      "entropy": 5.628123617172241,
      "epoch": 3.2501814223512335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003962410534230943,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2265948936343193,
      "num_tokens": 74353786.0,
      "step": 35830
    },
    {
      "entropy": 5.66396198272705,
      "epoch": 3.250634978229318,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039621379504738385,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21867869794368744,
      "num_tokens": 74363732.0,
      "step": 35835
    },
    {
      "entropy": 5.643849182128906,
      "epoch": 3.251088534107402,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039618653416494137,
      "loss": 4.9248,
      "mean_token_accuracy": 0.2202475681900978,
      "num_tokens": 74374077.0,
      "step": 35840
    },
    {
      "entropy": 5.642728424072265,
      "epoch": 3.2515420899854863,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039615927077633043,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21897670477628708,
      "num_tokens": 74384758.0,
      "step": 35845
    },
    {
      "entropy": 5.708169603347779,
      "epoch": 3.2519956458635706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039613200488211483,
      "loss": 4.9412,
      "mean_token_accuracy": 0.22317821085453032,
      "num_tokens": 74395811.0,
      "step": 35850
    },
    {
      "entropy": 5.646434593200683,
      "epoch": 3.2524492017416544,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039610473648285857,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21638157069683076,
      "num_tokens": 74407830.0,
      "step": 35855
    },
    {
      "entropy": 5.589155197143555,
      "epoch": 3.2529027576197387,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003960774655791255,
      "loss": 4.8662,
      "mean_token_accuracy": 0.21954732090234758,
      "num_tokens": 74417565.0,
      "step": 35860
    },
    {
      "entropy": 5.678437662124634,
      "epoch": 3.253356313497823,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039605019217147947,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22099993377923965,
      "num_tokens": 74428015.0,
      "step": 35865
    },
    {
      "entropy": 5.597277975082397,
      "epoch": 3.253809869375907,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039602291626048475,
      "loss": 4.9116,
      "mean_token_accuracy": 0.22099123448133468,
      "num_tokens": 74439523.0,
      "step": 35870
    },
    {
      "entropy": 5.691626119613647,
      "epoch": 3.2542634252539915,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039599563784670526,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20558119863271712,
      "num_tokens": 74449897.0,
      "step": 35875
    },
    {
      "entropy": 5.621711540222168,
      "epoch": 3.2547169811320753,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003959683569307051,
      "loss": 4.8267,
      "mean_token_accuracy": 0.22352193892002106,
      "num_tokens": 74459888.0,
      "step": 35880
    },
    {
      "entropy": 5.609740924835205,
      "epoch": 3.2551705370101596,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003959410735130485,
      "loss": 4.8138,
      "mean_token_accuracy": 0.21500419825315475,
      "num_tokens": 74470894.0,
      "step": 35885
    },
    {
      "entropy": 5.6058878898620605,
      "epoch": 3.255624092888244,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039591378759429974,
      "loss": 4.8099,
      "mean_token_accuracy": 0.23100608438253403,
      "num_tokens": 74481306.0,
      "step": 35890
    },
    {
      "entropy": 5.651317977905274,
      "epoch": 3.256077648766328,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000395886499175023,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2219666913151741,
      "num_tokens": 74491048.0,
      "step": 35895
    },
    {
      "entropy": 5.599463844299317,
      "epoch": 3.2565312046444124,
      "grad_norm": 1.125,
      "learning_rate": 0.00039585920825578266,
      "loss": 4.844,
      "mean_token_accuracy": 0.226677505671978,
      "num_tokens": 74501656.0,
      "step": 35900
    },
    {
      "entropy": 5.683964776992798,
      "epoch": 3.256984760522496,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003958319148371431,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21967142969369888,
      "num_tokens": 74513280.0,
      "step": 35905
    },
    {
      "entropy": 5.615320777893066,
      "epoch": 3.2574383164005805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003958046189196687,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2172779306769371,
      "num_tokens": 74522898.0,
      "step": 35910
    },
    {
      "entropy": 5.697880649566651,
      "epoch": 3.2578918722786647,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039577732050392395,
      "loss": 4.9373,
      "mean_token_accuracy": 0.21768889278173448,
      "num_tokens": 74532551.0,
      "step": 35915
    },
    {
      "entropy": 5.682796907424927,
      "epoch": 3.258345428156749,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003957500195904735,
      "loss": 4.9064,
      "mean_token_accuracy": 0.22344327419996263,
      "num_tokens": 74542285.0,
      "step": 35920
    },
    {
      "entropy": 5.689100170135498,
      "epoch": 3.2587989840348333,
      "grad_norm": 1.328125,
      "learning_rate": 0.000395722716179882,
      "loss": 4.9562,
      "mean_token_accuracy": 0.21521099656820297,
      "num_tokens": 74551982.0,
      "step": 35925
    },
    {
      "entropy": 5.661095523834229,
      "epoch": 3.259252539912917,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003956954102727138,
      "loss": 4.9128,
      "mean_token_accuracy": 0.21996078938245772,
      "num_tokens": 74562190.0,
      "step": 35930
    },
    {
      "entropy": 5.601325368881225,
      "epoch": 3.2597060957910013,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039566810186953376,
      "loss": 4.8368,
      "mean_token_accuracy": 0.2263374775648117,
      "num_tokens": 74572673.0,
      "step": 35935
    },
    {
      "entropy": 5.5878705978393555,
      "epoch": 3.2601596516690856,
      "grad_norm": 1.25,
      "learning_rate": 0.0003956407909709068,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22473883330821992,
      "num_tokens": 74582686.0,
      "step": 35940
    },
    {
      "entropy": 5.693186950683594,
      "epoch": 3.26061320754717,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039561347757739734,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21870440244674683,
      "num_tokens": 74593658.0,
      "step": 35945
    },
    {
      "entropy": 5.697020244598389,
      "epoch": 3.261066763425254,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039558616168957044,
      "loss": 4.9919,
      "mean_token_accuracy": 0.22000594884157182,
      "num_tokens": 74603585.0,
      "step": 35950
    },
    {
      "entropy": 5.626160335540772,
      "epoch": 3.2615203193033384,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003955588433079911,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21849820017814636,
      "num_tokens": 74612806.0,
      "step": 35955
    },
    {
      "entropy": 5.648509740829468,
      "epoch": 3.2619738751814222,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039553152243322414,
      "loss": 4.9054,
      "mean_token_accuracy": 0.2242448776960373,
      "num_tokens": 74623252.0,
      "step": 35960
    },
    {
      "entropy": 5.609339094161987,
      "epoch": 3.2624274310595065,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003955041990658346,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22297365814447404,
      "num_tokens": 74633555.0,
      "step": 35965
    },
    {
      "entropy": 5.628423833847046,
      "epoch": 3.2628809869375908,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003954768732063874,
      "loss": 4.9041,
      "mean_token_accuracy": 0.22183336913585663,
      "num_tokens": 74643322.0,
      "step": 35970
    },
    {
      "entropy": 5.622592973709106,
      "epoch": 3.263334542815675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003954495448554478,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22827915698289872,
      "num_tokens": 74653461.0,
      "step": 35975
    },
    {
      "entropy": 5.6619940280914305,
      "epoch": 3.263788098693759,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039542221401358104,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2237296834588051,
      "num_tokens": 74663470.0,
      "step": 35980
    },
    {
      "entropy": 5.5964216709136965,
      "epoch": 3.264241654571843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000395394880681352,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2268133893609047,
      "num_tokens": 74674074.0,
      "step": 35985
    },
    {
      "entropy": 5.696441793441773,
      "epoch": 3.2646952104499274,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039536754485932626,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21063876152038574,
      "num_tokens": 74684968.0,
      "step": 35990
    },
    {
      "entropy": 5.627312135696411,
      "epoch": 3.2651487663280117,
      "grad_norm": 1.25,
      "learning_rate": 0.00039534020654806906,
      "loss": 4.8134,
      "mean_token_accuracy": 0.23491248041391372,
      "num_tokens": 74694776.0,
      "step": 35995
    },
    {
      "entropy": 5.604242038726807,
      "epoch": 3.265602322206096,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039531286574814563,
      "loss": 4.8431,
      "mean_token_accuracy": 0.2202404573559761,
      "num_tokens": 74705506.0,
      "step": 36000
    },
    {
      "epoch": 3.265602322206096,
      "eval_entropy": 5.334039939043209,
      "eval_loss": 4.975577354431152,
      "eval_mean_token_accuracy": 0.22394091301003463,
      "eval_num_tokens": 74705506.0,
      "eval_runtime": 20.4906,
      "eval_samples_per_second": 1725.669,
      "eval_steps_per_second": 215.709,
      "step": 36000
    },
    {
      "entropy": 5.565581750869751,
      "epoch": 3.26605587808418,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003952855224601215,
      "loss": 4.8777,
      "mean_token_accuracy": 0.21996652632951735,
      "num_tokens": 74715677.0,
      "step": 36005
    },
    {
      "entropy": 5.679157495498657,
      "epoch": 3.266509433962264,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003952581766845622,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22630117535591127,
      "num_tokens": 74725296.0,
      "step": 36010
    },
    {
      "entropy": 5.629681348800659,
      "epoch": 3.2669629898403483,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003952308284220331,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22535340487957,
      "num_tokens": 74736124.0,
      "step": 36015
    },
    {
      "entropy": 5.534782409667969,
      "epoch": 3.2674165457184325,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039520347767309975,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21663733720779418,
      "num_tokens": 74746447.0,
      "step": 36020
    },
    {
      "entropy": 5.590958166122436,
      "epoch": 3.267870101596517,
      "grad_norm": 1.203125,
      "learning_rate": 0.000395176124438328,
      "loss": 4.9027,
      "mean_token_accuracy": 0.2173424854874611,
      "num_tokens": 74756947.0,
      "step": 36025
    },
    {
      "entropy": 5.637294006347656,
      "epoch": 3.2683236574746006,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00039514876871828326,
      "loss": 4.8274,
      "mean_token_accuracy": 0.22575157433748244,
      "num_tokens": 74767856.0,
      "step": 36030
    },
    {
      "entropy": 5.636200904846191,
      "epoch": 3.268777213352685,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039512141051353144,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22309077829122542,
      "num_tokens": 74778335.0,
      "step": 36035
    },
    {
      "entropy": 5.678413057327271,
      "epoch": 3.269230769230769,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003950940498246382,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2241508588194847,
      "num_tokens": 74788430.0,
      "step": 36040
    },
    {
      "entropy": 5.668745613098144,
      "epoch": 3.2696843251088534,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039506668665216944,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21801635771989822,
      "num_tokens": 74800193.0,
      "step": 36045
    },
    {
      "entropy": 5.665507125854492,
      "epoch": 3.2701378809869377,
      "grad_norm": 1.234375,
      "learning_rate": 0.000395039320996691,
      "loss": 4.9229,
      "mean_token_accuracy": 0.2133774757385254,
      "num_tokens": 74810368.0,
      "step": 36050
    },
    {
      "entropy": 5.698576593399048,
      "epoch": 3.270591436865022,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039501195285876874,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2263675943017006,
      "num_tokens": 74820687.0,
      "step": 36055
    },
    {
      "entropy": 5.624888944625854,
      "epoch": 3.271044992743106,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003949845822389688,
      "loss": 4.8449,
      "mean_token_accuracy": 0.2258884862065315,
      "num_tokens": 74830489.0,
      "step": 36060
    },
    {
      "entropy": 5.526579141616821,
      "epoch": 3.27149854862119,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000394957209137857,
      "loss": 4.824,
      "mean_token_accuracy": 0.22220639288425445,
      "num_tokens": 74840638.0,
      "step": 36065
    },
    {
      "entropy": 5.669620084762573,
      "epoch": 3.2719521044992743,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003949298335559997,
      "loss": 5.0269,
      "mean_token_accuracy": 0.2081297218799591,
      "num_tokens": 74850600.0,
      "step": 36070
    },
    {
      "entropy": 5.664458608627319,
      "epoch": 3.2724056603773586,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003949024554939628,
      "loss": 4.901,
      "mean_token_accuracy": 0.22155162692070007,
      "num_tokens": 74861177.0,
      "step": 36075
    },
    {
      "entropy": 5.6895064353942875,
      "epoch": 3.272859216255443,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039487507495231263,
      "loss": 4.9645,
      "mean_token_accuracy": 0.21687162518501282,
      "num_tokens": 74871747.0,
      "step": 36080
    },
    {
      "entropy": 5.610798788070679,
      "epoch": 3.2733127721335267,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003948476919316153,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21683374792337418,
      "num_tokens": 74882796.0,
      "step": 36085
    },
    {
      "entropy": 5.683705615997314,
      "epoch": 3.273766328011611,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003948203064324371,
      "loss": 4.8994,
      "mean_token_accuracy": 0.2152738094329834,
      "num_tokens": 74894123.0,
      "step": 36090
    },
    {
      "entropy": 5.676540040969849,
      "epoch": 3.274219883889695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039479291845534453,
      "loss": 4.9413,
      "mean_token_accuracy": 0.2130412369966507,
      "num_tokens": 74904787.0,
      "step": 36095
    },
    {
      "entropy": 5.667060089111328,
      "epoch": 3.2746734397677795,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039476552800090384,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20901102274656297,
      "num_tokens": 74915041.0,
      "step": 36100
    },
    {
      "entropy": 5.675154781341552,
      "epoch": 3.2751269956458637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039473813506968146,
      "loss": 4.87,
      "mean_token_accuracy": 0.21157475262880326,
      "num_tokens": 74926217.0,
      "step": 36105
    },
    {
      "entropy": 5.682069778442383,
      "epoch": 3.2755805515239476,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039471073966224404,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21418286114931107,
      "num_tokens": 74936616.0,
      "step": 36110
    },
    {
      "entropy": 5.6700982570648195,
      "epoch": 3.276034107402032,
      "grad_norm": 1.21875,
      "learning_rate": 0.000394683341779158,
      "loss": 4.9307,
      "mean_token_accuracy": 0.21193190813064575,
      "num_tokens": 74947443.0,
      "step": 36115
    },
    {
      "entropy": 5.658718061447144,
      "epoch": 3.276487663280116,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003946559414209899,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22059800773859023,
      "num_tokens": 74957618.0,
      "step": 36120
    },
    {
      "entropy": 5.628855991363525,
      "epoch": 3.2769412191582004,
      "grad_norm": 1.25,
      "learning_rate": 0.0003946285385883065,
      "loss": 4.836,
      "mean_token_accuracy": 0.2197231635451317,
      "num_tokens": 74967982.0,
      "step": 36125
    },
    {
      "entropy": 5.576105833053589,
      "epoch": 3.2773947750362846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039460113328167434,
      "loss": 4.883,
      "mean_token_accuracy": 0.22566833794116975,
      "num_tokens": 74978801.0,
      "step": 36130
    },
    {
      "entropy": 5.58773193359375,
      "epoch": 3.2778483309143684,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003945737255016603,
      "loss": 4.8462,
      "mean_token_accuracy": 0.22251618504524232,
      "num_tokens": 74989000.0,
      "step": 36135
    },
    {
      "entropy": 5.658890438079834,
      "epoch": 3.2783018867924527,
      "grad_norm": 1.125,
      "learning_rate": 0.00039454631524883123,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21813136637210845,
      "num_tokens": 74999664.0,
      "step": 36140
    },
    {
      "entropy": 5.649524211883545,
      "epoch": 3.278755442670537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003945189025237538,
      "loss": 4.895,
      "mean_token_accuracy": 0.2238043650984764,
      "num_tokens": 75010187.0,
      "step": 36145
    },
    {
      "entropy": 5.734669208526611,
      "epoch": 3.2792089985486212,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000394491487326995,
      "loss": 4.9618,
      "mean_token_accuracy": 0.2112666040658951,
      "num_tokens": 75021222.0,
      "step": 36150
    },
    {
      "entropy": 5.605030059814453,
      "epoch": 3.2796625544267055,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003944640696591219,
      "loss": 4.9083,
      "mean_token_accuracy": 0.2221206784248352,
      "num_tokens": 75031136.0,
      "step": 36155
    },
    {
      "entropy": 5.577616596221924,
      "epoch": 3.28011611030479,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039443664952070135,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22218406498432158,
      "num_tokens": 75041545.0,
      "step": 36160
    },
    {
      "entropy": 5.646094274520874,
      "epoch": 3.2805696661828736,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039440922691230044,
      "loss": 4.8937,
      "mean_token_accuracy": 0.224629907310009,
      "num_tokens": 75051805.0,
      "step": 36165
    },
    {
      "entropy": 5.644822311401367,
      "epoch": 3.281023222060958,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039438180183448625,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22180057913064957,
      "num_tokens": 75062551.0,
      "step": 36170
    },
    {
      "entropy": 5.616482400894165,
      "epoch": 3.281476777939042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000394354374287826,
      "loss": 4.8904,
      "mean_token_accuracy": 0.21719923317432405,
      "num_tokens": 75073040.0,
      "step": 36175
    },
    {
      "entropy": 5.65125732421875,
      "epoch": 3.2819303338171264,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039432694427288695,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2244493454694748,
      "num_tokens": 75082543.0,
      "step": 36180
    },
    {
      "entropy": 5.661353731155396,
      "epoch": 3.28238388969521,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039429951179023614,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21495041698217393,
      "num_tokens": 75092480.0,
      "step": 36185
    },
    {
      "entropy": 5.597170162200928,
      "epoch": 3.2828374455732945,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039427207684044116,
      "loss": 4.8071,
      "mean_token_accuracy": 0.22196868509054185,
      "num_tokens": 75102445.0,
      "step": 36190
    },
    {
      "entropy": 5.661404991149903,
      "epoch": 3.2832910014513788,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039424463942406917,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21748390942811965,
      "num_tokens": 75112793.0,
      "step": 36195
    },
    {
      "entropy": 5.7532491207122805,
      "epoch": 3.283744557329463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003942171995416877,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2161255359649658,
      "num_tokens": 75123312.0,
      "step": 36200
    },
    {
      "entropy": 5.621792316436768,
      "epoch": 3.2841981132075473,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039418975719386406,
      "loss": 4.8509,
      "mean_token_accuracy": 0.21901890635490417,
      "num_tokens": 75133584.0,
      "step": 36205
    },
    {
      "entropy": 5.596186828613281,
      "epoch": 3.2846516690856316,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003941623123811659,
      "loss": 4.8903,
      "mean_token_accuracy": 0.2165917530655861,
      "num_tokens": 75144344.0,
      "step": 36210
    },
    {
      "entropy": 5.690426015853882,
      "epoch": 3.2851052249637154,
      "grad_norm": 1.25,
      "learning_rate": 0.00039413486510416087,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21251555979251863,
      "num_tokens": 75155086.0,
      "step": 36215
    },
    {
      "entropy": 5.651762819290161,
      "epoch": 3.2855587808417996,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039410741536341634,
      "loss": 4.864,
      "mean_token_accuracy": 0.2203646719455719,
      "num_tokens": 75166046.0,
      "step": 36220
    },
    {
      "entropy": 5.593733596801758,
      "epoch": 3.286012336719884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039407996315950016,
      "loss": 4.8205,
      "mean_token_accuracy": 0.22276679277420045,
      "num_tokens": 75175388.0,
      "step": 36225
    },
    {
      "entropy": 5.631497001647949,
      "epoch": 3.286465892597968,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039405250849297997,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21909729689359664,
      "num_tokens": 75186213.0,
      "step": 36230
    },
    {
      "entropy": 5.592648220062256,
      "epoch": 3.2869194484760524,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003940250513644236,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22140010297298432,
      "num_tokens": 75196141.0,
      "step": 36235
    },
    {
      "entropy": 5.572011470794678,
      "epoch": 3.2873730043541363,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003939975917743988,
      "loss": 4.7585,
      "mean_token_accuracy": 0.23153959661722184,
      "num_tokens": 75205414.0,
      "step": 36240
    },
    {
      "entropy": 5.66672649383545,
      "epoch": 3.2878265602322205,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003939701297234735,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21441813707351684,
      "num_tokens": 75215860.0,
      "step": 36245
    },
    {
      "entropy": 5.642694282531738,
      "epoch": 3.288280116110305,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039394266521221565,
      "loss": 4.9186,
      "mean_token_accuracy": 0.2122296780347824,
      "num_tokens": 75226358.0,
      "step": 36250
    },
    {
      "entropy": 5.5598348617553714,
      "epoch": 3.288733671988389,
      "grad_norm": 1.46875,
      "learning_rate": 0.00039391519824119315,
      "loss": 4.7628,
      "mean_token_accuracy": 0.23325237333774568,
      "num_tokens": 75235779.0,
      "step": 36255
    },
    {
      "entropy": 5.686887407302857,
      "epoch": 3.2891872278664733,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039388772881097396,
      "loss": 5.0229,
      "mean_token_accuracy": 0.21225060522556305,
      "num_tokens": 75246417.0,
      "step": 36260
    },
    {
      "entropy": 5.665030384063721,
      "epoch": 3.289640783744557,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039386025692212637,
      "loss": 4.8662,
      "mean_token_accuracy": 0.22235198467969894,
      "num_tokens": 75257094.0,
      "step": 36265
    },
    {
      "entropy": 5.617471361160279,
      "epoch": 3.2900943396226414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003938327825752183,
      "loss": 4.8512,
      "mean_token_accuracy": 0.21995169371366502,
      "num_tokens": 75267374.0,
      "step": 36270
    },
    {
      "entropy": 5.5946238994598385,
      "epoch": 3.2905478955007257,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000393805305770818,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2270051583647728,
      "num_tokens": 75276936.0,
      "step": 36275
    },
    {
      "entropy": 5.5787183284759525,
      "epoch": 3.29100145137881,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003937778265094937,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22402942776679993,
      "num_tokens": 75286403.0,
      "step": 36280
    },
    {
      "entropy": 5.665550279617309,
      "epoch": 3.291455007256894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003937503447918137,
      "loss": 4.924,
      "mean_token_accuracy": 0.21691127717494965,
      "num_tokens": 75297421.0,
      "step": 36285
    },
    {
      "entropy": 5.678347826004028,
      "epoch": 3.291908563134978,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039372286061834633,
      "loss": 4.92,
      "mean_token_accuracy": 0.21643347889184952,
      "num_tokens": 75307955.0,
      "step": 36290
    },
    {
      "entropy": 5.6450341701507565,
      "epoch": 3.2923621190130623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039369537398966,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21890365928411484,
      "num_tokens": 75318739.0,
      "step": 36295
    },
    {
      "entropy": 5.593415307998657,
      "epoch": 3.2928156748911466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039366788490632286,
      "loss": 4.7794,
      "mean_token_accuracy": 0.2307997852563858,
      "num_tokens": 75328629.0,
      "step": 36300
    },
    {
      "entropy": 5.593215560913086,
      "epoch": 3.293269230769231,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039364039336890386,
      "loss": 4.8494,
      "mean_token_accuracy": 0.23059324324131011,
      "num_tokens": 75337986.0,
      "step": 36305
    },
    {
      "entropy": 5.574730110168457,
      "epoch": 3.293722786647315,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003936128993779712,
      "loss": 4.831,
      "mean_token_accuracy": 0.22380361407995225,
      "num_tokens": 75346169.0,
      "step": 36310
    },
    {
      "entropy": 5.653278160095215,
      "epoch": 3.2941763425253994,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003935854029340936,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22533187568187713,
      "num_tokens": 75356111.0,
      "step": 36315
    },
    {
      "entropy": 5.628912544250488,
      "epoch": 3.294629898403483,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039355790403783954,
      "loss": 4.8498,
      "mean_token_accuracy": 0.23083717226982117,
      "num_tokens": 75366679.0,
      "step": 36320
    },
    {
      "entropy": 5.55464539527893,
      "epoch": 3.2950834542815675,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039353040268977785,
      "loss": 4.7684,
      "mean_token_accuracy": 0.22916725873947144,
      "num_tokens": 75377500.0,
      "step": 36325
    },
    {
      "entropy": 5.587055969238281,
      "epoch": 3.2955370101596517,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003935028988904773,
      "loss": 4.8407,
      "mean_token_accuracy": 0.2247241660952568,
      "num_tokens": 75387115.0,
      "step": 36330
    },
    {
      "entropy": 5.611231470108033,
      "epoch": 3.295990566037736,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039347539264050654,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2118021786212921,
      "num_tokens": 75396330.0,
      "step": 36335
    },
    {
      "entropy": 5.604051113128662,
      "epoch": 3.29644412191582,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003934478839404345,
      "loss": 4.887,
      "mean_token_accuracy": 0.2271381363272667,
      "num_tokens": 75407168.0,
      "step": 36340
    },
    {
      "entropy": 5.657563877105713,
      "epoch": 3.296897677793904,
      "grad_norm": 1.25,
      "learning_rate": 0.00039342037279083006,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2176897868514061,
      "num_tokens": 75418117.0,
      "step": 36345
    },
    {
      "entropy": 5.691016626358032,
      "epoch": 3.2973512336719883,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003933928591922621,
      "loss": 5.0022,
      "mean_token_accuracy": 0.21178116798400878,
      "num_tokens": 75429026.0,
      "step": 36350
    },
    {
      "entropy": 5.675773763656617,
      "epoch": 3.2978047895500726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003933653431452996,
      "loss": 4.8494,
      "mean_token_accuracy": 0.22299687564373016,
      "num_tokens": 75439710.0,
      "step": 36355
    },
    {
      "entropy": 5.6629781246185305,
      "epoch": 3.298258345428157,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039333782465051173,
      "loss": 4.872,
      "mean_token_accuracy": 0.2245420053601265,
      "num_tokens": 75450086.0,
      "step": 36360
    },
    {
      "entropy": 5.61488094329834,
      "epoch": 3.298711901306241,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003933103037084674,
      "loss": 4.8808,
      "mean_token_accuracy": 0.21944738030433655,
      "num_tokens": 75460988.0,
      "step": 36365
    },
    {
      "entropy": 5.650631427764893,
      "epoch": 3.299165457184325,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039328278031973586,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2130212515592575,
      "num_tokens": 75471560.0,
      "step": 36370
    },
    {
      "entropy": 5.631032276153564,
      "epoch": 3.2996190130624092,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003932552544848863,
      "loss": 4.9299,
      "mean_token_accuracy": 0.214851413667202,
      "num_tokens": 75482124.0,
      "step": 36375
    },
    {
      "entropy": 5.634671640396118,
      "epoch": 3.3000725689404935,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000393227726204488,
      "loss": 4.8966,
      "mean_token_accuracy": 0.21661461889743805,
      "num_tokens": 75492351.0,
      "step": 36380
    },
    {
      "entropy": 5.631858205795288,
      "epoch": 3.3005261248185778,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039320019547911013,
      "loss": 4.8086,
      "mean_token_accuracy": 0.2223750203847885,
      "num_tokens": 75502475.0,
      "step": 36385
    },
    {
      "entropy": 5.586136293411255,
      "epoch": 3.3009796806966616,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003931726623093221,
      "loss": 4.8145,
      "mean_token_accuracy": 0.227279132604599,
      "num_tokens": 75512345.0,
      "step": 36390
    },
    {
      "entropy": 5.58604416847229,
      "epoch": 3.301433236574746,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039314512669569323,
      "loss": 4.7896,
      "mean_token_accuracy": 0.2345356822013855,
      "num_tokens": 75523011.0,
      "step": 36395
    },
    {
      "entropy": 5.632163619995117,
      "epoch": 3.30188679245283,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039311758863879316,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2150120258331299,
      "num_tokens": 75533775.0,
      "step": 36400
    },
    {
      "entropy": 5.642969274520874,
      "epoch": 3.3023403483309144,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039309004813919117,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20664936155080796,
      "num_tokens": 75544558.0,
      "step": 36405
    },
    {
      "entropy": 5.7243523597717285,
      "epoch": 3.3027939042089987,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039306250519745696,
      "loss": 4.9948,
      "mean_token_accuracy": 0.21097481548786162,
      "num_tokens": 75555599.0,
      "step": 36410
    },
    {
      "entropy": 5.6358414649963375,
      "epoch": 3.303247460087083,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039303495981416007,
      "loss": 4.8688,
      "mean_token_accuracy": 0.2211988389492035,
      "num_tokens": 75566320.0,
      "step": 36415
    },
    {
      "entropy": 5.636953592300415,
      "epoch": 3.3037010159651667,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003930074119898701,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21995309740304947,
      "num_tokens": 75576069.0,
      "step": 36420
    },
    {
      "entropy": 5.641560077667236,
      "epoch": 3.304154571843251,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003929798617251567,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21882313936948777,
      "num_tokens": 75586012.0,
      "step": 36425
    },
    {
      "entropy": 5.681649494171142,
      "epoch": 3.3046081277213353,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003929523090205898,
      "loss": 4.9705,
      "mean_token_accuracy": 0.211904639005661,
      "num_tokens": 75596499.0,
      "step": 36430
    },
    {
      "entropy": 5.583703279495239,
      "epoch": 3.3050616835994195,
      "grad_norm": 1.296875,
      "learning_rate": 0.000392924753876739,
      "loss": 4.8326,
      "mean_token_accuracy": 0.22787392288446426,
      "num_tokens": 75606329.0,
      "step": 36435
    },
    {
      "entropy": 5.628923606872559,
      "epoch": 3.305515239477504,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003928971962941743,
      "loss": 4.8966,
      "mean_token_accuracy": 0.21493573635816574,
      "num_tokens": 75617371.0,
      "step": 36440
    },
    {
      "entropy": 5.662423324584961,
      "epoch": 3.3059687953555876,
      "grad_norm": 1.25,
      "learning_rate": 0.0003928696362734655,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22077554017305373,
      "num_tokens": 75627949.0,
      "step": 36445
    },
    {
      "entropy": 5.643526458740235,
      "epoch": 3.306422351233672,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039284207381518266,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21727335900068284,
      "num_tokens": 75638642.0,
      "step": 36450
    },
    {
      "entropy": 5.585008001327514,
      "epoch": 3.306875907111756,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039281450891989555,
      "loss": 4.8713,
      "mean_token_accuracy": 0.222797192633152,
      "num_tokens": 75649997.0,
      "step": 36455
    },
    {
      "entropy": 5.601866149902344,
      "epoch": 3.3073294629898404,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003927869415881745,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22872352302074433,
      "num_tokens": 75660223.0,
      "step": 36460
    },
    {
      "entropy": 5.61545729637146,
      "epoch": 3.3077830188679247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003927593718205895,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22704134434461593,
      "num_tokens": 75670550.0,
      "step": 36465
    },
    {
      "entropy": 5.67523159980774,
      "epoch": 3.3082365747460085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039273179961771065,
      "loss": 5.0157,
      "mean_token_accuracy": 0.20890735536813737,
      "num_tokens": 75681631.0,
      "step": 36470
    },
    {
      "entropy": 5.622164487838745,
      "epoch": 3.308690130624093,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003927042249801082,
      "loss": 4.8493,
      "mean_token_accuracy": 0.22572583556175232,
      "num_tokens": 75691561.0,
      "step": 36475
    },
    {
      "entropy": 5.648086261749268,
      "epoch": 3.309143686502177,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039267664790835235,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22560212314128875,
      "num_tokens": 75701639.0,
      "step": 36480
    },
    {
      "entropy": 5.650271844863892,
      "epoch": 3.3095972423802613,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003926490684030134,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2257583260536194,
      "num_tokens": 75712823.0,
      "step": 36485
    },
    {
      "entropy": 5.62116756439209,
      "epoch": 3.3100507982583456,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039262148646466185,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21811418533325194,
      "num_tokens": 75723221.0,
      "step": 36490
    },
    {
      "entropy": 5.616157102584839,
      "epoch": 3.3105043541364294,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003925939020938679,
      "loss": 4.8892,
      "mean_token_accuracy": 0.22351250499486924,
      "num_tokens": 75733565.0,
      "step": 36495
    },
    {
      "entropy": 5.68096957206726,
      "epoch": 3.3109579100145137,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039256631529120206,
      "loss": 4.8707,
      "mean_token_accuracy": 0.2291053056716919,
      "num_tokens": 75743511.0,
      "step": 36500
    },
    {
      "entropy": 5.728188228607178,
      "epoch": 3.311411465892598,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003925387260572349,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2139803558588028,
      "num_tokens": 75754593.0,
      "step": 36505
    },
    {
      "entropy": 5.559893608093262,
      "epoch": 3.311865021770682,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039251113439253693,
      "loss": 4.8422,
      "mean_token_accuracy": 0.23125759959220887,
      "num_tokens": 75764534.0,
      "step": 36510
    },
    {
      "entropy": 5.576159286499023,
      "epoch": 3.3123185776487665,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003924835402976788,
      "loss": 4.836,
      "mean_token_accuracy": 0.22600491642951964,
      "num_tokens": 75774707.0,
      "step": 36515
    },
    {
      "entropy": 5.54131326675415,
      "epoch": 3.3127721335268507,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039245594377323106,
      "loss": 4.7895,
      "mean_token_accuracy": 0.22175977677106856,
      "num_tokens": 75784900.0,
      "step": 36520
    },
    {
      "entropy": 5.667115545272827,
      "epoch": 3.3132256894049346,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003924283448197645,
      "loss": 4.9923,
      "mean_token_accuracy": 0.21477226465940474,
      "num_tokens": 75795711.0,
      "step": 36525
    },
    {
      "entropy": 5.639719295501709,
      "epoch": 3.313679245283019,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003924007434378498,
      "loss": 4.8893,
      "mean_token_accuracy": 0.2188007965683937,
      "num_tokens": 75805951.0,
      "step": 36530
    },
    {
      "entropy": 5.620043039321899,
      "epoch": 3.314132801161103,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003923731396280579,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22319262176752092,
      "num_tokens": 75816023.0,
      "step": 36535
    },
    {
      "entropy": 5.684797716140747,
      "epoch": 3.3145863570391874,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039234553339095944,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21676614433526992,
      "num_tokens": 75827395.0,
      "step": 36540
    },
    {
      "entropy": 5.671935987472534,
      "epoch": 3.315039912917271,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039231792472712554,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22910070717334746,
      "num_tokens": 75836663.0,
      "step": 36545
    },
    {
      "entropy": 5.632699489593506,
      "epoch": 3.3154934687953554,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000392290313637127,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21638139188289643,
      "num_tokens": 75846686.0,
      "step": 36550
    },
    {
      "entropy": 5.59012188911438,
      "epoch": 3.3159470246734397,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039226270012153485,
      "loss": 4.932,
      "mean_token_accuracy": 0.21944591253995896,
      "num_tokens": 75857386.0,
      "step": 36555
    },
    {
      "entropy": 5.63440637588501,
      "epoch": 3.316400580551524,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003922350841809203,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21591539829969406,
      "num_tokens": 75867003.0,
      "step": 36560
    },
    {
      "entropy": 5.6424946784973145,
      "epoch": 3.3168541364296082,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003922074658158543,
      "loss": 4.9384,
      "mean_token_accuracy": 0.2222453013062477,
      "num_tokens": 75878155.0,
      "step": 36565
    },
    {
      "entropy": 5.596266889572144,
      "epoch": 3.3173076923076925,
      "grad_norm": 1.25,
      "learning_rate": 0.00039217984502690797,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22051066756248475,
      "num_tokens": 75888478.0,
      "step": 36570
    },
    {
      "entropy": 5.559034824371338,
      "epoch": 3.3177612481857763,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039215222181465263,
      "loss": 4.835,
      "mean_token_accuracy": 0.2222893238067627,
      "num_tokens": 75899541.0,
      "step": 36575
    },
    {
      "entropy": 5.6551636219024655,
      "epoch": 3.3182148040638606,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039212459617965956,
      "loss": 4.8968,
      "mean_token_accuracy": 0.2220456063747406,
      "num_tokens": 75910175.0,
      "step": 36580
    },
    {
      "entropy": 5.603334617614746,
      "epoch": 3.318668359941945,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003920969681224999,
      "loss": 4.8966,
      "mean_token_accuracy": 0.21762137711048127,
      "num_tokens": 75919713.0,
      "step": 36585
    },
    {
      "entropy": 5.625803756713867,
      "epoch": 3.319121915820029,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039206933764374515,
      "loss": 4.9697,
      "mean_token_accuracy": 0.21427001804113388,
      "num_tokens": 75929366.0,
      "step": 36590
    },
    {
      "entropy": 5.726399517059326,
      "epoch": 3.3195754716981134,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039204170474396656,
      "loss": 4.9993,
      "mean_token_accuracy": 0.2068703591823578,
      "num_tokens": 75940284.0,
      "step": 36595
    },
    {
      "entropy": 5.614320516586304,
      "epoch": 3.3200290275761972,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003920140694237358,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2233371004462242,
      "num_tokens": 75950752.0,
      "step": 36600
    },
    {
      "entropy": 5.6806182861328125,
      "epoch": 3.3204825834542815,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003919864316836242,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20822545737028123,
      "num_tokens": 75961308.0,
      "step": 36605
    },
    {
      "entropy": 5.6629071712493895,
      "epoch": 3.3209361393323658,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003919587915242034,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22195300310850144,
      "num_tokens": 75972317.0,
      "step": 36610
    },
    {
      "entropy": 5.610107803344727,
      "epoch": 3.32138969521045,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039193114894604504,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2312440127134323,
      "num_tokens": 75982769.0,
      "step": 36615
    },
    {
      "entropy": 5.613437271118164,
      "epoch": 3.3218432510885343,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039190350394972074,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2262508124113083,
      "num_tokens": 75992209.0,
      "step": 36620
    },
    {
      "entropy": 5.678586053848266,
      "epoch": 3.322296806966618,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003918758565358021,
      "loss": 4.9661,
      "mean_token_accuracy": 0.22328823804855347,
      "num_tokens": 76001749.0,
      "step": 36625
    },
    {
      "entropy": 5.695167112350464,
      "epoch": 3.3227503628447024,
      "grad_norm": 1.25,
      "learning_rate": 0.000391848206704861,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21765912771224977,
      "num_tokens": 76011326.0,
      "step": 36630
    },
    {
      "entropy": 5.573837375640869,
      "epoch": 3.3232039187227866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039182055445746917,
      "loss": 4.8169,
      "mean_token_accuracy": 0.22764573246240616,
      "num_tokens": 76021527.0,
      "step": 36635
    },
    {
      "entropy": 5.582305240631103,
      "epoch": 3.323657474600871,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039179289979419853,
      "loss": 4.7488,
      "mean_token_accuracy": 0.23210876286029816,
      "num_tokens": 76031217.0,
      "step": 36640
    },
    {
      "entropy": 5.630798864364624,
      "epoch": 3.324111030478955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039176524271562096,
      "loss": 5.0157,
      "mean_token_accuracy": 0.21261018216609956,
      "num_tokens": 76041936.0,
      "step": 36645
    },
    {
      "entropy": 5.641689872741699,
      "epoch": 3.324564586357039,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039173758322230847,
      "loss": 4.946,
      "mean_token_accuracy": 0.21993618458509445,
      "num_tokens": 76052431.0,
      "step": 36650
    },
    {
      "entropy": 5.634128427505493,
      "epoch": 3.3250181422351233,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003917099213148329,
      "loss": 4.8116,
      "mean_token_accuracy": 0.22751404643058776,
      "num_tokens": 76061589.0,
      "step": 36655
    },
    {
      "entropy": 5.702914667129517,
      "epoch": 3.3254716981132075,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039168225699376646,
      "loss": 4.9286,
      "mean_token_accuracy": 0.22597684264183043,
      "num_tokens": 76071642.0,
      "step": 36660
    },
    {
      "entropy": 5.602454423904419,
      "epoch": 3.325925253991292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039165459025968126,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2252756506204605,
      "num_tokens": 76081084.0,
      "step": 36665
    },
    {
      "entropy": 5.6210531234741214,
      "epoch": 3.326378809869376,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003916269211131494,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21773461997509003,
      "num_tokens": 76091125.0,
      "step": 36670
    },
    {
      "entropy": 5.716686582565307,
      "epoch": 3.3268323657474603,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003915992495547431,
      "loss": 4.9538,
      "mean_token_accuracy": 0.21376671344041825,
      "num_tokens": 76101452.0,
      "step": 36675
    },
    {
      "entropy": 5.633708572387695,
      "epoch": 3.327285921625544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039157157558503456,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21213718503713608,
      "num_tokens": 76112034.0,
      "step": 36680
    },
    {
      "entropy": 5.608516788482666,
      "epoch": 3.3277394775036284,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039154389920459615,
      "loss": 4.8846,
      "mean_token_accuracy": 0.222933891415596,
      "num_tokens": 76122395.0,
      "step": 36685
    },
    {
      "entropy": 5.673959445953369,
      "epoch": 3.3281930333817127,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003915162204140002,
      "loss": 4.8773,
      "mean_token_accuracy": 0.21718025356531143,
      "num_tokens": 76133622.0,
      "step": 36690
    },
    {
      "entropy": 5.685332250595093,
      "epoch": 3.328646589259797,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039148853921381916,
      "loss": 4.88,
      "mean_token_accuracy": 0.21922763735055922,
      "num_tokens": 76143757.0,
      "step": 36695
    },
    {
      "entropy": 5.780637454986572,
      "epoch": 3.3291001451378808,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003914608556046254,
      "loss": 5.1307,
      "mean_token_accuracy": 0.203074911236763,
      "num_tokens": 76154943.0,
      "step": 36700
    },
    {
      "entropy": 5.5843733787536625,
      "epoch": 3.329553701015965,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003914331695869914,
      "loss": 4.8975,
      "mean_token_accuracy": 0.21318686604499817,
      "num_tokens": 76165227.0,
      "step": 36705
    },
    {
      "entropy": 5.667214870452881,
      "epoch": 3.3300072568940493,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003914054811614899,
      "loss": 4.9146,
      "mean_token_accuracy": 0.2168612614274025,
      "num_tokens": 76174127.0,
      "step": 36710
    },
    {
      "entropy": 5.710438919067383,
      "epoch": 3.3304608127721336,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003913777903286934,
      "loss": 5.0059,
      "mean_token_accuracy": 0.21246084868907927,
      "num_tokens": 76183518.0,
      "step": 36715
    },
    {
      "entropy": 5.667507934570312,
      "epoch": 3.330914368650218,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003913500970891745,
      "loss": 4.934,
      "mean_token_accuracy": 0.21488931328058242,
      "num_tokens": 76193865.0,
      "step": 36720
    },
    {
      "entropy": 5.682605266571045,
      "epoch": 3.331367924528302,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003913224014435058,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21554100662469863,
      "num_tokens": 76204273.0,
      "step": 36725
    },
    {
      "entropy": 5.590853309631347,
      "epoch": 3.331821480406386,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039129470339226044,
      "loss": 4.8025,
      "mean_token_accuracy": 0.22912035286426544,
      "num_tokens": 76213706.0,
      "step": 36730
    },
    {
      "entropy": 5.658242321014404,
      "epoch": 3.33227503628447,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039126700293601086,
      "loss": 4.967,
      "mean_token_accuracy": 0.2194268748164177,
      "num_tokens": 76224484.0,
      "step": 36735
    },
    {
      "entropy": 5.638776445388794,
      "epoch": 3.3327285921625545,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003912393000753299,
      "loss": 4.9605,
      "mean_token_accuracy": 0.2113253653049469,
      "num_tokens": 76236004.0,
      "step": 36740
    },
    {
      "entropy": 5.688573789596558,
      "epoch": 3.3331821480406387,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039121159481079077,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22338373214006424,
      "num_tokens": 76245109.0,
      "step": 36745
    },
    {
      "entropy": 5.672269201278686,
      "epoch": 3.3336357039187225,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039118388714296626,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22094995826482772,
      "num_tokens": 76255154.0,
      "step": 36750
    },
    {
      "entropy": 5.626294565200806,
      "epoch": 3.334089259796807,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003911561770724292,
      "loss": 4.8462,
      "mean_token_accuracy": 0.21998923122882844,
      "num_tokens": 76266014.0,
      "step": 36755
    },
    {
      "entropy": 5.529028224945068,
      "epoch": 3.334542815674891,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000391128464599753,
      "loss": 4.8303,
      "mean_token_accuracy": 0.22455097436904908,
      "num_tokens": 76277593.0,
      "step": 36760
    },
    {
      "entropy": 5.557000541687012,
      "epoch": 3.3349963715529753,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003911007497255104,
      "loss": 4.7952,
      "mean_token_accuracy": 0.2335573747754097,
      "num_tokens": 76287816.0,
      "step": 36765
    },
    {
      "entropy": 5.657432889938354,
      "epoch": 3.3354499274310596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003910730324502748,
      "loss": 4.9343,
      "mean_token_accuracy": 0.21621690690517426,
      "num_tokens": 76298299.0,
      "step": 36770
    },
    {
      "entropy": 5.64734992980957,
      "epoch": 3.335903483309144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003910453127746192,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21185123175382614,
      "num_tokens": 76308747.0,
      "step": 36775
    },
    {
      "entropy": 5.633251667022705,
      "epoch": 3.3363570391872277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003910175906991171,
      "loss": 4.8352,
      "mean_token_accuracy": 0.2219248116016388,
      "num_tokens": 76318923.0,
      "step": 36780
    },
    {
      "entropy": 5.666672945022583,
      "epoch": 3.336810595065312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003909898662243415,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21398621648550034,
      "num_tokens": 76327985.0,
      "step": 36785
    },
    {
      "entropy": 5.648075580596924,
      "epoch": 3.3372641509433962,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039096213935086604,
      "loss": 5.001,
      "mean_token_accuracy": 0.21219079047441483,
      "num_tokens": 76338933.0,
      "step": 36790
    },
    {
      "entropy": 5.770057249069214,
      "epoch": 3.3377177068214805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039093441007926396,
      "loss": 5.0038,
      "mean_token_accuracy": 0.2134322240948677,
      "num_tokens": 76349601.0,
      "step": 36795
    },
    {
      "entropy": 5.660004901885986,
      "epoch": 3.3381712626995648,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039090667841010874,
      "loss": 4.8662,
      "mean_token_accuracy": 0.2266321823000908,
      "num_tokens": 76359968.0,
      "step": 36800
    },
    {
      "entropy": 5.547734928131104,
      "epoch": 3.3386248185776486,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003908789443439737,
      "loss": 4.8116,
      "mean_token_accuracy": 0.22933948487043382,
      "num_tokens": 76369864.0,
      "step": 36805
    },
    {
      "entropy": 5.577858543395996,
      "epoch": 3.339078374455733,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003908512078814329,
      "loss": 4.8123,
      "mean_token_accuracy": 0.22675940841436387,
      "num_tokens": 76379158.0,
      "step": 36810
    },
    {
      "entropy": 5.62724461555481,
      "epoch": 3.339531930333817,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039082346902305937,
      "loss": 4.8428,
      "mean_token_accuracy": 0.22314077764749526,
      "num_tokens": 76388798.0,
      "step": 36815
    },
    {
      "entropy": 5.570607900619507,
      "epoch": 3.3399854862119014,
      "grad_norm": 1.1875,
      "learning_rate": 0.000390795727769427,
      "loss": 4.8534,
      "mean_token_accuracy": 0.225554059445858,
      "num_tokens": 76399315.0,
      "step": 36820
    },
    {
      "entropy": 5.633226108551026,
      "epoch": 3.3404390420899857,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003907679841211095,
      "loss": 4.92,
      "mean_token_accuracy": 0.2147127464413643,
      "num_tokens": 76408841.0,
      "step": 36825
    },
    {
      "entropy": 5.6968847751617435,
      "epoch": 3.3408925979680695,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003907402380786806,
      "loss": 4.9618,
      "mean_token_accuracy": 0.2105090230703354,
      "num_tokens": 76420055.0,
      "step": 36830
    },
    {
      "entropy": 5.60674901008606,
      "epoch": 3.3413461538461537,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003907124896427141,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2286544471979141,
      "num_tokens": 76430921.0,
      "step": 36835
    },
    {
      "entropy": 5.626554727554321,
      "epoch": 3.341799709724238,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003906847388137837,
      "loss": 4.8723,
      "mean_token_accuracy": 0.21629425883293152,
      "num_tokens": 76441216.0,
      "step": 36840
    },
    {
      "entropy": 5.65028944015503,
      "epoch": 3.3422532656023223,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003906569855924636,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22278068959712982,
      "num_tokens": 76451940.0,
      "step": 36845
    },
    {
      "entropy": 5.672978687286377,
      "epoch": 3.3427068214804065,
      "grad_norm": 1.25,
      "learning_rate": 0.0003906292299793275,
      "loss": 4.9179,
      "mean_token_accuracy": 0.22534059882164,
      "num_tokens": 76461701.0,
      "step": 36850
    },
    {
      "entropy": 5.674586582183838,
      "epoch": 3.3431603773584904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039060147197494943,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20937626659870148,
      "num_tokens": 76473658.0,
      "step": 36855
    },
    {
      "entropy": 5.63720850944519,
      "epoch": 3.3436139332365746,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039057371157990347,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22110254019498826,
      "num_tokens": 76483821.0,
      "step": 36860
    },
    {
      "entropy": 5.616532707214356,
      "epoch": 3.344067489114659,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039054594879476374,
      "loss": 4.9134,
      "mean_token_accuracy": 0.2158988744020462,
      "num_tokens": 76494699.0,
      "step": 36865
    },
    {
      "entropy": 5.653545284271241,
      "epoch": 3.344521044992743,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039051818362010433,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22793588489294053,
      "num_tokens": 76504299.0,
      "step": 36870
    },
    {
      "entropy": 5.703910684585571,
      "epoch": 3.3449746008708274,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003904904160564994,
      "loss": 4.9537,
      "mean_token_accuracy": 0.21410917937755586,
      "num_tokens": 76514469.0,
      "step": 36875
    },
    {
      "entropy": 5.657998752593994,
      "epoch": 3.3454281567489117,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003904626461045232,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21513868421316146,
      "num_tokens": 76524666.0,
      "step": 36880
    },
    {
      "entropy": 5.553595781326294,
      "epoch": 3.3458817126269955,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039043487376475017,
      "loss": 4.8824,
      "mean_token_accuracy": 0.21571989506483077,
      "num_tokens": 76534787.0,
      "step": 36885
    },
    {
      "entropy": 5.626442337036133,
      "epoch": 3.34633526850508,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039040709903775446,
      "loss": 4.91,
      "mean_token_accuracy": 0.21576206535100936,
      "num_tokens": 76545055.0,
      "step": 36890
    },
    {
      "entropy": 5.584850931167603,
      "epoch": 3.346788824383164,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003903793219241106,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22780652195215226,
      "num_tokens": 76555458.0,
      "step": 36895
    },
    {
      "entropy": 5.597615909576416,
      "epoch": 3.3472423802612483,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039035154242439294,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22727704793214798,
      "num_tokens": 76566272.0,
      "step": 36900
    },
    {
      "entropy": 5.70754919052124,
      "epoch": 3.347695936139332,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003903237605391759,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2175618126988411,
      "num_tokens": 76575656.0,
      "step": 36905
    },
    {
      "entropy": 5.712425422668457,
      "epoch": 3.3481494920174164,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003902959762690342,
      "loss": 4.8443,
      "mean_token_accuracy": 0.2206042915582657,
      "num_tokens": 76586481.0,
      "step": 36910
    },
    {
      "entropy": 5.618427181243897,
      "epoch": 3.3486030478955007,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003902681896145423,
      "loss": 4.839,
      "mean_token_accuracy": 0.2296876609325409,
      "num_tokens": 76596023.0,
      "step": 36915
    },
    {
      "entropy": 5.53643536567688,
      "epoch": 3.349056603773585,
      "grad_norm": 1.125,
      "learning_rate": 0.00039024040057627487,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22770035415887832,
      "num_tokens": 76605887.0,
      "step": 36920
    },
    {
      "entropy": 5.583228921890258,
      "epoch": 3.349510159651669,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003902126091548066,
      "loss": 4.8174,
      "mean_token_accuracy": 0.23167879432439803,
      "num_tokens": 76616007.0,
      "step": 36925
    },
    {
      "entropy": 5.664948844909668,
      "epoch": 3.3499637155297535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039018481535071215,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21813850849866867,
      "num_tokens": 76626253.0,
      "step": 36930
    },
    {
      "entropy": 5.644758558273315,
      "epoch": 3.3504172714078373,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003901570191645665,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2226046070456505,
      "num_tokens": 76636341.0,
      "step": 36935
    },
    {
      "entropy": 5.6173115253448485,
      "epoch": 3.3508708272859216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039012922059694416,
      "loss": 4.8527,
      "mean_token_accuracy": 0.22423526346683503,
      "num_tokens": 76647616.0,
      "step": 36940
    },
    {
      "entropy": 5.617476511001587,
      "epoch": 3.351324383164006,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003901014196484203,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22635423094034196,
      "num_tokens": 76658261.0,
      "step": 36945
    },
    {
      "entropy": 5.600188827514648,
      "epoch": 3.35177793904209,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039007361631956976,
      "loss": 4.888,
      "mean_token_accuracy": 0.2141212910413742,
      "num_tokens": 76667985.0,
      "step": 36950
    },
    {
      "entropy": 5.6251383304595945,
      "epoch": 3.3522314949201744,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039004581061096745,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2227220669388771,
      "num_tokens": 76678538.0,
      "step": 36955
    },
    {
      "entropy": 5.6550829887390135,
      "epoch": 3.352685050798258,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003900180025231886,
      "loss": 4.791,
      "mean_token_accuracy": 0.22498877644538878,
      "num_tokens": 76688052.0,
      "step": 36960
    },
    {
      "entropy": 5.590238904953003,
      "epoch": 3.3531386066763424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038999019205680796,
      "loss": 4.8455,
      "mean_token_accuracy": 0.2230977326631546,
      "num_tokens": 76699241.0,
      "step": 36965
    },
    {
      "entropy": 5.609111690521241,
      "epoch": 3.3535921625544267,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000389962379212401,
      "loss": 4.8795,
      "mean_token_accuracy": 0.22016009092330932,
      "num_tokens": 76708979.0,
      "step": 36970
    },
    {
      "entropy": 5.651079273223877,
      "epoch": 3.354045718432511,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003899345639905426,
      "loss": 4.8535,
      "mean_token_accuracy": 0.2220097839832306,
      "num_tokens": 76719330.0,
      "step": 36975
    },
    {
      "entropy": 5.675118160247803,
      "epoch": 3.3544992743105952,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003899067463918083,
      "loss": 4.9969,
      "mean_token_accuracy": 0.2108915165066719,
      "num_tokens": 76730210.0,
      "step": 36980
    },
    {
      "entropy": 5.72794623374939,
      "epoch": 3.354952830188679,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038987892641677307,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21876634955406188,
      "num_tokens": 76739977.0,
      "step": 36985
    },
    {
      "entropy": 5.606538152694702,
      "epoch": 3.3554063860667633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003898511040660124,
      "loss": 4.8716,
      "mean_token_accuracy": 0.21951078474521638,
      "num_tokens": 76750200.0,
      "step": 36990
    },
    {
      "entropy": 5.6927406787872314,
      "epoch": 3.3558599419448476,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003898232793401017,
      "loss": 4.984,
      "mean_token_accuracy": 0.2083211973309517,
      "num_tokens": 76761420.0,
      "step": 36995
    },
    {
      "entropy": 5.610450220108032,
      "epoch": 3.356313497822932,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038979545223961627,
      "loss": 4.8559,
      "mean_token_accuracy": 0.22148122191429137,
      "num_tokens": 76771872.0,
      "step": 37000
    },
    {
      "entropy": 5.57882080078125,
      "epoch": 3.356767053701016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038976762276513163,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22403264045715332,
      "num_tokens": 76782303.0,
      "step": 37005
    },
    {
      "entropy": 5.676229524612427,
      "epoch": 3.3572206095791,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003897397909172232,
      "loss": 4.975,
      "mean_token_accuracy": 0.21418311446905136,
      "num_tokens": 76792595.0,
      "step": 37010
    },
    {
      "entropy": 5.688700151443482,
      "epoch": 3.3576741654571842,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003897119566964669,
      "loss": 4.93,
      "mean_token_accuracy": 0.2173720806837082,
      "num_tokens": 76802577.0,
      "step": 37015
    },
    {
      "entropy": 5.648487854003906,
      "epoch": 3.3581277213352685,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003896841201034379,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22972912043333055,
      "num_tokens": 76812686.0,
      "step": 37020
    },
    {
      "entropy": 5.552531290054321,
      "epoch": 3.3585812772133528,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003896562811387121,
      "loss": 4.7747,
      "mean_token_accuracy": 0.23097454458475114,
      "num_tokens": 76823641.0,
      "step": 37025
    },
    {
      "entropy": 5.656320858001709,
      "epoch": 3.359034833091437,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038962843980286523,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21978622525930405,
      "num_tokens": 76834060.0,
      "step": 37030
    },
    {
      "entropy": 5.717332077026367,
      "epoch": 3.3594883889695213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038960059609647306,
      "loss": 4.9709,
      "mean_token_accuracy": 0.2201485112309456,
      "num_tokens": 76845078.0,
      "step": 37035
    },
    {
      "entropy": 5.687024450302124,
      "epoch": 3.359941944847605,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003895727500201113,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21406721323728561,
      "num_tokens": 76856539.0,
      "step": 37040
    },
    {
      "entropy": 5.726544904708862,
      "epoch": 3.3603955007256894,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038954490157435583,
      "loss": 4.993,
      "mean_token_accuracy": 0.2162439852952957,
      "num_tokens": 76866941.0,
      "step": 37045
    },
    {
      "entropy": 5.67036452293396,
      "epoch": 3.3608490566037736,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038951705075978264,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2171493262052536,
      "num_tokens": 76878205.0,
      "step": 37050
    },
    {
      "entropy": 5.674322271347046,
      "epoch": 3.361302612481858,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003894891975769677,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20860136896371842,
      "num_tokens": 76888609.0,
      "step": 37055
    },
    {
      "entropy": 5.590348100662231,
      "epoch": 3.3617561683599417,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003894613420264869,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2163400486111641,
      "num_tokens": 76899637.0,
      "step": 37060
    },
    {
      "entropy": 5.644905233383179,
      "epoch": 3.362209724238026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038943348410891643,
      "loss": 4.9187,
      "mean_token_accuracy": 0.2158982515335083,
      "num_tokens": 76911186.0,
      "step": 37065
    },
    {
      "entropy": 5.672036266326904,
      "epoch": 3.3626632801161103,
      "grad_norm": 1.25,
      "learning_rate": 0.0003894056238248324,
      "loss": 4.8665,
      "mean_token_accuracy": 0.21881164312362672,
      "num_tokens": 76921704.0,
      "step": 37070
    },
    {
      "entropy": 5.587625932693482,
      "epoch": 3.3631168359941945,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003893777611748108,
      "loss": 4.863,
      "mean_token_accuracy": 0.22683814764022828,
      "num_tokens": 76932362.0,
      "step": 37075
    },
    {
      "entropy": 5.661067008972168,
      "epoch": 3.363570391872279,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038934989615942807,
      "loss": 5.0115,
      "mean_token_accuracy": 0.21478288173675536,
      "num_tokens": 76942581.0,
      "step": 37080
    },
    {
      "entropy": 5.655270290374756,
      "epoch": 3.364023947750363,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003893220287792603,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2186492696404457,
      "num_tokens": 76952912.0,
      "step": 37085
    },
    {
      "entropy": 5.681910562515259,
      "epoch": 3.364477503628447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038929415903488383,
      "loss": 4.9248,
      "mean_token_accuracy": 0.2178400918841362,
      "num_tokens": 76963320.0,
      "step": 37090
    },
    {
      "entropy": 5.613359355926514,
      "epoch": 3.364931059506531,
      "grad_norm": 1.25,
      "learning_rate": 0.000389266286926875,
      "loss": 4.7885,
      "mean_token_accuracy": 0.22660157829523087,
      "num_tokens": 76973448.0,
      "step": 37095
    },
    {
      "entropy": 5.589280605316162,
      "epoch": 3.3653846153846154,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038923841245581026,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22826192975044252,
      "num_tokens": 76983671.0,
      "step": 37100
    },
    {
      "entropy": 5.676054239273071,
      "epoch": 3.3658381712626997,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038921053562226607,
      "loss": 4.9991,
      "mean_token_accuracy": 0.20990085899829863,
      "num_tokens": 76994345.0,
      "step": 37105
    },
    {
      "entropy": 5.595503473281861,
      "epoch": 3.3662917271407835,
      "grad_norm": 1.125,
      "learning_rate": 0.00038918265642681884,
      "loss": 4.8031,
      "mean_token_accuracy": 0.22392067015171052,
      "num_tokens": 77004809.0,
      "step": 37110
    },
    {
      "entropy": 5.619469738006591,
      "epoch": 3.3667452830188678,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038915477487004525,
      "loss": 4.8884,
      "mean_token_accuracy": 0.21928932517766953,
      "num_tokens": 77015505.0,
      "step": 37115
    },
    {
      "entropy": 5.554884624481201,
      "epoch": 3.367198838896952,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038912689095252183,
      "loss": 4.7275,
      "mean_token_accuracy": 0.2352317303419113,
      "num_tokens": 77025572.0,
      "step": 37120
    },
    {
      "entropy": 5.71528992652893,
      "epoch": 3.3676523947750363,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003890990046748251,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20950431376695633,
      "num_tokens": 77037273.0,
      "step": 37125
    },
    {
      "entropy": 5.713378381729126,
      "epoch": 3.3681059506531206,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038907111603753196,
      "loss": 4.966,
      "mean_token_accuracy": 0.21572712510824205,
      "num_tokens": 77047724.0,
      "step": 37130
    },
    {
      "entropy": 5.64531569480896,
      "epoch": 3.368559506531205,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038904322504121916,
      "loss": 4.8981,
      "mean_token_accuracy": 0.22062236517667771,
      "num_tokens": 77057839.0,
      "step": 37135
    },
    {
      "entropy": 5.612880802154541,
      "epoch": 3.3690130624092887,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003890153316864632,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2150793269276619,
      "num_tokens": 77068505.0,
      "step": 37140
    },
    {
      "entropy": 5.6744067668914795,
      "epoch": 3.369466618287373,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038898743597384126,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21884512454271315,
      "num_tokens": 77078126.0,
      "step": 37145
    },
    {
      "entropy": 5.630244588851928,
      "epoch": 3.369920174165457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038895953790393014,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22241708636283875,
      "num_tokens": 77087802.0,
      "step": 37150
    },
    {
      "entropy": 5.657577037811279,
      "epoch": 3.3703737300435415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038893163747730664,
      "loss": 4.915,
      "mean_token_accuracy": 0.22626010477542877,
      "num_tokens": 77097901.0,
      "step": 37155
    },
    {
      "entropy": 5.631954956054687,
      "epoch": 3.3708272859216257,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038890373469454794,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22109645754098892,
      "num_tokens": 77107604.0,
      "step": 37160
    },
    {
      "entropy": 5.645776796340942,
      "epoch": 3.3712808417997095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038887582955623084,
      "loss": 4.8281,
      "mean_token_accuracy": 0.22884132117033004,
      "num_tokens": 77117410.0,
      "step": 37165
    },
    {
      "entropy": 5.676944065093994,
      "epoch": 3.371734397677794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003888479220629326,
      "loss": 4.9009,
      "mean_token_accuracy": 0.2230340287089348,
      "num_tokens": 77129546.0,
      "step": 37170
    },
    {
      "entropy": 5.606623029708862,
      "epoch": 3.372187953555878,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003888200122152304,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22369149178266526,
      "num_tokens": 77139217.0,
      "step": 37175
    },
    {
      "entropy": 5.678354024887085,
      "epoch": 3.3726415094339623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003887921000137013,
      "loss": 4.9946,
      "mean_token_accuracy": 0.21099012792110444,
      "num_tokens": 77149694.0,
      "step": 37180
    },
    {
      "entropy": 5.611155033111572,
      "epoch": 3.3730950653120466,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003887641854589226,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22674783021211625,
      "num_tokens": 77159291.0,
      "step": 37185
    },
    {
      "entropy": 5.594446086883545,
      "epoch": 3.3735486211901304,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003887362685514716,
      "loss": 4.8634,
      "mean_token_accuracy": 0.21796692460775374,
      "num_tokens": 77170507.0,
      "step": 37190
    },
    {
      "entropy": 5.64688925743103,
      "epoch": 3.3740021770682147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003887083492919254,
      "loss": 4.9072,
      "mean_token_accuracy": 0.2241587221622467,
      "num_tokens": 77179673.0,
      "step": 37195
    },
    {
      "entropy": 5.68870358467102,
      "epoch": 3.374455732946299,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003886804276808617,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22971894443035126,
      "num_tokens": 77190163.0,
      "step": 37200
    },
    {
      "entropy": 5.676418352127075,
      "epoch": 3.3749092888243832,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003886525037188578,
      "loss": 4.9028,
      "mean_token_accuracy": 0.22358739078044892,
      "num_tokens": 77201498.0,
      "step": 37205
    },
    {
      "entropy": 5.625469636917114,
      "epoch": 3.3753628447024675,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003886245774064911,
      "loss": 4.8477,
      "mean_token_accuracy": 0.2226455807685852,
      "num_tokens": 77211475.0,
      "step": 37210
    },
    {
      "entropy": 5.625943946838379,
      "epoch": 3.3758164005805513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038859664874433923,
      "loss": 4.9027,
      "mean_token_accuracy": 0.22737003713846207,
      "num_tokens": 77221972.0,
      "step": 37215
    },
    {
      "entropy": 5.6225871562957765,
      "epoch": 3.3762699564586356,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038856871773297965,
      "loss": 4.9151,
      "mean_token_accuracy": 0.2190821498632431,
      "num_tokens": 77232701.0,
      "step": 37220
    },
    {
      "entropy": 5.6404424667358395,
      "epoch": 3.37672351233672,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038854078437299005,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2315544843673706,
      "num_tokens": 77243242.0,
      "step": 37225
    },
    {
      "entropy": 5.617395973205566,
      "epoch": 3.377177068214804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003885128486649482,
      "loss": 4.8283,
      "mean_token_accuracy": 0.22328327894210814,
      "num_tokens": 77254279.0,
      "step": 37230
    },
    {
      "entropy": 5.627441310882569,
      "epoch": 3.3776306240928884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038848491060943164,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22122459560632707,
      "num_tokens": 77265483.0,
      "step": 37235
    },
    {
      "entropy": 5.5777356147766115,
      "epoch": 3.3780841799709727,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003884569702070182,
      "loss": 4.8363,
      "mean_token_accuracy": 0.21845719665288926,
      "num_tokens": 77275848.0,
      "step": 37240
    },
    {
      "entropy": 5.584741640090942,
      "epoch": 3.3785377358490565,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038842902745828573,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22069669812917708,
      "num_tokens": 77285943.0,
      "step": 37245
    },
    {
      "entropy": 5.608461809158325,
      "epoch": 3.3789912917271407,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003884010823638121,
      "loss": 4.8613,
      "mean_token_accuracy": 0.22333627939224243,
      "num_tokens": 77295276.0,
      "step": 37250
    },
    {
      "entropy": 5.679275941848755,
      "epoch": 3.379444847605225,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003883731349241751,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22171399742364883,
      "num_tokens": 77305713.0,
      "step": 37255
    },
    {
      "entropy": 5.650550603866577,
      "epoch": 3.3798984034833093,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003883451851399528,
      "loss": 4.9611,
      "mean_token_accuracy": 0.2204166069626808,
      "num_tokens": 77316717.0,
      "step": 37260
    },
    {
      "entropy": 5.595538806915283,
      "epoch": 3.380351959361393,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038831723301172325,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22316533625125884,
      "num_tokens": 77327763.0,
      "step": 37265
    },
    {
      "entropy": 5.670816946029663,
      "epoch": 3.3808055152394774,
      "grad_norm": 1.375,
      "learning_rate": 0.0003882892785400644,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21523924469947814,
      "num_tokens": 77337495.0,
      "step": 37270
    },
    {
      "entropy": 5.6860452651977536,
      "epoch": 3.3812590711175616,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038826132172555444,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21573468148708344,
      "num_tokens": 77348849.0,
      "step": 37275
    },
    {
      "entropy": 5.623383569717407,
      "epoch": 3.381712626995646,
      "grad_norm": 1.25,
      "learning_rate": 0.0003882333625687714,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21307836323976517,
      "num_tokens": 77359634.0,
      "step": 37280
    },
    {
      "entropy": 5.666058349609375,
      "epoch": 3.38216618287373,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038820540107029365,
      "loss": 4.8668,
      "mean_token_accuracy": 0.22210056483745574,
      "num_tokens": 77369342.0,
      "step": 37285
    },
    {
      "entropy": 5.697228813171387,
      "epoch": 3.3826197387518144,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003881774372306993,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22103191018104554,
      "num_tokens": 77379963.0,
      "step": 37290
    },
    {
      "entropy": 5.619308233261108,
      "epoch": 3.3830732946298983,
      "grad_norm": 1.375,
      "learning_rate": 0.0003881494710505667,
      "loss": 4.8362,
      "mean_token_accuracy": 0.2210730493068695,
      "num_tokens": 77390135.0,
      "step": 37295
    },
    {
      "entropy": 5.685325288772583,
      "epoch": 3.3835268505079825,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038812150253047427,
      "loss": 4.9901,
      "mean_token_accuracy": 0.21846597790718078,
      "num_tokens": 77400151.0,
      "step": 37300
    },
    {
      "entropy": 5.634002351760865,
      "epoch": 3.383980406386067,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003880935316710003,
      "loss": 4.8889,
      "mean_token_accuracy": 0.22444234043359756,
      "num_tokens": 77410723.0,
      "step": 37305
    },
    {
      "entropy": 5.598892545700073,
      "epoch": 3.384433962264151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003880655584727233,
      "loss": 4.8243,
      "mean_token_accuracy": 0.22736201137304307,
      "num_tokens": 77421213.0,
      "step": 37310
    },
    {
      "entropy": 5.646526718139649,
      "epoch": 3.384887518142235,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003880375829362217,
      "loss": 4.9293,
      "mean_token_accuracy": 0.2224885955452919,
      "num_tokens": 77431705.0,
      "step": 37315
    },
    {
      "entropy": 5.599805164337158,
      "epoch": 3.385341074020319,
      "grad_norm": 1.25,
      "learning_rate": 0.0003880096050620741,
      "loss": 4.8609,
      "mean_token_accuracy": 0.21971948593854904,
      "num_tokens": 77442120.0,
      "step": 37320
    },
    {
      "entropy": 5.636885929107666,
      "epoch": 3.3857946298984034,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003879816248508591,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22191519886255265,
      "num_tokens": 77451992.0,
      "step": 37325
    },
    {
      "entropy": 5.6995704650878904,
      "epoch": 3.3862481857764877,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038795364230315523,
      "loss": 5.0875,
      "mean_token_accuracy": 0.20254082083702088,
      "num_tokens": 77462830.0,
      "step": 37330
    },
    {
      "entropy": 5.675750303268432,
      "epoch": 3.386701741654572,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003879256574195414,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2236740231513977,
      "num_tokens": 77472617.0,
      "step": 37335
    },
    {
      "entropy": 5.562463808059692,
      "epoch": 3.387155297532656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038789767020059614,
      "loss": 4.8201,
      "mean_token_accuracy": 0.2271183893084526,
      "num_tokens": 77483430.0,
      "step": 37340
    },
    {
      "entropy": 5.562888717651367,
      "epoch": 3.38760885341074,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038786968064689816,
      "loss": 4.8098,
      "mean_token_accuracy": 0.2263078898191452,
      "num_tokens": 77493038.0,
      "step": 37345
    },
    {
      "entropy": 5.659874010086059,
      "epoch": 3.3880624092888243,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003878416887590265,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2145719274878502,
      "num_tokens": 77502816.0,
      "step": 37350
    },
    {
      "entropy": 5.581992673873901,
      "epoch": 3.3885159651669086,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038781369453756004,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2255415290594101,
      "num_tokens": 77513026.0,
      "step": 37355
    },
    {
      "entropy": 5.652221059799194,
      "epoch": 3.388969521044993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003877856979830775,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2155012905597687,
      "num_tokens": 77523217.0,
      "step": 37360
    },
    {
      "entropy": 5.615833377838134,
      "epoch": 3.389423076923077,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038775769909615805,
      "loss": 4.9712,
      "mean_token_accuracy": 0.22026692777872087,
      "num_tokens": 77533339.0,
      "step": 37365
    },
    {
      "entropy": 5.616430568695068,
      "epoch": 3.389876632801161,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038772969787738074,
      "loss": 4.856,
      "mean_token_accuracy": 0.2229970932006836,
      "num_tokens": 77543438.0,
      "step": 37370
    },
    {
      "entropy": 5.6887706279754635,
      "epoch": 3.390330188679245,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003877016943273244,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21426608860492707,
      "num_tokens": 77552825.0,
      "step": 37375
    },
    {
      "entropy": 5.635131931304931,
      "epoch": 3.3907837445573294,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038767368844656836,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21278969645500184,
      "num_tokens": 77563107.0,
      "step": 37380
    },
    {
      "entropy": 5.702635431289673,
      "epoch": 3.3912373004354137,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038764568023569177,
      "loss": 4.9433,
      "mean_token_accuracy": 0.21049367338418962,
      "num_tokens": 77573312.0,
      "step": 37385
    },
    {
      "entropy": 5.60852165222168,
      "epoch": 3.391690856313498,
      "grad_norm": 1.25,
      "learning_rate": 0.0003876176696952737,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21554463654756545,
      "num_tokens": 77583742.0,
      "step": 37390
    },
    {
      "entropy": 5.733423137664795,
      "epoch": 3.3921444121915822,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038758965682589357,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20895381271839142,
      "num_tokens": 77593667.0,
      "step": 37395
    },
    {
      "entropy": 5.642626810073852,
      "epoch": 3.392597968069666,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003875616416281307,
      "loss": 4.8424,
      "mean_token_accuracy": 0.21972652524709702,
      "num_tokens": 77604249.0,
      "step": 37400
    },
    {
      "entropy": 5.602638244628906,
      "epoch": 3.3930515239477503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038753362410256435,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22382428199052812,
      "num_tokens": 77615095.0,
      "step": 37405
    },
    {
      "entropy": 5.588194799423218,
      "epoch": 3.3935050798258346,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003875056042497739,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22604676187038422,
      "num_tokens": 77625182.0,
      "step": 37410
    },
    {
      "entropy": 5.643997526168823,
      "epoch": 3.393958635703919,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038747758207033903,
      "loss": 4.8591,
      "mean_token_accuracy": 0.22249070405960084,
      "num_tokens": 77635423.0,
      "step": 37415
    },
    {
      "entropy": 5.550887393951416,
      "epoch": 3.3944121915820027,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003874495575648391,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2261244237422943,
      "num_tokens": 77645262.0,
      "step": 37420
    },
    {
      "entropy": 5.6124796867370605,
      "epoch": 3.394865747460087,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038742153073385356,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21230219751596452,
      "num_tokens": 77657647.0,
      "step": 37425
    },
    {
      "entropy": 5.629698801040649,
      "epoch": 3.3953193033381712,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038739350157796214,
      "loss": 4.8956,
      "mean_token_accuracy": 0.2191387176513672,
      "num_tokens": 77668074.0,
      "step": 37430
    },
    {
      "entropy": 5.702228021621704,
      "epoch": 3.3957728592162555,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038736547009774445,
      "loss": 4.8502,
      "mean_token_accuracy": 0.21971208304166795,
      "num_tokens": 77678661.0,
      "step": 37435
    },
    {
      "entropy": 5.644606971740723,
      "epoch": 3.3962264150943398,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003873374362937803,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21349372565746308,
      "num_tokens": 77689206.0,
      "step": 37440
    },
    {
      "entropy": 5.646375417709351,
      "epoch": 3.396679970972424,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003873094001666493,
      "loss": 4.9692,
      "mean_token_accuracy": 0.2182112842798233,
      "num_tokens": 77699254.0,
      "step": 37445
    },
    {
      "entropy": 5.642358350753784,
      "epoch": 3.397133526850508,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003872813617169312,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2223755970597267,
      "num_tokens": 77709595.0,
      "step": 37450
    },
    {
      "entropy": 5.599517297744751,
      "epoch": 3.397587082728592,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038725332094520607,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22112639248371124,
      "num_tokens": 77719446.0,
      "step": 37455
    },
    {
      "entropy": 5.575517082214356,
      "epoch": 3.3980406386066764,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003872252778520536,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2274940252304077,
      "num_tokens": 77729723.0,
      "step": 37460
    },
    {
      "entropy": 5.590863561630249,
      "epoch": 3.3984941944847606,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003871972324380537,
      "loss": 4.8618,
      "mean_token_accuracy": 0.21867963522672654,
      "num_tokens": 77739380.0,
      "step": 37465
    },
    {
      "entropy": 5.644940614700317,
      "epoch": 3.3989477503628445,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003871691847037865,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22448876351118088,
      "num_tokens": 77750683.0,
      "step": 37470
    },
    {
      "entropy": 5.641706275939941,
      "epoch": 3.3994013062409287,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038714113464983205,
      "loss": 4.9579,
      "mean_token_accuracy": 0.2079791784286499,
      "num_tokens": 77761951.0,
      "step": 37475
    },
    {
      "entropy": 5.673253965377808,
      "epoch": 3.399854862119013,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003871130822767702,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21969587057828904,
      "num_tokens": 77772165.0,
      "step": 37480
    },
    {
      "entropy": 5.67796688079834,
      "epoch": 3.4003084179970973,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003870850275851814,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20794787406921386,
      "num_tokens": 77781916.0,
      "step": 37485
    },
    {
      "entropy": 5.62520580291748,
      "epoch": 3.4007619738751815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038705697057564556,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21550885438919068,
      "num_tokens": 77791397.0,
      "step": 37490
    },
    {
      "entropy": 5.560829067230225,
      "epoch": 3.401215529753266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000387028911248743,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22356773167848587,
      "num_tokens": 77801694.0,
      "step": 37495
    },
    {
      "entropy": 5.657276630401611,
      "epoch": 3.4016690856313496,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038700084960505414,
      "loss": 4.95,
      "mean_token_accuracy": 0.2195652574300766,
      "num_tokens": 77813199.0,
      "step": 37500
    },
    {
      "entropy": 5.709721088409424,
      "epoch": 3.402122641509434,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003869727856451591,
      "loss": 5.0204,
      "mean_token_accuracy": 0.2177110493183136,
      "num_tokens": 77823689.0,
      "step": 37505
    },
    {
      "entropy": 5.630706834793091,
      "epoch": 3.402576197387518,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038694471936963825,
      "loss": 4.858,
      "mean_token_accuracy": 0.2285211369395256,
      "num_tokens": 77834331.0,
      "step": 37510
    },
    {
      "entropy": 5.630411100387573,
      "epoch": 3.4030297532656024,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038691665077907213,
      "loss": 4.8143,
      "mean_token_accuracy": 0.22167232632637024,
      "num_tokens": 77843988.0,
      "step": 37515
    },
    {
      "entropy": 5.613130950927735,
      "epoch": 3.4034833091436867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003868885798740412,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2197393298149109,
      "num_tokens": 77854553.0,
      "step": 37520
    },
    {
      "entropy": 5.600230026245117,
      "epoch": 3.4039368650217705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038686050665512583,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22118595987558365,
      "num_tokens": 77864629.0,
      "step": 37525
    },
    {
      "entropy": 5.651218271255493,
      "epoch": 3.4043904208998548,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038683243112290673,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21115792095661162,
      "num_tokens": 77875465.0,
      "step": 37530
    },
    {
      "entropy": 5.677199268341065,
      "epoch": 3.404843976777939,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003868043532779645,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2210352823138237,
      "num_tokens": 77885337.0,
      "step": 37535
    },
    {
      "entropy": 5.660326862335205,
      "epoch": 3.4052975326560233,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038677627312087967,
      "loss": 4.8551,
      "mean_token_accuracy": 0.22385770529508592,
      "num_tokens": 77896465.0,
      "step": 37540
    },
    {
      "entropy": 5.636050510406494,
      "epoch": 3.4057510885341076,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003867481906522331,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2208396539092064,
      "num_tokens": 77906214.0,
      "step": 37545
    },
    {
      "entropy": 5.662158632278443,
      "epoch": 3.4062046444121914,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003867201058726054,
      "loss": 5.0489,
      "mean_token_accuracy": 0.2059251293540001,
      "num_tokens": 77917466.0,
      "step": 37550
    },
    {
      "entropy": 5.71481351852417,
      "epoch": 3.4066582002902757,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003866920187825775,
      "loss": 4.9659,
      "mean_token_accuracy": 0.21619915813207627,
      "num_tokens": 77927648.0,
      "step": 37555
    },
    {
      "entropy": 5.605722093582154,
      "epoch": 3.40711175616836,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038666392938273015,
      "loss": 4.8082,
      "mean_token_accuracy": 0.22768684923648835,
      "num_tokens": 77938028.0,
      "step": 37560
    },
    {
      "entropy": 5.6339057922363285,
      "epoch": 3.407565312046444,
      "grad_norm": 1.25,
      "learning_rate": 0.0003866358376736443,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22474474608898162,
      "num_tokens": 77947276.0,
      "step": 37565
    },
    {
      "entropy": 5.674513292312622,
      "epoch": 3.4080188679245285,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003866077436559009,
      "loss": 4.9112,
      "mean_token_accuracy": 0.2203516498208046,
      "num_tokens": 77956791.0,
      "step": 37570
    },
    {
      "entropy": 5.6811309337615965,
      "epoch": 3.4084724238026123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003865796473300809,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21620651185512543,
      "num_tokens": 77967402.0,
      "step": 37575
    },
    {
      "entropy": 5.742505884170532,
      "epoch": 3.4089259796806965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003865515486967653,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2149927243590355,
      "num_tokens": 77978000.0,
      "step": 37580
    },
    {
      "entropy": 5.5713791847229,
      "epoch": 3.409379535558781,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038652344775653535,
      "loss": 4.8047,
      "mean_token_accuracy": 0.23236246407032013,
      "num_tokens": 77987959.0,
      "step": 37585
    },
    {
      "entropy": 5.60419774055481,
      "epoch": 3.409833091436865,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003864953445099721,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22378165721893312,
      "num_tokens": 77997990.0,
      "step": 37590
    },
    {
      "entropy": 5.662144613265991,
      "epoch": 3.4102866473149493,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038646723895765654,
      "loss": 4.9063,
      "mean_token_accuracy": 0.21356898099184035,
      "num_tokens": 78008831.0,
      "step": 37595
    },
    {
      "entropy": 5.72682032585144,
      "epoch": 3.4107402031930336,
      "grad_norm": 1.25,
      "learning_rate": 0.0003864391311001702,
      "loss": 4.9732,
      "mean_token_accuracy": 0.21466707438230515,
      "num_tokens": 78019338.0,
      "step": 37600
    },
    {
      "entropy": 5.706204509735107,
      "epoch": 3.4111937590711174,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003864110209380942,
      "loss": 4.9681,
      "mean_token_accuracy": 0.2116099402308464,
      "num_tokens": 78030700.0,
      "step": 37605
    },
    {
      "entropy": 5.656766748428344,
      "epoch": 3.4116473149492017,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038638290847200997,
      "loss": 4.8578,
      "mean_token_accuracy": 0.23068244606256486,
      "num_tokens": 78040190.0,
      "step": 37610
    },
    {
      "entropy": 5.62103419303894,
      "epoch": 3.412100870827286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003863547937024987,
      "loss": 4.8721,
      "mean_token_accuracy": 0.23022119402885438,
      "num_tokens": 78050847.0,
      "step": 37615
    },
    {
      "entropy": 5.719326305389404,
      "epoch": 3.4125544267053702,
      "grad_norm": 1.34375,
      "learning_rate": 0.000386326676630142,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22499766647815705,
      "num_tokens": 78060584.0,
      "step": 37620
    },
    {
      "entropy": 5.591067981719971,
      "epoch": 3.413007982583454,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003862985572555212,
      "loss": 4.8011,
      "mean_token_accuracy": 0.22327739894390106,
      "num_tokens": 78071189.0,
      "step": 37625
    },
    {
      "entropy": 5.646454572677612,
      "epoch": 3.4134615384615383,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038627043557921785,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22277014404535295,
      "num_tokens": 78082054.0,
      "step": 37630
    },
    {
      "entropy": 5.597133588790894,
      "epoch": 3.4139150943396226,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003862423116018136,
      "loss": 4.8614,
      "mean_token_accuracy": 0.21728608012199402,
      "num_tokens": 78094032.0,
      "step": 37635
    },
    {
      "entropy": 5.642258834838867,
      "epoch": 3.414368650217707,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038621418532389004,
      "loss": 4.896,
      "mean_token_accuracy": 0.2204824224114418,
      "num_tokens": 78104724.0,
      "step": 37640
    },
    {
      "entropy": 5.672085237503052,
      "epoch": 3.414822206095791,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038618605674602867,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21467514485120773,
      "num_tokens": 78115927.0,
      "step": 37645
    },
    {
      "entropy": 5.667862319946289,
      "epoch": 3.4152757619738754,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038615792586881145,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21859412938356398,
      "num_tokens": 78126744.0,
      "step": 37650
    },
    {
      "entropy": 5.593398332595825,
      "epoch": 3.415729317851959,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003861297926928199,
      "loss": 4.8974,
      "mean_token_accuracy": 0.2184928685426712,
      "num_tokens": 78137625.0,
      "step": 37655
    },
    {
      "entropy": 5.5482312679290775,
      "epoch": 3.4161828737300435,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003861016572186359,
      "loss": 4.8356,
      "mean_token_accuracy": 0.22602216750383378,
      "num_tokens": 78147709.0,
      "step": 37660
    },
    {
      "entropy": 5.691802740097046,
      "epoch": 3.4166364296081277,
      "grad_norm": 1.25,
      "learning_rate": 0.00038607351944684135,
      "loss": 5.0243,
      "mean_token_accuracy": 0.21430206298828125,
      "num_tokens": 78158634.0,
      "step": 37665
    },
    {
      "entropy": 5.716847372055054,
      "epoch": 3.417089985486212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003860453793780181,
      "loss": 4.9888,
      "mean_token_accuracy": 0.2103051796555519,
      "num_tokens": 78170279.0,
      "step": 37670
    },
    {
      "entropy": 5.6639501571655275,
      "epoch": 3.417543541364296,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038601723701274813,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2191453233361244,
      "num_tokens": 78180209.0,
      "step": 37675
    },
    {
      "entropy": 5.652136945724488,
      "epoch": 3.41799709724238,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003859890923516134,
      "loss": 4.9572,
      "mean_token_accuracy": 0.2149360254406929,
      "num_tokens": 78189937.0,
      "step": 37680
    },
    {
      "entropy": 5.649466896057129,
      "epoch": 3.4184506531204644,
      "grad_norm": 1.265625,
      "learning_rate": 0.000385960945395196,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2210889995098114,
      "num_tokens": 78199224.0,
      "step": 37685
    },
    {
      "entropy": 5.642710781097412,
      "epoch": 3.4189042089985486,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003859327961440779,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22446128427982331,
      "num_tokens": 78209912.0,
      "step": 37690
    },
    {
      "entropy": 5.630396986007691,
      "epoch": 3.419357764876633,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038590464459884133,
      "loss": 4.9189,
      "mean_token_accuracy": 0.22270567715168,
      "num_tokens": 78220388.0,
      "step": 37695
    },
    {
      "entropy": 5.625854778289795,
      "epoch": 3.419811320754717,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038587649076006844,
      "loss": 5.0233,
      "mean_token_accuracy": 0.21242413073778152,
      "num_tokens": 78231094.0,
      "step": 37700
    },
    {
      "entropy": 5.647361135482788,
      "epoch": 3.420264876632801,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003858483346283415,
      "loss": 4.8719,
      "mean_token_accuracy": 0.21633055359125136,
      "num_tokens": 78241111.0,
      "step": 37705
    },
    {
      "entropy": 5.706854677200317,
      "epoch": 3.4207184325108853,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003858201762042428,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22021221965551377,
      "num_tokens": 78251036.0,
      "step": 37710
    },
    {
      "entropy": 5.633130359649658,
      "epoch": 3.4211719883889695,
      "grad_norm": 1.125,
      "learning_rate": 0.0003857920154883545,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21122273206710815,
      "num_tokens": 78261840.0,
      "step": 37715
    },
    {
      "entropy": 5.622582292556762,
      "epoch": 3.421625544267054,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003857638524812591,
      "loss": 4.8648,
      "mean_token_accuracy": 0.21868070662021638,
      "num_tokens": 78272539.0,
      "step": 37720
    },
    {
      "entropy": 5.677495765686035,
      "epoch": 3.422079100145138,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038573568718353914,
      "loss": 4.8743,
      "mean_token_accuracy": 0.22061379998922348,
      "num_tokens": 78283849.0,
      "step": 37725
    },
    {
      "entropy": 5.671757221221924,
      "epoch": 3.422532656023222,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003857075195957768,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21955160051584244,
      "num_tokens": 78293640.0,
      "step": 37730
    },
    {
      "entropy": 5.611025142669678,
      "epoch": 3.422986211901306,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003856793497185548,
      "loss": 4.8976,
      "mean_token_accuracy": 0.22343799024820327,
      "num_tokens": 78303980.0,
      "step": 37735
    },
    {
      "entropy": 5.62617678642273,
      "epoch": 3.4234397677793904,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003856511775524556,
      "loss": 4.8798,
      "mean_token_accuracy": 0.21981031596660613,
      "num_tokens": 78314233.0,
      "step": 37740
    },
    {
      "entropy": 5.665864038467407,
      "epoch": 3.4238933236574747,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038562300309806193,
      "loss": 4.9021,
      "mean_token_accuracy": 0.2266559988260269,
      "num_tokens": 78324449.0,
      "step": 37745
    },
    {
      "entropy": 5.68809814453125,
      "epoch": 3.424346879535559,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003855948263559563,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2170569509267807,
      "num_tokens": 78334357.0,
      "step": 37750
    },
    {
      "entropy": 5.653690385818481,
      "epoch": 3.424800435413643,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038556664732672147,
      "loss": 4.869,
      "mean_token_accuracy": 0.2198845401406288,
      "num_tokens": 78344221.0,
      "step": 37755
    },
    {
      "entropy": 5.611452341079712,
      "epoch": 3.425253991291727,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038553846601094033,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22702255994081497,
      "num_tokens": 78354097.0,
      "step": 37760
    },
    {
      "entropy": 5.751009273529053,
      "epoch": 3.4257075471698113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003855102824091954,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2186056852340698,
      "num_tokens": 78363797.0,
      "step": 37765
    },
    {
      "entropy": 5.64217963218689,
      "epoch": 3.4261611030478956,
      "grad_norm": 1.25,
      "learning_rate": 0.0003854820965220697,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2167496368288994,
      "num_tokens": 78374084.0,
      "step": 37770
    },
    {
      "entropy": 5.643146705627442,
      "epoch": 3.42661465892598,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038545390835014614,
      "loss": 4.8841,
      "mean_token_accuracy": 0.2159845620393753,
      "num_tokens": 78385346.0,
      "step": 37775
    },
    {
      "entropy": 5.630850076675415,
      "epoch": 3.4270682148040637,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038542571789400754,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22179464995861053,
      "num_tokens": 78395498.0,
      "step": 37780
    },
    {
      "entropy": 5.603688526153564,
      "epoch": 3.427521770682148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038539752515423704,
      "loss": 4.8549,
      "mean_token_accuracy": 0.21432248651981353,
      "num_tokens": 78405792.0,
      "step": 37785
    },
    {
      "entropy": 5.6770118236541744,
      "epoch": 3.427975326560232,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003853693301314175,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2215485543012619,
      "num_tokens": 78415672.0,
      "step": 37790
    },
    {
      "entropy": 5.59925947189331,
      "epoch": 3.4284288824383164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003853411328261321,
      "loss": 4.8308,
      "mean_token_accuracy": 0.22380381524562837,
      "num_tokens": 78426937.0,
      "step": 37795
    },
    {
      "entropy": 5.6452069759368895,
      "epoch": 3.4288824383164007,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000385312933238964,
      "loss": 4.9181,
      "mean_token_accuracy": 0.2225162133574486,
      "num_tokens": 78436817.0,
      "step": 37800
    },
    {
      "entropy": 5.721516561508179,
      "epoch": 3.429335994194485,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003852847313704963,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2177321195602417,
      "num_tokens": 78447348.0,
      "step": 37805
    },
    {
      "entropy": 5.684627103805542,
      "epoch": 3.429789550072569,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003852565272213123,
      "loss": 4.9271,
      "mean_token_accuracy": 0.22226367443799971,
      "num_tokens": 78457631.0,
      "step": 37810
    },
    {
      "entropy": 5.566251230239868,
      "epoch": 3.430243105950653,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003852283207919951,
      "loss": 4.83,
      "mean_token_accuracy": 0.23379361480474473,
      "num_tokens": 78468542.0,
      "step": 37815
    },
    {
      "entropy": 5.68494291305542,
      "epoch": 3.4306966618287373,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003852001120831282,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21776785999536513,
      "num_tokens": 78478580.0,
      "step": 37820
    },
    {
      "entropy": 5.70031852722168,
      "epoch": 3.4311502177068216,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038517190109529497,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21715081185102464,
      "num_tokens": 78488477.0,
      "step": 37825
    },
    {
      "entropy": 5.665575456619263,
      "epoch": 3.4316037735849054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038514368782907864,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22204332798719406,
      "num_tokens": 78498887.0,
      "step": 37830
    },
    {
      "entropy": 5.611326694488525,
      "epoch": 3.4320573294629897,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038511547228506284,
      "loss": 4.8621,
      "mean_token_accuracy": 0.22409311830997466,
      "num_tokens": 78509030.0,
      "step": 37835
    },
    {
      "entropy": 5.543166065216065,
      "epoch": 3.432510885341074,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038508725446383097,
      "loss": 4.7634,
      "mean_token_accuracy": 0.22847527712583543,
      "num_tokens": 78518254.0,
      "step": 37840
    },
    {
      "entropy": 5.6645081520080565,
      "epoch": 3.4329644412191582,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003850590343659667,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22517662942409516,
      "num_tokens": 78527966.0,
      "step": 37845
    },
    {
      "entropy": 5.656971406936646,
      "epoch": 3.4334179970972425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038503081199205353,
      "loss": 4.962,
      "mean_token_accuracy": 0.21889282315969466,
      "num_tokens": 78537932.0,
      "step": 37850
    },
    {
      "entropy": 5.703976011276245,
      "epoch": 3.4338715529753268,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038500258734267505,
      "loss": 4.9925,
      "mean_token_accuracy": 0.21560388505458833,
      "num_tokens": 78549213.0,
      "step": 37855
    },
    {
      "entropy": 5.613211822509766,
      "epoch": 3.4343251088534106,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038497436041841517,
      "loss": 4.8276,
      "mean_token_accuracy": 0.21863780468702315,
      "num_tokens": 78558990.0,
      "step": 37860
    },
    {
      "entropy": 5.653682136535645,
      "epoch": 3.434778664731495,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038494613121985735,
      "loss": 4.9126,
      "mean_token_accuracy": 0.21797598749399186,
      "num_tokens": 78568496.0,
      "step": 37865
    },
    {
      "entropy": 5.6276038646697994,
      "epoch": 3.435232220609579,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003849178997475856,
      "loss": 4.7791,
      "mean_token_accuracy": 0.23200920671224595,
      "num_tokens": 78578386.0,
      "step": 37870
    },
    {
      "entropy": 5.575590467453003,
      "epoch": 3.4356857764876634,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038488966600218365,
      "loss": 4.8315,
      "mean_token_accuracy": 0.22493296712636948,
      "num_tokens": 78588855.0,
      "step": 37875
    },
    {
      "entropy": 5.571306371688843,
      "epoch": 3.4361393323657476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038486142998423535,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2306309998035431,
      "num_tokens": 78599573.0,
      "step": 37880
    },
    {
      "entropy": 5.71565842628479,
      "epoch": 3.4365928882438315,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003848331916943248,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21993970721960068,
      "num_tokens": 78609793.0,
      "step": 37885
    },
    {
      "entropy": 5.5871048927307125,
      "epoch": 3.4370464441219157,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003848049511330358,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22129549980163574,
      "num_tokens": 78619637.0,
      "step": 37890
    },
    {
      "entropy": 5.685829067230225,
      "epoch": 3.4375,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003847767083009525,
      "loss": 4.9228,
      "mean_token_accuracy": 0.22472057044506072,
      "num_tokens": 78629889.0,
      "step": 37895
    },
    {
      "entropy": 5.587181615829468,
      "epoch": 3.4379535558780843,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038474846319865876,
      "loss": 4.8551,
      "mean_token_accuracy": 0.2252215027809143,
      "num_tokens": 78639853.0,
      "step": 37900
    },
    {
      "entropy": 5.6437427520751955,
      "epoch": 3.4384071117561685,
      "grad_norm": 1.25,
      "learning_rate": 0.0003847202158267389,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22473836243152617,
      "num_tokens": 78649732.0,
      "step": 37905
    },
    {
      "entropy": 5.520084428787231,
      "epoch": 3.4388606676342524,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038469196618577694,
      "loss": 4.8011,
      "mean_token_accuracy": 0.22158783823251724,
      "num_tokens": 78659794.0,
      "step": 37910
    },
    {
      "entropy": 5.700807809829712,
      "epoch": 3.4393142235123366,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003846637142763573,
      "loss": 4.9931,
      "mean_token_accuracy": 0.21339887082576753,
      "num_tokens": 78670526.0,
      "step": 37915
    },
    {
      "entropy": 5.652344036102295,
      "epoch": 3.439767779390421,
      "grad_norm": 1.234375,
      "learning_rate": 0.000384635460099064,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22571070194244386,
      "num_tokens": 78680177.0,
      "step": 37920
    },
    {
      "entropy": 5.694871664047241,
      "epoch": 3.440221335268505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038460720365448154,
      "loss": 5.0473,
      "mean_token_accuracy": 0.20870906710624695,
      "num_tokens": 78690958.0,
      "step": 37925
    },
    {
      "entropy": 5.712219142913819,
      "epoch": 3.4406748911465894,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003845789449431941,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2156258851289749,
      "num_tokens": 78700904.0,
      "step": 37930
    },
    {
      "entropy": 5.7467200756073,
      "epoch": 3.4411284470246732,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003845506839657862,
      "loss": 4.9743,
      "mean_token_accuracy": 0.216835817694664,
      "num_tokens": 78712085.0,
      "step": 37935
    },
    {
      "entropy": 5.594562244415283,
      "epoch": 3.4415820029027575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003845224207228422,
      "loss": 4.8619,
      "mean_token_accuracy": 0.2226394772529602,
      "num_tokens": 78722161.0,
      "step": 37940
    },
    {
      "entropy": 5.660646104812622,
      "epoch": 3.4420355587808418,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038449415521494667,
      "loss": 4.8902,
      "mean_token_accuracy": 0.2206542745232582,
      "num_tokens": 78732485.0,
      "step": 37945
    },
    {
      "entropy": 5.646375036239624,
      "epoch": 3.442489114658926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003844658874426841,
      "loss": 4.9088,
      "mean_token_accuracy": 0.21391827762126922,
      "num_tokens": 78744564.0,
      "step": 37950
    },
    {
      "entropy": 5.672148942947388,
      "epoch": 3.4429426705370103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000384437617406639,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22410544902086257,
      "num_tokens": 78755227.0,
      "step": 37955
    },
    {
      "entropy": 5.642614650726318,
      "epoch": 3.4433962264150946,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003844093451073962,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21557254791259767,
      "num_tokens": 78765393.0,
      "step": 37960
    },
    {
      "entropy": 5.648634147644043,
      "epoch": 3.4438497822931784,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038438107054554027,
      "loss": 4.8616,
      "mean_token_accuracy": 0.22499302625656128,
      "num_tokens": 78775587.0,
      "step": 37965
    },
    {
      "entropy": 5.581699228286743,
      "epoch": 3.4443033381712627,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003843527937216558,
      "loss": 4.8086,
      "mean_token_accuracy": 0.23461911380290984,
      "num_tokens": 78785098.0,
      "step": 37970
    },
    {
      "entropy": 5.587944221496582,
      "epoch": 3.444756894049347,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003843245146363278,
      "loss": 4.8467,
      "mean_token_accuracy": 0.2272839516401291,
      "num_tokens": 78795348.0,
      "step": 37975
    },
    {
      "entropy": 5.7159576416015625,
      "epoch": 3.445210449927431,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038429623329014096,
      "loss": 4.9953,
      "mean_token_accuracy": 0.2142575204372406,
      "num_tokens": 78805348.0,
      "step": 37980
    },
    {
      "entropy": 5.5927667140960695,
      "epoch": 3.445664005805515,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038426794968368015,
      "loss": 4.8264,
      "mean_token_accuracy": 0.22185047566890717,
      "num_tokens": 78814765.0,
      "step": 37985
    },
    {
      "entropy": 5.663602828979492,
      "epoch": 3.4461175616835993,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003842396638175303,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22366278171539306,
      "num_tokens": 78824745.0,
      "step": 37990
    },
    {
      "entropy": 5.60273699760437,
      "epoch": 3.4465711175616836,
      "grad_norm": 1.25,
      "learning_rate": 0.0003842113756922763,
      "loss": 4.9467,
      "mean_token_accuracy": 0.22094383835792542,
      "num_tokens": 78835101.0,
      "step": 37995
    },
    {
      "entropy": 5.701382303237915,
      "epoch": 3.447024673439768,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038418308530850333,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21216570138931273,
      "num_tokens": 78845881.0,
      "step": 38000
    },
    {
      "entropy": 5.66708722114563,
      "epoch": 3.447478229317852,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038415479266679624,
      "loss": 4.91,
      "mean_token_accuracy": 0.21548085063695907,
      "num_tokens": 78856536.0,
      "step": 38005
    },
    {
      "entropy": 5.622337532043457,
      "epoch": 3.4479317851959363,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003841264977677403,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22257553935050964,
      "num_tokens": 78866887.0,
      "step": 38010
    },
    {
      "entropy": 5.603978872299194,
      "epoch": 3.44838534107402,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038409820061192045,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22752683758735656,
      "num_tokens": 78877032.0,
      "step": 38015
    },
    {
      "entropy": 5.587260484695435,
      "epoch": 3.4488388969521044,
      "grad_norm": 1.4375,
      "learning_rate": 0.00038406990119992215,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22937188297510147,
      "num_tokens": 78887237.0,
      "step": 38020
    },
    {
      "entropy": 5.536665725708008,
      "epoch": 3.4492924528301887,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038404159953233047,
      "loss": 4.7907,
      "mean_token_accuracy": 0.23267127424478531,
      "num_tokens": 78897436.0,
      "step": 38025
    },
    {
      "entropy": 5.649680042266846,
      "epoch": 3.449746008708273,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003840132956097307,
      "loss": 4.883,
      "mean_token_accuracy": 0.22053813934326172,
      "num_tokens": 78907691.0,
      "step": 38030
    },
    {
      "entropy": 5.6220708847045895,
      "epoch": 3.450199564586357,
      "grad_norm": 1.25,
      "learning_rate": 0.00038398498943270816,
      "loss": 4.9182,
      "mean_token_accuracy": 0.22164131551980973,
      "num_tokens": 78917517.0,
      "step": 38035
    },
    {
      "entropy": 5.714004182815552,
      "epoch": 3.450653120464441,
      "grad_norm": 1.25,
      "learning_rate": 0.0003839566810018483,
      "loss": 4.8762,
      "mean_token_accuracy": 0.21921992003917695,
      "num_tokens": 78927835.0,
      "step": 38040
    },
    {
      "entropy": 5.589601755142212,
      "epoch": 3.4511066763425253,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038392837031773656,
      "loss": 4.8081,
      "mean_token_accuracy": 0.22518548220396042,
      "num_tokens": 78938853.0,
      "step": 38045
    },
    {
      "entropy": 5.601043081283569,
      "epoch": 3.4515602322206096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003839000573809583,
      "loss": 4.8653,
      "mean_token_accuracy": 0.23010605573654175,
      "num_tokens": 78948508.0,
      "step": 38050
    },
    {
      "entropy": 5.674191284179687,
      "epoch": 3.452013788098694,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038387174219209916,
      "loss": 4.8766,
      "mean_token_accuracy": 0.22717321664094925,
      "num_tokens": 78959002.0,
      "step": 38055
    },
    {
      "entropy": 5.660146427154541,
      "epoch": 3.452467343976778,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003838434247517446,
      "loss": 4.8865,
      "mean_token_accuracy": 0.22319020330905914,
      "num_tokens": 78970358.0,
      "step": 38060
    },
    {
      "entropy": 5.666746377944946,
      "epoch": 3.452920899854862,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003838151050604803,
      "loss": 4.8127,
      "mean_token_accuracy": 0.227305606007576,
      "num_tokens": 78979408.0,
      "step": 38065
    },
    {
      "entropy": 5.527356481552124,
      "epoch": 3.453374455732946,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038378678311889184,
      "loss": 4.8247,
      "mean_token_accuracy": 0.22888706773519515,
      "num_tokens": 78990411.0,
      "step": 38070
    },
    {
      "entropy": 5.630104684829712,
      "epoch": 3.4538280116110305,
      "grad_norm": 1.234375,
      "learning_rate": 0.000383758458927565,
      "loss": 4.963,
      "mean_token_accuracy": 0.22190680354833603,
      "num_tokens": 79001734.0,
      "step": 38075
    },
    {
      "entropy": 5.685411882400513,
      "epoch": 3.4542815674891147,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003837301324870856,
      "loss": 4.8173,
      "mean_token_accuracy": 0.22820084542036057,
      "num_tokens": 79011381.0,
      "step": 38080
    },
    {
      "entropy": 5.667036437988282,
      "epoch": 3.454735123367199,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038370180379803927,
      "loss": 4.9524,
      "mean_token_accuracy": 0.2172626495361328,
      "num_tokens": 79021124.0,
      "step": 38085
    },
    {
      "entropy": 5.673640251159668,
      "epoch": 3.455188679245283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038367347286101197,
      "loss": 5.0146,
      "mean_token_accuracy": 0.20204024016857147,
      "num_tokens": 79033105.0,
      "step": 38090
    },
    {
      "entropy": 5.663362169265747,
      "epoch": 3.455642235123367,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003836451396765896,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21427410691976548,
      "num_tokens": 79043458.0,
      "step": 38095
    },
    {
      "entropy": 5.667144441604615,
      "epoch": 3.4560957910014514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038361680424535795,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2220745712518692,
      "num_tokens": 79055186.0,
      "step": 38100
    },
    {
      "entropy": 5.67755913734436,
      "epoch": 3.4565493468795356,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003835884665679032,
      "loss": 4.9217,
      "mean_token_accuracy": 0.2195105254650116,
      "num_tokens": 79065743.0,
      "step": 38105
    },
    {
      "entropy": 5.67501859664917,
      "epoch": 3.45700290275762,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003835601266448112,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21161009669303893,
      "num_tokens": 79077285.0,
      "step": 38110
    },
    {
      "entropy": 5.6570065975189205,
      "epoch": 3.457456458635704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003835317844766682,
      "loss": 4.9018,
      "mean_token_accuracy": 0.22252228409051894,
      "num_tokens": 79087885.0,
      "step": 38115
    },
    {
      "entropy": 5.68416395187378,
      "epoch": 3.457910014513788,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038350344006406016,
      "loss": 4.9344,
      "mean_token_accuracy": 0.22118688821792604,
      "num_tokens": 79097907.0,
      "step": 38120
    },
    {
      "entropy": 5.712332153320313,
      "epoch": 3.4583635703918723,
      "grad_norm": 1.25,
      "learning_rate": 0.00038347509340757343,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20990787148475648,
      "num_tokens": 79108478.0,
      "step": 38125
    },
    {
      "entropy": 5.711361885070801,
      "epoch": 3.4588171262699565,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003834467445077941,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21531004458665848,
      "num_tokens": 79117988.0,
      "step": 38130
    },
    {
      "entropy": 5.720006656646729,
      "epoch": 3.459270682148041,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003834183933653084,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22780102044343947,
      "num_tokens": 79127786.0,
      "step": 38135
    },
    {
      "entropy": 5.6404743671417235,
      "epoch": 3.4597242380261246,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003833900399807027,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22621534764766693,
      "num_tokens": 79137775.0,
      "step": 38140
    },
    {
      "entropy": 5.547027111053467,
      "epoch": 3.460177793904209,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00038336168435456345,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2237204611301422,
      "num_tokens": 79147803.0,
      "step": 38145
    },
    {
      "entropy": 5.5697732925415036,
      "epoch": 3.460631349782293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038333332648747684,
      "loss": 4.8049,
      "mean_token_accuracy": 0.22736114859580994,
      "num_tokens": 79158377.0,
      "step": 38150
    },
    {
      "entropy": 5.634149217605591,
      "epoch": 3.4610849056603774,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003833049663800295,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21759562343358993,
      "num_tokens": 79168514.0,
      "step": 38155
    },
    {
      "entropy": 5.695762872695923,
      "epoch": 3.4615384615384617,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003832766040328078,
      "loss": 5.0109,
      "mean_token_accuracy": 0.21513009667396546,
      "num_tokens": 79177588.0,
      "step": 38160
    },
    {
      "entropy": 5.643855905532837,
      "epoch": 3.461992017416546,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003832482394463984,
      "loss": 4.871,
      "mean_token_accuracy": 0.22111586779356002,
      "num_tokens": 79188128.0,
      "step": 38165
    },
    {
      "entropy": 5.580990886688232,
      "epoch": 3.4624455732946298,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003832198726213878,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22148845195770264,
      "num_tokens": 79198928.0,
      "step": 38170
    },
    {
      "entropy": 5.557035636901856,
      "epoch": 3.462899129172714,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038319150355836266,
      "loss": 4.812,
      "mean_token_accuracy": 0.23224353790283203,
      "num_tokens": 79209751.0,
      "step": 38175
    },
    {
      "entropy": 5.627704668045044,
      "epoch": 3.4633526850507983,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038316313225790965,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2158006548881531,
      "num_tokens": 79220210.0,
      "step": 38180
    },
    {
      "entropy": 5.638386583328247,
      "epoch": 3.4638062409288826,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003831347587206155,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2244991734623909,
      "num_tokens": 79231530.0,
      "step": 38185
    },
    {
      "entropy": 5.596593141555786,
      "epoch": 3.4642597968069664,
      "grad_norm": 1.203125,
      "learning_rate": 0.000383106382947067,
      "loss": 4.8621,
      "mean_token_accuracy": 0.22113476693630219,
      "num_tokens": 79242771.0,
      "step": 38190
    },
    {
      "entropy": 5.664237976074219,
      "epoch": 3.4647133526850507,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038307800493785093,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22361240535974503,
      "num_tokens": 79253713.0,
      "step": 38195
    },
    {
      "entropy": 5.638049554824829,
      "epoch": 3.465166908563135,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038304962469355415,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21809325367212296,
      "num_tokens": 79264087.0,
      "step": 38200
    },
    {
      "entropy": 5.6401184558868405,
      "epoch": 3.465620464441219,
      "grad_norm": 1.25,
      "learning_rate": 0.0003830212422147637,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2231224626302719,
      "num_tokens": 79274859.0,
      "step": 38205
    },
    {
      "entropy": 5.6572778701782225,
      "epoch": 3.4660740203193035,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003829928575020663,
      "loss": 4.92,
      "mean_token_accuracy": 0.22196309119462967,
      "num_tokens": 79284778.0,
      "step": 38210
    },
    {
      "entropy": 5.679198789596557,
      "epoch": 3.4665275761973877,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038296447055604916,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2135196566581726,
      "num_tokens": 79295034.0,
      "step": 38215
    },
    {
      "entropy": 5.58616452217102,
      "epoch": 3.4669811320754715,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003829360813772992,
      "loss": 4.8246,
      "mean_token_accuracy": 0.23288164883852006,
      "num_tokens": 79306013.0,
      "step": 38220
    },
    {
      "entropy": 5.626765632629395,
      "epoch": 3.467434687953556,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038290768996640366,
      "loss": 4.985,
      "mean_token_accuracy": 0.21295868456363679,
      "num_tokens": 79317590.0,
      "step": 38225
    },
    {
      "entropy": 5.69479808807373,
      "epoch": 3.46788824383164,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038287929632394957,
      "loss": 4.9308,
      "mean_token_accuracy": 0.218781079351902,
      "num_tokens": 79328669.0,
      "step": 38230
    },
    {
      "entropy": 5.561502122879029,
      "epoch": 3.4683417997097243,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003828509004505241,
      "loss": 4.8252,
      "mean_token_accuracy": 0.23009409755468369,
      "num_tokens": 79339336.0,
      "step": 38235
    },
    {
      "entropy": 5.716666030883789,
      "epoch": 3.4687953555878086,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003828225023467146,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21630705446004866,
      "num_tokens": 79350199.0,
      "step": 38240
    },
    {
      "entropy": 5.6421808242797855,
      "epoch": 3.4692489114658924,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003827941020131082,
      "loss": 4.873,
      "mean_token_accuracy": 0.23125556707382203,
      "num_tokens": 79359906.0,
      "step": 38245
    },
    {
      "entropy": 5.59387354850769,
      "epoch": 3.4697024673439767,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038276569945029225,
      "loss": 4.8239,
      "mean_token_accuracy": 0.21477842926979065,
      "num_tokens": 79371395.0,
      "step": 38250
    },
    {
      "entropy": 5.629295349121094,
      "epoch": 3.470156023222061,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003827372946588543,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22268614470958709,
      "num_tokens": 79381808.0,
      "step": 38255
    },
    {
      "entropy": 5.596413707733154,
      "epoch": 3.4706095791001452,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003827088876393816,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22246748208999634,
      "num_tokens": 79391906.0,
      "step": 38260
    },
    {
      "entropy": 5.610882520675659,
      "epoch": 3.4710631349782295,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038268047839246165,
      "loss": 4.8316,
      "mean_token_accuracy": 0.2280489668250084,
      "num_tokens": 79402522.0,
      "step": 38265
    },
    {
      "entropy": 5.5595237731933596,
      "epoch": 3.4715166908563133,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038265206691868195,
      "loss": 4.805,
      "mean_token_accuracy": 0.22312385439872742,
      "num_tokens": 79412751.0,
      "step": 38270
    },
    {
      "entropy": 5.716089820861816,
      "epoch": 3.4719702467343976,
      "grad_norm": 1.25,
      "learning_rate": 0.0003826236532186301,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21648291796445845,
      "num_tokens": 79424512.0,
      "step": 38275
    },
    {
      "entropy": 5.779439544677734,
      "epoch": 3.472423802612482,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003825952372928936,
      "loss": 5.0593,
      "mean_token_accuracy": 0.20616847276687622,
      "num_tokens": 79434082.0,
      "step": 38280
    },
    {
      "entropy": 5.676422500610352,
      "epoch": 3.472877358490566,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038256681914206026,
      "loss": 4.8864,
      "mean_token_accuracy": 0.22302726060152053,
      "num_tokens": 79444469.0,
      "step": 38285
    },
    {
      "entropy": 5.568297004699707,
      "epoch": 3.4733309143686504,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038253839876671765,
      "loss": 4.8829,
      "mean_token_accuracy": 0.21630964428186417,
      "num_tokens": 79455036.0,
      "step": 38290
    },
    {
      "entropy": 5.640959596633911,
      "epoch": 3.473784470246734,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003825099761674535,
      "loss": 4.874,
      "mean_token_accuracy": 0.21984777450561524,
      "num_tokens": 79464600.0,
      "step": 38295
    },
    {
      "entropy": 5.655915832519531,
      "epoch": 3.4742380261248185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003824815513448558,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22141298651695251,
      "num_tokens": 79474912.0,
      "step": 38300
    },
    {
      "entropy": 5.5332403659820555,
      "epoch": 3.4746915820029027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038245312429951214,
      "loss": 4.7691,
      "mean_token_accuracy": 0.22960416227579117,
      "num_tokens": 79485283.0,
      "step": 38305
    },
    {
      "entropy": 5.501477861404419,
      "epoch": 3.475145137880987,
      "grad_norm": 1.25,
      "learning_rate": 0.0003824246950320105,
      "loss": 4.7585,
      "mean_token_accuracy": 0.22108541131019593,
      "num_tokens": 79495323.0,
      "step": 38310
    },
    {
      "entropy": 5.611756324768066,
      "epoch": 3.4755986937590713,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003823962635429388,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22048257440328597,
      "num_tokens": 79505144.0,
      "step": 38315
    },
    {
      "entropy": 5.695401239395141,
      "epoch": 3.4760522496371555,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003823678298328849,
      "loss": 4.9467,
      "mean_token_accuracy": 0.2139294594526291,
      "num_tokens": 79515402.0,
      "step": 38320
    },
    {
      "entropy": 5.639161014556885,
      "epoch": 3.4765058055152394,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038233939390243697,
      "loss": 4.8169,
      "mean_token_accuracy": 0.23429738283157348,
      "num_tokens": 79525528.0,
      "step": 38325
    },
    {
      "entropy": 5.6395810604095455,
      "epoch": 3.4769593613933236,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038231095575218303,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2202695742249489,
      "num_tokens": 79535991.0,
      "step": 38330
    },
    {
      "entropy": 5.69022421836853,
      "epoch": 3.477412917271408,
      "grad_norm": 1.25,
      "learning_rate": 0.0003822825153827112,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2145649865269661,
      "num_tokens": 79545890.0,
      "step": 38335
    },
    {
      "entropy": 5.689422178268432,
      "epoch": 3.477866473149492,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038225407279460964,
      "loss": 4.9418,
      "mean_token_accuracy": 0.22076304703950883,
      "num_tokens": 79556363.0,
      "step": 38340
    },
    {
      "entropy": 5.641707849502564,
      "epoch": 3.478320029027576,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003822256279884665,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21994021832942962,
      "num_tokens": 79566924.0,
      "step": 38345
    },
    {
      "entropy": 5.601212692260742,
      "epoch": 3.4787735849056602,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038219718096487007,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22500191777944564,
      "num_tokens": 79577322.0,
      "step": 38350
    },
    {
      "entropy": 5.741864824295044,
      "epoch": 3.4792271407837445,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038216873172440855,
      "loss": 4.9917,
      "mean_token_accuracy": 0.21258788257837297,
      "num_tokens": 79588436.0,
      "step": 38355
    },
    {
      "entropy": 5.727066087722778,
      "epoch": 3.4796806966618288,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003821402802676704,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21798423528671265,
      "num_tokens": 79599015.0,
      "step": 38360
    },
    {
      "entropy": 5.618892860412598,
      "epoch": 3.480134252539913,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003821118265952439,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22861704528331755,
      "num_tokens": 79608443.0,
      "step": 38365
    },
    {
      "entropy": 5.582566738128662,
      "epoch": 3.4805878084179973,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003820833707077176,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22075275033712388,
      "num_tokens": 79619010.0,
      "step": 38370
    },
    {
      "entropy": 5.678500986099243,
      "epoch": 3.481041364296081,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003820549126056799,
      "loss": 4.9112,
      "mean_token_accuracy": 0.21913174241781236,
      "num_tokens": 79629665.0,
      "step": 38375
    },
    {
      "entropy": 5.55227255821228,
      "epoch": 3.4814949201741654,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003820264522897193,
      "loss": 4.7838,
      "mean_token_accuracy": 0.2277619481086731,
      "num_tokens": 79640578.0,
      "step": 38380
    },
    {
      "entropy": 5.661253356933594,
      "epoch": 3.4819484760522497,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003819979897604243,
      "loss": 4.8551,
      "mean_token_accuracy": 0.21649368107318878,
      "num_tokens": 79649576.0,
      "step": 38385
    },
    {
      "entropy": 5.646604108810425,
      "epoch": 3.482402031930334,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038196952501838376,
      "loss": 4.9126,
      "mean_token_accuracy": 0.21030827462673188,
      "num_tokens": 79659926.0,
      "step": 38390
    },
    {
      "entropy": 5.710169744491577,
      "epoch": 3.4828555878084178,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038194105806418617,
      "loss": 5.0227,
      "mean_token_accuracy": 0.20706610530614852,
      "num_tokens": 79669969.0,
      "step": 38395
    },
    {
      "entropy": 5.640405654907227,
      "epoch": 3.483309143686502,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003819125888984202,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2249169424176216,
      "num_tokens": 79679706.0,
      "step": 38400
    },
    {
      "entropy": 5.59892201423645,
      "epoch": 3.4837626995645863,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038188411752167466,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22400908172130585,
      "num_tokens": 79691543.0,
      "step": 38405
    },
    {
      "entropy": 5.636819362640381,
      "epoch": 3.4842162554426706,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003818556439345383,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2234833225607872,
      "num_tokens": 79701657.0,
      "step": 38410
    },
    {
      "entropy": 5.609713792800903,
      "epoch": 3.484669811320755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038182716813760004,
      "loss": 4.9105,
      "mean_token_accuracy": 0.22530845254659654,
      "num_tokens": 79711615.0,
      "step": 38415
    },
    {
      "entropy": 5.646623134613037,
      "epoch": 3.485123367198839,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003817986901314487,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22565362751483917,
      "num_tokens": 79721427.0,
      "step": 38420
    },
    {
      "entropy": 5.665906095504761,
      "epoch": 3.485576923076923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003817702099166732,
      "loss": 4.9213,
      "mean_token_accuracy": 0.22388988435268403,
      "num_tokens": 79732565.0,
      "step": 38425
    },
    {
      "entropy": 5.579575443267823,
      "epoch": 3.486030478955007,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003817417274938626,
      "loss": 4.7562,
      "mean_token_accuracy": 0.22740735560655595,
      "num_tokens": 79744496.0,
      "step": 38430
    },
    {
      "entropy": 5.6460710048675535,
      "epoch": 3.4864840348330914,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003817132428636058,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21806008368730545,
      "num_tokens": 79755626.0,
      "step": 38435
    },
    {
      "entropy": 5.616807222366333,
      "epoch": 3.4869375907111757,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038168475602649203,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21947404146194457,
      "num_tokens": 79765447.0,
      "step": 38440
    },
    {
      "entropy": 5.653316926956177,
      "epoch": 3.48739114658926,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003816562669831103,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21821122616529465,
      "num_tokens": 79776047.0,
      "step": 38445
    },
    {
      "entropy": 5.6446298122406,
      "epoch": 3.487844702467344,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038162777573404975,
      "loss": 4.8493,
      "mean_token_accuracy": 0.22709202617406846,
      "num_tokens": 79786061.0,
      "step": 38450
    },
    {
      "entropy": 5.602183818817139,
      "epoch": 3.488298258345428,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038159928227989964,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21845054179430007,
      "num_tokens": 79796844.0,
      "step": 38455
    },
    {
      "entropy": 5.664546823501587,
      "epoch": 3.4887518142235123,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003815707866212493,
      "loss": 4.8739,
      "mean_token_accuracy": 0.22813622355461122,
      "num_tokens": 79808120.0,
      "step": 38460
    },
    {
      "entropy": 5.685071992874145,
      "epoch": 3.4892053701015966,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003815422887586878,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22445545196533204,
      "num_tokens": 79818901.0,
      "step": 38465
    },
    {
      "entropy": 5.683806324005127,
      "epoch": 3.489658925979681,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003815137886928047,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2158046245574951,
      "num_tokens": 79829090.0,
      "step": 38470
    },
    {
      "entropy": 5.72820200920105,
      "epoch": 3.490112481857765,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003814852864241892,
      "loss": 4.9456,
      "mean_token_accuracy": 0.21561352610588075,
      "num_tokens": 79838751.0,
      "step": 38475
    },
    {
      "entropy": 5.61675877571106,
      "epoch": 3.490566037735849,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038145678195343095,
      "loss": 4.9922,
      "mean_token_accuracy": 0.21416785269975663,
      "num_tokens": 79849410.0,
      "step": 38480
    },
    {
      "entropy": 5.731597661972046,
      "epoch": 3.491019593613933,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003814282752811193,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21647255271673202,
      "num_tokens": 79859279.0,
      "step": 38485
    },
    {
      "entropy": 5.722654104232788,
      "epoch": 3.4914731494920175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003813997664078438,
      "loss": 4.852,
      "mean_token_accuracy": 0.2274712398648262,
      "num_tokens": 79869535.0,
      "step": 38490
    },
    {
      "entropy": 5.68737759590149,
      "epoch": 3.4919267053701017,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038137125533419403,
      "loss": 5.0326,
      "mean_token_accuracy": 0.2029233917593956,
      "num_tokens": 79881396.0,
      "step": 38495
    },
    {
      "entropy": 5.5814307689666744,
      "epoch": 3.4923802612481856,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003813427420607596,
      "loss": 4.8459,
      "mean_token_accuracy": 0.21493924409151077,
      "num_tokens": 79891250.0,
      "step": 38500
    },
    {
      "entropy": 5.601520586013794,
      "epoch": 3.49283381712627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038131422658813004,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22096751630306244,
      "num_tokens": 79901982.0,
      "step": 38505
    },
    {
      "entropy": 5.641486120223999,
      "epoch": 3.493287373004354,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038128570891689527,
      "loss": 4.9594,
      "mean_token_accuracy": 0.214323490858078,
      "num_tokens": 79912480.0,
      "step": 38510
    },
    {
      "entropy": 5.674976253509522,
      "epoch": 3.4937409288824384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000381257189047645,
      "loss": 4.9106,
      "mean_token_accuracy": 0.22532671242952346,
      "num_tokens": 79923793.0,
      "step": 38515
    },
    {
      "entropy": 5.661871242523193,
      "epoch": 3.4941944847605226,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038122866698096894,
      "loss": 4.9055,
      "mean_token_accuracy": 0.21955492198467255,
      "num_tokens": 79933858.0,
      "step": 38520
    },
    {
      "entropy": 5.698384428024292,
      "epoch": 3.494648040638607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038120014271745693,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21500428915023803,
      "num_tokens": 79944093.0,
      "step": 38525
    },
    {
      "entropy": 5.662602519989013,
      "epoch": 3.4951015965166907,
      "grad_norm": 1.28125,
      "learning_rate": 0.000381171616257699,
      "loss": 4.9229,
      "mean_token_accuracy": 0.21361628472805022,
      "num_tokens": 79955072.0,
      "step": 38530
    },
    {
      "entropy": 5.669543743133545,
      "epoch": 3.495555152394775,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038114308760228493,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22173127233982087,
      "num_tokens": 79966949.0,
      "step": 38535
    },
    {
      "entropy": 5.601763439178467,
      "epoch": 3.4960087082728593,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003811145567518048,
      "loss": 4.857,
      "mean_token_accuracy": 0.21989901810884477,
      "num_tokens": 79977109.0,
      "step": 38540
    },
    {
      "entropy": 5.684397029876709,
      "epoch": 3.4964622641509435,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038108602370684854,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21701878756284715,
      "num_tokens": 79987094.0,
      "step": 38545
    },
    {
      "entropy": 5.673355150222778,
      "epoch": 3.4969158200290273,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003810574884680063,
      "loss": 4.9932,
      "mean_token_accuracy": 0.22414342314004898,
      "num_tokens": 79997503.0,
      "step": 38550
    },
    {
      "entropy": 5.63979287147522,
      "epoch": 3.4973693759071116,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038102895103586817,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2210952639579773,
      "num_tokens": 80007745.0,
      "step": 38555
    },
    {
      "entropy": 5.678734254837036,
      "epoch": 3.497822931785196,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003810004114110243,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21331811547279358,
      "num_tokens": 80019330.0,
      "step": 38560
    },
    {
      "entropy": 5.68154501914978,
      "epoch": 3.49827648766328,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003809718695940649,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21800312548875808,
      "num_tokens": 80029732.0,
      "step": 38565
    },
    {
      "entropy": 5.632097339630127,
      "epoch": 3.4987300435413644,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003809433255855803,
      "loss": 4.8033,
      "mean_token_accuracy": 0.22539930045604706,
      "num_tokens": 80041052.0,
      "step": 38570
    },
    {
      "entropy": 5.662900590896607,
      "epoch": 3.4991835994194487,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038091477938616075,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2210698828101158,
      "num_tokens": 80051929.0,
      "step": 38575
    },
    {
      "entropy": 5.693867301940918,
      "epoch": 3.4996371552975325,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038088623099639653,
      "loss": 4.9791,
      "mean_token_accuracy": 0.21416809111833573,
      "num_tokens": 80061153.0,
      "step": 38580
    },
    {
      "entropy": 5.686135196685791,
      "epoch": 3.5000907111756168,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000380857680416878,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22171344608068466,
      "num_tokens": 80070560.0,
      "step": 38585
    },
    {
      "entropy": 5.8162181854248045,
      "epoch": 3.500544267053701,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003808291276481957,
      "loss": 4.9502,
      "mean_token_accuracy": 0.2174808606505394,
      "num_tokens": 80081141.0,
      "step": 38590
    },
    {
      "entropy": 5.607828569412232,
      "epoch": 3.5009978229317853,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003808005726909401,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23037008792161942,
      "num_tokens": 80092114.0,
      "step": 38595
    },
    {
      "entropy": 5.577740430831909,
      "epoch": 3.501451378809869,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038077201554570175,
      "loss": 4.837,
      "mean_token_accuracy": 0.22557417750358583,
      "num_tokens": 80101966.0,
      "step": 38600
    },
    {
      "entropy": 5.690110683441162,
      "epoch": 3.5019049346879534,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003807434562130711,
      "loss": 5.0054,
      "mean_token_accuracy": 0.22089681923389434,
      "num_tokens": 80113420.0,
      "step": 38605
    },
    {
      "entropy": 5.634019756317139,
      "epoch": 3.5023584905660377,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003807148946936387,
      "loss": 4.9145,
      "mean_token_accuracy": 0.21413621455430984,
      "num_tokens": 80123398.0,
      "step": 38610
    },
    {
      "entropy": 5.657048320770263,
      "epoch": 3.502812046444122,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003806863309879956,
      "loss": 4.861,
      "mean_token_accuracy": 0.23065851628780365,
      "num_tokens": 80133885.0,
      "step": 38615
    },
    {
      "entropy": 5.625788354873658,
      "epoch": 3.503265602322206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003806577650967321,
      "loss": 4.8103,
      "mean_token_accuracy": 0.23959063440561296,
      "num_tokens": 80144168.0,
      "step": 38620
    },
    {
      "entropy": 5.592288446426392,
      "epoch": 3.5037191582002905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003806291970204391,
      "loss": 4.8829,
      "mean_token_accuracy": 0.21960345953702926,
      "num_tokens": 80153956.0,
      "step": 38625
    },
    {
      "entropy": 5.677044010162353,
      "epoch": 3.5041727140783747,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038060062675970743,
      "loss": 4.9372,
      "mean_token_accuracy": 0.20915284007787704,
      "num_tokens": 80163617.0,
      "step": 38630
    },
    {
      "entropy": 5.6456911087036135,
      "epoch": 3.5046262699564585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003805720543151278,
      "loss": 4.8231,
      "mean_token_accuracy": 0.22625830322504042,
      "num_tokens": 80174049.0,
      "step": 38635
    },
    {
      "entropy": 5.7088581085205075,
      "epoch": 3.505079825834543,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003805434796872912,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22086495161056519,
      "num_tokens": 80185043.0,
      "step": 38640
    },
    {
      "entropy": 5.60680456161499,
      "epoch": 3.505533381712627,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038051490287678856,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22150878459215165,
      "num_tokens": 80194810.0,
      "step": 38645
    },
    {
      "entropy": 5.592621088027954,
      "epoch": 3.5059869375907113,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003804863238842109,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2285303995013237,
      "num_tokens": 80204861.0,
      "step": 38650
    },
    {
      "entropy": 5.63749737739563,
      "epoch": 3.506440493468795,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003804577427101491,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23063274323940278,
      "num_tokens": 80214447.0,
      "step": 38655
    },
    {
      "entropy": 5.617111396789551,
      "epoch": 3.5068940493468794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003804291593551944,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22542436122894288,
      "num_tokens": 80224213.0,
      "step": 38660
    },
    {
      "entropy": 5.641775226593017,
      "epoch": 3.5073476052249637,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003804005738199377,
      "loss": 4.9112,
      "mean_token_accuracy": 0.21927156448364257,
      "num_tokens": 80234161.0,
      "step": 38665
    },
    {
      "entropy": 5.638116979598999,
      "epoch": 3.507801161103048,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038037198610497037,
      "loss": 4.8778,
      "mean_token_accuracy": 0.2222333073616028,
      "num_tokens": 80243788.0,
      "step": 38670
    },
    {
      "entropy": 5.644446611404419,
      "epoch": 3.5082547169811322,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003803433962108835,
      "loss": 4.8976,
      "mean_token_accuracy": 0.21956930160522461,
      "num_tokens": 80254658.0,
      "step": 38675
    },
    {
      "entropy": 5.686088132858276,
      "epoch": 3.5087082728592165,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003803148041382683,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22225405722856523,
      "num_tokens": 80264820.0,
      "step": 38680
    },
    {
      "entropy": 5.57348403930664,
      "epoch": 3.5091618287373003,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038028620988771617,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23009472638368605,
      "num_tokens": 80274639.0,
      "step": 38685
    },
    {
      "entropy": 5.588067197799683,
      "epoch": 3.5096153846153846,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003802576134598183,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22792890667915344,
      "num_tokens": 80284542.0,
      "step": 38690
    },
    {
      "entropy": 5.5946413516998295,
      "epoch": 3.510068940493469,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00038022901485516625,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2267189532518387,
      "num_tokens": 80294405.0,
      "step": 38695
    },
    {
      "entropy": 5.627056837081909,
      "epoch": 3.510522496371553,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038020041407435133,
      "loss": 4.8573,
      "mean_token_accuracy": 0.21651632934808732,
      "num_tokens": 80305326.0,
      "step": 38700
    },
    {
      "entropy": 5.610388708114624,
      "epoch": 3.510976052249637,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038017181111796506,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2218819737434387,
      "num_tokens": 80315423.0,
      "step": 38705
    },
    {
      "entropy": 5.629005765914917,
      "epoch": 3.511429608127721,
      "grad_norm": 1.25,
      "learning_rate": 0.0003801432059865989,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21443607211112975,
      "num_tokens": 80325372.0,
      "step": 38710
    },
    {
      "entropy": 5.668954610824585,
      "epoch": 3.5118831640058055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038011459868084446,
      "loss": 4.7988,
      "mean_token_accuracy": 0.22460764348506929,
      "num_tokens": 80335467.0,
      "step": 38715
    },
    {
      "entropy": 5.629906511306762,
      "epoch": 3.5123367198838897,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003800859892012933,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22232987880706787,
      "num_tokens": 80345420.0,
      "step": 38720
    },
    {
      "entropy": 5.599252700805664,
      "epoch": 3.512790275761974,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003800573775485372,
      "loss": 4.8779,
      "mean_token_accuracy": 0.23185952752828598,
      "num_tokens": 80354369.0,
      "step": 38725
    },
    {
      "entropy": 5.680689811706543,
      "epoch": 3.5132438316400583,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038002876372316774,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2172993987798691,
      "num_tokens": 80365292.0,
      "step": 38730
    },
    {
      "entropy": 5.709176874160766,
      "epoch": 3.513697387518142,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038000014772577657,
      "loss": 4.9048,
      "mean_token_accuracy": 0.22354436218738555,
      "num_tokens": 80376340.0,
      "step": 38735
    },
    {
      "entropy": 5.565360164642334,
      "epoch": 3.5141509433962264,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003799715295569557,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2194734290242195,
      "num_tokens": 80386193.0,
      "step": 38740
    },
    {
      "entropy": 5.725466680526734,
      "epoch": 3.5146044992743106,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003799429092172968,
      "loss": 5.0602,
      "mean_token_accuracy": 0.2148408219218254,
      "num_tokens": 80397055.0,
      "step": 38745
    },
    {
      "entropy": 5.669391775131226,
      "epoch": 3.515058055152395,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003799142867073918,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2132392093539238,
      "num_tokens": 80407375.0,
      "step": 38750
    },
    {
      "entropy": 5.705404949188233,
      "epoch": 3.5155116110304787,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037988566202783256,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21610867232084274,
      "num_tokens": 80417942.0,
      "step": 38755
    },
    {
      "entropy": 5.573472309112549,
      "epoch": 3.515965166908563,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003798570351792112,
      "loss": 4.7923,
      "mean_token_accuracy": 0.22393308877944945,
      "num_tokens": 80428716.0,
      "step": 38760
    },
    {
      "entropy": 5.609170150756836,
      "epoch": 3.5164187227866472,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037982840616211964,
      "loss": 4.9538,
      "mean_token_accuracy": 0.21630744487047196,
      "num_tokens": 80439688.0,
      "step": 38765
    },
    {
      "entropy": 5.702987575531006,
      "epoch": 3.5168722786647315,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037979977497714984,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21656095534563063,
      "num_tokens": 80450434.0,
      "step": 38770
    },
    {
      "entropy": 5.6568663120269775,
      "epoch": 3.5173258345428158,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037977114162489404,
      "loss": 4.9633,
      "mean_token_accuracy": 0.215074360370636,
      "num_tokens": 80462187.0,
      "step": 38775
    },
    {
      "entropy": 5.640180349349976,
      "epoch": 3.5177793904209,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003797425061059443,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22144632190465927,
      "num_tokens": 80473049.0,
      "step": 38780
    },
    {
      "entropy": 5.601675081253052,
      "epoch": 3.5182329462989843,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037971386842089287,
      "loss": 4.8609,
      "mean_token_accuracy": 0.2269955411553383,
      "num_tokens": 80484119.0,
      "step": 38785
    },
    {
      "entropy": 5.605462980270386,
      "epoch": 3.518686502177068,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003796852285703319,
      "loss": 4.7845,
      "mean_token_accuracy": 0.23113539516925813,
      "num_tokens": 80494882.0,
      "step": 38790
    },
    {
      "entropy": 5.674931955337525,
      "epoch": 3.5191400580551524,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003796565865548538,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22131285667419434,
      "num_tokens": 80505362.0,
      "step": 38795
    },
    {
      "entropy": 5.6500050067901615,
      "epoch": 3.5195936139332367,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003796279423750508,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21475103944540025,
      "num_tokens": 80516027.0,
      "step": 38800
    },
    {
      "entropy": 5.69084734916687,
      "epoch": 3.5200471698113205,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003795992960315153,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21867989450693132,
      "num_tokens": 80527045.0,
      "step": 38805
    },
    {
      "entropy": 5.63895173072815,
      "epoch": 3.5205007256894048,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037957064752483966,
      "loss": 4.8184,
      "mean_token_accuracy": 0.22899862825870515,
      "num_tokens": 80537373.0,
      "step": 38810
    },
    {
      "entropy": 5.601835632324219,
      "epoch": 3.520954281567489,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003795419968556164,
      "loss": 4.8911,
      "mean_token_accuracy": 0.21184316128492356,
      "num_tokens": 80546632.0,
      "step": 38815
    },
    {
      "entropy": 5.6124248027801515,
      "epoch": 3.5214078374455733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003795133440244381,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22231239527463914,
      "num_tokens": 80556694.0,
      "step": 38820
    },
    {
      "entropy": 5.683526468276978,
      "epoch": 3.5218613933236576,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003794846890318971,
      "loss": 4.9578,
      "mean_token_accuracy": 0.2106945186853409,
      "num_tokens": 80567834.0,
      "step": 38825
    },
    {
      "entropy": 5.659314966201782,
      "epoch": 3.522314949201742,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003794560318785861,
      "loss": 4.8884,
      "mean_token_accuracy": 0.22470719814300538,
      "num_tokens": 80578299.0,
      "step": 38830
    },
    {
      "entropy": 5.6747547626495365,
      "epoch": 3.522768505079826,
      "grad_norm": 1.25,
      "learning_rate": 0.00037942737256509786,
      "loss": 4.9191,
      "mean_token_accuracy": 0.21745397597551347,
      "num_tokens": 80588620.0,
      "step": 38835
    },
    {
      "entropy": 5.798318910598755,
      "epoch": 3.52322206095791,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003793987110920248,
      "loss": 5.0857,
      "mean_token_accuracy": 0.20557896494865419,
      "num_tokens": 80599076.0,
      "step": 38840
    },
    {
      "entropy": 5.731644535064698,
      "epoch": 3.523675616835994,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037937004745995987,
      "loss": 5.0045,
      "mean_token_accuracy": 0.2126277893781662,
      "num_tokens": 80609440.0,
      "step": 38845
    },
    {
      "entropy": 5.76024899482727,
      "epoch": 3.5241291727140784,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003793413816694957,
      "loss": 4.983,
      "mean_token_accuracy": 0.21231929510831832,
      "num_tokens": 80619169.0,
      "step": 38850
    },
    {
      "entropy": 5.6157317638397215,
      "epoch": 3.5245827285921627,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003793127137212252,
      "loss": 4.8537,
      "mean_token_accuracy": 0.224270823597908,
      "num_tokens": 80628719.0,
      "step": 38855
    },
    {
      "entropy": 5.658299541473388,
      "epoch": 3.5250362844702465,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037928404361574125,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21427015513181685,
      "num_tokens": 80638391.0,
      "step": 38860
    },
    {
      "entropy": 5.655683851242065,
      "epoch": 3.525489840348331,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003792553713536367,
      "loss": 4.9472,
      "mean_token_accuracy": 0.21779898703098297,
      "num_tokens": 80647720.0,
      "step": 38865
    },
    {
      "entropy": 5.5854710102081295,
      "epoch": 3.525943396226415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003792266969355044,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22293582260608674,
      "num_tokens": 80658841.0,
      "step": 38870
    },
    {
      "entropy": 5.651483535766602,
      "epoch": 3.5263969521044993,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037919802036193753,
      "loss": 4.9098,
      "mean_token_accuracy": 0.221801295876503,
      "num_tokens": 80668735.0,
      "step": 38875
    },
    {
      "entropy": 5.685948181152344,
      "epoch": 3.5268505079825836,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037916934163352903,
      "loss": 4.8661,
      "mean_token_accuracy": 0.22489866465330124,
      "num_tokens": 80680757.0,
      "step": 38880
    },
    {
      "entropy": 5.62222261428833,
      "epoch": 3.527304063860668,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000379140660750872,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2200845703482628,
      "num_tokens": 80689833.0,
      "step": 38885
    },
    {
      "entropy": 5.570931053161621,
      "epoch": 3.5277576197387517,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003791119777145595,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22644873857498168,
      "num_tokens": 80700347.0,
      "step": 38890
    },
    {
      "entropy": 5.530778455734253,
      "epoch": 3.528211175616836,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003790832925251849,
      "loss": 4.7328,
      "mean_token_accuracy": 0.2385459452867508,
      "num_tokens": 80710559.0,
      "step": 38895
    },
    {
      "entropy": 5.658862543106079,
      "epoch": 3.52866473149492,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003790546051833412,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21817902177572251,
      "num_tokens": 80720805.0,
      "step": 38900
    },
    {
      "entropy": 5.662647724151611,
      "epoch": 3.5291182873730045,
      "grad_norm": 1.25,
      "learning_rate": 0.0003790259156896217,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2043817952275276,
      "num_tokens": 80730937.0,
      "step": 38905
    },
    {
      "entropy": 5.669439125061035,
      "epoch": 3.5295718432510883,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037899722404461983,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22262938022613527,
      "num_tokens": 80741597.0,
      "step": 38910
    },
    {
      "entropy": 5.6773701190948485,
      "epoch": 3.5300253991291726,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003789685302489289,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2141635835170746,
      "num_tokens": 80752496.0,
      "step": 38915
    },
    {
      "entropy": 5.675675678253174,
      "epoch": 3.530478955007257,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037893983430314226,
      "loss": 4.8775,
      "mean_token_accuracy": 0.21867239624261856,
      "num_tokens": 80762329.0,
      "step": 38920
    },
    {
      "entropy": 5.618783092498779,
      "epoch": 3.530932510885341,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037891113620785327,
      "loss": 4.9669,
      "mean_token_accuracy": 0.21754271388053895,
      "num_tokens": 80773152.0,
      "step": 38925
    },
    {
      "entropy": 5.6123795986175535,
      "epoch": 3.5313860667634254,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037888243596365557,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22067504674196242,
      "num_tokens": 80783502.0,
      "step": 38930
    },
    {
      "entropy": 5.67082371711731,
      "epoch": 3.5318396226415096,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037885373357114265,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22149303406476975,
      "num_tokens": 80792644.0,
      "step": 38935
    },
    {
      "entropy": 5.670018529891967,
      "epoch": 3.5322931785195935,
      "grad_norm": 1.203125,
      "learning_rate": 0.000378825029030908,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2167245164513588,
      "num_tokens": 80803184.0,
      "step": 38940
    },
    {
      "entropy": 5.623274326324463,
      "epoch": 3.5327467343976777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003787963223435453,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22082321494817733,
      "num_tokens": 80813353.0,
      "step": 38945
    },
    {
      "entropy": 5.654191780090332,
      "epoch": 3.533200290275762,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003787676135096482,
      "loss": 4.9854,
      "mean_token_accuracy": 0.21453527063131334,
      "num_tokens": 80824663.0,
      "step": 38950
    },
    {
      "entropy": 5.766034555435181,
      "epoch": 3.5336538461538463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003787389025298104,
      "loss": 5.0151,
      "mean_token_accuracy": 0.21007631421089173,
      "num_tokens": 80835507.0,
      "step": 38955
    },
    {
      "entropy": 5.666920852661133,
      "epoch": 3.53410740203193,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003787101894046257,
      "loss": 4.8238,
      "mean_token_accuracy": 0.2341285303235054,
      "num_tokens": 80845035.0,
      "step": 38960
    },
    {
      "entropy": 5.6574183940887455,
      "epoch": 3.5345609579100143,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037868147413468785,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2206186130642891,
      "num_tokens": 80854992.0,
      "step": 38965
    },
    {
      "entropy": 5.663049173355103,
      "epoch": 3.5350145137880986,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003786527567205907,
      "loss": 4.9981,
      "mean_token_accuracy": 0.21088607609272003,
      "num_tokens": 80864658.0,
      "step": 38970
    },
    {
      "entropy": 5.687146949768066,
      "epoch": 3.535468069666183,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037862403716292805,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2201855942606926,
      "num_tokens": 80873964.0,
      "step": 38975
    },
    {
      "entropy": 5.673787450790405,
      "epoch": 3.535921625544267,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037859531546229405,
      "loss": 4.9207,
      "mean_token_accuracy": 0.2163935199379921,
      "num_tokens": 80883752.0,
      "step": 38980
    },
    {
      "entropy": 5.684512710571289,
      "epoch": 3.5363751814223514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003785665916192824,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21969943046569823,
      "num_tokens": 80894265.0,
      "step": 38985
    },
    {
      "entropy": 5.585261106491089,
      "epoch": 3.5368287373004357,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037853786563448724,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2253129854798317,
      "num_tokens": 80904246.0,
      "step": 38990
    },
    {
      "entropy": 5.6701438426971436,
      "epoch": 3.5372822931785195,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003785091375085027,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21751925051212312,
      "num_tokens": 80913595.0,
      "step": 38995
    },
    {
      "entropy": 5.628238582611084,
      "epoch": 3.5377358490566038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003784804072419228,
      "loss": 4.898,
      "mean_token_accuracy": 0.21578398197889329,
      "num_tokens": 80924754.0,
      "step": 39000
    },
    {
      "epoch": 3.5377358490566038,
      "eval_entropy": 5.393534546632033,
      "eval_loss": 4.970121383666992,
      "eval_mean_token_accuracy": 0.2253137246330534,
      "eval_num_tokens": 80924754.0,
      "eval_runtime": 20.6395,
      "eval_samples_per_second": 1713.22,
      "eval_steps_per_second": 214.153,
      "step": 39000
    },
    {
      "entropy": 5.616163158416748,
      "epoch": 3.538189404934688,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003784516748353417,
      "loss": 4.8214,
      "mean_token_accuracy": 0.23074238002300262,
      "num_tokens": 80935318.0,
      "step": 39005
    },
    {
      "entropy": 5.646154832839966,
      "epoch": 3.5386429608127723,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003784229402893536,
      "loss": 4.9478,
      "mean_token_accuracy": 0.22214530706405639,
      "num_tokens": 80944886.0,
      "step": 39010
    },
    {
      "entropy": 5.6769355773925785,
      "epoch": 3.539096516690856,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003783942036045528,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21518024057149887,
      "num_tokens": 80955566.0,
      "step": 39015
    },
    {
      "entropy": 5.695873641967774,
      "epoch": 3.5395500725689404,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037836546478153344,
      "loss": 4.9562,
      "mean_token_accuracy": 0.21601938009262084,
      "num_tokens": 80965285.0,
      "step": 39020
    },
    {
      "entropy": 5.591855764389038,
      "epoch": 3.5400036284470247,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003783367238208899,
      "loss": 4.8684,
      "mean_token_accuracy": 0.2274763986468315,
      "num_tokens": 80976402.0,
      "step": 39025
    },
    {
      "entropy": 5.690197992324829,
      "epoch": 3.540457184325109,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037830798072321665,
      "loss": 4.9723,
      "mean_token_accuracy": 0.2182393953204155,
      "num_tokens": 80988408.0,
      "step": 39030
    },
    {
      "entropy": 5.676499557495117,
      "epoch": 3.540910740203193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037827923548910806,
      "loss": 4.8945,
      "mean_token_accuracy": 0.21863435804843903,
      "num_tokens": 80999099.0,
      "step": 39035
    },
    {
      "entropy": 5.608412170410157,
      "epoch": 3.5413642960812775,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037825048811915854,
      "loss": 4.8043,
      "mean_token_accuracy": 0.22520106583833693,
      "num_tokens": 81009350.0,
      "step": 39040
    },
    {
      "entropy": 5.629689168930054,
      "epoch": 3.5418178519593613,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037822173861396256,
      "loss": 4.8056,
      "mean_token_accuracy": 0.2192219853401184,
      "num_tokens": 81019851.0,
      "step": 39045
    },
    {
      "entropy": 5.646512365341186,
      "epoch": 3.5422714078374455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037819298697411475,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21510945856571198,
      "num_tokens": 81028987.0,
      "step": 39050
    },
    {
      "entropy": 5.6185201644897464,
      "epoch": 3.54272496371553,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037816423320020964,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2241588994860649,
      "num_tokens": 81038081.0,
      "step": 39055
    },
    {
      "entropy": 5.67468957901001,
      "epoch": 3.543178519593614,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037813547729284195,
      "loss": 4.8676,
      "mean_token_accuracy": 0.21889322996139526,
      "num_tokens": 81048066.0,
      "step": 39060
    },
    {
      "entropy": 5.519023180007935,
      "epoch": 3.543632075471698,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037810671925260633,
      "loss": 4.7136,
      "mean_token_accuracy": 0.24082041233778,
      "num_tokens": 81057421.0,
      "step": 39065
    },
    {
      "entropy": 5.582855892181397,
      "epoch": 3.544085631349782,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037807795908009745,
      "loss": 4.8999,
      "mean_token_accuracy": 0.22361056208610536,
      "num_tokens": 81068435.0,
      "step": 39070
    },
    {
      "entropy": 5.671843242645264,
      "epoch": 3.5445391872278664,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037804919677591006,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22070489674806595,
      "num_tokens": 81079378.0,
      "step": 39075
    },
    {
      "entropy": 5.615456247329712,
      "epoch": 3.5449927431059507,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003780204323406391,
      "loss": 4.9062,
      "mean_token_accuracy": 0.21721493005752562,
      "num_tokens": 81089872.0,
      "step": 39080
    },
    {
      "entropy": 5.607994937896729,
      "epoch": 3.545446298984035,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003779916657748793,
      "loss": 4.8182,
      "mean_token_accuracy": 0.2268304780125618,
      "num_tokens": 81099653.0,
      "step": 39085
    },
    {
      "entropy": 5.625764989852906,
      "epoch": 3.5458998548621192,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003779628970792256,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2192773386836052,
      "num_tokens": 81109840.0,
      "step": 39090
    },
    {
      "entropy": 5.593977785110473,
      "epoch": 3.546353410740203,
      "grad_norm": 1.296875,
      "learning_rate": 0.000377934126254273,
      "loss": 4.8158,
      "mean_token_accuracy": 0.23031169921159744,
      "num_tokens": 81120991.0,
      "step": 39095
    },
    {
      "entropy": 5.686555814743042,
      "epoch": 3.5468069666182873,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037790535330061635,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22146696597337723,
      "num_tokens": 81129831.0,
      "step": 39100
    },
    {
      "entropy": 5.645819234848022,
      "epoch": 3.5472605224963716,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003778765782188508,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22617004215717315,
      "num_tokens": 81140604.0,
      "step": 39105
    },
    {
      "entropy": 5.654625272750854,
      "epoch": 3.547714078374456,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037784780100957145,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2243119955062866,
      "num_tokens": 81150591.0,
      "step": 39110
    },
    {
      "entropy": 5.722482585906983,
      "epoch": 3.5481676342525397,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003778190216733733,
      "loss": 4.962,
      "mean_token_accuracy": 0.2143990233540535,
      "num_tokens": 81161444.0,
      "step": 39115
    },
    {
      "entropy": 5.667954444885254,
      "epoch": 3.548621190130624,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037779024021085167,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22865815758705138,
      "num_tokens": 81171216.0,
      "step": 39120
    },
    {
      "entropy": 5.659074115753174,
      "epoch": 3.549074746008708,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003777614566226016,
      "loss": 4.9169,
      "mean_token_accuracy": 0.2257553979754448,
      "num_tokens": 81181665.0,
      "step": 39125
    },
    {
      "entropy": 5.543111944198609,
      "epoch": 3.5495283018867925,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003777326709092184,
      "loss": 4.8479,
      "mean_token_accuracy": 0.22232834100723267,
      "num_tokens": 81191683.0,
      "step": 39130
    },
    {
      "entropy": 5.6455193042755125,
      "epoch": 3.5499818577648767,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037770388307129747,
      "loss": 4.8572,
      "mean_token_accuracy": 0.22266591638326644,
      "num_tokens": 81201211.0,
      "step": 39135
    },
    {
      "entropy": 5.657996082305909,
      "epoch": 3.550435413642961,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037767509310943407,
      "loss": 4.8664,
      "mean_token_accuracy": 0.2244148924946785,
      "num_tokens": 81210668.0,
      "step": 39140
    },
    {
      "entropy": 5.608302307128906,
      "epoch": 3.5508889695210453,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003776463010242235,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22493915557861327,
      "num_tokens": 81221073.0,
      "step": 39145
    },
    {
      "entropy": 5.624004936218261,
      "epoch": 3.551342525399129,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037761750681626134,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21426354199647904,
      "num_tokens": 81231687.0,
      "step": 39150
    },
    {
      "entropy": 5.6993824481964115,
      "epoch": 3.5517960812772134,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037758871048614295,
      "loss": 4.9392,
      "mean_token_accuracy": 0.213796067237854,
      "num_tokens": 81242323.0,
      "step": 39155
    },
    {
      "entropy": 5.620159149169922,
      "epoch": 3.5522496371552976,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003775599120344638,
      "loss": 4.7702,
      "mean_token_accuracy": 0.23432543128728867,
      "num_tokens": 81252543.0,
      "step": 39160
    },
    {
      "entropy": 5.641052341461181,
      "epoch": 3.5527031930333814,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037753111146181966,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2142824798822403,
      "num_tokens": 81262180.0,
      "step": 39165
    },
    {
      "entropy": 5.758733129501342,
      "epoch": 3.5531567489114657,
      "grad_norm": 1.296875,
      "learning_rate": 0.000377502308768806,
      "loss": 5.0995,
      "mean_token_accuracy": 0.20128827691078185,
      "num_tokens": 81272088.0,
      "step": 39170
    },
    {
      "entropy": 5.653423547744751,
      "epoch": 3.55361030478955,
      "grad_norm": 1.25,
      "learning_rate": 0.0003774735039560184,
      "loss": 4.8662,
      "mean_token_accuracy": 0.2229309692978859,
      "num_tokens": 81281723.0,
      "step": 39175
    },
    {
      "entropy": 5.686999320983887,
      "epoch": 3.5540638606676342,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037744469702405273,
      "loss": 4.9482,
      "mean_token_accuracy": 0.22228230834007262,
      "num_tokens": 81292030.0,
      "step": 39180
    },
    {
      "entropy": 5.600500440597534,
      "epoch": 3.5545174165457185,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003774158879735045,
      "loss": 4.8584,
      "mean_token_accuracy": 0.22498147338628768,
      "num_tokens": 81302283.0,
      "step": 39185
    },
    {
      "entropy": 5.672839069366455,
      "epoch": 3.5549709724238028,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003773870768049697,
      "loss": 4.9564,
      "mean_token_accuracy": 0.21361616998910904,
      "num_tokens": 81313058.0,
      "step": 39190
    },
    {
      "entropy": 5.657902336120605,
      "epoch": 3.555424528301887,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000377358263519044,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22400048673152922,
      "num_tokens": 81323827.0,
      "step": 39195
    },
    {
      "entropy": 5.687113046646118,
      "epoch": 3.555878084179971,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003773294481163234,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21485089361667634,
      "num_tokens": 81333106.0,
      "step": 39200
    },
    {
      "entropy": 5.612260866165161,
      "epoch": 3.556331640058055,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003773006305974037,
      "loss": 4.873,
      "mean_token_accuracy": 0.22215779721736909,
      "num_tokens": 81343245.0,
      "step": 39205
    },
    {
      "entropy": 5.617501878738404,
      "epoch": 3.5567851959361394,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037727181096288077,
      "loss": 4.8802,
      "mean_token_accuracy": 0.2294653609395027,
      "num_tokens": 81352917.0,
      "step": 39210
    },
    {
      "entropy": 5.574107694625854,
      "epoch": 3.5572387518142237,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003772429892133509,
      "loss": 4.8353,
      "mean_token_accuracy": 0.21991948038339615,
      "num_tokens": 81362838.0,
      "step": 39215
    },
    {
      "entropy": 5.614162254333496,
      "epoch": 3.5576923076923075,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003772141653494099,
      "loss": 4.8148,
      "mean_token_accuracy": 0.23064998388290406,
      "num_tokens": 81373395.0,
      "step": 39220
    },
    {
      "entropy": 5.610518884658814,
      "epoch": 3.5581458635703918,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003771853393716538,
      "loss": 4.8592,
      "mean_token_accuracy": 0.2208454817533493,
      "num_tokens": 81383976.0,
      "step": 39225
    },
    {
      "entropy": 5.629348039627075,
      "epoch": 3.558599419448476,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000377156511280679,
      "loss": 4.8791,
      "mean_token_accuracy": 0.21517060101032257,
      "num_tokens": 81394328.0,
      "step": 39230
    },
    {
      "entropy": 5.668675231933594,
      "epoch": 3.5590529753265603,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003771276810770815,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2194511726498604,
      "num_tokens": 81405451.0,
      "step": 39235
    },
    {
      "entropy": 5.673462009429931,
      "epoch": 3.5595065312046446,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003770988487614575,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20736407488584518,
      "num_tokens": 81416589.0,
      "step": 39240
    },
    {
      "entropy": 5.561825180053711,
      "epoch": 3.559960087082729,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003770700143344033,
      "loss": 4.8358,
      "mean_token_accuracy": 0.22503251880407332,
      "num_tokens": 81428553.0,
      "step": 39245
    },
    {
      "entropy": 5.771281576156616,
      "epoch": 3.5604136429608126,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037704117779651514,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21511709988117217,
      "num_tokens": 81438720.0,
      "step": 39250
    },
    {
      "entropy": 5.612032556533814,
      "epoch": 3.560867198838897,
      "grad_norm": 1.484375,
      "learning_rate": 0.00037701233914838947,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21700548082590104,
      "num_tokens": 81448142.0,
      "step": 39255
    },
    {
      "entropy": 5.654911708831787,
      "epoch": 3.561320754716981,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003769834983906226,
      "loss": 4.9557,
      "mean_token_accuracy": 0.218122760951519,
      "num_tokens": 81459448.0,
      "step": 39260
    },
    {
      "entropy": 5.653246259689331,
      "epoch": 3.5617743105950654,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037695465552381105,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2210347518324852,
      "num_tokens": 81470356.0,
      "step": 39265
    },
    {
      "entropy": 5.641185569763183,
      "epoch": 3.5622278664731493,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037692581054855125,
      "loss": 4.8519,
      "mean_token_accuracy": 0.2232051119208336,
      "num_tokens": 81480472.0,
      "step": 39270
    },
    {
      "entropy": 5.608519697189331,
      "epoch": 3.5626814223512335,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037689696346543964,
      "loss": 4.836,
      "mean_token_accuracy": 0.22760785669088363,
      "num_tokens": 81489949.0,
      "step": 39275
    },
    {
      "entropy": 5.577717256546021,
      "epoch": 3.563134978229318,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037686811427507284,
      "loss": 4.7935,
      "mean_token_accuracy": 0.22302671521902084,
      "num_tokens": 81500276.0,
      "step": 39280
    },
    {
      "entropy": 5.613933420181274,
      "epoch": 3.563588534107402,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037683926297804756,
      "loss": 4.793,
      "mean_token_accuracy": 0.2225412830710411,
      "num_tokens": 81510431.0,
      "step": 39285
    },
    {
      "entropy": 5.603700447082519,
      "epoch": 3.5640420899854863,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003768104095749603,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21796686202287674,
      "num_tokens": 81520718.0,
      "step": 39290
    },
    {
      "entropy": 5.636259508132935,
      "epoch": 3.5644956458635706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003767815540664079,
      "loss": 4.9678,
      "mean_token_accuracy": 0.2167618215084076,
      "num_tokens": 81531904.0,
      "step": 39295
    },
    {
      "entropy": 5.639812469482422,
      "epoch": 3.5649492017416544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037675269645298697,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21602933704853058,
      "num_tokens": 81542093.0,
      "step": 39300
    },
    {
      "entropy": 5.633946657180786,
      "epoch": 3.5654027576197387,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003767238367352944,
      "loss": 4.823,
      "mean_token_accuracy": 0.22708417177200318,
      "num_tokens": 81552020.0,
      "step": 39305
    },
    {
      "entropy": 5.569300031661987,
      "epoch": 3.565856313497823,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037669497491392683,
      "loss": 4.7981,
      "mean_token_accuracy": 0.22968486398458482,
      "num_tokens": 81561644.0,
      "step": 39310
    },
    {
      "entropy": 5.619177913665771,
      "epoch": 3.566309869375907,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037666611098948144,
      "loss": 4.9191,
      "mean_token_accuracy": 0.22486685812473298,
      "num_tokens": 81572158.0,
      "step": 39315
    },
    {
      "entropy": 5.718697118759155,
      "epoch": 3.566763425253991,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003766372449625549,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21216964423656465,
      "num_tokens": 81582335.0,
      "step": 39320
    },
    {
      "entropy": 5.743814134597779,
      "epoch": 3.5672169811320753,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003766083768337442,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2155804678797722,
      "num_tokens": 81593004.0,
      "step": 39325
    },
    {
      "entropy": 5.66798586845398,
      "epoch": 3.5676705370101596,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003765795066036464,
      "loss": 4.9041,
      "mean_token_accuracy": 0.2164766550064087,
      "num_tokens": 81603656.0,
      "step": 39330
    },
    {
      "entropy": 5.622368669509887,
      "epoch": 3.568124092888244,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037655063427285854,
      "loss": 4.917,
      "mean_token_accuracy": 0.21528741270303725,
      "num_tokens": 81613039.0,
      "step": 39335
    },
    {
      "entropy": 5.5999902248382565,
      "epoch": 3.568577648766328,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037652175984197764,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22598878890275956,
      "num_tokens": 81624410.0,
      "step": 39340
    },
    {
      "entropy": 5.6626465797424315,
      "epoch": 3.5690312046444124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037649288331160086,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22347572296857834,
      "num_tokens": 81635160.0,
      "step": 39345
    },
    {
      "entropy": 5.633468198776245,
      "epoch": 3.5694847605224966,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037646400468232546,
      "loss": 4.8654,
      "mean_token_accuracy": 0.21862493455410004,
      "num_tokens": 81645686.0,
      "step": 39350
    },
    {
      "entropy": 5.613604497909546,
      "epoch": 3.5699383164005805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037643512395474853,
      "loss": 4.8254,
      "mean_token_accuracy": 0.22539474368095397,
      "num_tokens": 81656330.0,
      "step": 39355
    },
    {
      "entropy": 5.603452110290528,
      "epoch": 3.5703918722786647,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003764062411294674,
      "loss": 4.8814,
      "mean_token_accuracy": 0.2166259169578552,
      "num_tokens": 81665895.0,
      "step": 39360
    },
    {
      "entropy": 5.710248613357544,
      "epoch": 3.570845428156749,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037637735620707935,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2193511113524437,
      "num_tokens": 81676616.0,
      "step": 39365
    },
    {
      "entropy": 5.651619720458984,
      "epoch": 3.5712989840348333,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003763484691881818,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2194437175989151,
      "num_tokens": 81687169.0,
      "step": 39370
    },
    {
      "entropy": 5.664397573471069,
      "epoch": 3.571752539912917,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037631958007337195,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22131420522928238,
      "num_tokens": 81698291.0,
      "step": 39375
    },
    {
      "entropy": 5.685623979568481,
      "epoch": 3.5722060957910013,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003762906888632475,
      "loss": 4.8989,
      "mean_token_accuracy": 0.21665120124816895,
      "num_tokens": 81708750.0,
      "step": 39380
    },
    {
      "entropy": 5.585685205459595,
      "epoch": 3.5726596516690856,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003762617955584058,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22218581289052963,
      "num_tokens": 81719474.0,
      "step": 39385
    },
    {
      "entropy": 5.735635280609131,
      "epoch": 3.57311320754717,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003762329001594443,
      "loss": 4.9617,
      "mean_token_accuracy": 0.21766179352998732,
      "num_tokens": 81730092.0,
      "step": 39390
    },
    {
      "entropy": 5.6638270854949955,
      "epoch": 3.573566763425254,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003762040026669606,
      "loss": 4.8913,
      "mean_token_accuracy": 0.21569544225931167,
      "num_tokens": 81741235.0,
      "step": 39395
    },
    {
      "entropy": 5.594498157501221,
      "epoch": 3.5740203193033384,
      "grad_norm": 1.25,
      "learning_rate": 0.0003761751030815524,
      "loss": 4.8583,
      "mean_token_accuracy": 0.22641670405864717,
      "num_tokens": 81751434.0,
      "step": 39400
    },
    {
      "entropy": 5.646251058578491,
      "epoch": 3.5744738751814222,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003761462014038173,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21071959733963014,
      "num_tokens": 81763935.0,
      "step": 39405
    },
    {
      "entropy": 5.635839939117432,
      "epoch": 3.5749274310595065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003761172976343528,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21976048350334168,
      "num_tokens": 81775007.0,
      "step": 39410
    },
    {
      "entropy": 5.639729356765747,
      "epoch": 3.5753809869375908,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037608839177375705,
      "loss": 4.9216,
      "mean_token_accuracy": 0.21952951699495316,
      "num_tokens": 81785690.0,
      "step": 39415
    },
    {
      "entropy": 5.6427809715271,
      "epoch": 3.575834542815675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003760594838226275,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2263173833489418,
      "num_tokens": 81796143.0,
      "step": 39420
    },
    {
      "entropy": 5.7831244468688965,
      "epoch": 3.576288098693759,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003760305737815621,
      "loss": 4.9887,
      "mean_token_accuracy": 0.2179051697254181,
      "num_tokens": 81806607.0,
      "step": 39425
    },
    {
      "entropy": 5.615176486968994,
      "epoch": 3.576741654571843,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003760016616511586,
      "loss": 4.8235,
      "mean_token_accuracy": 0.2272084042429924,
      "num_tokens": 81817216.0,
      "step": 39430
    },
    {
      "entropy": 5.616209554672241,
      "epoch": 3.5771952104499274,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037597274743201507,
      "loss": 4.899,
      "mean_token_accuracy": 0.2211093619465828,
      "num_tokens": 81827211.0,
      "step": 39435
    },
    {
      "entropy": 5.635285043716431,
      "epoch": 3.5776487663280117,
      "grad_norm": 1.25,
      "learning_rate": 0.0003759438311247294,
      "loss": 4.8416,
      "mean_token_accuracy": 0.22919540703296662,
      "num_tokens": 81837365.0,
      "step": 39440
    },
    {
      "entropy": 5.632224273681641,
      "epoch": 3.578102322206096,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003759149127298995,
      "loss": 4.842,
      "mean_token_accuracy": 0.22837208211421967,
      "num_tokens": 81847803.0,
      "step": 39445
    },
    {
      "entropy": 5.606670904159546,
      "epoch": 3.57855587808418,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037588599224812363,
      "loss": 4.8328,
      "mean_token_accuracy": 0.22794439643621445,
      "num_tokens": 81858123.0,
      "step": 39450
    },
    {
      "entropy": 5.581077671051025,
      "epoch": 3.579009433962264,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003758570696799997,
      "loss": 4.8392,
      "mean_token_accuracy": 0.23114202916622162,
      "num_tokens": 81868668.0,
      "step": 39455
    },
    {
      "entropy": 5.608233213424683,
      "epoch": 3.5794629898403483,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037582814502612575,
      "loss": 4.8166,
      "mean_token_accuracy": 0.23261297196149827,
      "num_tokens": 81877482.0,
      "step": 39460
    },
    {
      "entropy": 5.554647159576416,
      "epoch": 3.5799165457184325,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003757992182871001,
      "loss": 4.7392,
      "mean_token_accuracy": 0.23605061024427415,
      "num_tokens": 81886700.0,
      "step": 39465
    },
    {
      "entropy": 5.631014251708985,
      "epoch": 3.580370101596517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037577028946352107,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22622024714946748,
      "num_tokens": 81896446.0,
      "step": 39470
    },
    {
      "entropy": 5.662693023681641,
      "epoch": 3.5808236574746006,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003757413585559866,
      "loss": 4.887,
      "mean_token_accuracy": 0.23129415065050124,
      "num_tokens": 81907418.0,
      "step": 39475
    },
    {
      "entropy": 5.674208879470825,
      "epoch": 3.581277213352685,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037571242556509526,
      "loss": 4.8532,
      "mean_token_accuracy": 0.225468248128891,
      "num_tokens": 81917940.0,
      "step": 39480
    },
    {
      "entropy": 5.57749605178833,
      "epoch": 3.581730769230769,
      "grad_norm": 1.25,
      "learning_rate": 0.0003756834904914453,
      "loss": 4.8428,
      "mean_token_accuracy": 0.22322502583265305,
      "num_tokens": 81928158.0,
      "step": 39485
    },
    {
      "entropy": 5.674011611938477,
      "epoch": 3.5821843251088534,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003756545533356351,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21963465213775635,
      "num_tokens": 81938111.0,
      "step": 39490
    },
    {
      "entropy": 5.67154631614685,
      "epoch": 3.5826378809869377,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000375625614098263,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2254747986793518,
      "num_tokens": 81948798.0,
      "step": 39495
    },
    {
      "entropy": 5.686433744430542,
      "epoch": 3.583091436865022,
      "grad_norm": 1.25,
      "learning_rate": 0.00037559667277992765,
      "loss": 4.9108,
      "mean_token_accuracy": 0.22242577821016313,
      "num_tokens": 81959240.0,
      "step": 39500
    },
    {
      "entropy": 5.6343300342559814,
      "epoch": 3.5835449927431062,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003755677293812274,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22503917515277863,
      "num_tokens": 81969929.0,
      "step": 39505
    },
    {
      "entropy": 5.698253870010376,
      "epoch": 3.58399854862119,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003755387839027609,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2120912715792656,
      "num_tokens": 81981007.0,
      "step": 39510
    },
    {
      "entropy": 5.707571220397949,
      "epoch": 3.5844521044992743,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037550983634512683,
      "loss": 5.0122,
      "mean_token_accuracy": 0.21395166218280792,
      "num_tokens": 81991172.0,
      "step": 39515
    },
    {
      "entropy": 5.613654708862304,
      "epoch": 3.5849056603773586,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003754808867089236,
      "loss": 4.7976,
      "mean_token_accuracy": 0.22859591394662857,
      "num_tokens": 82001282.0,
      "step": 39520
    },
    {
      "entropy": 5.630348110198975,
      "epoch": 3.5853592162554424,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037545193499475003,
      "loss": 4.8871,
      "mean_token_accuracy": 0.22027432918548584,
      "num_tokens": 82011929.0,
      "step": 39525
    },
    {
      "entropy": 5.719523286819458,
      "epoch": 3.5858127721335267,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003754229812032049,
      "loss": 4.9207,
      "mean_token_accuracy": 0.2286134496331215,
      "num_tokens": 82021663.0,
      "step": 39530
    },
    {
      "entropy": 5.676588821411133,
      "epoch": 3.586266328011611,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037539402533488694,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21083466708660126,
      "num_tokens": 82031547.0,
      "step": 39535
    },
    {
      "entropy": 5.696749544143676,
      "epoch": 3.586719883889695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003753650673903949,
      "loss": 4.9323,
      "mean_token_accuracy": 0.22018466740846634,
      "num_tokens": 82041722.0,
      "step": 39540
    },
    {
      "entropy": 5.668599414825439,
      "epoch": 3.5871734397677795,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037533610737032773,
      "loss": 4.9133,
      "mean_token_accuracy": 0.22211923450231552,
      "num_tokens": 82052715.0,
      "step": 39545
    },
    {
      "entropy": 5.65114541053772,
      "epoch": 3.5876269956458637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037530714527528433,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22731748223304749,
      "num_tokens": 82062709.0,
      "step": 39550
    },
    {
      "entropy": 5.684697771072388,
      "epoch": 3.588080551523948,
      "grad_norm": 1.25,
      "learning_rate": 0.00037527818110586354,
      "loss": 4.911,
      "mean_token_accuracy": 0.2186839610338211,
      "num_tokens": 82072731.0,
      "step": 39555
    },
    {
      "entropy": 5.662290382385254,
      "epoch": 3.588534107402032,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003752492148626644,
      "loss": 4.8167,
      "mean_token_accuracy": 0.2284128338098526,
      "num_tokens": 82083216.0,
      "step": 39560
    },
    {
      "entropy": 5.676496124267578,
      "epoch": 3.588987663280116,
      "grad_norm": 1.171875,
      "learning_rate": 0.000375220246546286,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2155966728925705,
      "num_tokens": 82094482.0,
      "step": 39565
    },
    {
      "entropy": 5.659044170379639,
      "epoch": 3.5894412191582004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003751912761573273,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2198419004678726,
      "num_tokens": 82104988.0,
      "step": 39570
    },
    {
      "entropy": 5.6291591167449955,
      "epoch": 3.5898947750362846,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037516230369638755,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21424392610788345,
      "num_tokens": 82114455.0,
      "step": 39575
    },
    {
      "entropy": 5.654451990127564,
      "epoch": 3.5903483309143684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003751333291640658,
      "loss": 4.9036,
      "mean_token_accuracy": 0.22112831622362136,
      "num_tokens": 82125001.0,
      "step": 39580
    },
    {
      "entropy": 5.662765979766846,
      "epoch": 3.5908018867924527,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037510435256096124,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2205198049545288,
      "num_tokens": 82134586.0,
      "step": 39585
    },
    {
      "entropy": 5.602188396453857,
      "epoch": 3.591255442670537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037507537388767326,
      "loss": 4.847,
      "mean_token_accuracy": 0.22516326755285263,
      "num_tokens": 82144612.0,
      "step": 39590
    },
    {
      "entropy": 5.585591554641724,
      "epoch": 3.5917089985486212,
      "grad_norm": 1.265625,
      "learning_rate": 0.000375046393144801,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22928515672683716,
      "num_tokens": 82154913.0,
      "step": 39595
    },
    {
      "entropy": 5.655853939056397,
      "epoch": 3.5921625544267055,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003750174103329438,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21962901204824448,
      "num_tokens": 82164469.0,
      "step": 39600
    },
    {
      "entropy": 5.683607482910157,
      "epoch": 3.59261611030479,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037498842545270105,
      "loss": 4.9139,
      "mean_token_accuracy": 0.2225888878107071,
      "num_tokens": 82174575.0,
      "step": 39605
    },
    {
      "entropy": 5.692196655273437,
      "epoch": 3.5930696661828736,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037495943850467226,
      "loss": 4.893,
      "mean_token_accuracy": 0.2226075068116188,
      "num_tokens": 82184661.0,
      "step": 39610
    },
    {
      "entropy": 5.712615299224853,
      "epoch": 3.593523222060958,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037493044948945677,
      "loss": 4.91,
      "mean_token_accuracy": 0.21913286596536635,
      "num_tokens": 82194877.0,
      "step": 39615
    },
    {
      "entropy": 5.643481254577637,
      "epoch": 3.593976777939042,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037490145840765414,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2230779528617859,
      "num_tokens": 82206278.0,
      "step": 39620
    },
    {
      "entropy": 5.62686014175415,
      "epoch": 3.5944303338171264,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037487246525986384,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2190272852778435,
      "num_tokens": 82216140.0,
      "step": 39625
    },
    {
      "entropy": 5.7096781730651855,
      "epoch": 3.59488388969521,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003748434700466856,
      "loss": 4.9463,
      "mean_token_accuracy": 0.22008083909749984,
      "num_tokens": 82226239.0,
      "step": 39630
    },
    {
      "entropy": 5.674899291992188,
      "epoch": 3.5953374455732945,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003748144727687188,
      "loss": 4.868,
      "mean_token_accuracy": 0.2250571444630623,
      "num_tokens": 82235795.0,
      "step": 39635
    },
    {
      "entropy": 5.600372695922852,
      "epoch": 3.5957910014513788,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037478547342656345,
      "loss": 4.8218,
      "mean_token_accuracy": 0.23095702081918718,
      "num_tokens": 82245644.0,
      "step": 39640
    },
    {
      "entropy": 5.620706701278687,
      "epoch": 3.596244557329463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000374756472020819,
      "loss": 4.8631,
      "mean_token_accuracy": 0.2238653853535652,
      "num_tokens": 82255608.0,
      "step": 39645
    },
    {
      "entropy": 5.62995138168335,
      "epoch": 3.5966981132075473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037472746855208525,
      "loss": 4.854,
      "mean_token_accuracy": 0.22334845811128617,
      "num_tokens": 82265488.0,
      "step": 39650
    },
    {
      "entropy": 5.63012580871582,
      "epoch": 3.5971516690856316,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003746984630209621,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2154064193367958,
      "num_tokens": 82276866.0,
      "step": 39655
    },
    {
      "entropy": 5.635176706314087,
      "epoch": 3.5976052249637154,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00037466945542804933,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22262948304414748,
      "num_tokens": 82287125.0,
      "step": 39660
    },
    {
      "entropy": 5.699474716186524,
      "epoch": 3.5980587808417996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037464044577394676,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21946818977594376,
      "num_tokens": 82296887.0,
      "step": 39665
    },
    {
      "entropy": 5.672021770477295,
      "epoch": 3.598512336719884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003746114340592544,
      "loss": 4.9788,
      "mean_token_accuracy": 0.21009893864393234,
      "num_tokens": 82306928.0,
      "step": 39670
    },
    {
      "entropy": 5.638008117675781,
      "epoch": 3.598965892597968,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037458242028457216,
      "loss": 4.9134,
      "mean_token_accuracy": 0.22398993670940398,
      "num_tokens": 82317873.0,
      "step": 39675
    },
    {
      "entropy": 5.581651830673218,
      "epoch": 3.599419448476052,
      "grad_norm": 1.125,
      "learning_rate": 0.00037455340445050023,
      "loss": 4.7702,
      "mean_token_accuracy": 0.23053000420331954,
      "num_tokens": 82328184.0,
      "step": 39680
    },
    {
      "entropy": 5.611350727081299,
      "epoch": 3.5998730043541363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037452438655763843,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22242008596658708,
      "num_tokens": 82338191.0,
      "step": 39685
    },
    {
      "entropy": 5.628508281707764,
      "epoch": 3.6003265602322205,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037449536660658696,
      "loss": 4.9667,
      "mean_token_accuracy": 0.2164360001683235,
      "num_tokens": 82348095.0,
      "step": 39690
    },
    {
      "entropy": 5.626165437698364,
      "epoch": 3.600780116110305,
      "grad_norm": 1.265625,
      "learning_rate": 0.000374466344597946,
      "loss": 4.8003,
      "mean_token_accuracy": 0.2291667327284813,
      "num_tokens": 82358216.0,
      "step": 39695
    },
    {
      "entropy": 5.589766788482666,
      "epoch": 3.601233671988389,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003744373205323157,
      "loss": 4.8246,
      "mean_token_accuracy": 0.22677585035562514,
      "num_tokens": 82368581.0,
      "step": 39700
    },
    {
      "entropy": 5.613586521148681,
      "epoch": 3.6016872278664733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003744082944102962,
      "loss": 4.8805,
      "mean_token_accuracy": 0.21822649836540223,
      "num_tokens": 82379609.0,
      "step": 39705
    },
    {
      "entropy": 5.672139596939087,
      "epoch": 3.6021407837445576,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003743792662324878,
      "loss": 4.9214,
      "mean_token_accuracy": 0.2171761378645897,
      "num_tokens": 82390589.0,
      "step": 39710
    },
    {
      "entropy": 5.582696962356567,
      "epoch": 3.6025943396226414,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037435023599949086,
      "loss": 4.8515,
      "mean_token_accuracy": 0.21978304535150528,
      "num_tokens": 82400915.0,
      "step": 39715
    },
    {
      "entropy": 5.622618770599365,
      "epoch": 3.6030478955007257,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037432120371190585,
      "loss": 4.8763,
      "mean_token_accuracy": 0.2261260449886322,
      "num_tokens": 82411544.0,
      "step": 39720
    },
    {
      "entropy": 5.617326927185059,
      "epoch": 3.60350145137881,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037429216937033294,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2225077897310257,
      "num_tokens": 82422391.0,
      "step": 39725
    },
    {
      "entropy": 5.619288444519043,
      "epoch": 3.6039550072568938,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037426313297537263,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22646329551935196,
      "num_tokens": 82433564.0,
      "step": 39730
    },
    {
      "entropy": 5.6200549602508545,
      "epoch": 3.604408563134978,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003742340945276255,
      "loss": 4.9144,
      "mean_token_accuracy": 0.2234766185283661,
      "num_tokens": 82445005.0,
      "step": 39735
    },
    {
      "entropy": 5.632007837295532,
      "epoch": 3.6048621190130623,
      "grad_norm": 1.28125,
      "learning_rate": 0.000374205054027692,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21421051472425462,
      "num_tokens": 82455998.0,
      "step": 39740
    },
    {
      "entropy": 5.645858478546143,
      "epoch": 3.6053156748911466,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037417601147617265,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22270803153514862,
      "num_tokens": 82466789.0,
      "step": 39745
    },
    {
      "entropy": 5.718984889984131,
      "epoch": 3.605769230769231,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037414696687366816,
      "loss": 5.0094,
      "mean_token_accuracy": 0.21109584867954254,
      "num_tokens": 82478181.0,
      "step": 39750
    },
    {
      "entropy": 5.650729942321777,
      "epoch": 3.606222786647315,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037411792022077917,
      "loss": 4.8413,
      "mean_token_accuracy": 0.21976753324270248,
      "num_tokens": 82488452.0,
      "step": 39755
    },
    {
      "entropy": 5.690230798721314,
      "epoch": 3.6066763425253994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003740888715181063,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22004005908966065,
      "num_tokens": 82498179.0,
      "step": 39760
    },
    {
      "entropy": 5.61546139717102,
      "epoch": 3.607129898403483,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037405982076625026,
      "loss": 4.9439,
      "mean_token_accuracy": 0.22491095960140228,
      "num_tokens": 82508242.0,
      "step": 39765
    },
    {
      "entropy": 5.690432977676392,
      "epoch": 3.6075834542815675,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037403076796581195,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21453270614147185,
      "num_tokens": 82519363.0,
      "step": 39770
    },
    {
      "entropy": 5.7164936542510985,
      "epoch": 3.6080370101596517,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003740017131173921,
      "loss": 4.921,
      "mean_token_accuracy": 0.21915996670722962,
      "num_tokens": 82529745.0,
      "step": 39775
    },
    {
      "entropy": 5.705504751205444,
      "epoch": 3.608490566037736,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003739726562215916,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21702400892972945,
      "num_tokens": 82539573.0,
      "step": 39780
    },
    {
      "entropy": 5.679179286956787,
      "epoch": 3.60894412191582,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037394359727901136,
      "loss": 4.9528,
      "mean_token_accuracy": 0.22171331495046614,
      "num_tokens": 82551313.0,
      "step": 39785
    },
    {
      "entropy": 5.619750356674194,
      "epoch": 3.609397677793904,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003739145362902523,
      "loss": 4.8853,
      "mean_token_accuracy": 0.2150670364499092,
      "num_tokens": 82561283.0,
      "step": 39790
    },
    {
      "entropy": 5.618160009384155,
      "epoch": 3.6098512336719883,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003738854732559153,
      "loss": 4.8098,
      "mean_token_accuracy": 0.2296390116214752,
      "num_tokens": 82571118.0,
      "step": 39795
    },
    {
      "entropy": 5.6271984577178955,
      "epoch": 3.6103047895500726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003738564081766017,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22267504632472992,
      "num_tokens": 82581663.0,
      "step": 39800
    },
    {
      "entropy": 5.610398149490356,
      "epoch": 3.610758345428157,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037382734105291234,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2219679594039917,
      "num_tokens": 82592593.0,
      "step": 39805
    },
    {
      "entropy": 5.62231936454773,
      "epoch": 3.611211901306241,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037379827188544835,
      "loss": 4.9707,
      "mean_token_accuracy": 0.21576057970523835,
      "num_tokens": 82603047.0,
      "step": 39810
    },
    {
      "entropy": 5.708199548721313,
      "epoch": 3.611665457184325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037376920067481095,
      "loss": 4.9169,
      "mean_token_accuracy": 0.2200032129883766,
      "num_tokens": 82612789.0,
      "step": 39815
    },
    {
      "entropy": 5.644432163238525,
      "epoch": 3.6121190130624092,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003737401274216012,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22033663392066954,
      "num_tokens": 82622881.0,
      "step": 39820
    },
    {
      "entropy": 5.630075168609619,
      "epoch": 3.6125725689404935,
      "grad_norm": 1.375,
      "learning_rate": 0.0003737110521264206,
      "loss": 4.8365,
      "mean_token_accuracy": 0.21775075048208237,
      "num_tokens": 82633933.0,
      "step": 39825
    },
    {
      "entropy": 5.644597721099854,
      "epoch": 3.6130261248185778,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003736819747898702,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2244173228740692,
      "num_tokens": 82645069.0,
      "step": 39830
    },
    {
      "entropy": 5.686211824417114,
      "epoch": 3.6134796806966616,
      "grad_norm": 1.25,
      "learning_rate": 0.00037365289541255154,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2205202490091324,
      "num_tokens": 82655152.0,
      "step": 39835
    },
    {
      "entropy": 5.601044797897339,
      "epoch": 3.613933236574746,
      "grad_norm": 1.25,
      "learning_rate": 0.00037362381399506575,
      "loss": 4.8633,
      "mean_token_accuracy": 0.21666934192180634,
      "num_tokens": 82665260.0,
      "step": 39840
    },
    {
      "entropy": 5.623809194564819,
      "epoch": 3.61438679245283,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037359473053801435,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22947915643453598,
      "num_tokens": 82676287.0,
      "step": 39845
    },
    {
      "entropy": 5.657204723358154,
      "epoch": 3.6148403483309144,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003735656450419988,
      "loss": 4.9833,
      "mean_token_accuracy": 0.21008455753326416,
      "num_tokens": 82686958.0,
      "step": 39850
    },
    {
      "entropy": 5.677321052551269,
      "epoch": 3.6152939042089987,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037353655750762063,
      "loss": 5.0309,
      "mean_token_accuracy": 0.20693982839584352,
      "num_tokens": 82698623.0,
      "step": 39855
    },
    {
      "entropy": 5.68940691947937,
      "epoch": 3.615747460087083,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037350746793548135,
      "loss": 4.9575,
      "mean_token_accuracy": 0.21703476309776307,
      "num_tokens": 82709146.0,
      "step": 39860
    },
    {
      "entropy": 5.647206783294678,
      "epoch": 3.6162010159651667,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037347837632618246,
      "loss": 4.8297,
      "mean_token_accuracy": 0.22628388404846192,
      "num_tokens": 82719323.0,
      "step": 39865
    },
    {
      "entropy": 5.681375932693482,
      "epoch": 3.616654571843251,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037344928268032574,
      "loss": 4.9514,
      "mean_token_accuracy": 0.2160400152206421,
      "num_tokens": 82730068.0,
      "step": 39870
    },
    {
      "entropy": 5.665633058547973,
      "epoch": 3.6171081277213353,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003734201869985127,
      "loss": 4.9398,
      "mean_token_accuracy": 0.22275419533252716,
      "num_tokens": 82740460.0,
      "step": 39875
    },
    {
      "entropy": 5.627183532714843,
      "epoch": 3.6175616835994195,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003733910892813451,
      "loss": 4.8583,
      "mean_token_accuracy": 0.21603755205869674,
      "num_tokens": 82751488.0,
      "step": 39880
    },
    {
      "entropy": 5.540359878540039,
      "epoch": 3.6180152394775034,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037336198952942477,
      "loss": 4.7839,
      "mean_token_accuracy": 0.2355194568634033,
      "num_tokens": 82762061.0,
      "step": 39885
    },
    {
      "entropy": 5.6118522644042965,
      "epoch": 3.6184687953555876,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003733328877433534,
      "loss": 4.8433,
      "mean_token_accuracy": 0.23038543462753297,
      "num_tokens": 82773421.0,
      "step": 39890
    },
    {
      "entropy": 5.57828311920166,
      "epoch": 3.618922351233672,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037330378392373285,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2281870111823082,
      "num_tokens": 82783342.0,
      "step": 39895
    },
    {
      "entropy": 5.63250994682312,
      "epoch": 3.619375907111756,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037327467807116507,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21426529735326766,
      "num_tokens": 82793885.0,
      "step": 39900
    },
    {
      "entropy": 5.650369358062744,
      "epoch": 3.6198294629898404,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037324557018625186,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21836500912904738,
      "num_tokens": 82803746.0,
      "step": 39905
    },
    {
      "entropy": 5.652484273910522,
      "epoch": 3.6202830188679247,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003732164602695952,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22193869054317475,
      "num_tokens": 82813858.0,
      "step": 39910
    },
    {
      "entropy": 5.707403230667114,
      "epoch": 3.620736574746009,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037318734832179705,
      "loss": 4.935,
      "mean_token_accuracy": 0.21966660618782044,
      "num_tokens": 82824358.0,
      "step": 39915
    },
    {
      "entropy": 5.676679849624634,
      "epoch": 3.621190130624093,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037315823434345963,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2315877005457878,
      "num_tokens": 82835801.0,
      "step": 39920
    },
    {
      "entropy": 5.567750787734985,
      "epoch": 3.621643686502177,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003731291183351848,
      "loss": 4.8228,
      "mean_token_accuracy": 0.2311795875430107,
      "num_tokens": 82845818.0,
      "step": 39925
    },
    {
      "entropy": 5.599418592453003,
      "epoch": 3.6220972423802613,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037310000029757486,
      "loss": 4.8589,
      "mean_token_accuracy": 0.22515048682689667,
      "num_tokens": 82856920.0,
      "step": 39930
    },
    {
      "entropy": 5.6821064949035645,
      "epoch": 3.6225507982583456,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003730708802312319,
      "loss": 4.961,
      "mean_token_accuracy": 0.21929134726524352,
      "num_tokens": 82867220.0,
      "step": 39935
    },
    {
      "entropy": 5.63896107673645,
      "epoch": 3.6230043541364294,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003730417581367582,
      "loss": 4.8841,
      "mean_token_accuracy": 0.22628148794174194,
      "num_tokens": 82877541.0,
      "step": 39940
    },
    {
      "entropy": 5.690818405151367,
      "epoch": 3.6234579100145137,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003730126340147559,
      "loss": 4.9302,
      "mean_token_accuracy": 0.22232334017753602,
      "num_tokens": 82888346.0,
      "step": 39945
    },
    {
      "entropy": 5.709954929351807,
      "epoch": 3.623911465892598,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003729835078658273,
      "loss": 4.9666,
      "mean_token_accuracy": 0.22023930847644807,
      "num_tokens": 82899721.0,
      "step": 39950
    },
    {
      "entropy": 5.662163877487183,
      "epoch": 3.624365021770682,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037295437969057493,
      "loss": 4.8809,
      "mean_token_accuracy": 0.22382118105888366,
      "num_tokens": 82910149.0,
      "step": 39955
    },
    {
      "entropy": 5.654334306716919,
      "epoch": 3.6248185776487665,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037292524948960093,
      "loss": 4.9096,
      "mean_token_accuracy": 0.22447170913219452,
      "num_tokens": 82920883.0,
      "step": 39960
    },
    {
      "entropy": 5.646292591094971,
      "epoch": 3.6252721335268507,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003728961172635078,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21913919001817703,
      "num_tokens": 82931091.0,
      "step": 39965
    },
    {
      "entropy": 5.651579666137695,
      "epoch": 3.6257256894049346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037286698301289804,
      "loss": 4.8075,
      "mean_token_accuracy": 0.22505465894937515,
      "num_tokens": 82941108.0,
      "step": 39970
    },
    {
      "entropy": 5.613724613189698,
      "epoch": 3.626179245283019,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003728378467383742,
      "loss": 4.8716,
      "mean_token_accuracy": 0.22475457191467285,
      "num_tokens": 82951943.0,
      "step": 39975
    },
    {
      "entropy": 5.696922016143799,
      "epoch": 3.626632801161103,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003728087084405387,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22236319184303283,
      "num_tokens": 82961495.0,
      "step": 39980
    },
    {
      "entropy": 5.6964035987854,
      "epoch": 3.6270863570391874,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037277956811999414,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22365380823612213,
      "num_tokens": 82971858.0,
      "step": 39985
    },
    {
      "entropy": 5.661520957946777,
      "epoch": 3.627539912917271,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003727504257773433,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22139185965061187,
      "num_tokens": 82982805.0,
      "step": 39990
    },
    {
      "entropy": 5.646217393875122,
      "epoch": 3.6279934687953554,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037272128141318866,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21410252898931503,
      "num_tokens": 82993446.0,
      "step": 39995
    },
    {
      "entropy": 5.619251155853272,
      "epoch": 3.6284470246734397,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000372692135028133,
      "loss": 4.8053,
      "mean_token_accuracy": 0.22217184603214263,
      "num_tokens": 83003378.0,
      "step": 40000
    },
    {
      "entropy": 5.636310052871704,
      "epoch": 3.628900580551524,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037266298662277914,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22047396302223204,
      "num_tokens": 83014223.0,
      "step": 40005
    },
    {
      "entropy": 5.6078626155853275,
      "epoch": 3.6293541364296082,
      "grad_norm": 1.25,
      "learning_rate": 0.00037263383619772986,
      "loss": 4.9216,
      "mean_token_accuracy": 0.22089984714984895,
      "num_tokens": 83024691.0,
      "step": 40010
    },
    {
      "entropy": 5.651585388183594,
      "epoch": 3.6298076923076925,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037260468375358794,
      "loss": 4.9009,
      "mean_token_accuracy": 0.2201688289642334,
      "num_tokens": 83035704.0,
      "step": 40015
    },
    {
      "entropy": 5.693813705444336,
      "epoch": 3.6302612481857763,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003725755292909562,
      "loss": 4.9901,
      "mean_token_accuracy": 0.21433429718017577,
      "num_tokens": 83047173.0,
      "step": 40020
    },
    {
      "entropy": 5.709642553329468,
      "epoch": 3.6307148040638606,
      "grad_norm": 1.375,
      "learning_rate": 0.0003725463728104378,
      "loss": 4.9315,
      "mean_token_accuracy": 0.21617625653743744,
      "num_tokens": 83056399.0,
      "step": 40025
    },
    {
      "entropy": 5.623583221435547,
      "epoch": 3.631168359941945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003725172143126354,
      "loss": 4.8387,
      "mean_token_accuracy": 0.21872634142637254,
      "num_tokens": 83066413.0,
      "step": 40030
    },
    {
      "entropy": 5.620371294021607,
      "epoch": 3.631621915820029,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037248805379815227,
      "loss": 4.9161,
      "mean_token_accuracy": 0.21911440640687943,
      "num_tokens": 83077843.0,
      "step": 40035
    },
    {
      "entropy": 5.700660181045532,
      "epoch": 3.632075471698113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003724588912675913,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21824240535497666,
      "num_tokens": 83088585.0,
      "step": 40040
    },
    {
      "entropy": 5.731334829330445,
      "epoch": 3.6325290275761972,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003724297267215557,
      "loss": 5.0125,
      "mean_token_accuracy": 0.22148914635181427,
      "num_tokens": 83099295.0,
      "step": 40045
    },
    {
      "entropy": 5.600619697570801,
      "epoch": 3.6329825834542815,
      "grad_norm": 1.375,
      "learning_rate": 0.00037240056016064844,
      "loss": 4.8074,
      "mean_token_accuracy": 0.2316656768321991,
      "num_tokens": 83108530.0,
      "step": 40050
    },
    {
      "entropy": 5.610688352584839,
      "epoch": 3.6334361393323658,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003723713915854728,
      "loss": 4.8933,
      "mean_token_accuracy": 0.21968873739242553,
      "num_tokens": 83119703.0,
      "step": 40055
    },
    {
      "entropy": 5.667976999282837,
      "epoch": 3.63388969521045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003723422209966319,
      "loss": 4.873,
      "mean_token_accuracy": 0.22223049104213716,
      "num_tokens": 83129323.0,
      "step": 40060
    },
    {
      "entropy": 5.653559017181396,
      "epoch": 3.6343432510885343,
      "grad_norm": 1.25,
      "learning_rate": 0.0003723130483947292,
      "loss": 4.8419,
      "mean_token_accuracy": 0.2281960129737854,
      "num_tokens": 83139339.0,
      "step": 40065
    },
    {
      "entropy": 5.606348609924316,
      "epoch": 3.6347968069666186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003722838737803678,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22200387716293335,
      "num_tokens": 83149845.0,
      "step": 40070
    },
    {
      "entropy": 5.68199348449707,
      "epoch": 3.6352503628447024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037225469715415106,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21887548118829728,
      "num_tokens": 83160332.0,
      "step": 40075
    },
    {
      "entropy": 5.687393379211426,
      "epoch": 3.6357039187227866,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037222551851668245,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21995497047901152,
      "num_tokens": 83170892.0,
      "step": 40080
    },
    {
      "entropy": 5.6071295738220215,
      "epoch": 3.636157474600871,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003721963378685652,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22541280686855317,
      "num_tokens": 83181035.0,
      "step": 40085
    },
    {
      "entropy": 5.6039036273956295,
      "epoch": 3.6366110304789547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037216715521040314,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22605569362640382,
      "num_tokens": 83191886.0,
      "step": 40090
    },
    {
      "entropy": 5.643513298034668,
      "epoch": 3.637064586357039,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003721379705427994,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22598879784345627,
      "num_tokens": 83202356.0,
      "step": 40095
    },
    {
      "entropy": 5.663507986068725,
      "epoch": 3.6375181422351233,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037210878386635774,
      "loss": 4.9216,
      "mean_token_accuracy": 0.2278437063097954,
      "num_tokens": 83212909.0,
      "step": 40100
    },
    {
      "entropy": 5.6779200553894045,
      "epoch": 3.6379716981132075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037207959518168165,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2168870598077774,
      "num_tokens": 83223523.0,
      "step": 40105
    },
    {
      "entropy": 5.7612841606140135,
      "epoch": 3.638425253991292,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037205040448937485,
      "loss": 5.0123,
      "mean_token_accuracy": 0.21286313831806183,
      "num_tokens": 83233665.0,
      "step": 40110
    },
    {
      "entropy": 5.625516033172607,
      "epoch": 3.638878809869376,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003720212117900409,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23000577092170715,
      "num_tokens": 83243563.0,
      "step": 40115
    },
    {
      "entropy": 5.634356927871704,
      "epoch": 3.6393323657474603,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037199201708428356,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22218847423791885,
      "num_tokens": 83254246.0,
      "step": 40120
    },
    {
      "entropy": 5.66231861114502,
      "epoch": 3.639785921625544,
      "grad_norm": 1.25,
      "learning_rate": 0.0003719628203727066,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2198367714881897,
      "num_tokens": 83264361.0,
      "step": 40125
    },
    {
      "entropy": 5.692606019973755,
      "epoch": 3.6402394775036284,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003719336216559138,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22032452821731568,
      "num_tokens": 83274193.0,
      "step": 40130
    },
    {
      "entropy": 5.678310585021973,
      "epoch": 3.6406930333817127,
      "grad_norm": 1.34375,
      "learning_rate": 0.000371904420934509,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22529397606849672,
      "num_tokens": 83284593.0,
      "step": 40135
    },
    {
      "entropy": 5.596283674240112,
      "epoch": 3.641146589259797,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003718752182090961,
      "loss": 4.8619,
      "mean_token_accuracy": 0.2219424530863762,
      "num_tokens": 83294602.0,
      "step": 40140
    },
    {
      "entropy": 5.638093137741089,
      "epoch": 3.6416001451378808,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000371846013480279,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2213722974061966,
      "num_tokens": 83305399.0,
      "step": 40145
    },
    {
      "entropy": 5.576687622070312,
      "epoch": 3.642053701015965,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037181680674866165,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22188881933689117,
      "num_tokens": 83315696.0,
      "step": 40150
    },
    {
      "entropy": 5.712692785263061,
      "epoch": 3.6425072568940493,
      "grad_norm": 1.25,
      "learning_rate": 0.0003717875980148481,
      "loss": 5.0185,
      "mean_token_accuracy": 0.21227151155471802,
      "num_tokens": 83326686.0,
      "step": 40155
    },
    {
      "entropy": 5.677354240417481,
      "epoch": 3.6429608127721336,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003717583872794423,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2321374773979187,
      "num_tokens": 83336902.0,
      "step": 40160
    },
    {
      "entropy": 5.6812585353851315,
      "epoch": 3.643414368650218,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003717291745430484,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22088903337717056,
      "num_tokens": 83347324.0,
      "step": 40165
    },
    {
      "entropy": 5.57171950340271,
      "epoch": 3.643867924528302,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037169995980627056,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22425670623779298,
      "num_tokens": 83358458.0,
      "step": 40170
    },
    {
      "entropy": 5.580731678009033,
      "epoch": 3.644321480406386,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037167074306971285,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21660740971565245,
      "num_tokens": 83369047.0,
      "step": 40175
    },
    {
      "entropy": 5.6755071640014645,
      "epoch": 3.64477503628447,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037164152433397964,
      "loss": 4.8816,
      "mean_token_accuracy": 0.22275365889072418,
      "num_tokens": 83378891.0,
      "step": 40180
    },
    {
      "entropy": 5.674152517318726,
      "epoch": 3.6452285921625545,
      "grad_norm": 1.359375,
      "learning_rate": 0.000371612303599675,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2193396806716919,
      "num_tokens": 83389289.0,
      "step": 40185
    },
    {
      "entropy": 5.715043783187866,
      "epoch": 3.6456821480406387,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037158308086740336,
      "loss": 4.9654,
      "mean_token_accuracy": 0.222805917263031,
      "num_tokens": 83399316.0,
      "step": 40190
    },
    {
      "entropy": 5.6228443622589115,
      "epoch": 3.6461357039187225,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003715538561377689,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22524963915348054,
      "num_tokens": 83409507.0,
      "step": 40195
    },
    {
      "entropy": 5.710901165008545,
      "epoch": 3.646589259796807,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003715246294113762,
      "loss": 4.9905,
      "mean_token_accuracy": 0.20930583477020265,
      "num_tokens": 83419687.0,
      "step": 40200
    },
    {
      "entropy": 5.762218332290649,
      "epoch": 3.647042815674891,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037149540068882945,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21536059230566024,
      "num_tokens": 83429711.0,
      "step": 40205
    },
    {
      "entropy": 5.62012300491333,
      "epoch": 3.6474963715529753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037146616997073327,
      "loss": 4.8466,
      "mean_token_accuracy": 0.21585071533918382,
      "num_tokens": 83440116.0,
      "step": 40210
    },
    {
      "entropy": 5.641993999481201,
      "epoch": 3.6479499274310596,
      "grad_norm": 1.25,
      "learning_rate": 0.00037143693725769214,
      "loss": 4.8787,
      "mean_token_accuracy": 0.22562815397977828,
      "num_tokens": 83450877.0,
      "step": 40215
    },
    {
      "entropy": 5.590087413787842,
      "epoch": 3.648403483309144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037140770255031066,
      "loss": 4.793,
      "mean_token_accuracy": 0.23063643127679825,
      "num_tokens": 83460080.0,
      "step": 40220
    },
    {
      "entropy": 5.63749942779541,
      "epoch": 3.6488570391872277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037137846584919316,
      "loss": 4.945,
      "mean_token_accuracy": 0.2169843077659607,
      "num_tokens": 83469933.0,
      "step": 40225
    },
    {
      "entropy": 5.632030296325683,
      "epoch": 3.649310595065312,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003713492271549445,
      "loss": 4.8577,
      "mean_token_accuracy": 0.22371985465288163,
      "num_tokens": 83480477.0,
      "step": 40230
    },
    {
      "entropy": 5.632258319854737,
      "epoch": 3.6497641509433962,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003713199864681693,
      "loss": 4.8867,
      "mean_token_accuracy": 0.221931654214859,
      "num_tokens": 83491022.0,
      "step": 40235
    },
    {
      "entropy": 5.662606525421142,
      "epoch": 3.6502177068214805,
      "grad_norm": 1.25,
      "learning_rate": 0.00037129074378947216,
      "loss": 4.9415,
      "mean_token_accuracy": 0.22252005636692046,
      "num_tokens": 83501643.0,
      "step": 40240
    },
    {
      "entropy": 5.687870359420776,
      "epoch": 3.6506712626995643,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037126149911945796,
      "loss": 4.9414,
      "mean_token_accuracy": 0.2132761538028717,
      "num_tokens": 83511948.0,
      "step": 40245
    },
    {
      "entropy": 5.677402496337891,
      "epoch": 3.6511248185776486,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037123225245873137,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21967084407806398,
      "num_tokens": 83521939.0,
      "step": 40250
    },
    {
      "entropy": 5.723622798919678,
      "epoch": 3.651578374455733,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037120300380789727,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2197018086910248,
      "num_tokens": 83532098.0,
      "step": 40255
    },
    {
      "entropy": 5.658861398696899,
      "epoch": 3.652031930333817,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037117375316756063,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2199383035302162,
      "num_tokens": 83543238.0,
      "step": 40260
    },
    {
      "entropy": 5.6958314895629885,
      "epoch": 3.6524854862119014,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003711445005383263,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2177405059337616,
      "num_tokens": 83552969.0,
      "step": 40265
    },
    {
      "entropy": 5.697970724105835,
      "epoch": 3.6529390420899857,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037111524592079896,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2078731656074524,
      "num_tokens": 83564569.0,
      "step": 40270
    },
    {
      "entropy": 5.679282093048096,
      "epoch": 3.65339259796807,
      "grad_norm": 1.234375,
      "learning_rate": 0.000371085989315584,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21783456355333328,
      "num_tokens": 83575613.0,
      "step": 40275
    },
    {
      "entropy": 5.6941159725189205,
      "epoch": 3.6538461538461537,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003710567307232864,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21702758520841597,
      "num_tokens": 83586311.0,
      "step": 40280
    },
    {
      "entropy": 5.707441282272339,
      "epoch": 3.654299709724238,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037102747014451095,
      "loss": 4.9599,
      "mean_token_accuracy": 0.2187309056520462,
      "num_tokens": 83596660.0,
      "step": 40285
    },
    {
      "entropy": 5.697196102142334,
      "epoch": 3.6547532656023223,
      "grad_norm": 1.359375,
      "learning_rate": 0.000370998207579863,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21857304722070695,
      "num_tokens": 83606882.0,
      "step": 40290
    },
    {
      "entropy": 5.704786157608032,
      "epoch": 3.6552068214804065,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003709689430299477,
      "loss": 4.9485,
      "mean_token_accuracy": 0.22320753186941147,
      "num_tokens": 83617380.0,
      "step": 40295
    },
    {
      "entropy": 5.661019563674927,
      "epoch": 3.6556603773584904,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003709396764953701,
      "loss": 4.9165,
      "mean_token_accuracy": 0.21984029114246367,
      "num_tokens": 83627961.0,
      "step": 40300
    },
    {
      "entropy": 5.528314256668091,
      "epoch": 3.6561139332365746,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003709104079767357,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2310820922255516,
      "num_tokens": 83637646.0,
      "step": 40305
    },
    {
      "entropy": 5.648510265350342,
      "epoch": 3.656567489114659,
      "grad_norm": 1.25,
      "learning_rate": 0.00037088113747464955,
      "loss": 4.9057,
      "mean_token_accuracy": 0.22455067485570906,
      "num_tokens": 83647807.0,
      "step": 40310
    },
    {
      "entropy": 5.658698081970215,
      "epoch": 3.657021044992743,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000370851864989717,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21397348046302794,
      "num_tokens": 83658002.0,
      "step": 40315
    },
    {
      "entropy": 5.629558563232422,
      "epoch": 3.6574746008708274,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003708225905225435,
      "loss": 4.8635,
      "mean_token_accuracy": 0.23341089189052583,
      "num_tokens": 83667875.0,
      "step": 40320
    },
    {
      "entropy": 5.6465513706207275,
      "epoch": 3.6579281567489117,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003707933140737345,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2244977906346321,
      "num_tokens": 83678376.0,
      "step": 40325
    },
    {
      "entropy": 5.675090217590332,
      "epoch": 3.6583817126269955,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037076403564389516,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2119140088558197,
      "num_tokens": 83688318.0,
      "step": 40330
    },
    {
      "entropy": 5.6525876998901365,
      "epoch": 3.65883526850508,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003707347552336313,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2170547917485237,
      "num_tokens": 83698935.0,
      "step": 40335
    },
    {
      "entropy": 5.684643268585205,
      "epoch": 3.659288824383164,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003707054728435482,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21953823864459993,
      "num_tokens": 83709382.0,
      "step": 40340
    },
    {
      "entropy": 5.627509021759034,
      "epoch": 3.6597423802612483,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003706761884742516,
      "loss": 4.81,
      "mean_token_accuracy": 0.23386919349431992,
      "num_tokens": 83719661.0,
      "step": 40345
    },
    {
      "entropy": 5.763510084152221,
      "epoch": 3.660195936139332,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037064690212634707,
      "loss": 4.9724,
      "mean_token_accuracy": 0.2139635980129242,
      "num_tokens": 83730679.0,
      "step": 40350
    },
    {
      "entropy": 5.58924617767334,
      "epoch": 3.6606494920174164,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003706176138004401,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22291076332330703,
      "num_tokens": 83740595.0,
      "step": 40355
    },
    {
      "entropy": 5.609420347213745,
      "epoch": 3.6611030478955007,
      "grad_norm": 1.25,
      "learning_rate": 0.00037058832349713667,
      "loss": 4.958,
      "mean_token_accuracy": 0.21693361401557923,
      "num_tokens": 83751605.0,
      "step": 40360
    },
    {
      "entropy": 5.71450662612915,
      "epoch": 3.661556603773585,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003705590312170422,
      "loss": 5.0041,
      "mean_token_accuracy": 0.21410494595766066,
      "num_tokens": 83763780.0,
      "step": 40365
    },
    {
      "entropy": 5.648506307601929,
      "epoch": 3.662010159651669,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037052973696076274,
      "loss": 4.8402,
      "mean_token_accuracy": 0.23000605702400206,
      "num_tokens": 83773800.0,
      "step": 40370
    },
    {
      "entropy": 5.716043519973755,
      "epoch": 3.6624637155297535,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037050044072890395,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20661567896604538,
      "num_tokens": 83784402.0,
      "step": 40375
    },
    {
      "entropy": 5.627768611907959,
      "epoch": 3.6629172714078373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003704711425220717,
      "loss": 4.8629,
      "mean_token_accuracy": 0.21692775338888168,
      "num_tokens": 83794852.0,
      "step": 40380
    },
    {
      "entropy": 5.673155164718628,
      "epoch": 3.6633708272859216,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003704418423408718,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2184833139181137,
      "num_tokens": 83805673.0,
      "step": 40385
    },
    {
      "entropy": 5.649004077911377,
      "epoch": 3.663824383164006,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003704125401859104,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2189710557460785,
      "num_tokens": 83815779.0,
      "step": 40390
    },
    {
      "entropy": 5.671184062957764,
      "epoch": 3.66427793904209,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037038323605779333,
      "loss": 4.9096,
      "mean_token_accuracy": 0.22570633590221406,
      "num_tokens": 83826804.0,
      "step": 40395
    },
    {
      "entropy": 5.680573463439941,
      "epoch": 3.664731494920174,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003703539299571266,
      "loss": 4.9218,
      "mean_token_accuracy": 0.22130713760852813,
      "num_tokens": 83836500.0,
      "step": 40400
    },
    {
      "entropy": 5.682557439804077,
      "epoch": 3.665185050798258,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003703246218845163,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2224021315574646,
      "num_tokens": 83846647.0,
      "step": 40405
    },
    {
      "entropy": 5.595215797424316,
      "epoch": 3.6656386066763424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003702953118405686,
      "loss": 4.8513,
      "mean_token_accuracy": 0.22574926912784576,
      "num_tokens": 83858088.0,
      "step": 40410
    },
    {
      "entropy": 5.673348903656006,
      "epoch": 3.6660921625544267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003702659998258895,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2200544461607933,
      "num_tokens": 83868936.0,
      "step": 40415
    },
    {
      "entropy": 5.739882135391236,
      "epoch": 3.666545718432511,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037023668584108523,
      "loss": 5.0251,
      "mean_token_accuracy": 0.20719054192304612,
      "num_tokens": 83879813.0,
      "step": 40420
    },
    {
      "entropy": 5.606545877456665,
      "epoch": 3.6669992743105952,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037020736988676206,
      "loss": 4.8575,
      "mean_token_accuracy": 0.21960405111312867,
      "num_tokens": 83891619.0,
      "step": 40425
    },
    {
      "entropy": 5.722917604446411,
      "epoch": 3.6674528301886795,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037017805196352623,
      "loss": 4.9924,
      "mean_token_accuracy": 0.21911616027355194,
      "num_tokens": 83901905.0,
      "step": 40430
    },
    {
      "entropy": 5.625254201889038,
      "epoch": 3.6679063860667633,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000370148732071984,
      "loss": 4.8196,
      "mean_token_accuracy": 0.2237980470061302,
      "num_tokens": 83912827.0,
      "step": 40435
    },
    {
      "entropy": 5.5971778392791744,
      "epoch": 3.6683599419448476,
      "grad_norm": 1.25,
      "learning_rate": 0.0003701194102127418,
      "loss": 4.8096,
      "mean_token_accuracy": 0.23318384289741517,
      "num_tokens": 83923719.0,
      "step": 40440
    },
    {
      "entropy": 5.539970636367798,
      "epoch": 3.668813497822932,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003700900863864059,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22861595153808595,
      "num_tokens": 83933715.0,
      "step": 40445
    },
    {
      "entropy": 5.668402290344238,
      "epoch": 3.6692670537010157,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003700607605935828,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21911519765853882,
      "num_tokens": 83943946.0,
      "step": 40450
    },
    {
      "entropy": 5.710284471511841,
      "epoch": 3.6697206095791,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003700314328348789,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2219032183289528,
      "num_tokens": 83953414.0,
      "step": 40455
    },
    {
      "entropy": 5.631934976577758,
      "epoch": 3.6701741654571842,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003700021031109008,
      "loss": 4.8656,
      "mean_token_accuracy": 0.2187546521425247,
      "num_tokens": 83963717.0,
      "step": 40460
    },
    {
      "entropy": 5.616801500320435,
      "epoch": 3.6706277213352685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036997277142225506,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2211845651268959,
      "num_tokens": 83973467.0,
      "step": 40465
    },
    {
      "entropy": 5.637853670120239,
      "epoch": 3.6710812772133528,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036994343776954806,
      "loss": 4.9159,
      "mean_token_accuracy": 0.22129099667072297,
      "num_tokens": 83984013.0,
      "step": 40470
    },
    {
      "entropy": 5.646801471710205,
      "epoch": 3.671534833091437,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003699141021533867,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2273121953010559,
      "num_tokens": 83994349.0,
      "step": 40475
    },
    {
      "entropy": 5.622933959960937,
      "epoch": 3.6719883889695213,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003698847645743775,
      "loss": 4.8092,
      "mean_token_accuracy": 0.21950105279684068,
      "num_tokens": 84004525.0,
      "step": 40480
    },
    {
      "entropy": 5.612362957000732,
      "epoch": 3.672441944847605,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003698554250331271,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21936506181955337,
      "num_tokens": 84014286.0,
      "step": 40485
    },
    {
      "entropy": 5.599745082855224,
      "epoch": 3.6728955007256894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003698260835302424,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22524648159742355,
      "num_tokens": 84024826.0,
      "step": 40490
    },
    {
      "entropy": 5.679998874664307,
      "epoch": 3.6733490566037736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003697967400663302,
      "loss": 4.912,
      "mean_token_accuracy": 0.219159072637558,
      "num_tokens": 84035726.0,
      "step": 40495
    },
    {
      "entropy": 5.60289421081543,
      "epoch": 3.673802612481858,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036976739464199715,
      "loss": 4.7954,
      "mean_token_accuracy": 0.22538332641124725,
      "num_tokens": 84045573.0,
      "step": 40500
    },
    {
      "entropy": 5.6782999515533445,
      "epoch": 3.6742561683599417,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036973804725785024,
      "loss": 4.8997,
      "mean_token_accuracy": 0.21584745347499848,
      "num_tokens": 84055983.0,
      "step": 40505
    },
    {
      "entropy": 5.703087425231933,
      "epoch": 3.674709724238026,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036970869791449644,
      "loss": 4.9807,
      "mean_token_accuracy": 0.219372983276844,
      "num_tokens": 84066905.0,
      "step": 40510
    },
    {
      "entropy": 5.609340000152588,
      "epoch": 3.6751632801161103,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036967934661254263,
      "loss": 4.8019,
      "mean_token_accuracy": 0.234575517475605,
      "num_tokens": 84076861.0,
      "step": 40515
    },
    {
      "entropy": 5.671474361419678,
      "epoch": 3.6756168359941945,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003696499933525957,
      "loss": 4.8625,
      "mean_token_accuracy": 0.2270957425236702,
      "num_tokens": 84086167.0,
      "step": 40520
    },
    {
      "entropy": 5.596030426025391,
      "epoch": 3.676070391872279,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036962063813526287,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2241722896695137,
      "num_tokens": 84095306.0,
      "step": 40525
    },
    {
      "entropy": 5.602426958084107,
      "epoch": 3.676523947750363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003695912809611511,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21810519695281982,
      "num_tokens": 84106505.0,
      "step": 40530
    },
    {
      "entropy": 5.69519681930542,
      "epoch": 3.676977503628447,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003695619218308676,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2240533635020256,
      "num_tokens": 84116393.0,
      "step": 40535
    },
    {
      "entropy": 5.655912303924561,
      "epoch": 3.677431059506531,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036953256074501937,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2188106283545494,
      "num_tokens": 84125890.0,
      "step": 40540
    },
    {
      "entropy": 5.672696495056153,
      "epoch": 3.6778846153846154,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003695031977042138,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2242170676589012,
      "num_tokens": 84137736.0,
      "step": 40545
    },
    {
      "entropy": 5.63420524597168,
      "epoch": 3.6783381712626997,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036947383270905795,
      "loss": 4.9287,
      "mean_token_accuracy": 0.22295952886343,
      "num_tokens": 84148974.0,
      "step": 40550
    },
    {
      "entropy": 5.650039434432983,
      "epoch": 3.6787917271407835,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036944446576015914,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21645791977643966,
      "num_tokens": 84159041.0,
      "step": 40555
    },
    {
      "entropy": 5.616870880126953,
      "epoch": 3.6792452830188678,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036941509685812476,
      "loss": 4.8108,
      "mean_token_accuracy": 0.228526309132576,
      "num_tokens": 84169432.0,
      "step": 40560
    },
    {
      "entropy": 5.617069959640503,
      "epoch": 3.679698838896952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036938572600356213,
      "loss": 4.8546,
      "mean_token_accuracy": 0.21958910375833512,
      "num_tokens": 84179964.0,
      "step": 40565
    },
    {
      "entropy": 5.640071535110474,
      "epoch": 3.6801523947750363,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036935635319707856,
      "loss": 4.9314,
      "mean_token_accuracy": 0.2178154930472374,
      "num_tokens": 84189496.0,
      "step": 40570
    },
    {
      "entropy": 5.641722106933594,
      "epoch": 3.6806059506531206,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003693269784392816,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22978959679603578,
      "num_tokens": 84200396.0,
      "step": 40575
    },
    {
      "entropy": 5.590877389907837,
      "epoch": 3.681059506531205,
      "grad_norm": 1.5,
      "learning_rate": 0.00036929760173077863,
      "loss": 4.8264,
      "mean_token_accuracy": 0.22643833309412004,
      "num_tokens": 84210114.0,
      "step": 40580
    },
    {
      "entropy": 5.639663124084473,
      "epoch": 3.6815130624092887,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003692682230721773,
      "loss": 4.869,
      "mean_token_accuracy": 0.2270136922597885,
      "num_tokens": 84220472.0,
      "step": 40585
    },
    {
      "entropy": 5.664499092102051,
      "epoch": 3.681966618287373,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003692388424640851,
      "loss": 4.92,
      "mean_token_accuracy": 0.2198185846209526,
      "num_tokens": 84230718.0,
      "step": 40590
    },
    {
      "entropy": 5.685943126678467,
      "epoch": 3.682420174165457,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036920945990710964,
      "loss": 4.9009,
      "mean_token_accuracy": 0.2216021552681923,
      "num_tokens": 84241736.0,
      "step": 40595
    },
    {
      "entropy": 5.546687650680542,
      "epoch": 3.6828737300435415,
      "grad_norm": 1.25,
      "learning_rate": 0.00036918007540185846,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2299594685435295,
      "num_tokens": 84251770.0,
      "step": 40600
    },
    {
      "entropy": 5.652056455612183,
      "epoch": 3.6833272859216253,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003691506889489394,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2200792133808136,
      "num_tokens": 84263355.0,
      "step": 40605
    },
    {
      "entropy": 5.6606934547424315,
      "epoch": 3.6837808417997095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036912130054896005,
      "loss": 4.9661,
      "mean_token_accuracy": 0.21096354573965073,
      "num_tokens": 84274835.0,
      "step": 40610
    },
    {
      "entropy": 5.7257383346557615,
      "epoch": 3.684234397677794,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036909191020252825,
      "loss": 4.9946,
      "mean_token_accuracy": 0.21024842858314513,
      "num_tokens": 84286099.0,
      "step": 40615
    },
    {
      "entropy": 5.701558589935303,
      "epoch": 3.684687953555878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003690625179102517,
      "loss": 4.9598,
      "mean_token_accuracy": 0.222568841278553,
      "num_tokens": 84296168.0,
      "step": 40620
    },
    {
      "entropy": 5.767237567901612,
      "epoch": 3.6851415094339623,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003690331236727384,
      "loss": 5.0209,
      "mean_token_accuracy": 0.20820657461881636,
      "num_tokens": 84307352.0,
      "step": 40625
    },
    {
      "entropy": 5.671647262573242,
      "epoch": 3.6855950653120466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003690037274905961,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21867855489253998,
      "num_tokens": 84318496.0,
      "step": 40630
    },
    {
      "entropy": 5.59604959487915,
      "epoch": 3.686048621190131,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003689743293644327,
      "loss": 4.8455,
      "mean_token_accuracy": 0.2231295496225357,
      "num_tokens": 84329101.0,
      "step": 40635
    },
    {
      "entropy": 5.60073299407959,
      "epoch": 3.6865021770682147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003689449292948563,
      "loss": 4.8666,
      "mean_token_accuracy": 0.22625523507595063,
      "num_tokens": 84339143.0,
      "step": 40640
    },
    {
      "entropy": 5.697188520431519,
      "epoch": 3.686955732946299,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036891552728247483,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21939027756452562,
      "num_tokens": 84349693.0,
      "step": 40645
    },
    {
      "entropy": 5.692017459869385,
      "epoch": 3.6874092888243832,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036888612332789636,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21815599650144576,
      "num_tokens": 84359799.0,
      "step": 40650
    },
    {
      "entropy": 5.6600888729095455,
      "epoch": 3.6878628447024675,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036885671743172875,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22099415957927704,
      "num_tokens": 84370695.0,
      "step": 40655
    },
    {
      "entropy": 5.643673467636108,
      "epoch": 3.6883164005805513,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003688273095945805,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21444448083639145,
      "num_tokens": 84380205.0,
      "step": 40660
    },
    {
      "entropy": 5.670331954956055,
      "epoch": 3.6887699564586356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036879789981705954,
      "loss": 4.9002,
      "mean_token_accuracy": 0.22777147591114044,
      "num_tokens": 84391034.0,
      "step": 40665
    },
    {
      "entropy": 5.713764095306397,
      "epoch": 3.68922351233672,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036876848809977404,
      "loss": 4.9783,
      "mean_token_accuracy": 0.2094938188791275,
      "num_tokens": 84401452.0,
      "step": 40670
    },
    {
      "entropy": 5.621707773208618,
      "epoch": 3.689677068214804,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036873907444333234,
      "loss": 4.8322,
      "mean_token_accuracy": 0.22599519193172454,
      "num_tokens": 84411126.0,
      "step": 40675
    },
    {
      "entropy": 5.659043836593628,
      "epoch": 3.6901306240928884,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036870965884834274,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22305284291505814,
      "num_tokens": 84422740.0,
      "step": 40680
    },
    {
      "entropy": 5.653793621063232,
      "epoch": 3.6905841799709727,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036868024131541337,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2174427926540375,
      "num_tokens": 84432831.0,
      "step": 40685
    },
    {
      "entropy": 5.703464508056641,
      "epoch": 3.6910377358490565,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036865082184515296,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22220373898744583,
      "num_tokens": 84443176.0,
      "step": 40690
    },
    {
      "entropy": 5.716496276855469,
      "epoch": 3.6914912917271407,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036862140043816957,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21485623866319656,
      "num_tokens": 84453185.0,
      "step": 40695
    },
    {
      "entropy": 5.6939490795135494,
      "epoch": 3.691944847605225,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003685919770950718,
      "loss": 4.8893,
      "mean_token_accuracy": 0.22202895432710648,
      "num_tokens": 84464303.0,
      "step": 40700
    },
    {
      "entropy": 5.671566581726074,
      "epoch": 3.6923984034833093,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036856255181646797,
      "loss": 4.9214,
      "mean_token_accuracy": 0.2199340984225273,
      "num_tokens": 84474463.0,
      "step": 40705
    },
    {
      "entropy": 5.623949384689331,
      "epoch": 3.692851959361393,
      "grad_norm": 1.25,
      "learning_rate": 0.0003685331246029668,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22054630070924758,
      "num_tokens": 84484682.0,
      "step": 40710
    },
    {
      "entropy": 5.663768720626831,
      "epoch": 3.6933055152394774,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036850369545517685,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2293693780899048,
      "num_tokens": 84494830.0,
      "step": 40715
    },
    {
      "entropy": 5.617914724349975,
      "epoch": 3.6937590711175616,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003684742643737065,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2238822638988495,
      "num_tokens": 84505449.0,
      "step": 40720
    },
    {
      "entropy": 5.698879337310791,
      "epoch": 3.694212626995646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036844483135916467,
      "loss": 4.9536,
      "mean_token_accuracy": 0.21640323400497435,
      "num_tokens": 84515833.0,
      "step": 40725
    },
    {
      "entropy": 5.69665560722351,
      "epoch": 3.69466618287373,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003684153964121599,
      "loss": 4.9538,
      "mean_token_accuracy": 0.21794192343950272,
      "num_tokens": 84526677.0,
      "step": 40730
    },
    {
      "entropy": 5.642237329483033,
      "epoch": 3.6951197387518144,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036838595953330095,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22644708156585694,
      "num_tokens": 84536656.0,
      "step": 40735
    },
    {
      "entropy": 5.6743004322052,
      "epoch": 3.6955732946298983,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036835652072319646,
      "loss": 4.8349,
      "mean_token_accuracy": 0.22813600599765776,
      "num_tokens": 84546506.0,
      "step": 40740
    },
    {
      "entropy": 5.677061223983765,
      "epoch": 3.6960268505079825,
      "grad_norm": 1.25,
      "learning_rate": 0.0003683270799824554,
      "loss": 4.9527,
      "mean_token_accuracy": 0.22106551826000215,
      "num_tokens": 84556921.0,
      "step": 40745
    },
    {
      "entropy": 5.6613450050354,
      "epoch": 3.696480406386067,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003682976373116865,
      "loss": 4.886,
      "mean_token_accuracy": 0.2261745363473892,
      "num_tokens": 84566325.0,
      "step": 40750
    },
    {
      "entropy": 5.607352590560913,
      "epoch": 3.696933962264151,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036826819271149877,
      "loss": 4.8543,
      "mean_token_accuracy": 0.22172658294439315,
      "num_tokens": 84577167.0,
      "step": 40755
    },
    {
      "entropy": 5.617634248733521,
      "epoch": 3.697387518142235,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036823874618250105,
      "loss": 4.8654,
      "mean_token_accuracy": 0.21913614869117737,
      "num_tokens": 84586764.0,
      "step": 40760
    },
    {
      "entropy": 5.60825924873352,
      "epoch": 3.697841074020319,
      "grad_norm": 1.25,
      "learning_rate": 0.00036820929772530224,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2206513375043869,
      "num_tokens": 84598903.0,
      "step": 40765
    },
    {
      "entropy": 5.630033016204834,
      "epoch": 3.6982946298984034,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003681798473405114,
      "loss": 4.8388,
      "mean_token_accuracy": 0.22656822353601455,
      "num_tokens": 84608695.0,
      "step": 40770
    },
    {
      "entropy": 5.584740781784058,
      "epoch": 3.6987481857764877,
      "grad_norm": 1.25,
      "learning_rate": 0.00036815039502873756,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22289512306451797,
      "num_tokens": 84619377.0,
      "step": 40775
    },
    {
      "entropy": 5.61126298904419,
      "epoch": 3.699201741654572,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036812094079058985,
      "loss": 4.8394,
      "mean_token_accuracy": 0.21988830715417862,
      "num_tokens": 84629716.0,
      "step": 40780
    },
    {
      "entropy": 5.5905218601226805,
      "epoch": 3.699655297532656,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003680914846266773,
      "loss": 4.7794,
      "mean_token_accuracy": 0.23450044095516204,
      "num_tokens": 84640214.0,
      "step": 40785
    },
    {
      "entropy": 5.663218450546265,
      "epoch": 3.7001088534107405,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003680620265376092,
      "loss": 4.9678,
      "mean_token_accuracy": 0.21970319896936416,
      "num_tokens": 84650403.0,
      "step": 40790
    },
    {
      "entropy": 5.652617311477661,
      "epoch": 3.7005624092888243,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036803256652399456,
      "loss": 4.8534,
      "mean_token_accuracy": 0.22251996397972107,
      "num_tokens": 84660912.0,
      "step": 40795
    },
    {
      "entropy": 5.661609745025634,
      "epoch": 3.7010159651669086,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036800310458644285,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21118397265672684,
      "num_tokens": 84672194.0,
      "step": 40800
    },
    {
      "entropy": 5.681054401397705,
      "epoch": 3.701469521044993,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036797364072556316,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21239921152591706,
      "num_tokens": 84681850.0,
      "step": 40805
    },
    {
      "entropy": 5.685785484313965,
      "epoch": 3.7019230769230766,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000367944174941965,
      "loss": 4.8975,
      "mean_token_accuracy": 0.2270340785384178,
      "num_tokens": 84691271.0,
      "step": 40810
    },
    {
      "entropy": 5.606936073303222,
      "epoch": 3.702376632801161,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036791470723625746,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21239540725946426,
      "num_tokens": 84702189.0,
      "step": 40815
    },
    {
      "entropy": 5.6548752784729,
      "epoch": 3.702830188679245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003678852376090502,
      "loss": 4.8096,
      "mean_token_accuracy": 0.23135502338409425,
      "num_tokens": 84712399.0,
      "step": 40820
    },
    {
      "entropy": 5.691664934158325,
      "epoch": 3.7032837445573294,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003678557660609525,
      "loss": 5.0124,
      "mean_token_accuracy": 0.21444610953330995,
      "num_tokens": 84723174.0,
      "step": 40825
    },
    {
      "entropy": 5.6580400466918945,
      "epoch": 3.7037373004354137,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000367826292592574,
      "loss": 4.8737,
      "mean_token_accuracy": 0.230885548889637,
      "num_tokens": 84732498.0,
      "step": 40830
    },
    {
      "entropy": 5.679837846755982,
      "epoch": 3.704190856313498,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003677968172045241,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2171540766954422,
      "num_tokens": 84742362.0,
      "step": 40835
    },
    {
      "entropy": 5.642715358734131,
      "epoch": 3.7046444121915822,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003677673398974122,
      "loss": 4.904,
      "mean_token_accuracy": 0.22081730961799623,
      "num_tokens": 84753379.0,
      "step": 40840
    },
    {
      "entropy": 5.69362325668335,
      "epoch": 3.705097968069666,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003677378606718482,
      "loss": 4.9904,
      "mean_token_accuracy": 0.2137149155139923,
      "num_tokens": 84763245.0,
      "step": 40845
    },
    {
      "entropy": 5.667326545715332,
      "epoch": 3.7055515239477503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003677083795284416,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21358240991830826,
      "num_tokens": 84773929.0,
      "step": 40850
    },
    {
      "entropy": 5.728064107894897,
      "epoch": 3.7060050798258346,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036767889646780204,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21249008625745774,
      "num_tokens": 84784733.0,
      "step": 40855
    },
    {
      "entropy": 5.663216829299927,
      "epoch": 3.706458635703919,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003676494114905394,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21417641639709473,
      "num_tokens": 84794254.0,
      "step": 40860
    },
    {
      "entropy": 5.705105829238891,
      "epoch": 3.7069121915820027,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036761992459726325,
      "loss": 4.9004,
      "mean_token_accuracy": 0.21380765736103058,
      "num_tokens": 84804361.0,
      "step": 40865
    },
    {
      "entropy": 5.648979663848877,
      "epoch": 3.707365747460087,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036759043578858344,
      "loss": 4.89,
      "mean_token_accuracy": 0.2239452138543129,
      "num_tokens": 84814497.0,
      "step": 40870
    },
    {
      "entropy": 5.688341093063355,
      "epoch": 3.7078193033381712,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036756094506510984,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22103255838155747,
      "num_tokens": 84826871.0,
      "step": 40875
    },
    {
      "entropy": 5.666487121582032,
      "epoch": 3.7082728592162555,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036753145242745226,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22331540137529374,
      "num_tokens": 84837053.0,
      "step": 40880
    },
    {
      "entropy": 5.564202928543091,
      "epoch": 3.7087264150943398,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003675019578762208,
      "loss": 4.8367,
      "mean_token_accuracy": 0.2255967691540718,
      "num_tokens": 84846681.0,
      "step": 40885
    },
    {
      "entropy": 5.595720052719116,
      "epoch": 3.709179970972424,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003674724614120251,
      "loss": 4.8211,
      "mean_token_accuracy": 0.2218451529741287,
      "num_tokens": 84856651.0,
      "step": 40890
    },
    {
      "entropy": 5.6585493087768555,
      "epoch": 3.709633526850508,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036744296303547547,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21268591433763503,
      "num_tokens": 84867531.0,
      "step": 40895
    },
    {
      "entropy": 5.624187850952149,
      "epoch": 3.710087082728592,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003674134627471818,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2233106791973114,
      "num_tokens": 84879132.0,
      "step": 40900
    },
    {
      "entropy": 5.660195302963257,
      "epoch": 3.7105406386066764,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000367383960547754,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22991647124290465,
      "num_tokens": 84888873.0,
      "step": 40905
    },
    {
      "entropy": 5.641703701019287,
      "epoch": 3.7109941944847606,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036735445643780256,
      "loss": 4.928,
      "mean_token_accuracy": 0.21705437898635865,
      "num_tokens": 84898926.0,
      "step": 40910
    },
    {
      "entropy": 5.639666366577148,
      "epoch": 3.7114477503628445,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003673249504179374,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22094259560108184,
      "num_tokens": 84908502.0,
      "step": 40915
    },
    {
      "entropy": 5.652897644042969,
      "epoch": 3.7119013062409287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003672954424887687,
      "loss": 4.904,
      "mean_token_accuracy": 0.2173845425248146,
      "num_tokens": 84919681.0,
      "step": 40920
    },
    {
      "entropy": 5.5717450141906735,
      "epoch": 3.712354862119013,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003672659326509066,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22530381232500077,
      "num_tokens": 84930149.0,
      "step": 40925
    },
    {
      "entropy": 5.5976927280426025,
      "epoch": 3.7128084179970973,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036723642090496166,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2250242367386818,
      "num_tokens": 84940086.0,
      "step": 40930
    },
    {
      "entropy": 5.705440855026245,
      "epoch": 3.7132619738751815,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036720690725154404,
      "loss": 4.933,
      "mean_token_accuracy": 0.22080663591623306,
      "num_tokens": 84949248.0,
      "step": 40935
    },
    {
      "entropy": 5.633121633529663,
      "epoch": 3.713715529753266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036717739169126395,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22151263058185577,
      "num_tokens": 84960892.0,
      "step": 40940
    },
    {
      "entropy": 5.641955423355102,
      "epoch": 3.7141690856313496,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036714787422473206,
      "loss": 4.892,
      "mean_token_accuracy": 0.2165308937430382,
      "num_tokens": 84971024.0,
      "step": 40945
    },
    {
      "entropy": 5.598126220703125,
      "epoch": 3.714622641509434,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003671183548525586,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22717070132493972,
      "num_tokens": 84980945.0,
      "step": 40950
    },
    {
      "entropy": 5.671842193603515,
      "epoch": 3.715076197387518,
      "grad_norm": 1.25,
      "learning_rate": 0.000367088833575354,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21502626687288284,
      "num_tokens": 84992055.0,
      "step": 40955
    },
    {
      "entropy": 5.67706332206726,
      "epoch": 3.7155297532656024,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036705931039372904,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22320076823234558,
      "num_tokens": 85002619.0,
      "step": 40960
    },
    {
      "entropy": 5.673063468933106,
      "epoch": 3.7159833091436862,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003670297853082939,
      "loss": 4.8952,
      "mean_token_accuracy": 0.22195317447185517,
      "num_tokens": 85013317.0,
      "step": 40965
    },
    {
      "entropy": 5.670144891738891,
      "epoch": 3.7164368650217705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036700025831965937,
      "loss": 4.8743,
      "mean_token_accuracy": 0.22089784741401672,
      "num_tokens": 85023457.0,
      "step": 40970
    },
    {
      "entropy": 5.651346683502197,
      "epoch": 3.7168904208998548,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036697072942843616,
      "loss": 4.8801,
      "mean_token_accuracy": 0.21573939472436904,
      "num_tokens": 85033498.0,
      "step": 40975
    },
    {
      "entropy": 5.76431360244751,
      "epoch": 3.717343976777939,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003669411986352348,
      "loss": 5.0889,
      "mean_token_accuracy": 0.21457834392786027,
      "num_tokens": 85043687.0,
      "step": 40980
    },
    {
      "entropy": 5.689324522018433,
      "epoch": 3.7177975326560233,
      "grad_norm": 1.0625,
      "learning_rate": 0.000366911665940666,
      "loss": 4.9419,
      "mean_token_accuracy": 0.21958806365728378,
      "num_tokens": 85055045.0,
      "step": 40985
    },
    {
      "entropy": 5.653619813919067,
      "epoch": 3.7182510885341076,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036688213134534044,
      "loss": 4.8518,
      "mean_token_accuracy": 0.23805569857358932,
      "num_tokens": 85064625.0,
      "step": 40990
    },
    {
      "entropy": 5.739334964752198,
      "epoch": 3.718704644412192,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003668525948498691,
      "loss": 5.0565,
      "mean_token_accuracy": 0.20217333137989044,
      "num_tokens": 85075249.0,
      "step": 40995
    },
    {
      "entropy": 5.657452583312988,
      "epoch": 3.7191582002902757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036682305645486266,
      "loss": 4.896,
      "mean_token_accuracy": 0.22502252757549285,
      "num_tokens": 85085900.0,
      "step": 41000
    },
    {
      "entropy": 5.664230537414551,
      "epoch": 3.71961175616836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036679351616093194,
      "loss": 4.9421,
      "mean_token_accuracy": 0.22187916785478592,
      "num_tokens": 85095772.0,
      "step": 41005
    },
    {
      "entropy": 5.638325023651123,
      "epoch": 3.720065312046444,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036676397396868797,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22622692584991455,
      "num_tokens": 85105653.0,
      "step": 41010
    },
    {
      "entropy": 5.716484880447387,
      "epoch": 3.7205188679245285,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003667344298787417,
      "loss": 4.9685,
      "mean_token_accuracy": 0.2144452527165413,
      "num_tokens": 85115432.0,
      "step": 41015
    },
    {
      "entropy": 5.729081010818481,
      "epoch": 3.7209724238026123,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036670488389170395,
      "loss": 5.0824,
      "mean_token_accuracy": 0.20987560600042343,
      "num_tokens": 85126572.0,
      "step": 41020
    },
    {
      "entropy": 5.587710952758789,
      "epoch": 3.7214259796806965,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036667533600818586,
      "loss": 4.844,
      "mean_token_accuracy": 0.22681321948766708,
      "num_tokens": 85136809.0,
      "step": 41025
    },
    {
      "entropy": 5.850367546081543,
      "epoch": 3.721879535558781,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003666457862287985,
      "loss": 5.0696,
      "mean_token_accuracy": 0.21007951498031616,
      "num_tokens": 85148346.0,
      "step": 41030
    },
    {
      "entropy": 5.661215829849243,
      "epoch": 3.722333091436865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036661623455415296,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21990771293640138,
      "num_tokens": 85158807.0,
      "step": 41035
    },
    {
      "entropy": 5.688000917434692,
      "epoch": 3.7227866473149493,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003665866809848602,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20994648486375808,
      "num_tokens": 85170752.0,
      "step": 41040
    },
    {
      "entropy": 5.625526237487793,
      "epoch": 3.7232402031930336,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036655712552153164,
      "loss": 4.9025,
      "mean_token_accuracy": 0.21547161787748337,
      "num_tokens": 85180779.0,
      "step": 41045
    },
    {
      "entropy": 5.685053062438965,
      "epoch": 3.7236937590711174,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003665275681647784,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22153379768133163,
      "num_tokens": 85191391.0,
      "step": 41050
    },
    {
      "entropy": 5.693622303009033,
      "epoch": 3.7241473149492017,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003664980089152116,
      "loss": 4.8891,
      "mean_token_accuracy": 0.2162279024720192,
      "num_tokens": 85201357.0,
      "step": 41055
    },
    {
      "entropy": 5.6578082084655765,
      "epoch": 3.724600870827286,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036646844777344284,
      "loss": 4.9137,
      "mean_token_accuracy": 0.2227029785513878,
      "num_tokens": 85212193.0,
      "step": 41060
    },
    {
      "entropy": 5.664473676681519,
      "epoch": 3.7250544267053702,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036643888474008313,
      "loss": 4.9283,
      "mean_token_accuracy": 0.2193174734711647,
      "num_tokens": 85223809.0,
      "step": 41065
    },
    {
      "entropy": 5.651267194747925,
      "epoch": 3.725507982583454,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003664093198157441,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22259458750486374,
      "num_tokens": 85234992.0,
      "step": 41070
    },
    {
      "entropy": 5.728595781326294,
      "epoch": 3.7259615384615383,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003663797530010369,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2119491770863533,
      "num_tokens": 85244157.0,
      "step": 41075
    },
    {
      "entropy": 5.613103628158569,
      "epoch": 3.7264150943396226,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003663501842965732,
      "loss": 4.867,
      "mean_token_accuracy": 0.22966155111789704,
      "num_tokens": 85255946.0,
      "step": 41080
    },
    {
      "entropy": 5.610699987411499,
      "epoch": 3.726868650217707,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003663206137029645,
      "loss": 4.868,
      "mean_token_accuracy": 0.22856574654579162,
      "num_tokens": 85265273.0,
      "step": 41085
    },
    {
      "entropy": 5.573185014724731,
      "epoch": 3.727322206095791,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036629104122082215,
      "loss": 4.7211,
      "mean_token_accuracy": 0.2303519532084465,
      "num_tokens": 85275794.0,
      "step": 41090
    },
    {
      "entropy": 5.6412163257598875,
      "epoch": 3.7277757619738754,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036626146685075775,
      "loss": 4.9906,
      "mean_token_accuracy": 0.2108025535941124,
      "num_tokens": 85287334.0,
      "step": 41095
    },
    {
      "entropy": 5.67722520828247,
      "epoch": 3.728229317851959,
      "grad_norm": 1.3125,
      "learning_rate": 0.000366231890593383,
      "loss": 4.9124,
      "mean_token_accuracy": 0.2216177687048912,
      "num_tokens": 85297320.0,
      "step": 41100
    },
    {
      "entropy": 5.601094150543213,
      "epoch": 3.7286828737300435,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003662023124493095,
      "loss": 4.8274,
      "mean_token_accuracy": 0.2279417484998703,
      "num_tokens": 85308071.0,
      "step": 41105
    },
    {
      "entropy": 5.68747844696045,
      "epoch": 3.7291364296081277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000366172732419149,
      "loss": 4.9257,
      "mean_token_accuracy": 0.21675894409418106,
      "num_tokens": 85317790.0,
      "step": 41110
    },
    {
      "entropy": 5.7154566764831545,
      "epoch": 3.729589985486212,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036614315050351317,
      "loss": 4.984,
      "mean_token_accuracy": 0.21242806017398835,
      "num_tokens": 85329008.0,
      "step": 41115
    },
    {
      "entropy": 5.7604231357574465,
      "epoch": 3.730043541364296,
      "grad_norm": 1.25,
      "learning_rate": 0.0003661135667030137,
      "loss": 4.9941,
      "mean_token_accuracy": 0.21173521876335144,
      "num_tokens": 85339095.0,
      "step": 41120
    },
    {
      "entropy": 5.699007987976074,
      "epoch": 3.73049709724238,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003660839810182624,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21794274002313613,
      "num_tokens": 85349833.0,
      "step": 41125
    },
    {
      "entropy": 5.721101188659668,
      "epoch": 3.7309506531204644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036605439344987125,
      "loss": 4.9711,
      "mean_token_accuracy": 0.21819814294576645,
      "num_tokens": 85359512.0,
      "step": 41130
    },
    {
      "entropy": 5.70348858833313,
      "epoch": 3.7314042089985486,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000366024803998452,
      "loss": 4.9481,
      "mean_token_accuracy": 0.224931900203228,
      "num_tokens": 85370053.0,
      "step": 41135
    },
    {
      "entropy": 5.659979963302613,
      "epoch": 3.731857764876633,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036599521266461653,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22135833352804185,
      "num_tokens": 85380020.0,
      "step": 41140
    },
    {
      "entropy": 5.663234090805053,
      "epoch": 3.732311320754717,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036596561944897703,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22230454683303832,
      "num_tokens": 85390232.0,
      "step": 41145
    },
    {
      "entropy": 5.6056296825408936,
      "epoch": 3.7327648766328014,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003659360243521453,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23131534904241563,
      "num_tokens": 85400401.0,
      "step": 41150
    },
    {
      "entropy": 5.6432029724121096,
      "epoch": 3.7332184325108853,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003659064273747334,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21713484227657318,
      "num_tokens": 85411749.0,
      "step": 41155
    },
    {
      "entropy": 5.706897354125976,
      "epoch": 3.7336719883889695,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036587682851735336,
      "loss": 4.9946,
      "mean_token_accuracy": 0.2157077744603157,
      "num_tokens": 85424065.0,
      "step": 41160
    },
    {
      "entropy": 5.707071161270141,
      "epoch": 3.734125544267054,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036584722778061743,
      "loss": 4.9301,
      "mean_token_accuracy": 0.22047340571880342,
      "num_tokens": 85434412.0,
      "step": 41165
    },
    {
      "entropy": 5.6794806003570555,
      "epoch": 3.7345791001451376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003658176251651376,
      "loss": 4.9031,
      "mean_token_accuracy": 0.2253032073378563,
      "num_tokens": 85444489.0,
      "step": 41170
    },
    {
      "entropy": 5.6987415790557865,
      "epoch": 3.735032656023222,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036578802067152615,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22232849448919295,
      "num_tokens": 85454556.0,
      "step": 41175
    },
    {
      "entropy": 5.59938268661499,
      "epoch": 3.735486211901306,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003657584143003954,
      "loss": 4.8317,
      "mean_token_accuracy": 0.2245876133441925,
      "num_tokens": 85464701.0,
      "step": 41180
    },
    {
      "entropy": 5.653997182846069,
      "epoch": 3.7359397677793904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003657288060523574,
      "loss": 4.9656,
      "mean_token_accuracy": 0.22018109858036042,
      "num_tokens": 85475657.0,
      "step": 41185
    },
    {
      "entropy": 5.709797286987305,
      "epoch": 3.7363933236574747,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036569919592802466,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22522761374711991,
      "num_tokens": 85486358.0,
      "step": 41190
    },
    {
      "entropy": 5.742954969406128,
      "epoch": 3.736846879535559,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003656695839280093,
      "loss": 4.9078,
      "mean_token_accuracy": 0.21693163514137268,
      "num_tokens": 85497913.0,
      "step": 41195
    },
    {
      "entropy": 5.653684663772583,
      "epoch": 3.737300435413643,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036563997005292396,
      "loss": 4.9034,
      "mean_token_accuracy": 0.22670360803604125,
      "num_tokens": 85508314.0,
      "step": 41200
    },
    {
      "entropy": 5.603851556777954,
      "epoch": 3.737753991291727,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036561035430338084,
      "loss": 4.8967,
      "mean_token_accuracy": 0.21857466250658036,
      "num_tokens": 85518142.0,
      "step": 41205
    },
    {
      "entropy": 5.656486845016479,
      "epoch": 3.7382075471698113,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036558073667999257,
      "loss": 4.784,
      "mean_token_accuracy": 0.22903891950845717,
      "num_tokens": 85526854.0,
      "step": 41210
    },
    {
      "entropy": 5.594369220733642,
      "epoch": 3.7386611030478956,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036555111718337166,
      "loss": 4.7946,
      "mean_token_accuracy": 0.2322963684797287,
      "num_tokens": 85537153.0,
      "step": 41215
    },
    {
      "entropy": 5.708690786361695,
      "epoch": 3.73911465892598,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003655214958141305,
      "loss": 4.9969,
      "mean_token_accuracy": 0.21064232140779496,
      "num_tokens": 85548583.0,
      "step": 41220
    },
    {
      "entropy": 5.6763382911682125,
      "epoch": 3.7395682148040637,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036549187257288164,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2300586834549904,
      "num_tokens": 85559228.0,
      "step": 41225
    },
    {
      "entropy": 5.707980489730835,
      "epoch": 3.740021770682148,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036546224746023795,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22223663926124573,
      "num_tokens": 85569757.0,
      "step": 41230
    },
    {
      "entropy": 5.649372529983521,
      "epoch": 3.740475326560232,
      "grad_norm": 1.125,
      "learning_rate": 0.00036543262047681186,
      "loss": 4.8983,
      "mean_token_accuracy": 0.23110807985067366,
      "num_tokens": 85582013.0,
      "step": 41235
    },
    {
      "entropy": 5.652613115310669,
      "epoch": 3.7409288824383164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003654029916232161,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2358803018927574,
      "num_tokens": 85594049.0,
      "step": 41240
    },
    {
      "entropy": 5.665416145324707,
      "epoch": 3.7413824383164007,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036537336090006347,
      "loss": 4.922,
      "mean_token_accuracy": 0.21921512335538865,
      "num_tokens": 85605253.0,
      "step": 41245
    },
    {
      "entropy": 5.620599222183228,
      "epoch": 3.741835994194485,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036534372830796674,
      "loss": 4.8463,
      "mean_token_accuracy": 0.22598191499710082,
      "num_tokens": 85615032.0,
      "step": 41250
    },
    {
      "entropy": 5.65076756477356,
      "epoch": 3.742289550072569,
      "grad_norm": 1.25,
      "learning_rate": 0.0003653140938475386,
      "loss": 4.881,
      "mean_token_accuracy": 0.23063501119613647,
      "num_tokens": 85625497.0,
      "step": 41255
    },
    {
      "entropy": 5.659329271316528,
      "epoch": 3.742743105950653,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036528445751939206,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21988602578639985,
      "num_tokens": 85636058.0,
      "step": 41260
    },
    {
      "entropy": 5.658809995651245,
      "epoch": 3.7431966618287373,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036525481932413997,
      "loss": 4.918,
      "mean_token_accuracy": 0.216049624979496,
      "num_tokens": 85646505.0,
      "step": 41265
    },
    {
      "entropy": 5.718615961074829,
      "epoch": 3.7436502177068216,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036522517926239513,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2189261719584465,
      "num_tokens": 85656864.0,
      "step": 41270
    },
    {
      "entropy": 5.6087257862091064,
      "epoch": 3.7441037735849054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036519553733477055,
      "loss": 4.8361,
      "mean_token_accuracy": 0.2275415152311325,
      "num_tokens": 85666783.0,
      "step": 41275
    },
    {
      "entropy": 5.552579784393311,
      "epoch": 3.7445573294629897,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003651658935418794,
      "loss": 4.7818,
      "mean_token_accuracy": 0.2315801203250885,
      "num_tokens": 85677102.0,
      "step": 41280
    },
    {
      "entropy": 5.592778873443604,
      "epoch": 3.745010885341074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036513624788433454,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23277545720338821,
      "num_tokens": 85687048.0,
      "step": 41285
    },
    {
      "entropy": 5.626360464096069,
      "epoch": 3.7454644412191582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003651066003627491,
      "loss": 4.8936,
      "mean_token_accuracy": 0.21562793254852294,
      "num_tokens": 85698314.0,
      "step": 41290
    },
    {
      "entropy": 5.711394214630127,
      "epoch": 3.7459179970972425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003650769509777362,
      "loss": 4.9573,
      "mean_token_accuracy": 0.21696987748146057,
      "num_tokens": 85708039.0,
      "step": 41295
    },
    {
      "entropy": 5.645800971984864,
      "epoch": 3.7463715529753268,
      "grad_norm": 1.484375,
      "learning_rate": 0.00036504729972990904,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23273114264011383,
      "num_tokens": 85717741.0,
      "step": 41300
    },
    {
      "entropy": 5.683948278427124,
      "epoch": 3.7468251088534106,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003650176466198807,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22189702838659286,
      "num_tokens": 85728286.0,
      "step": 41305
    },
    {
      "entropy": 5.649855327606201,
      "epoch": 3.747278664731495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036498799164826453,
      "loss": 4.8684,
      "mean_token_accuracy": 0.224978768825531,
      "num_tokens": 85738970.0,
      "step": 41310
    },
    {
      "entropy": 5.607178974151611,
      "epoch": 3.747732220609579,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036495833481567375,
      "loss": 4.8389,
      "mean_token_accuracy": 0.22328358590602876,
      "num_tokens": 85749284.0,
      "step": 41315
    },
    {
      "entropy": 5.694136428833008,
      "epoch": 3.7481857764876634,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036492867612272176,
      "loss": 4.9487,
      "mean_token_accuracy": 0.2162918210029602,
      "num_tokens": 85759595.0,
      "step": 41320
    },
    {
      "entropy": 5.728177928924561,
      "epoch": 3.748639332365747,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003648990155700217,
      "loss": 5.0014,
      "mean_token_accuracy": 0.21251268386840821,
      "num_tokens": 85771049.0,
      "step": 41325
    },
    {
      "entropy": 5.670376825332641,
      "epoch": 3.7490928882438315,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036486935315818713,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2176561549305916,
      "num_tokens": 85781656.0,
      "step": 41330
    },
    {
      "entropy": 5.579206848144532,
      "epoch": 3.7495464441219157,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036483968888783154,
      "loss": 4.7526,
      "mean_token_accuracy": 0.2277541771531105,
      "num_tokens": 85790922.0,
      "step": 41335
    },
    {
      "entropy": 5.634003162384033,
      "epoch": 3.75,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036481002275956813,
      "loss": 4.8398,
      "mean_token_accuracy": 0.22941983342170716,
      "num_tokens": 85799928.0,
      "step": 41340
    },
    {
      "entropy": 5.667712116241455,
      "epoch": 3.7504535558780843,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003647803547740106,
      "loss": 4.943,
      "mean_token_accuracy": 0.21469739377498626,
      "num_tokens": 85810594.0,
      "step": 41345
    },
    {
      "entropy": 5.636851453781128,
      "epoch": 3.7509071117561685,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036475068493177254,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21709463149309158,
      "num_tokens": 85820690.0,
      "step": 41350
    },
    {
      "entropy": 5.6454047679901125,
      "epoch": 3.751360667634253,
      "grad_norm": 1.25,
      "learning_rate": 0.0003647210132334674,
      "loss": 4.851,
      "mean_token_accuracy": 0.22203511595726014,
      "num_tokens": 85830314.0,
      "step": 41355
    },
    {
      "entropy": 5.652485418319702,
      "epoch": 3.7518142235123366,
      "grad_norm": 1.25,
      "learning_rate": 0.0003646913396797088,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20690085887908935,
      "num_tokens": 85841099.0,
      "step": 41360
    },
    {
      "entropy": 5.707526826858521,
      "epoch": 3.752267779390421,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036466166427111046,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2213337689638138,
      "num_tokens": 85852813.0,
      "step": 41365
    },
    {
      "entropy": 5.665508699417114,
      "epoch": 3.752721335268505,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036463198700828616,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22168927043676376,
      "num_tokens": 85862778.0,
      "step": 41370
    },
    {
      "entropy": 5.616978883743286,
      "epoch": 3.7531748911465894,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003646023078918494,
      "loss": 4.921,
      "mean_token_accuracy": 0.222724349796772,
      "num_tokens": 85872729.0,
      "step": 41375
    },
    {
      "entropy": 5.659740447998047,
      "epoch": 3.7536284470246732,
      "grad_norm": 1.25,
      "learning_rate": 0.0003645726269224141,
      "loss": 4.9018,
      "mean_token_accuracy": 0.22810726314783097,
      "num_tokens": 85883289.0,
      "step": 41380
    },
    {
      "entropy": 5.580375623703003,
      "epoch": 3.7540820029027575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036454294410059407,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22975561618804932,
      "num_tokens": 85892318.0,
      "step": 41385
    },
    {
      "entropy": 5.601224660873413,
      "epoch": 3.7545355587808418,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003645132594270032,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22175518572330474,
      "num_tokens": 85902719.0,
      "step": 41390
    },
    {
      "entropy": 5.624606275558472,
      "epoch": 3.754989114658926,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036448357290225523,
      "loss": 4.8629,
      "mean_token_accuracy": 0.21994414627552034,
      "num_tokens": 85913271.0,
      "step": 41395
    },
    {
      "entropy": 5.588603734970093,
      "epoch": 3.7554426705370103,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003644538845269642,
      "loss": 4.7926,
      "mean_token_accuracy": 0.2242666721343994,
      "num_tokens": 85922397.0,
      "step": 41400
    },
    {
      "entropy": 5.67548303604126,
      "epoch": 3.7558962264150946,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000364424194301744,
      "loss": 4.928,
      "mean_token_accuracy": 0.2190377652645111,
      "num_tokens": 85933261.0,
      "step": 41405
    },
    {
      "entropy": 5.698658609390259,
      "epoch": 3.7563497822931784,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003643945022272087,
      "loss": 4.9563,
      "mean_token_accuracy": 0.221746763586998,
      "num_tokens": 85943860.0,
      "step": 41410
    },
    {
      "entropy": 5.687657260894776,
      "epoch": 3.7568033381712627,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003643648083039724,
      "loss": 4.9582,
      "mean_token_accuracy": 0.2167433977127075,
      "num_tokens": 85953984.0,
      "step": 41415
    },
    {
      "entropy": 5.673469400405883,
      "epoch": 3.757256894049347,
      "grad_norm": 1.328125,
      "learning_rate": 0.000364335112532649,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2184562787413597,
      "num_tokens": 85964771.0,
      "step": 41420
    },
    {
      "entropy": 5.677586269378662,
      "epoch": 3.757710449927431,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003643054149138527,
      "loss": 4.9413,
      "mean_token_accuracy": 0.22036302238702773,
      "num_tokens": 85975859.0,
      "step": 41425
    },
    {
      "entropy": 5.755876684188843,
      "epoch": 3.758164005805515,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036427571544819765,
      "loss": 4.9778,
      "mean_token_accuracy": 0.2161107450723648,
      "num_tokens": 85987022.0,
      "step": 41430
    },
    {
      "entropy": 5.611931705474854,
      "epoch": 3.7586175616835993,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003642460141362981,
      "loss": 4.838,
      "mean_token_accuracy": 0.22265738248825073,
      "num_tokens": 85997056.0,
      "step": 41435
    },
    {
      "entropy": 5.672608661651611,
      "epoch": 3.7590711175616836,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036421631097876823,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21579169929027558,
      "num_tokens": 86008853.0,
      "step": 41440
    },
    {
      "entropy": 5.573115348815918,
      "epoch": 3.759524673439768,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003641866059762223,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21726031601428986,
      "num_tokens": 86020358.0,
      "step": 41445
    },
    {
      "entropy": 5.662569046020508,
      "epoch": 3.759978229317852,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003641568991292747,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21406012922525405,
      "num_tokens": 86030706.0,
      "step": 41450
    },
    {
      "entropy": 5.730230045318604,
      "epoch": 3.7604317851959363,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036412719043853956,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21748934388160707,
      "num_tokens": 86041964.0,
      "step": 41455
    },
    {
      "entropy": 5.690208196640015,
      "epoch": 3.76088534107402,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036409747990463146,
      "loss": 4.9176,
      "mean_token_accuracy": 0.22287604659795762,
      "num_tokens": 86052213.0,
      "step": 41460
    },
    {
      "entropy": 5.619532918930053,
      "epoch": 3.7613388969521044,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003640677675281649,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21440352350473404,
      "num_tokens": 86061419.0,
      "step": 41465
    },
    {
      "entropy": 5.715872955322266,
      "epoch": 3.7617924528301887,
      "grad_norm": 1.375,
      "learning_rate": 0.0003640380533097541,
      "loss": 4.91,
      "mean_token_accuracy": 0.2179046869277954,
      "num_tokens": 86071585.0,
      "step": 41470
    },
    {
      "entropy": 5.674550628662109,
      "epoch": 3.762246008708273,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036400833725001364,
      "loss": 4.8957,
      "mean_token_accuracy": 0.21819409877061843,
      "num_tokens": 86081175.0,
      "step": 41475
    },
    {
      "entropy": 5.658315801620484,
      "epoch": 3.762699564586357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003639786193495581,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21548353731632233,
      "num_tokens": 86092407.0,
      "step": 41480
    },
    {
      "entropy": 5.723626661300659,
      "epoch": 3.763153120464441,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036394889960900215,
      "loss": 4.9427,
      "mean_token_accuracy": 0.21736511886119841,
      "num_tokens": 86102218.0,
      "step": 41485
    },
    {
      "entropy": 5.734164619445801,
      "epoch": 3.7636066763425253,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003639191780289602,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22272278666496276,
      "num_tokens": 86112631.0,
      "step": 41490
    },
    {
      "entropy": 5.700323009490967,
      "epoch": 3.7640602322206096,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003638894546100469,
      "loss": 4.9654,
      "mean_token_accuracy": 0.20971276760101318,
      "num_tokens": 86123675.0,
      "step": 41495
    },
    {
      "entropy": 5.590743112564087,
      "epoch": 3.764513788098694,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036385972935287715,
      "loss": 4.8425,
      "mean_token_accuracy": 0.222859163582325,
      "num_tokens": 86133909.0,
      "step": 41500
    },
    {
      "entropy": 5.65754976272583,
      "epoch": 3.764967343976778,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003638300022580656,
      "loss": 4.9293,
      "mean_token_accuracy": 0.21580233424901962,
      "num_tokens": 86144560.0,
      "step": 41505
    },
    {
      "entropy": 5.755287218093872,
      "epoch": 3.7654208998548624,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036380027332622684,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21797743290662766,
      "num_tokens": 86155140.0,
      "step": 41510
    },
    {
      "entropy": 5.713537216186523,
      "epoch": 3.765874455732946,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036377054255797584,
      "loss": 4.9351,
      "mean_token_accuracy": 0.2162645310163498,
      "num_tokens": 86165555.0,
      "step": 41515
    },
    {
      "entropy": 5.670260858535767,
      "epoch": 3.7663280116110305,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036374080995392737,
      "loss": 4.897,
      "mean_token_accuracy": 0.22103025317192077,
      "num_tokens": 86176204.0,
      "step": 41520
    },
    {
      "entropy": 5.681659317016601,
      "epoch": 3.7667815674891147,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003637110755146964,
      "loss": 4.9827,
      "mean_token_accuracy": 0.2189554899930954,
      "num_tokens": 86186888.0,
      "step": 41525
    },
    {
      "entropy": 5.6775728225708,
      "epoch": 3.7672351233671986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036368133924089766,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22728827893733977,
      "num_tokens": 86196654.0,
      "step": 41530
    },
    {
      "entropy": 5.649123334884644,
      "epoch": 3.767688679245283,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003636516011331464,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2250275954604149,
      "num_tokens": 86207045.0,
      "step": 41535
    },
    {
      "entropy": 5.693004417419433,
      "epoch": 3.768142235123367,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036362186119205734,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21175480037927627,
      "num_tokens": 86216906.0,
      "step": 41540
    },
    {
      "entropy": 5.618576431274414,
      "epoch": 3.7685957910014514,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003635921194182457,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2223455309867859,
      "num_tokens": 86227380.0,
      "step": 41545
    },
    {
      "entropy": 5.6646894931793215,
      "epoch": 3.7690493468795356,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003635623758123263,
      "loss": 5.0194,
      "mean_token_accuracy": 0.22061097025871276,
      "num_tokens": 86236948.0,
      "step": 41550
    },
    {
      "entropy": 5.676881885528564,
      "epoch": 3.76950290275762,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036353263037491445,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21786512285470963,
      "num_tokens": 86246784.0,
      "step": 41555
    },
    {
      "entropy": 5.631447172164917,
      "epoch": 3.769956458635704,
      "grad_norm": 1.25,
      "learning_rate": 0.0003635028831066253,
      "loss": 4.837,
      "mean_token_accuracy": 0.2246741160750389,
      "num_tokens": 86257381.0,
      "step": 41560
    },
    {
      "entropy": 5.611772966384888,
      "epoch": 3.770410014513788,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003634731340080739,
      "loss": 4.85,
      "mean_token_accuracy": 0.22241034358739853,
      "num_tokens": 86268117.0,
      "step": 41565
    },
    {
      "entropy": 5.7618732929229735,
      "epoch": 3.7708635703918723,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003634433830798756,
      "loss": 5.0387,
      "mean_token_accuracy": 0.21439319401979445,
      "num_tokens": 86279195.0,
      "step": 41570
    },
    {
      "entropy": 5.574959754943848,
      "epoch": 3.7713171262699565,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003634136303226457,
      "loss": 4.7418,
      "mean_token_accuracy": 0.23419877290725707,
      "num_tokens": 86288154.0,
      "step": 41575
    },
    {
      "entropy": 5.659147024154663,
      "epoch": 3.771770682148041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003633838757369992,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21908959299325942,
      "num_tokens": 86297524.0,
      "step": 41580
    },
    {
      "entropy": 5.652503728866577,
      "epoch": 3.7722242380261246,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003633541193235517,
      "loss": 4.9802,
      "mean_token_accuracy": 0.21118232756853103,
      "num_tokens": 86307031.0,
      "step": 41585
    },
    {
      "entropy": 5.637276840209961,
      "epoch": 3.772677793904209,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003633243610829185,
      "loss": 4.9387,
      "mean_token_accuracy": 0.21906437277793883,
      "num_tokens": 86317629.0,
      "step": 41590
    },
    {
      "entropy": 5.6781778812408445,
      "epoch": 3.773131349782293,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036329460101571496,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21658796221017837,
      "num_tokens": 86328243.0,
      "step": 41595
    },
    {
      "entropy": 5.757309007644653,
      "epoch": 3.7735849056603774,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003632648391225568,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2148237019777298,
      "num_tokens": 86338905.0,
      "step": 41600
    },
    {
      "entropy": 5.680874967575074,
      "epoch": 3.7740384615384617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036323507540405903,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22035324573516846,
      "num_tokens": 86350433.0,
      "step": 41605
    },
    {
      "entropy": 5.6957090377807615,
      "epoch": 3.774492017416546,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003632053098608374,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2112775757908821,
      "num_tokens": 86360664.0,
      "step": 41610
    },
    {
      "entropy": 5.683890295028687,
      "epoch": 3.7749455732946298,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036317554249350767,
      "loss": 4.972,
      "mean_token_accuracy": 0.22332140058279037,
      "num_tokens": 86372101.0,
      "step": 41615
    },
    {
      "entropy": 5.66197657585144,
      "epoch": 3.775399129172714,
      "grad_norm": 1.125,
      "learning_rate": 0.0003631457733026851,
      "loss": 4.817,
      "mean_token_accuracy": 0.2249531030654907,
      "num_tokens": 86382494.0,
      "step": 41620
    },
    {
      "entropy": 5.764450597763061,
      "epoch": 3.7758526850507983,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036311600228898554,
      "loss": 5.0635,
      "mean_token_accuracy": 0.20831287354230882,
      "num_tokens": 86393292.0,
      "step": 41625
    },
    {
      "entropy": 5.5949243068695065,
      "epoch": 3.7763062409288826,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036308622945302455,
      "loss": 4.837,
      "mean_token_accuracy": 0.21961106061935426,
      "num_tokens": 86402940.0,
      "step": 41630
    },
    {
      "entropy": 5.595632410049438,
      "epoch": 3.7767597968069664,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000363056454795418,
      "loss": 4.8665,
      "mean_token_accuracy": 0.21814990639686585,
      "num_tokens": 86413382.0,
      "step": 41635
    },
    {
      "entropy": 5.69773497581482,
      "epoch": 3.7772133526850507,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003630266783167814,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21711234897375106,
      "num_tokens": 86423866.0,
      "step": 41640
    },
    {
      "entropy": 5.7516913414001465,
      "epoch": 3.777666908563135,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003629969000177307,
      "loss": 5.0128,
      "mean_token_accuracy": 0.21043222546577453,
      "num_tokens": 86434908.0,
      "step": 41645
    },
    {
      "entropy": 5.560062265396118,
      "epoch": 3.778120464441219,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003629671198988817,
      "loss": 4.8308,
      "mean_token_accuracy": 0.23035694360733033,
      "num_tokens": 86445604.0,
      "step": 41650
    },
    {
      "entropy": 5.58459701538086,
      "epoch": 3.7785740203193035,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003629373379608502,
      "loss": 4.7759,
      "mean_token_accuracy": 0.23566643595695497,
      "num_tokens": 86455813.0,
      "step": 41655
    },
    {
      "entropy": 5.559374570846558,
      "epoch": 3.7790275761973877,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036290755420425214,
      "loss": 4.7502,
      "mean_token_accuracy": 0.23715393990278244,
      "num_tokens": 86465256.0,
      "step": 41660
    },
    {
      "entropy": 5.628467178344726,
      "epoch": 3.7794811320754715,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036287776862970356,
      "loss": 4.8824,
      "mean_token_accuracy": 0.22413792759180068,
      "num_tokens": 86474605.0,
      "step": 41665
    },
    {
      "entropy": 5.671857595443726,
      "epoch": 3.779934687953556,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036284798123782024,
      "loss": 4.8962,
      "mean_token_accuracy": 0.22032371461391448,
      "num_tokens": 86484886.0,
      "step": 41670
    },
    {
      "entropy": 5.624969005584717,
      "epoch": 3.78038824383164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003628181920292182,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21418100744485855,
      "num_tokens": 86495816.0,
      "step": 41675
    },
    {
      "entropy": 5.632235431671143,
      "epoch": 3.7808417997097243,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003627884010045138,
      "loss": 4.8622,
      "mean_token_accuracy": 0.22076003402471542,
      "num_tokens": 86506319.0,
      "step": 41680
    },
    {
      "entropy": 5.716420412063599,
      "epoch": 3.781295355587808,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003627586081643227,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21986618638038635,
      "num_tokens": 86516952.0,
      "step": 41685
    },
    {
      "entropy": 5.71786470413208,
      "epoch": 3.7817489114658924,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036272881350926135,
      "loss": 5.0165,
      "mean_token_accuracy": 0.21247045546770096,
      "num_tokens": 86527327.0,
      "step": 41690
    },
    {
      "entropy": 5.6497101306915285,
      "epoch": 3.7822024673439767,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003626990170399457,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22317089140415192,
      "num_tokens": 86538281.0,
      "step": 41695
    },
    {
      "entropy": 5.675183296203613,
      "epoch": 3.782656023222061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036266921875699205,
      "loss": 4.9733,
      "mean_token_accuracy": 0.21008704006671905,
      "num_tokens": 86549034.0,
      "step": 41700
    },
    {
      "entropy": 5.6818126201629635,
      "epoch": 3.7831095791001452,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003626394186610166,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21852248311042785,
      "num_tokens": 86559306.0,
      "step": 41705
    },
    {
      "entropy": 5.730512619018555,
      "epoch": 3.7835631349782295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003626096167526356,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21690935641527176,
      "num_tokens": 86569414.0,
      "step": 41710
    },
    {
      "entropy": 5.557330083847046,
      "epoch": 3.7840166908563138,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003625798130324655,
      "loss": 4.7823,
      "mean_token_accuracy": 0.2280845656991005,
      "num_tokens": 86578554.0,
      "step": 41715
    },
    {
      "entropy": 5.6679039478302,
      "epoch": 3.7844702467343976,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036255000750112256,
      "loss": 4.8968,
      "mean_token_accuracy": 0.2222454845905304,
      "num_tokens": 86588624.0,
      "step": 41720
    },
    {
      "entropy": 5.717407751083374,
      "epoch": 3.784923802612482,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036252020015922304,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21759973019361495,
      "num_tokens": 86599266.0,
      "step": 41725
    },
    {
      "entropy": 5.726531982421875,
      "epoch": 3.785377358490566,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036249039100738357,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20948874354362487,
      "num_tokens": 86609853.0,
      "step": 41730
    },
    {
      "entropy": 5.58441195487976,
      "epoch": 3.7858309143686504,
      "grad_norm": 1.25,
      "learning_rate": 0.0003624605800462205,
      "loss": 4.8637,
      "mean_token_accuracy": 0.2207888185977936,
      "num_tokens": 86619927.0,
      "step": 41735
    },
    {
      "entropy": 5.710479736328125,
      "epoch": 3.786284470246734,
      "grad_norm": 1.375,
      "learning_rate": 0.00036243076727635027,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2182391494512558,
      "num_tokens": 86630079.0,
      "step": 41740
    },
    {
      "entropy": 5.670481777191162,
      "epoch": 3.7867380261248185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036240095269838955,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22344246208667756,
      "num_tokens": 86639744.0,
      "step": 41745
    },
    {
      "entropy": 5.626803636550903,
      "epoch": 3.7871915820029027,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036237113631295485,
      "loss": 4.9574,
      "mean_token_accuracy": 0.21240945160388947,
      "num_tokens": 86650866.0,
      "step": 41750
    },
    {
      "entropy": 5.632354879379273,
      "epoch": 3.787645137880987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036234131812066286,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2288187637925148,
      "num_tokens": 86659864.0,
      "step": 41755
    },
    {
      "entropy": 5.583402919769287,
      "epoch": 3.7880986937590713,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036231149812213006,
      "loss": 4.8275,
      "mean_token_accuracy": 0.22766820192337037,
      "num_tokens": 86670039.0,
      "step": 41760
    },
    {
      "entropy": 5.62681941986084,
      "epoch": 3.7885522496371555,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003622816763179732,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22567262798547744,
      "num_tokens": 86680434.0,
      "step": 41765
    },
    {
      "entropy": 5.676084518432617,
      "epoch": 3.7890058055152394,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003622518527088091,
      "loss": 4.909,
      "mean_token_accuracy": 0.21579646021127702,
      "num_tokens": 86691026.0,
      "step": 41770
    },
    {
      "entropy": 5.701051568984985,
      "epoch": 3.7894593613933236,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036222202729525453,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21668500155210496,
      "num_tokens": 86701323.0,
      "step": 41775
    },
    {
      "entropy": 5.6415246486663815,
      "epoch": 3.789912917271408,
      "grad_norm": 1.203125,
      "learning_rate": 0.000362192200077926,
      "loss": 4.852,
      "mean_token_accuracy": 0.23028887510299684,
      "num_tokens": 86711152.0,
      "step": 41780
    },
    {
      "entropy": 5.700291538238526,
      "epoch": 3.790366473149492,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036216237105744076,
      "loss": 5.0067,
      "mean_token_accuracy": 0.220675265789032,
      "num_tokens": 86720814.0,
      "step": 41785
    },
    {
      "entropy": 5.67734055519104,
      "epoch": 3.790820029027576,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003621325402344154,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2294454827904701,
      "num_tokens": 86731229.0,
      "step": 41790
    },
    {
      "entropy": 5.644534921646118,
      "epoch": 3.7912735849056602,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036210270760946684,
      "loss": 4.8778,
      "mean_token_accuracy": 0.21905871629714965,
      "num_tokens": 86741440.0,
      "step": 41795
    },
    {
      "entropy": 5.730336475372314,
      "epoch": 3.7917271407837445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003620728731832122,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2130439653992653,
      "num_tokens": 86752022.0,
      "step": 41800
    },
    {
      "entropy": 5.710660171508789,
      "epoch": 3.7921806966618288,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003620430369562683,
      "loss": 4.8631,
      "mean_token_accuracy": 0.21816670000553132,
      "num_tokens": 86763090.0,
      "step": 41805
    },
    {
      "entropy": 5.650964021682739,
      "epoch": 3.792634252539913,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036201319892925214,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22272211015224458,
      "num_tokens": 86773357.0,
      "step": 41810
    },
    {
      "entropy": 5.626104068756104,
      "epoch": 3.7930878084179973,
      "grad_norm": 1.328125,
      "learning_rate": 0.000361983359102781,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22403311729431152,
      "num_tokens": 86784402.0,
      "step": 41815
    },
    {
      "entropy": 5.678772974014282,
      "epoch": 3.793541364296081,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036195351747747177,
      "loss": 4.9386,
      "mean_token_accuracy": 0.2217594116926193,
      "num_tokens": 86794945.0,
      "step": 41820
    },
    {
      "entropy": 5.651041793823242,
      "epoch": 3.7939949201741654,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036192367405394157,
      "loss": 4.8993,
      "mean_token_accuracy": 0.2199644699692726,
      "num_tokens": 86806936.0,
      "step": 41825
    },
    {
      "entropy": 5.737751197814942,
      "epoch": 3.7944484760522497,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003618938288328077,
      "loss": 5.003,
      "mean_token_accuracy": 0.21363526582717896,
      "num_tokens": 86817776.0,
      "step": 41830
    },
    {
      "entropy": 5.613715744018554,
      "epoch": 3.794902031930334,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003618639818146873,
      "loss": 4.7739,
      "mean_token_accuracy": 0.22628486156463623,
      "num_tokens": 86828054.0,
      "step": 41835
    },
    {
      "entropy": 5.6177269458770756,
      "epoch": 3.7953555878084178,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036183413300019767,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22060526460409163,
      "num_tokens": 86838132.0,
      "step": 41840
    },
    {
      "entropy": 5.589739084243774,
      "epoch": 3.795809143686502,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036180428238995595,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21964032500982283,
      "num_tokens": 86848219.0,
      "step": 41845
    },
    {
      "entropy": 5.607968997955322,
      "epoch": 3.7962626995645863,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003617744299845796,
      "loss": 4.8288,
      "mean_token_accuracy": 0.22354236543178557,
      "num_tokens": 86858559.0,
      "step": 41850
    },
    {
      "entropy": 5.620931577682495,
      "epoch": 3.7967162554426706,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003617445757846859,
      "loss": 4.9301,
      "mean_token_accuracy": 0.22050397843122482,
      "num_tokens": 86868573.0,
      "step": 41855
    },
    {
      "entropy": 5.728419017791748,
      "epoch": 3.797169811320755,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036171471979089227,
      "loss": 4.9092,
      "mean_token_accuracy": 0.2237102523446083,
      "num_tokens": 86878641.0,
      "step": 41860
    },
    {
      "entropy": 5.71133828163147,
      "epoch": 3.797623367198839,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003616848620038162,
      "loss": 4.948,
      "mean_token_accuracy": 0.2154051721096039,
      "num_tokens": 86888644.0,
      "step": 41865
    },
    {
      "entropy": 5.737832403182983,
      "epoch": 3.7980769230769234,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000361655002424075,
      "loss": 5.0507,
      "mean_token_accuracy": 0.2099604532122612,
      "num_tokens": 86899428.0,
      "step": 41870
    },
    {
      "entropy": 5.71644082069397,
      "epoch": 3.798530478955007,
      "grad_norm": 1.25,
      "learning_rate": 0.00036162514105228635,
      "loss": 4.9277,
      "mean_token_accuracy": 0.21976384818553923,
      "num_tokens": 86910155.0,
      "step": 41875
    },
    {
      "entropy": 5.626809692382812,
      "epoch": 3.7989840348330914,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003615952778890676,
      "loss": 4.8824,
      "mean_token_accuracy": 0.22089843600988388,
      "num_tokens": 86919783.0,
      "step": 41880
    },
    {
      "entropy": 5.661923933029175,
      "epoch": 3.7994375907111757,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036156541293503653,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21564303934574128,
      "num_tokens": 86929647.0,
      "step": 41885
    },
    {
      "entropy": 5.676453304290772,
      "epoch": 3.7998911465892595,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036153554619081064,
      "loss": 4.865,
      "mean_token_accuracy": 0.22050565481185913,
      "num_tokens": 86940113.0,
      "step": 41890
    },
    {
      "entropy": 5.697487878799438,
      "epoch": 3.800344702467344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003615056776570076,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22345710247755052,
      "num_tokens": 86949950.0,
      "step": 41895
    },
    {
      "entropy": 5.605305290222168,
      "epoch": 3.800798258345428,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036147580733424513,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22983558177948,
      "num_tokens": 86959286.0,
      "step": 41900
    },
    {
      "entropy": 5.629342365264892,
      "epoch": 3.8012518142235123,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003614459352231409,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22396614104509355,
      "num_tokens": 86970165.0,
      "step": 41905
    },
    {
      "entropy": 5.638010835647583,
      "epoch": 3.8017053701015966,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003614160613243127,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22057184129953383,
      "num_tokens": 86980331.0,
      "step": 41910
    },
    {
      "entropy": 5.755900764465332,
      "epoch": 3.802158925979681,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003613861856383783,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21341018676757811,
      "num_tokens": 86991790.0,
      "step": 41915
    },
    {
      "entropy": 5.7179975509643555,
      "epoch": 3.802612481857765,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003613563081659556,
      "loss": 5.0289,
      "mean_token_accuracy": 0.22142690569162368,
      "num_tokens": 87001220.0,
      "step": 41920
    },
    {
      "entropy": 5.681937408447266,
      "epoch": 3.803066037735849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003613264289076625,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22190347462892532,
      "num_tokens": 87012720.0,
      "step": 41925
    },
    {
      "entropy": 5.678753805160523,
      "epoch": 3.803519593613933,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003612965478641167,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21755171120166777,
      "num_tokens": 87021891.0,
      "step": 41930
    },
    {
      "entropy": 5.746618986129761,
      "epoch": 3.8039731494920175,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036126666503593643,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20899759232997894,
      "num_tokens": 87032546.0,
      "step": 41935
    },
    {
      "entropy": 5.675739765167236,
      "epoch": 3.8044267053701017,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003612367804237395,
      "loss": 4.868,
      "mean_token_accuracy": 0.21791332811117173,
      "num_tokens": 87042173.0,
      "step": 41940
    },
    {
      "entropy": 5.702556896209717,
      "epoch": 3.8048802612481856,
      "grad_norm": 1.25,
      "learning_rate": 0.0003612068940281439,
      "loss": 4.9548,
      "mean_token_accuracy": 0.22178468108177185,
      "num_tokens": 87052475.0,
      "step": 41945
    },
    {
      "entropy": 5.666600227355957,
      "epoch": 3.80533381712627,
      "grad_norm": 1.25,
      "learning_rate": 0.0003611770058497679,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21802892088890075,
      "num_tokens": 87062698.0,
      "step": 41950
    },
    {
      "entropy": 5.644330978393555,
      "epoch": 3.805787373004354,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036114711588922945,
      "loss": 4.8914,
      "mean_token_accuracy": 0.20887335687875747,
      "num_tokens": 87073168.0,
      "step": 41955
    },
    {
      "entropy": 5.706921482086182,
      "epoch": 3.8062409288824384,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003611172241471466,
      "loss": 4.981,
      "mean_token_accuracy": 0.21309708058834076,
      "num_tokens": 87084270.0,
      "step": 41960
    },
    {
      "entropy": 5.641605520248413,
      "epoch": 3.8066944847605226,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003610873306241376,
      "loss": 4.9279,
      "mean_token_accuracy": 0.2200206845998764,
      "num_tokens": 87093869.0,
      "step": 41965
    },
    {
      "entropy": 5.635382652282715,
      "epoch": 3.807148040638607,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036105743532082075,
      "loss": 4.8631,
      "mean_token_accuracy": 0.21979954987764358,
      "num_tokens": 87103371.0,
      "step": 41970
    },
    {
      "entropy": 5.677100706100464,
      "epoch": 3.8076015965166907,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003610275382378142,
      "loss": 4.8946,
      "mean_token_accuracy": 0.22230157405138015,
      "num_tokens": 87113384.0,
      "step": 41975
    },
    {
      "entropy": 5.748106241226196,
      "epoch": 3.808055152394775,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003609976393757362,
      "loss": 4.9635,
      "mean_token_accuracy": 0.21906560957431792,
      "num_tokens": 87123925.0,
      "step": 41980
    },
    {
      "entropy": 5.659660196304321,
      "epoch": 3.8085087082728593,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036096773873520513,
      "loss": 4.873,
      "mean_token_accuracy": 0.21941274851560594,
      "num_tokens": 87134608.0,
      "step": 41985
    },
    {
      "entropy": 5.662461137771606,
      "epoch": 3.8089622641509435,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036093783631683927,
      "loss": 4.89,
      "mean_token_accuracy": 0.21935552656650542,
      "num_tokens": 87145184.0,
      "step": 41990
    },
    {
      "entropy": 5.639924192428589,
      "epoch": 3.8094158200290273,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003609079321212571,
      "loss": 4.7957,
      "mean_token_accuracy": 0.22748257219791412,
      "num_tokens": 87155483.0,
      "step": 41995
    },
    {
      "entropy": 5.656929588317871,
      "epoch": 3.8098693759071116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036087802614907696,
      "loss": 4.989,
      "mean_token_accuracy": 0.21253972947597505,
      "num_tokens": 87166615.0,
      "step": 42000
    },
    {
      "epoch": 3.8098693759071116,
      "eval_entropy": 5.384581031950351,
      "eval_loss": 4.967734336853027,
      "eval_mean_token_accuracy": 0.22687854531528723,
      "eval_num_tokens": 87166615.0,
      "eval_runtime": 20.6387,
      "eval_samples_per_second": 1713.285,
      "eval_steps_per_second": 214.161,
      "step": 42000
    },
    {
      "entropy": 5.680204439163208,
      "epoch": 3.810322931785196,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036084811840091736,
      "loss": 4.8975,
      "mean_token_accuracy": 0.2192764014005661,
      "num_tokens": 87177151.0,
      "step": 42005
    },
    {
      "entropy": 5.660609245300293,
      "epoch": 3.81077648766328,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003608182088773969,
      "loss": 4.889,
      "mean_token_accuracy": 0.22212653309106828,
      "num_tokens": 87188085.0,
      "step": 42010
    },
    {
      "entropy": 5.658230924606324,
      "epoch": 3.8112300435413644,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003607882975791338,
      "loss": 4.9598,
      "mean_token_accuracy": 0.21465011090040206,
      "num_tokens": 87197240.0,
      "step": 42015
    },
    {
      "entropy": 5.644859695434571,
      "epoch": 3.8116835994194487,
      "grad_norm": 1.21875,
      "learning_rate": 0.000360758384506747,
      "loss": 4.8793,
      "mean_token_accuracy": 0.228286075592041,
      "num_tokens": 87208064.0,
      "step": 42020
    },
    {
      "entropy": 5.653703308105468,
      "epoch": 3.8121371552975325,
      "grad_norm": 1.25,
      "learning_rate": 0.00036072846966085496,
      "loss": 4.8079,
      "mean_token_accuracy": 0.22089341282844543,
      "num_tokens": 87218386.0,
      "step": 42025
    },
    {
      "entropy": 5.61584358215332,
      "epoch": 3.8125907111756168,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036069855304207634,
      "loss": 4.8014,
      "mean_token_accuracy": 0.22561967074871064,
      "num_tokens": 87228459.0,
      "step": 42030
    },
    {
      "entropy": 5.676889228820801,
      "epoch": 3.813044267053701,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003606686346510297,
      "loss": 4.9929,
      "mean_token_accuracy": 0.21037752330303192,
      "num_tokens": 87240317.0,
      "step": 42035
    },
    {
      "entropy": 5.607076740264892,
      "epoch": 3.8134978229317853,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000360638714488334,
      "loss": 4.7968,
      "mean_token_accuracy": 0.22523918747901917,
      "num_tokens": 87250197.0,
      "step": 42040
    },
    {
      "entropy": 5.61420693397522,
      "epoch": 3.813951378809869,
      "grad_norm": 1.25,
      "learning_rate": 0.00036060879255460766,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2207457348704338,
      "num_tokens": 87261266.0,
      "step": 42045
    },
    {
      "entropy": 5.591328716278076,
      "epoch": 3.8144049346879534,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003605788688504699,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23129039853811265,
      "num_tokens": 87270490.0,
      "step": 42050
    },
    {
      "entropy": 5.662823438644409,
      "epoch": 3.8148584905660377,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003605489433765392,
      "loss": 4.9538,
      "mean_token_accuracy": 0.22233954071998596,
      "num_tokens": 87281260.0,
      "step": 42055
    },
    {
      "entropy": 5.630814409255981,
      "epoch": 3.815312046444122,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003605190161334345,
      "loss": 4.8174,
      "mean_token_accuracy": 0.22971397191286086,
      "num_tokens": 87290577.0,
      "step": 42060
    },
    {
      "entropy": 5.596176338195801,
      "epoch": 3.815765602322206,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003604890871217748,
      "loss": 4.7494,
      "mean_token_accuracy": 0.23253820985555648,
      "num_tokens": 87300185.0,
      "step": 42065
    },
    {
      "entropy": 5.679486799240112,
      "epoch": 3.8162191582002905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003604591563421791,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21951487958431243,
      "num_tokens": 87310088.0,
      "step": 42070
    },
    {
      "entropy": 5.665472793579101,
      "epoch": 3.8166727140783747,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003604292237952662,
      "loss": 4.9221,
      "mean_token_accuracy": 0.2151668593287468,
      "num_tokens": 87319942.0,
      "step": 42075
    },
    {
      "entropy": 5.676095628738404,
      "epoch": 3.8171262699564585,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003603992894816551,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22059175372123718,
      "num_tokens": 87329175.0,
      "step": 42080
    },
    {
      "entropy": 5.585857772827149,
      "epoch": 3.817579825834543,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000360369353401965,
      "loss": 4.8033,
      "mean_token_accuracy": 0.2286943018436432,
      "num_tokens": 87338555.0,
      "step": 42085
    },
    {
      "entropy": 5.6433518409729,
      "epoch": 3.818033381712627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036033941555681486,
      "loss": 4.8635,
      "mean_token_accuracy": 0.23093942701816558,
      "num_tokens": 87348424.0,
      "step": 42090
    },
    {
      "entropy": 5.626125717163086,
      "epoch": 3.8184869375907113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036030947594682395,
      "loss": 4.8088,
      "mean_token_accuracy": 0.22642666101455688,
      "num_tokens": 87359724.0,
      "step": 42095
    },
    {
      "entropy": 5.644570446014404,
      "epoch": 3.818940493468795,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003602795345726113,
      "loss": 4.9697,
      "mean_token_accuracy": 0.2167738288640976,
      "num_tokens": 87371324.0,
      "step": 42100
    },
    {
      "entropy": 5.666535758972168,
      "epoch": 3.8193940493468794,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003602495914347962,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22643609046936036,
      "num_tokens": 87380770.0,
      "step": 42105
    },
    {
      "entropy": 5.637687110900879,
      "epoch": 3.8198476052249637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036021964653399773,
      "loss": 4.9316,
      "mean_token_accuracy": 0.21992146223783493,
      "num_tokens": 87391123.0,
      "step": 42110
    },
    {
      "entropy": 5.687300491333008,
      "epoch": 3.820301161103048,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036018969987083524,
      "loss": 4.9386,
      "mean_token_accuracy": 0.217915178835392,
      "num_tokens": 87401357.0,
      "step": 42115
    },
    {
      "entropy": 5.672944498062134,
      "epoch": 3.8207547169811322,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003601597514459281,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22395591288805008,
      "num_tokens": 87411373.0,
      "step": 42120
    },
    {
      "entropy": 5.622548007965088,
      "epoch": 3.8212082728592165,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036012980125989567,
      "loss": 4.8599,
      "mean_token_accuracy": 0.23098938763141633,
      "num_tokens": 87421915.0,
      "step": 42125
    },
    {
      "entropy": 5.660698080062867,
      "epoch": 3.8216618287373003,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003600998493133571,
      "loss": 4.8916,
      "mean_token_accuracy": 0.22224850803613663,
      "num_tokens": 87432148.0,
      "step": 42130
    },
    {
      "entropy": 5.650674295425415,
      "epoch": 3.8221153846153846,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003600698956069321,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21943789422512056,
      "num_tokens": 87442605.0,
      "step": 42135
    },
    {
      "entropy": 5.660288906097412,
      "epoch": 3.822568940493469,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003600399401412399,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22572477757930756,
      "num_tokens": 87452482.0,
      "step": 42140
    },
    {
      "entropy": 5.573111867904663,
      "epoch": 3.823022496371553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036000998291690005,
      "loss": 4.7588,
      "mean_token_accuracy": 0.22543646544218063,
      "num_tokens": 87462074.0,
      "step": 42145
    },
    {
      "entropy": 5.69090576171875,
      "epoch": 3.823476052249637,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003599800239345321,
      "loss": 4.9031,
      "mean_token_accuracy": 0.22086234837770463,
      "num_tokens": 87472377.0,
      "step": 42150
    },
    {
      "entropy": 5.644957590103149,
      "epoch": 3.823929608127721,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003599500631947557,
      "loss": 4.8868,
      "mean_token_accuracy": 0.22008109092712402,
      "num_tokens": 87483436.0,
      "step": 42155
    },
    {
      "entropy": 5.709145784378052,
      "epoch": 3.8243831640058055,
      "grad_norm": 1.375,
      "learning_rate": 0.00035992010069819025,
      "loss": 4.8721,
      "mean_token_accuracy": 0.22492895126342774,
      "num_tokens": 87492231.0,
      "step": 42160
    },
    {
      "entropy": 5.769653940200806,
      "epoch": 3.8248367198838897,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035989013644545555,
      "loss": 4.9738,
      "mean_token_accuracy": 0.2219487965106964,
      "num_tokens": 87503416.0,
      "step": 42165
    },
    {
      "entropy": 5.642581176757813,
      "epoch": 3.825290275761974,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035986017043717114,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2276483952999115,
      "num_tokens": 87513737.0,
      "step": 42170
    },
    {
      "entropy": 5.695849800109864,
      "epoch": 3.8257438316400583,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003598302026739568,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21451593339443206,
      "num_tokens": 87523532.0,
      "step": 42175
    },
    {
      "entropy": 5.6309166431427,
      "epoch": 3.826197387518142,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035980023315643217,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2308505430817604,
      "num_tokens": 87534066.0,
      "step": 42180
    },
    {
      "entropy": 5.718373584747314,
      "epoch": 3.8266509433962264,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003597702618852172,
      "loss": 5.0245,
      "mean_token_accuracy": 0.20748225301504136,
      "num_tokens": 87544520.0,
      "step": 42185
    },
    {
      "entropy": 5.6563128471374515,
      "epoch": 3.8271044992743106,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003597402888609316,
      "loss": 4.9061,
      "mean_token_accuracy": 0.22385678589344024,
      "num_tokens": 87553913.0,
      "step": 42190
    },
    {
      "entropy": 5.679483699798584,
      "epoch": 3.827558055152395,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035971031408419524,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22020772099494934,
      "num_tokens": 87563434.0,
      "step": 42195
    },
    {
      "entropy": 5.674253368377686,
      "epoch": 3.8280116110304787,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035968033755562795,
      "loss": 4.9175,
      "mean_token_accuracy": 0.22293412834405898,
      "num_tokens": 87573137.0,
      "step": 42200
    },
    {
      "entropy": 5.723353528976441,
      "epoch": 3.828465166908563,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003596503592758497,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22234390676021576,
      "num_tokens": 87582610.0,
      "step": 42205
    },
    {
      "entropy": 5.6488790035247805,
      "epoch": 3.8289187227866472,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003596203792454805,
      "loss": 4.8644,
      "mean_token_accuracy": 0.22284764498472215,
      "num_tokens": 87593062.0,
      "step": 42210
    },
    {
      "entropy": 5.674526309967041,
      "epoch": 3.8293722786647315,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003595903974651403,
      "loss": 4.9626,
      "mean_token_accuracy": 0.22136088013648986,
      "num_tokens": 87604250.0,
      "step": 42215
    },
    {
      "entropy": 5.7232660293579105,
      "epoch": 3.8298258345428158,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003595604139354491,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21748800873756408,
      "num_tokens": 87615434.0,
      "step": 42220
    },
    {
      "entropy": 5.675612068176269,
      "epoch": 3.8302793904209,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000359530428657027,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2279006376862526,
      "num_tokens": 87625005.0,
      "step": 42225
    },
    {
      "entropy": 5.623678255081177,
      "epoch": 3.8307329462989843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003595004416304941,
      "loss": 4.8601,
      "mean_token_accuracy": 0.22245027869939804,
      "num_tokens": 87635486.0,
      "step": 42230
    },
    {
      "entropy": 5.668291807174683,
      "epoch": 3.831186502177068,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035947045285647045,
      "loss": 4.8906,
      "mean_token_accuracy": 0.2254578799009323,
      "num_tokens": 87645771.0,
      "step": 42235
    },
    {
      "entropy": 5.754446744918823,
      "epoch": 3.8316400580551524,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003594404623355765,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2226011261343956,
      "num_tokens": 87656116.0,
      "step": 42240
    },
    {
      "entropy": 5.8198717594146725,
      "epoch": 3.8320936139332367,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003594104700684321,
      "loss": 5.093,
      "mean_token_accuracy": 0.21064186543226243,
      "num_tokens": 87666476.0,
      "step": 42245
    },
    {
      "entropy": 5.6665606021881105,
      "epoch": 3.8325471698113205,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035938047605565764,
      "loss": 4.9014,
      "mean_token_accuracy": 0.23099779933691025,
      "num_tokens": 87677406.0,
      "step": 42250
    },
    {
      "entropy": 5.614005374908447,
      "epoch": 3.8330007256894048,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035935048029787356,
      "loss": 4.8268,
      "mean_token_accuracy": 0.22654626965522767,
      "num_tokens": 87687461.0,
      "step": 42255
    },
    {
      "entropy": 5.650856590270996,
      "epoch": 3.833454281567489,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003593204827957,
      "loss": 4.8506,
      "mean_token_accuracy": 0.2269226223230362,
      "num_tokens": 87696470.0,
      "step": 42260
    },
    {
      "entropy": 5.659826469421387,
      "epoch": 3.8339078374455733,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003592904835497574,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21572404354810715,
      "num_tokens": 87706598.0,
      "step": 42265
    },
    {
      "entropy": 5.769496011734009,
      "epoch": 3.8343613933236576,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003592604825606661,
      "loss": 5.0127,
      "mean_token_accuracy": 0.21079034954309464,
      "num_tokens": 87717749.0,
      "step": 42270
    },
    {
      "entropy": 5.6562755584716795,
      "epoch": 3.834814949201742,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003592304798290465,
      "loss": 4.8361,
      "mean_token_accuracy": 0.22981930524110794,
      "num_tokens": 87727893.0,
      "step": 42275
    },
    {
      "entropy": 5.607091474533081,
      "epoch": 3.835268505079826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003592004753555191,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22092060148715972,
      "num_tokens": 87737980.0,
      "step": 42280
    },
    {
      "entropy": 5.6526471138000485,
      "epoch": 3.83572206095791,
      "grad_norm": 1.25,
      "learning_rate": 0.0003591704691407045,
      "loss": 4.8936,
      "mean_token_accuracy": 0.223111492395401,
      "num_tokens": 87749271.0,
      "step": 42285
    },
    {
      "entropy": 5.649973440170288,
      "epoch": 3.836175616835994,
      "grad_norm": 1.375,
      "learning_rate": 0.00035914046118522313,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22665930539369583,
      "num_tokens": 87760523.0,
      "step": 42290
    },
    {
      "entropy": 5.648125171661377,
      "epoch": 3.8366291727140784,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035911045148969556,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22139427959918975,
      "num_tokens": 87770084.0,
      "step": 42295
    },
    {
      "entropy": 5.736617374420166,
      "epoch": 3.8370827285921627,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035908044005474244,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21465548723936081,
      "num_tokens": 87780163.0,
      "step": 42300
    },
    {
      "entropy": 5.579972600936889,
      "epoch": 3.8375362844702465,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003590504268809843,
      "loss": 4.8322,
      "mean_token_accuracy": 0.2264435574412346,
      "num_tokens": 87791382.0,
      "step": 42305
    },
    {
      "entropy": 5.541245746612549,
      "epoch": 3.837989840348331,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035902041196904204,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22245441973209382,
      "num_tokens": 87802143.0,
      "step": 42310
    },
    {
      "entropy": 5.649176836013794,
      "epoch": 3.838443396226415,
      "grad_norm": 1.375,
      "learning_rate": 0.0003589903953195362,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22229018360376357,
      "num_tokens": 87812255.0,
      "step": 42315
    },
    {
      "entropy": 5.713312864303589,
      "epoch": 3.8388969521044993,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035896037693308757,
      "loss": 4.908,
      "mean_token_accuracy": 0.21780331134796144,
      "num_tokens": 87822744.0,
      "step": 42320
    },
    {
      "entropy": 5.758703374862671,
      "epoch": 3.8393505079825836,
      "grad_norm": 1.390625,
      "learning_rate": 0.000358930356810317,
      "loss": 4.9789,
      "mean_token_accuracy": 0.21666617393493653,
      "num_tokens": 87833701.0,
      "step": 42325
    },
    {
      "entropy": 5.697555065155029,
      "epoch": 3.839804063860668,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003589003349518452,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21528303176164626,
      "num_tokens": 87844215.0,
      "step": 42330
    },
    {
      "entropy": 5.710298538208008,
      "epoch": 3.8402576197387517,
      "grad_norm": 1.25,
      "learning_rate": 0.00035887031135829305,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21628482341766359,
      "num_tokens": 87854281.0,
      "step": 42335
    },
    {
      "entropy": 5.710420608520508,
      "epoch": 3.840711175616836,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003588402860302816,
      "loss": 4.976,
      "mean_token_accuracy": 0.21514319777488708,
      "num_tokens": 87866003.0,
      "step": 42340
    },
    {
      "entropy": 5.681373405456543,
      "epoch": 3.84116473149492,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035881025896843165,
      "loss": 4.8567,
      "mean_token_accuracy": 0.22183520942926407,
      "num_tokens": 87875854.0,
      "step": 42345
    },
    {
      "entropy": 5.726830339431762,
      "epoch": 3.8416182873730045,
      "grad_norm": 1.25,
      "learning_rate": 0.00035878023017336415,
      "loss": 5.0196,
      "mean_token_accuracy": 0.21679281890392305,
      "num_tokens": 87887242.0,
      "step": 42350
    },
    {
      "entropy": 5.711236095428466,
      "epoch": 3.8420718432510883,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003587501996457001,
      "loss": 4.912,
      "mean_token_accuracy": 0.2161334663629532,
      "num_tokens": 87897859.0,
      "step": 42355
    },
    {
      "entropy": 5.659146785736084,
      "epoch": 3.8425253991291726,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035872016738606067,
      "loss": 4.8732,
      "mean_token_accuracy": 0.22292605489492417,
      "num_tokens": 87908945.0,
      "step": 42360
    },
    {
      "entropy": 5.668125963211059,
      "epoch": 3.842978955007257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035869013339506674,
      "loss": 4.8743,
      "mean_token_accuracy": 0.22772708535194397,
      "num_tokens": 87919572.0,
      "step": 42365
    },
    {
      "entropy": 5.69610104560852,
      "epoch": 3.843432510885341,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035866009767333964,
      "loss": 4.9561,
      "mean_token_accuracy": 0.21872315108776091,
      "num_tokens": 87929337.0,
      "step": 42370
    },
    {
      "entropy": 5.728446769714355,
      "epoch": 3.8438860667634254,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035863006022150034,
      "loss": 4.9921,
      "mean_token_accuracy": 0.21361604034900666,
      "num_tokens": 87940513.0,
      "step": 42375
    },
    {
      "entropy": 5.659322309494018,
      "epoch": 3.8443396226415096,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035860002104017014,
      "loss": 4.8422,
      "mean_token_accuracy": 0.22852765172719955,
      "num_tokens": 87951181.0,
      "step": 42380
    },
    {
      "entropy": 5.671182870864868,
      "epoch": 3.8447931785195935,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003585699801299701,
      "loss": 4.8999,
      "mean_token_accuracy": 0.22048140913248063,
      "num_tokens": 87961925.0,
      "step": 42385
    },
    {
      "entropy": 5.642671585083008,
      "epoch": 3.8452467343976777,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035853993749152165,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22414727210998536,
      "num_tokens": 87972233.0,
      "step": 42390
    },
    {
      "entropy": 5.685314893722534,
      "epoch": 3.845700290275762,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035850989312544596,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21532087177038192,
      "num_tokens": 87983131.0,
      "step": 42395
    },
    {
      "entropy": 5.648875999450683,
      "epoch": 3.8461538461538463,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003584798470323644,
      "loss": 4.8728,
      "mean_token_accuracy": 0.2225247800350189,
      "num_tokens": 87993926.0,
      "step": 42400
    },
    {
      "entropy": 5.745465564727783,
      "epoch": 3.84660740203193,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003584497992128984,
      "loss": 5.0254,
      "mean_token_accuracy": 0.21157946884632112,
      "num_tokens": 88004403.0,
      "step": 42405
    },
    {
      "entropy": 5.661701297760009,
      "epoch": 3.8470609579100143,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003584197496676692,
      "loss": 4.8528,
      "mean_token_accuracy": 0.22661157548427582,
      "num_tokens": 88015289.0,
      "step": 42410
    },
    {
      "entropy": 5.581419181823731,
      "epoch": 3.8475145137880986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035838969839729846,
      "loss": 4.847,
      "mean_token_accuracy": 0.22180329710245134,
      "num_tokens": 88025724.0,
      "step": 42415
    },
    {
      "entropy": 5.611706209182739,
      "epoch": 3.847968069666183,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003583596454024073,
      "loss": 4.9558,
      "mean_token_accuracy": 0.21425333470106125,
      "num_tokens": 88036444.0,
      "step": 42420
    },
    {
      "entropy": 5.755955171585083,
      "epoch": 3.848421625544267,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003583295906836176,
      "loss": 4.9257,
      "mean_token_accuracy": 0.22433597296476365,
      "num_tokens": 88047334.0,
      "step": 42425
    },
    {
      "entropy": 5.714289331436158,
      "epoch": 3.8488751814223514,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035829953424155065,
      "loss": 4.879,
      "mean_token_accuracy": 0.22144878804683685,
      "num_tokens": 88057268.0,
      "step": 42430
    },
    {
      "entropy": 5.589236450195313,
      "epoch": 3.8493287373004357,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003582694760768281,
      "loss": 4.8613,
      "mean_token_accuracy": 0.22484455555677413,
      "num_tokens": 88067615.0,
      "step": 42435
    },
    {
      "entropy": 5.668517780303955,
      "epoch": 3.8497822931785195,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035823941619007155,
      "loss": 4.9813,
      "mean_token_accuracy": 0.21915120780467987,
      "num_tokens": 88077754.0,
      "step": 42440
    },
    {
      "entropy": 5.637988471984864,
      "epoch": 3.8502358490566038,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035820935458190275,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21835225373506545,
      "num_tokens": 88088346.0,
      "step": 42445
    },
    {
      "entropy": 5.607458400726318,
      "epoch": 3.850689404934688,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035817929125294323,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22551659047603606,
      "num_tokens": 88098534.0,
      "step": 42450
    },
    {
      "entropy": 5.629177379608154,
      "epoch": 3.8511429608127723,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003581492262038148,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21963482201099396,
      "num_tokens": 88109264.0,
      "step": 42455
    },
    {
      "entropy": 5.691736936569214,
      "epoch": 3.851596516690856,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003581191594351392,
      "loss": 4.9263,
      "mean_token_accuracy": 0.2220442771911621,
      "num_tokens": 88120407.0,
      "step": 42460
    },
    {
      "entropy": 5.71118311882019,
      "epoch": 3.8520500725689404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003580890909475381,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21738689988851548,
      "num_tokens": 88131908.0,
      "step": 42465
    },
    {
      "entropy": 5.6135190486907955,
      "epoch": 3.8525036284470247,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035805902074163354,
      "loss": 4.8501,
      "mean_token_accuracy": 0.22048432677984237,
      "num_tokens": 88142559.0,
      "step": 42470
    },
    {
      "entropy": 5.638924598693848,
      "epoch": 3.852957184325109,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003580289488180472,
      "loss": 4.9273,
      "mean_token_accuracy": 0.22537946105003356,
      "num_tokens": 88153441.0,
      "step": 42475
    },
    {
      "entropy": 5.669911479949951,
      "epoch": 3.853410740203193,
      "grad_norm": 1.890625,
      "learning_rate": 0.00035799887517740103,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2249554842710495,
      "num_tokens": 88164153.0,
      "step": 42480
    },
    {
      "entropy": 5.690925455093383,
      "epoch": 3.8538642960812775,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000357968799820317,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21847398728132247,
      "num_tokens": 88174138.0,
      "step": 42485
    },
    {
      "entropy": 5.676265048980713,
      "epoch": 3.8543178519593613,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003579387227474171,
      "loss": 4.8933,
      "mean_token_accuracy": 0.21850620657205583,
      "num_tokens": 88184704.0,
      "step": 42490
    },
    {
      "entropy": 5.758743667602539,
      "epoch": 3.8547714078374455,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003579086439593232,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21315994560718537,
      "num_tokens": 88194225.0,
      "step": 42495
    },
    {
      "entropy": 5.711723804473877,
      "epoch": 3.85522496371553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003578785634566574,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21807560622692107,
      "num_tokens": 88206125.0,
      "step": 42500
    },
    {
      "entropy": 5.616908168792724,
      "epoch": 3.855678519593614,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035784848124004187,
      "loss": 4.7939,
      "mean_token_accuracy": 0.22949125617742538,
      "num_tokens": 88217085.0,
      "step": 42505
    },
    {
      "entropy": 5.633609485626221,
      "epoch": 3.856132075471698,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003578183973100986,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2189400628209114,
      "num_tokens": 88228126.0,
      "step": 42510
    },
    {
      "entropy": 5.655173254013062,
      "epoch": 3.856585631349782,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035778831166744966,
      "loss": 4.99,
      "mean_token_accuracy": 0.2155231013894081,
      "num_tokens": 88238829.0,
      "step": 42515
    },
    {
      "entropy": 5.663262462615966,
      "epoch": 3.8570391872278664,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003577582243127175,
      "loss": 4.919,
      "mean_token_accuracy": 0.21680222153663636,
      "num_tokens": 88249821.0,
      "step": 42520
    },
    {
      "entropy": 5.665457344055175,
      "epoch": 3.8574927431059507,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003577281352465241,
      "loss": 4.8713,
      "mean_token_accuracy": 0.22815054953098296,
      "num_tokens": 88260126.0,
      "step": 42525
    },
    {
      "entropy": 5.673862886428833,
      "epoch": 3.857946298984035,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035769804446949177,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22627848386764526,
      "num_tokens": 88271492.0,
      "step": 42530
    },
    {
      "entropy": 5.739221715927124,
      "epoch": 3.8583998548621192,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003576679519822428,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21619007140398025,
      "num_tokens": 88281966.0,
      "step": 42535
    },
    {
      "entropy": 5.568761157989502,
      "epoch": 3.858853410740203,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035763785778539963,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23384695053100585,
      "num_tokens": 88292805.0,
      "step": 42540
    },
    {
      "entropy": 5.632377243041992,
      "epoch": 3.8593069666182873,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003576077618795844,
      "loss": 4.847,
      "mean_token_accuracy": 0.22488627582788467,
      "num_tokens": 88302727.0,
      "step": 42545
    },
    {
      "entropy": 5.546494102478027,
      "epoch": 3.8597605224963716,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003575776642654196,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22535971701145172,
      "num_tokens": 88313150.0,
      "step": 42550
    },
    {
      "entropy": 5.656926202774048,
      "epoch": 3.860214078374456,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035754756494352767,
      "loss": 4.8302,
      "mean_token_accuracy": 0.22973676770925522,
      "num_tokens": 88323060.0,
      "step": 42555
    },
    {
      "entropy": 5.748113536834717,
      "epoch": 3.8606676342525397,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003575174639145311,
      "loss": 4.8918,
      "mean_token_accuracy": 0.22511588782072067,
      "num_tokens": 88332641.0,
      "step": 42560
    },
    {
      "entropy": 5.621601104736328,
      "epoch": 3.861121190130624,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003574873611790523,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22298148274421692,
      "num_tokens": 88342807.0,
      "step": 42565
    },
    {
      "entropy": 5.563461112976074,
      "epoch": 3.861574746008708,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035745725673771387,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23116808235645295,
      "num_tokens": 88352958.0,
      "step": 42570
    },
    {
      "entropy": 5.685440301895142,
      "epoch": 3.8620283018867925,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035742715059113836,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22002824544906616,
      "num_tokens": 88363184.0,
      "step": 42575
    },
    {
      "entropy": 5.586659955978393,
      "epoch": 3.8624818577648767,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003573970427399484,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2285648599267006,
      "num_tokens": 88373991.0,
      "step": 42580
    },
    {
      "entropy": 5.691594791412354,
      "epoch": 3.862935413642961,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035736693318476646,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21916747093200684,
      "num_tokens": 88384518.0,
      "step": 42585
    },
    {
      "entropy": 5.57020583152771,
      "epoch": 3.8633889695210453,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035733682192621555,
      "loss": 4.7822,
      "mean_token_accuracy": 0.22705867737531663,
      "num_tokens": 88394506.0,
      "step": 42590
    },
    {
      "entropy": 5.663253259658814,
      "epoch": 3.863842525399129,
      "grad_norm": 1.375,
      "learning_rate": 0.0003573067089649181,
      "loss": 4.882,
      "mean_token_accuracy": 0.2241256892681122,
      "num_tokens": 88404625.0,
      "step": 42595
    },
    {
      "entropy": 5.701590538024902,
      "epoch": 3.8642960812772134,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00035727659430149686,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22270999997854232,
      "num_tokens": 88413754.0,
      "step": 42600
    },
    {
      "entropy": 5.729970026016235,
      "epoch": 3.8647496371552976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003572464779365748,
      "loss": 5.0055,
      "mean_token_accuracy": 0.2104557454586029,
      "num_tokens": 88424613.0,
      "step": 42605
    },
    {
      "entropy": 5.682272052764892,
      "epoch": 3.8652031930333814,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035721635987077455,
      "loss": 4.8925,
      "mean_token_accuracy": 0.22691932320594788,
      "num_tokens": 88435248.0,
      "step": 42610
    },
    {
      "entropy": 5.6208305835723875,
      "epoch": 3.8656567489114657,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035718624010471904,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23039418011903762,
      "num_tokens": 88444802.0,
      "step": 42615
    },
    {
      "entropy": 5.6283266067504885,
      "epoch": 3.86611030478955,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003571561186390311,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2139541745185852,
      "num_tokens": 88455335.0,
      "step": 42620
    },
    {
      "entropy": 5.703613805770874,
      "epoch": 3.8665638606676342,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003571259954743338,
      "loss": 4.9219,
      "mean_token_accuracy": 0.2262326166033745,
      "num_tokens": 88465857.0,
      "step": 42625
    },
    {
      "entropy": 5.664879179000854,
      "epoch": 3.8670174165457185,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035709587061124985,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21216832846403122,
      "num_tokens": 88475930.0,
      "step": 42630
    },
    {
      "entropy": 5.673430395126343,
      "epoch": 3.8674709724238028,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035706574405040245,
      "loss": 4.9157,
      "mean_token_accuracy": 0.22543097287416458,
      "num_tokens": 88486732.0,
      "step": 42635
    },
    {
      "entropy": 5.698744487762451,
      "epoch": 3.867924528301887,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003570356157924145,
      "loss": 4.907,
      "mean_token_accuracy": 0.2204982176423073,
      "num_tokens": 88497017.0,
      "step": 42640
    },
    {
      "entropy": 5.63511905670166,
      "epoch": 3.868378084179971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003570054858379092,
      "loss": 4.7828,
      "mean_token_accuracy": 0.235060678422451,
      "num_tokens": 88507423.0,
      "step": 42645
    },
    {
      "entropy": 5.6305540084838865,
      "epoch": 3.868831640058055,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035697535418750955,
      "loss": 4.8917,
      "mean_token_accuracy": 0.2246344804763794,
      "num_tokens": 88517795.0,
      "step": 42650
    },
    {
      "entropy": 5.526465225219726,
      "epoch": 3.8692851959361394,
      "grad_norm": 1.25,
      "learning_rate": 0.00035694522084183864,
      "loss": 4.7891,
      "mean_token_accuracy": 0.22717912048101424,
      "num_tokens": 88527641.0,
      "step": 42655
    },
    {
      "entropy": 5.6489733219146725,
      "epoch": 3.8697387518142237,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035691508580151966,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21360521465539933,
      "num_tokens": 88537874.0,
      "step": 42660
    },
    {
      "entropy": 5.659334278106689,
      "epoch": 3.8701923076923075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003568849490671759,
      "loss": 4.9148,
      "mean_token_accuracy": 0.22205095142126083,
      "num_tokens": 88548108.0,
      "step": 42665
    },
    {
      "entropy": 5.632326650619507,
      "epoch": 3.8706458635703918,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003568548106394305,
      "loss": 4.7962,
      "mean_token_accuracy": 0.22812678813934326,
      "num_tokens": 88558284.0,
      "step": 42670
    },
    {
      "entropy": 5.7024312019348145,
      "epoch": 3.871099419448476,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003568246705189067,
      "loss": 4.9122,
      "mean_token_accuracy": 0.2200780212879181,
      "num_tokens": 88568635.0,
      "step": 42675
    },
    {
      "entropy": 5.660861492156982,
      "epoch": 3.8715529753265603,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003567945287062279,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2169419765472412,
      "num_tokens": 88577968.0,
      "step": 42680
    },
    {
      "entropy": 5.722889089584351,
      "epoch": 3.8720065312046446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035676438520201745,
      "loss": 4.9524,
      "mean_token_accuracy": 0.22100013941526414,
      "num_tokens": 88590562.0,
      "step": 42685
    },
    {
      "entropy": 5.661634492874145,
      "epoch": 3.872460087082729,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035673424000689857,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2160722002387047,
      "num_tokens": 88601187.0,
      "step": 42690
    },
    {
      "entropy": 5.599388647079468,
      "epoch": 3.8729136429608126,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035670409312149495,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22525851279497147,
      "num_tokens": 88611459.0,
      "step": 42695
    },
    {
      "entropy": 5.6537691116333,
      "epoch": 3.873367198838897,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035667394454642963,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22465277314186097,
      "num_tokens": 88621652.0,
      "step": 42700
    },
    {
      "entropy": 5.664369297027588,
      "epoch": 3.873820754716981,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003566437942823265,
      "loss": 4.7913,
      "mean_token_accuracy": 0.22261609137058258,
      "num_tokens": 88632425.0,
      "step": 42705
    },
    {
      "entropy": 5.586523389816284,
      "epoch": 3.8742743105950654,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035661364232980884,
      "loss": 4.8056,
      "mean_token_accuracy": 0.23191829174757003,
      "num_tokens": 88641263.0,
      "step": 42710
    },
    {
      "entropy": 5.605869436264038,
      "epoch": 3.8747278664731493,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035658348868950027,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21858488917350768,
      "num_tokens": 88651665.0,
      "step": 42715
    },
    {
      "entropy": 5.633283805847168,
      "epoch": 3.8751814223512335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035655333336202436,
      "loss": 4.8292,
      "mean_token_accuracy": 0.22767774611711503,
      "num_tokens": 88660791.0,
      "step": 42720
    },
    {
      "entropy": 5.575465106964112,
      "epoch": 3.875634978229318,
      "grad_norm": 1.375,
      "learning_rate": 0.0003565231763480047,
      "loss": 4.8078,
      "mean_token_accuracy": 0.2342938542366028,
      "num_tokens": 88671341.0,
      "step": 42725
    },
    {
      "entropy": 5.62753381729126,
      "epoch": 3.876088534107402,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003564930176480651,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21765490919351577,
      "num_tokens": 88682500.0,
      "step": 42730
    },
    {
      "entropy": 5.699588394165039,
      "epoch": 3.8765420899854863,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035646285726282896,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21494409292936326,
      "num_tokens": 88693513.0,
      "step": 42735
    },
    {
      "entropy": 5.715431070327758,
      "epoch": 3.8769956458635706,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035643269519292027,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21727383583784105,
      "num_tokens": 88703498.0,
      "step": 42740
    },
    {
      "entropy": 5.62911696434021,
      "epoch": 3.8774492017416544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003564025314389627,
      "loss": 4.8915,
      "mean_token_accuracy": 0.21947107315063477,
      "num_tokens": 88714217.0,
      "step": 42745
    },
    {
      "entropy": 5.644966650009155,
      "epoch": 3.8779027576197387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035637236600157994,
      "loss": 4.9286,
      "mean_token_accuracy": 0.21845815926790238,
      "num_tokens": 88725006.0,
      "step": 42750
    },
    {
      "entropy": 5.6078290939331055,
      "epoch": 3.878356313497823,
      "grad_norm": 1.296875,
      "learning_rate": 0.000356342198881396,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22952368557453157,
      "num_tokens": 88735253.0,
      "step": 42755
    },
    {
      "entropy": 5.676779317855835,
      "epoch": 3.878809869375907,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035631203007903465,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21298106908798217,
      "num_tokens": 88746313.0,
      "step": 42760
    },
    {
      "entropy": 5.6253883838653564,
      "epoch": 3.879263425253991,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003562818595951198,
      "loss": 4.8485,
      "mean_token_accuracy": 0.2282028838992119,
      "num_tokens": 88757448.0,
      "step": 42765
    },
    {
      "entropy": 5.6675474643707275,
      "epoch": 3.8797169811320753,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003562516874302754,
      "loss": 4.8949,
      "mean_token_accuracy": 0.2274181440472603,
      "num_tokens": 88767426.0,
      "step": 42770
    },
    {
      "entropy": 5.667445850372315,
      "epoch": 3.8801705370101596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003562215135851253,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2196241796016693,
      "num_tokens": 88778188.0,
      "step": 42775
    },
    {
      "entropy": 5.774208354949951,
      "epoch": 3.880624092888244,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035619133806029374,
      "loss": 4.9521,
      "mean_token_accuracy": 0.22391708791255951,
      "num_tokens": 88788927.0,
      "step": 42780
    },
    {
      "entropy": 5.74346227645874,
      "epoch": 3.881077648766328,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035616116085640455,
      "loss": 4.9479,
      "mean_token_accuracy": 0.2183104395866394,
      "num_tokens": 88798809.0,
      "step": 42785
    },
    {
      "entropy": 5.55405740737915,
      "epoch": 3.8815312046444124,
      "grad_norm": 1.25,
      "learning_rate": 0.0003561309819740819,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23459322154521942,
      "num_tokens": 88808688.0,
      "step": 42790
    },
    {
      "entropy": 5.644283056259155,
      "epoch": 3.8819847605224966,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035610080141394983,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22634931802749633,
      "num_tokens": 88818400.0,
      "step": 42795
    },
    {
      "entropy": 5.626582670211792,
      "epoch": 3.8824383164005805,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035607061917663254,
      "loss": 4.8922,
      "mean_token_accuracy": 0.2229297026991844,
      "num_tokens": 88828032.0,
      "step": 42800
    },
    {
      "entropy": 5.667214679718017,
      "epoch": 3.8828918722786647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035604043526275413,
      "loss": 4.9817,
      "mean_token_accuracy": 0.21971650868654252,
      "num_tokens": 88839856.0,
      "step": 42805
    },
    {
      "entropy": 5.693911695480347,
      "epoch": 3.883345428156749,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000356010249672939,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22619104236364365,
      "num_tokens": 88851016.0,
      "step": 42810
    },
    {
      "entropy": 5.647712564468383,
      "epoch": 3.8837989840348333,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003559800624078111,
      "loss": 4.9196,
      "mean_token_accuracy": 0.22074661105871202,
      "num_tokens": 88859795.0,
      "step": 42815
    },
    {
      "entropy": 5.6466803550720215,
      "epoch": 3.884252539912917,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003559498734679949,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2239233672618866,
      "num_tokens": 88869595.0,
      "step": 42820
    },
    {
      "entropy": 5.725682020187378,
      "epoch": 3.8847060957910013,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003559196828541148,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21362386643886566,
      "num_tokens": 88880857.0,
      "step": 42825
    },
    {
      "entropy": 5.660222005844116,
      "epoch": 3.8851596516690856,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000355889490566795,
      "loss": 4.8326,
      "mean_token_accuracy": 0.22948523759841918,
      "num_tokens": 88890687.0,
      "step": 42830
    },
    {
      "entropy": 5.650462865829468,
      "epoch": 3.88561320754717,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035585929660665984,
      "loss": 4.8593,
      "mean_token_accuracy": 0.22506752610206604,
      "num_tokens": 88899600.0,
      "step": 42835
    },
    {
      "entropy": 5.67701907157898,
      "epoch": 3.886066763425254,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003558291009743339,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2272772118449211,
      "num_tokens": 88909521.0,
      "step": 42840
    },
    {
      "entropy": 5.650293779373169,
      "epoch": 3.8865203193033384,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035579890367044153,
      "loss": 4.9409,
      "mean_token_accuracy": 0.2181972533464432,
      "num_tokens": 88920061.0,
      "step": 42845
    },
    {
      "entropy": 5.588614845275879,
      "epoch": 3.8869738751814222,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003557687046956073,
      "loss": 4.7929,
      "mean_token_accuracy": 0.2298422113060951,
      "num_tokens": 88929504.0,
      "step": 42850
    },
    {
      "entropy": 5.759730815887451,
      "epoch": 3.8874274310595065,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035573850405045554,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20895724892616271,
      "num_tokens": 88940653.0,
      "step": 42855
    },
    {
      "entropy": 5.745896053314209,
      "epoch": 3.8878809869375908,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003557083017356111,
      "loss": 4.9674,
      "mean_token_accuracy": 0.21899197846651078,
      "num_tokens": 88952133.0,
      "step": 42860
    },
    {
      "entropy": 5.62541823387146,
      "epoch": 3.888334542815675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003556780977516983,
      "loss": 4.8535,
      "mean_token_accuracy": 0.2242206260561943,
      "num_tokens": 88963190.0,
      "step": 42865
    },
    {
      "entropy": 5.565022134780884,
      "epoch": 3.888788098693759,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035564789209934183,
      "loss": 4.8013,
      "mean_token_accuracy": 0.22476990520954132,
      "num_tokens": 88973034.0,
      "step": 42870
    },
    {
      "entropy": 5.6414652347564695,
      "epoch": 3.889241654571843,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035561768477916655,
      "loss": 4.7931,
      "mean_token_accuracy": 0.23250749409198762,
      "num_tokens": 88982809.0,
      "step": 42875
    },
    {
      "entropy": 5.656826543807983,
      "epoch": 3.8896952104499274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000355587475791797,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21689132004976272,
      "num_tokens": 88993563.0,
      "step": 42880
    },
    {
      "entropy": 5.5386251449584964,
      "epoch": 3.8901487663280117,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035555726513785785,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23060047179460524,
      "num_tokens": 89002408.0,
      "step": 42885
    },
    {
      "entropy": 5.603125715255738,
      "epoch": 3.890602322206096,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003555270528179739,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2251296490430832,
      "num_tokens": 89013060.0,
      "step": 42890
    },
    {
      "entropy": 5.705453968048095,
      "epoch": 3.89105587808418,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035549683883277,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2228753462433815,
      "num_tokens": 89023406.0,
      "step": 42895
    },
    {
      "entropy": 5.701925945281983,
      "epoch": 3.891509433962264,
      "grad_norm": 1.28125,
      "learning_rate": 0.000355466623182871,
      "loss": 4.9299,
      "mean_token_accuracy": 0.22259570062160491,
      "num_tokens": 89033800.0,
      "step": 42900
    },
    {
      "entropy": 5.751665115356445,
      "epoch": 3.8919629898403483,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003554364058689017,
      "loss": 4.909,
      "mean_token_accuracy": 0.2214192718267441,
      "num_tokens": 89044601.0,
      "step": 42905
    },
    {
      "entropy": 5.721828031539917,
      "epoch": 3.8924165457184325,
      "grad_norm": 1.203125,
      "learning_rate": 0.000355406186891487,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21884151101112365,
      "num_tokens": 89056138.0,
      "step": 42910
    },
    {
      "entropy": 5.640722942352295,
      "epoch": 3.892870101596517,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003553759662512519,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2239087402820587,
      "num_tokens": 89065776.0,
      "step": 42915
    },
    {
      "entropy": 5.780185270309448,
      "epoch": 3.8933236574746006,
      "grad_norm": 1.25,
      "learning_rate": 0.00035534574394882135,
      "loss": 4.992,
      "mean_token_accuracy": 0.20792556554079056,
      "num_tokens": 89076926.0,
      "step": 42920
    },
    {
      "entropy": 5.657666540145874,
      "epoch": 3.893777213352685,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003553155199848203,
      "loss": 4.91,
      "mean_token_accuracy": 0.2116399422287941,
      "num_tokens": 89088036.0,
      "step": 42925
    },
    {
      "entropy": 5.668568706512451,
      "epoch": 3.894230769230769,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003552852943598739,
      "loss": 4.858,
      "mean_token_accuracy": 0.22347044944763184,
      "num_tokens": 89098606.0,
      "step": 42930
    },
    {
      "entropy": 5.635056400299073,
      "epoch": 3.8946843251088534,
      "grad_norm": 1.328125,
      "learning_rate": 0.000355255067074607,
      "loss": 4.8139,
      "mean_token_accuracy": 0.23013464957475663,
      "num_tokens": 89107733.0,
      "step": 42935
    },
    {
      "entropy": 5.635632848739624,
      "epoch": 3.8951378809869377,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035522483812964496,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2248557135462761,
      "num_tokens": 89119094.0,
      "step": 42940
    },
    {
      "entropy": 5.626153039932251,
      "epoch": 3.895591436865022,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035519460752561285,
      "loss": 4.7711,
      "mean_token_accuracy": 0.2281900867819786,
      "num_tokens": 89129254.0,
      "step": 42945
    },
    {
      "entropy": 5.657587242126465,
      "epoch": 3.8960449927431062,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003551643752631358,
      "loss": 4.835,
      "mean_token_accuracy": 0.22820369005203248,
      "num_tokens": 89139609.0,
      "step": 42950
    },
    {
      "entropy": 5.665063571929932,
      "epoch": 3.89649854862119,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003551341413428391,
      "loss": 4.8604,
      "mean_token_accuracy": 0.21357677280902862,
      "num_tokens": 89149504.0,
      "step": 42955
    },
    {
      "entropy": 5.643418216705323,
      "epoch": 3.8969521044992743,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003551039057653478,
      "loss": 4.861,
      "mean_token_accuracy": 0.22563812881708145,
      "num_tokens": 89160036.0,
      "step": 42960
    },
    {
      "entropy": 5.656509733200073,
      "epoch": 3.8974056603773586,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003550736685312874,
      "loss": 4.9213,
      "mean_token_accuracy": 0.2197621926665306,
      "num_tokens": 89170960.0,
      "step": 42965
    },
    {
      "entropy": 5.651523494720459,
      "epoch": 3.8978592162554424,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003550434296412831,
      "loss": 4.8525,
      "mean_token_accuracy": 0.23290955722332002,
      "num_tokens": 89179630.0,
      "step": 42970
    },
    {
      "entropy": 5.647391891479492,
      "epoch": 3.8983127721335267,
      "grad_norm": 1.25,
      "learning_rate": 0.0003550131890959603,
      "loss": 4.8833,
      "mean_token_accuracy": 0.2271190032362938,
      "num_tokens": 89188880.0,
      "step": 42975
    },
    {
      "entropy": 5.698987054824829,
      "epoch": 3.898766328011611,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035498294689594435,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21896457374095918,
      "num_tokens": 89199585.0,
      "step": 42980
    },
    {
      "entropy": 5.67002501487732,
      "epoch": 3.899219883889695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035495270304186074,
      "loss": 4.8646,
      "mean_token_accuracy": 0.21500124335289,
      "num_tokens": 89209812.0,
      "step": 42985
    },
    {
      "entropy": 5.635092163085938,
      "epoch": 3.8996734397677795,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003549224575343347,
      "loss": 4.8685,
      "mean_token_accuracy": 0.22597844451665877,
      "num_tokens": 89220244.0,
      "step": 42990
    },
    {
      "entropy": 5.643746662139892,
      "epoch": 3.9001269956458637,
      "grad_norm": 1.375,
      "learning_rate": 0.00035489221037399203,
      "loss": 4.9277,
      "mean_token_accuracy": 0.21631388664245604,
      "num_tokens": 89230618.0,
      "step": 42995
    },
    {
      "entropy": 5.665724611282348,
      "epoch": 3.900580551523948,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035486196156145805,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22363721281290055,
      "num_tokens": 89242041.0,
      "step": 43000
    },
    {
      "entropy": 5.699419355392456,
      "epoch": 3.901034107402032,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003548317110973583,
      "loss": 4.9104,
      "mean_token_accuracy": 0.22254988551139832,
      "num_tokens": 89252817.0,
      "step": 43005
    },
    {
      "entropy": 5.6780070781707765,
      "epoch": 3.901487663280116,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003548014589823185,
      "loss": 4.9337,
      "mean_token_accuracy": 0.22347624599933624,
      "num_tokens": 89263702.0,
      "step": 43010
    },
    {
      "entropy": 5.643161582946777,
      "epoch": 3.9019412191582004,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035477120521696417,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2279886230826378,
      "num_tokens": 89273822.0,
      "step": 43015
    },
    {
      "entropy": 5.725532913208008,
      "epoch": 3.9023947750362846,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003547409498019209,
      "loss": 4.9987,
      "mean_token_accuracy": 0.21075901985168458,
      "num_tokens": 89283973.0,
      "step": 43020
    },
    {
      "entropy": 5.7172383785247805,
      "epoch": 3.9028483309143684,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035471069273781454,
      "loss": 4.9354,
      "mean_token_accuracy": 0.22291269749403,
      "num_tokens": 89294059.0,
      "step": 43025
    },
    {
      "entropy": 5.664397382736206,
      "epoch": 3.9033018867924527,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003546804340252707,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22493355721235275,
      "num_tokens": 89304049.0,
      "step": 43030
    },
    {
      "entropy": 5.6013750553131105,
      "epoch": 3.903755442670537,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035465017366491527,
      "loss": 4.78,
      "mean_token_accuracy": 0.23305300623178482,
      "num_tokens": 89314468.0,
      "step": 43035
    },
    {
      "entropy": 5.578039312362671,
      "epoch": 3.9042089985486212,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035461991165737386,
      "loss": 4.852,
      "mean_token_accuracy": 0.22753328382968901,
      "num_tokens": 89324580.0,
      "step": 43040
    },
    {
      "entropy": 5.721569490432739,
      "epoch": 3.9046625544267055,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035458964800327245,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2118929833173752,
      "num_tokens": 89334976.0,
      "step": 43045
    },
    {
      "entropy": 5.673843050003052,
      "epoch": 3.90511611030479,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035455938270323684,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21832153350114822,
      "num_tokens": 89345210.0,
      "step": 43050
    },
    {
      "entropy": 5.638338375091553,
      "epoch": 3.9055696661828736,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003545291157578928,
      "loss": 4.7758,
      "mean_token_accuracy": 0.23181362450122833,
      "num_tokens": 89354623.0,
      "step": 43055
    },
    {
      "entropy": 5.645547008514404,
      "epoch": 3.906023222060958,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035449884716786653,
      "loss": 4.8549,
      "mean_token_accuracy": 0.22735803127288817,
      "num_tokens": 89364684.0,
      "step": 43060
    },
    {
      "entropy": 5.648725605010986,
      "epoch": 3.906476777939042,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035446857693378377,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21632350236177444,
      "num_tokens": 89375415.0,
      "step": 43065
    },
    {
      "entropy": 5.688320207595825,
      "epoch": 3.9069303338171264,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003544383050562706,
      "loss": 4.9678,
      "mean_token_accuracy": 0.22451281994581224,
      "num_tokens": 89385791.0,
      "step": 43070
    },
    {
      "entropy": 5.691095590591431,
      "epoch": 3.90738388969521,
      "grad_norm": 1.28125,
      "learning_rate": 0.000354408031535953,
      "loss": 5.0067,
      "mean_token_accuracy": 0.22069910615682603,
      "num_tokens": 89395345.0,
      "step": 43075
    },
    {
      "entropy": 5.637933921813965,
      "epoch": 3.9078374455732945,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035437775637345715,
      "loss": 4.846,
      "mean_token_accuracy": 0.22864942252635956,
      "num_tokens": 89405413.0,
      "step": 43080
    },
    {
      "entropy": 5.668398857116699,
      "epoch": 3.9082910014513788,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00035434747956940906,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22741466164588928,
      "num_tokens": 89415168.0,
      "step": 43085
    },
    {
      "entropy": 5.605725193023682,
      "epoch": 3.908744557329463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035431720112443475,
      "loss": 4.8606,
      "mean_token_accuracy": 0.228644897043705,
      "num_tokens": 89426438.0,
      "step": 43090
    },
    {
      "entropy": 5.614774513244629,
      "epoch": 3.9091981132075473,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035428692103916067,
      "loss": 4.8277,
      "mean_token_accuracy": 0.22236608415842057,
      "num_tokens": 89437454.0,
      "step": 43095
    },
    {
      "entropy": 5.678123807907104,
      "epoch": 3.9096516690856316,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003542566393142128,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22578066885471343,
      "num_tokens": 89448082.0,
      "step": 43100
    },
    {
      "entropy": 5.581694507598877,
      "epoch": 3.9101052249637154,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035422635595021745,
      "loss": 4.8434,
      "mean_token_accuracy": 0.2237669587135315,
      "num_tokens": 89458227.0,
      "step": 43105
    },
    {
      "entropy": 5.631004858016968,
      "epoch": 3.9105587808417996,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035419607094780084,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22126545161008834,
      "num_tokens": 89468374.0,
      "step": 43110
    },
    {
      "entropy": 5.649475145339966,
      "epoch": 3.911012336719884,
      "grad_norm": 1.375,
      "learning_rate": 0.00035416578430758934,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2207956552505493,
      "num_tokens": 89477728.0,
      "step": 43115
    },
    {
      "entropy": 5.700003051757813,
      "epoch": 3.911465892597968,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035413549603020915,
      "loss": 4.9009,
      "mean_token_accuracy": 0.22394519299268723,
      "num_tokens": 89487107.0,
      "step": 43120
    },
    {
      "entropy": 5.716149759292603,
      "epoch": 3.911919448476052,
      "grad_norm": 1.375,
      "learning_rate": 0.00035410520611628676,
      "loss": 4.9136,
      "mean_token_accuracy": 0.2304343596100807,
      "num_tokens": 89497075.0,
      "step": 43125
    },
    {
      "entropy": 5.561764574050903,
      "epoch": 3.9123730043541363,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003540749145664486,
      "loss": 4.8342,
      "mean_token_accuracy": 0.222140933573246,
      "num_tokens": 89508068.0,
      "step": 43130
    },
    {
      "entropy": 5.649159574508667,
      "epoch": 3.9128265602322205,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003540446213813211,
      "loss": 4.8288,
      "mean_token_accuracy": 0.2314961478114128,
      "num_tokens": 89518371.0,
      "step": 43135
    },
    {
      "entropy": 5.6770055294036865,
      "epoch": 3.913280116110305,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035401432656153057,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2243675783276558,
      "num_tokens": 89529024.0,
      "step": 43140
    },
    {
      "entropy": 5.598103141784668,
      "epoch": 3.913733671988389,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035398403010770377,
      "loss": 4.7841,
      "mean_token_accuracy": 0.2303077444434166,
      "num_tokens": 89540331.0,
      "step": 43145
    },
    {
      "entropy": 5.549307489395142,
      "epoch": 3.9141872278664733,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000353953732020467,
      "loss": 4.793,
      "mean_token_accuracy": 0.23020783811807632,
      "num_tokens": 89550385.0,
      "step": 43150
    },
    {
      "entropy": 5.679307174682617,
      "epoch": 3.9146407837445576,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035392343230044687,
      "loss": 4.9601,
      "mean_token_accuracy": 0.21727267056703567,
      "num_tokens": 89561480.0,
      "step": 43155
    },
    {
      "entropy": 5.696958780288696,
      "epoch": 3.9150943396226414,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035389313094827017,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22030405551195145,
      "num_tokens": 89571897.0,
      "step": 43160
    },
    {
      "entropy": 5.656802701950073,
      "epoch": 3.9155478955007257,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035386282796456343,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2238266125321388,
      "num_tokens": 89582019.0,
      "step": 43165
    },
    {
      "entropy": 5.622116374969482,
      "epoch": 3.91600145137881,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003538325233499532,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22821482121944428,
      "num_tokens": 89591806.0,
      "step": 43170
    },
    {
      "entropy": 5.617962169647217,
      "epoch": 3.9164550072568938,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035380221710506636,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22724343836307526,
      "num_tokens": 89601845.0,
      "step": 43175
    },
    {
      "entropy": 5.690812969207764,
      "epoch": 3.916908563134978,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035377190923052964,
      "loss": 4.9593,
      "mean_token_accuracy": 0.22130796015262605,
      "num_tokens": 89612331.0,
      "step": 43180
    },
    {
      "entropy": 5.694004774093628,
      "epoch": 3.9173621190130623,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035374159972696966,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2182876229286194,
      "num_tokens": 89622631.0,
      "step": 43185
    },
    {
      "entropy": 5.684329986572266,
      "epoch": 3.9178156748911466,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003537112885950133,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2195568785071373,
      "num_tokens": 89633065.0,
      "step": 43190
    },
    {
      "entropy": 5.696789741516113,
      "epoch": 3.918269230769231,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035368097583528756,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2251499831676483,
      "num_tokens": 89643451.0,
      "step": 43195
    },
    {
      "entropy": 5.771214723587036,
      "epoch": 3.918722786647315,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003536506614484192,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22323312610387802,
      "num_tokens": 89654786.0,
      "step": 43200
    },
    {
      "entropy": 5.7175980567932125,
      "epoch": 3.9191763425253994,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003536203454350351,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21938296854496003,
      "num_tokens": 89665459.0,
      "step": 43205
    },
    {
      "entropy": 5.628254461288452,
      "epoch": 3.919629898403483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003535900277957621,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22421299070119857,
      "num_tokens": 89675149.0,
      "step": 43210
    },
    {
      "entropy": 5.638657760620117,
      "epoch": 3.9200834542815675,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035355970853122743,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2270113617181778,
      "num_tokens": 89684994.0,
      "step": 43215
    },
    {
      "entropy": 5.749697017669678,
      "epoch": 3.9205370101596517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035352938764205793,
      "loss": 5.0245,
      "mean_token_accuracy": 0.2057173654437065,
      "num_tokens": 89694751.0,
      "step": 43220
    },
    {
      "entropy": 5.684262371063232,
      "epoch": 3.920990566037736,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003534990651288807,
      "loss": 4.8611,
      "mean_token_accuracy": 0.2194393590092659,
      "num_tokens": 89704976.0,
      "step": 43225
    },
    {
      "entropy": 5.644000768661499,
      "epoch": 3.92144412191582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035346874099232274,
      "loss": 4.959,
      "mean_token_accuracy": 0.21563856154680253,
      "num_tokens": 89715863.0,
      "step": 43230
    },
    {
      "entropy": 5.663571834564209,
      "epoch": 3.921897677793904,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035343841523301125,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2198469892144203,
      "num_tokens": 89725856.0,
      "step": 43235
    },
    {
      "entropy": 5.684086847305298,
      "epoch": 3.9223512336719883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003534080878515733,
      "loss": 4.937,
      "mean_token_accuracy": 0.21983996629714966,
      "num_tokens": 89736345.0,
      "step": 43240
    },
    {
      "entropy": 5.7095537185668945,
      "epoch": 3.9228047895500726,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003533777588486362,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2216563880443573,
      "num_tokens": 89746960.0,
      "step": 43245
    },
    {
      "entropy": 5.665602207183838,
      "epoch": 3.923258345428157,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000353347428224827,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22010509073734283,
      "num_tokens": 89757373.0,
      "step": 43250
    },
    {
      "entropy": 5.673526763916016,
      "epoch": 3.923711901306241,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035331709598077304,
      "loss": 4.9132,
      "mean_token_accuracy": 0.21877120584249496,
      "num_tokens": 89767895.0,
      "step": 43255
    },
    {
      "entropy": 5.681426954269409,
      "epoch": 3.924165457184325,
      "grad_norm": 1.25,
      "learning_rate": 0.0003532867621171016,
      "loss": 4.8839,
      "mean_token_accuracy": 0.2201533630490303,
      "num_tokens": 89777512.0,
      "step": 43260
    },
    {
      "entropy": 5.67314715385437,
      "epoch": 3.9246190130624092,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035325642663443994,
      "loss": 4.9112,
      "mean_token_accuracy": 0.2216028168797493,
      "num_tokens": 89787308.0,
      "step": 43265
    },
    {
      "entropy": 5.690020370483398,
      "epoch": 3.9250725689404935,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003532260895334155,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22531746327877045,
      "num_tokens": 89798384.0,
      "step": 43270
    },
    {
      "entropy": 5.710465955734253,
      "epoch": 3.9255261248185778,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035319575081465556,
      "loss": 4.9566,
      "mean_token_accuracy": 0.22179538905620574,
      "num_tokens": 89808757.0,
      "step": 43275
    },
    {
      "entropy": 5.711591720581055,
      "epoch": 3.9259796806966616,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035316541047878753,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21368564814329147,
      "num_tokens": 89818925.0,
      "step": 43280
    },
    {
      "entropy": 5.713443040847778,
      "epoch": 3.926433236574746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003531350685264389,
      "loss": 4.9397,
      "mean_token_accuracy": 0.22394015789031982,
      "num_tokens": 89828647.0,
      "step": 43285
    },
    {
      "entropy": 5.769444322586059,
      "epoch": 3.92688679245283,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035310472495823713,
      "loss": 4.9953,
      "mean_token_accuracy": 0.21058460026979448,
      "num_tokens": 89838784.0,
      "step": 43290
    },
    {
      "entropy": 5.673107624053955,
      "epoch": 3.9273403483309144,
      "grad_norm": 1.25,
      "learning_rate": 0.0003530743797748098,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21748734265565872,
      "num_tokens": 89848464.0,
      "step": 43295
    },
    {
      "entropy": 5.610161161422729,
      "epoch": 3.9277939042089987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003530440329767844,
      "loss": 4.8669,
      "mean_token_accuracy": 0.21795052140951157,
      "num_tokens": 89859594.0,
      "step": 43300
    },
    {
      "entropy": 5.659009504318237,
      "epoch": 3.928247460087083,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003530136845647885,
      "loss": 4.8458,
      "mean_token_accuracy": 0.2207550436258316,
      "num_tokens": 89869869.0,
      "step": 43305
    },
    {
      "entropy": 5.746286869049072,
      "epoch": 3.9287010159651667,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035298333453944976,
      "loss": 4.9395,
      "mean_token_accuracy": 0.2237633764743805,
      "num_tokens": 89879604.0,
      "step": 43310
    },
    {
      "entropy": 5.574654054641724,
      "epoch": 3.929154571843251,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003529529829013957,
      "loss": 4.7503,
      "mean_token_accuracy": 0.232356995344162,
      "num_tokens": 89889200.0,
      "step": 43315
    },
    {
      "entropy": 5.669319725036621,
      "epoch": 3.9296081277213353,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003529226296512541,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2268512010574341,
      "num_tokens": 89898525.0,
      "step": 43320
    },
    {
      "entropy": 5.622125482559204,
      "epoch": 3.9300616835994195,
      "grad_norm": 1.4375,
      "learning_rate": 0.00035289227478965267,
      "loss": 4.9405,
      "mean_token_accuracy": 0.2150866523385048,
      "num_tokens": 89909053.0,
      "step": 43325
    },
    {
      "entropy": 5.593583869934082,
      "epoch": 3.9305152394775034,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003528619183172192,
      "loss": 4.8277,
      "mean_token_accuracy": 0.22917849868535994,
      "num_tokens": 89918375.0,
      "step": 43330
    },
    {
      "entropy": 5.759978628158569,
      "epoch": 3.9309687953555876,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003528315602345814,
      "loss": 4.956,
      "mean_token_accuracy": 0.22752957791090012,
      "num_tokens": 89928252.0,
      "step": 43335
    },
    {
      "entropy": 5.651673460006714,
      "epoch": 3.931422351233672,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003528012005423671,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21788123697042466,
      "num_tokens": 89938613.0,
      "step": 43340
    },
    {
      "entropy": 5.723757791519165,
      "epoch": 3.931875907111756,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035277083924120406,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22394309490919112,
      "num_tokens": 89948781.0,
      "step": 43345
    },
    {
      "entropy": 5.651911830902099,
      "epoch": 3.9323294629898404,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035274047633172037,
      "loss": 4.9213,
      "mean_token_accuracy": 0.21476895809173585,
      "num_tokens": 89959928.0,
      "step": 43350
    },
    {
      "entropy": 5.717375087738037,
      "epoch": 3.9327830188679247,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003527101118145437,
      "loss": 4.941,
      "mean_token_accuracy": 0.21978844404220582,
      "num_tokens": 89970756.0,
      "step": 43355
    },
    {
      "entropy": 5.69220814704895,
      "epoch": 3.933236574746009,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003526797456903022,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2195417046546936,
      "num_tokens": 89981332.0,
      "step": 43360
    },
    {
      "entropy": 5.636444473266602,
      "epoch": 3.933690130624093,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035264937795962374,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22601353675127028,
      "num_tokens": 89991427.0,
      "step": 43365
    },
    {
      "entropy": 5.666295671463013,
      "epoch": 3.934143686502177,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003526190086231364,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22677917182445526,
      "num_tokens": 90001599.0,
      "step": 43370
    },
    {
      "entropy": 5.683712816238403,
      "epoch": 3.9345972423802613,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035258863768146816,
      "loss": 4.9014,
      "mean_token_accuracy": 0.22446888238191604,
      "num_tokens": 90011932.0,
      "step": 43375
    },
    {
      "entropy": 5.6849061965942385,
      "epoch": 3.9350507982583456,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003525582651352471,
      "loss": 4.9835,
      "mean_token_accuracy": 0.22281498908996583,
      "num_tokens": 90022123.0,
      "step": 43380
    },
    {
      "entropy": 5.595370864868164,
      "epoch": 3.9355043541364294,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003525278909851013,
      "loss": 4.8297,
      "mean_token_accuracy": 0.23288505971431733,
      "num_tokens": 90033024.0,
      "step": 43385
    },
    {
      "entropy": 5.658297157287597,
      "epoch": 3.9359579100145137,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000352497515231659,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22497707605361938,
      "num_tokens": 90043742.0,
      "step": 43390
    },
    {
      "entropy": 5.640840816497803,
      "epoch": 3.936411465892598,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003524671378755484,
      "loss": 4.8663,
      "mean_token_accuracy": 0.22248139083385468,
      "num_tokens": 90054508.0,
      "step": 43395
    },
    {
      "entropy": 5.677954006195068,
      "epoch": 3.936865021770682,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003524367589173976,
      "loss": 4.9122,
      "mean_token_accuracy": 0.2262275755405426,
      "num_tokens": 90065058.0,
      "step": 43400
    },
    {
      "entropy": 5.664167881011963,
      "epoch": 3.9373185776487665,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035240637835783485,
      "loss": 4.9229,
      "mean_token_accuracy": 0.22082047611474992,
      "num_tokens": 90075813.0,
      "step": 43405
    },
    {
      "entropy": 5.687118291854858,
      "epoch": 3.9377721335268507,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035237599619748853,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2143964573740959,
      "num_tokens": 90087064.0,
      "step": 43410
    },
    {
      "entropy": 5.656747436523437,
      "epoch": 3.9382256894049346,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003523456124369868,
      "loss": 4.8342,
      "mean_token_accuracy": 0.23068393021821976,
      "num_tokens": 90096668.0,
      "step": 43415
    },
    {
      "entropy": 5.613543844223022,
      "epoch": 3.938679245283019,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003523152270769582,
      "loss": 4.7683,
      "mean_token_accuracy": 0.23296170085668563,
      "num_tokens": 90105569.0,
      "step": 43420
    },
    {
      "entropy": 5.647732496261597,
      "epoch": 3.939132801161103,
      "grad_norm": 1.296875,
      "learning_rate": 0.000352284840118031,
      "loss": 4.9275,
      "mean_token_accuracy": 0.22127506881952286,
      "num_tokens": 90116040.0,
      "step": 43425
    },
    {
      "entropy": 5.67569146156311,
      "epoch": 3.9395863570391874,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003522544515608335,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2222743347287178,
      "num_tokens": 90125929.0,
      "step": 43430
    },
    {
      "entropy": 5.78849425315857,
      "epoch": 3.940039912917271,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003522240614059944,
      "loss": 4.9747,
      "mean_token_accuracy": 0.22025911659002304,
      "num_tokens": 90136251.0,
      "step": 43435
    },
    {
      "entropy": 5.67856650352478,
      "epoch": 3.9404934687953554,
      "grad_norm": 1.28125,
      "learning_rate": 0.000352193669654142,
      "loss": 4.8473,
      "mean_token_accuracy": 0.22875205427408218,
      "num_tokens": 90146480.0,
      "step": 43440
    },
    {
      "entropy": 5.670795869827271,
      "epoch": 3.9409470246734397,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035216327630590476,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2215244874358177,
      "num_tokens": 90157192.0,
      "step": 43445
    },
    {
      "entropy": 5.684213590621948,
      "epoch": 3.941400580551524,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035213288136191136,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21882251352071763,
      "num_tokens": 90167120.0,
      "step": 43450
    },
    {
      "entropy": 5.676723480224609,
      "epoch": 3.9418541364296082,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035210248482279034,
      "loss": 4.9294,
      "mean_token_accuracy": 0.2198939338326454,
      "num_tokens": 90176276.0,
      "step": 43455
    },
    {
      "entropy": 5.659224128723144,
      "epoch": 3.9423076923076925,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035207208668917026,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22431150525808335,
      "num_tokens": 90187150.0,
      "step": 43460
    },
    {
      "entropy": 5.602003812789917,
      "epoch": 3.9427612481857763,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035204168696167975,
      "loss": 4.8263,
      "mean_token_accuracy": 0.22026303112506868,
      "num_tokens": 90198112.0,
      "step": 43465
    },
    {
      "entropy": 5.671118688583374,
      "epoch": 3.9432148040638606,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035201128564094753,
      "loss": 4.9382,
      "mean_token_accuracy": 0.21844212859869003,
      "num_tokens": 90208940.0,
      "step": 43470
    },
    {
      "entropy": 5.6670006275177,
      "epoch": 3.943668359941945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003519808827276023,
      "loss": 4.89,
      "mean_token_accuracy": 0.2276556134223938,
      "num_tokens": 90220073.0,
      "step": 43475
    },
    {
      "entropy": 5.65039587020874,
      "epoch": 3.944121915820029,
      "grad_norm": 1.25,
      "learning_rate": 0.00035195047822227283,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21646596640348434,
      "num_tokens": 90230059.0,
      "step": 43480
    },
    {
      "entropy": 5.617595767974853,
      "epoch": 3.944575471698113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035192007212558787,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22388505637645723,
      "num_tokens": 90240486.0,
      "step": 43485
    },
    {
      "entropy": 5.744507598876953,
      "epoch": 3.9450290275761972,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003518896644381762,
      "loss": 4.9613,
      "mean_token_accuracy": 0.21825908869504929,
      "num_tokens": 90250014.0,
      "step": 43490
    },
    {
      "entropy": 5.660885000228882,
      "epoch": 3.9454825834542815,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003518592551606666,
      "loss": 4.8715,
      "mean_token_accuracy": 0.21523140668869017,
      "num_tokens": 90261040.0,
      "step": 43495
    },
    {
      "entropy": 5.700621795654297,
      "epoch": 3.9459361393323658,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003518288442936881,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2216187447309494,
      "num_tokens": 90271243.0,
      "step": 43500
    },
    {
      "entropy": 5.673984575271606,
      "epoch": 3.94638969521045,
      "grad_norm": 1.375,
      "learning_rate": 0.0003517984318378695,
      "loss": 4.8944,
      "mean_token_accuracy": 0.22099606990814208,
      "num_tokens": 90281176.0,
      "step": 43505
    },
    {
      "entropy": 5.754955577850342,
      "epoch": 3.9468432510885343,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003517680177938398,
      "loss": 4.976,
      "mean_token_accuracy": 0.21949897557497025,
      "num_tokens": 90292465.0,
      "step": 43510
    },
    {
      "entropy": 5.696611213684082,
      "epoch": 3.9472968069666186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003517376021622279,
      "loss": 4.8325,
      "mean_token_accuracy": 0.22815789133310319,
      "num_tokens": 90303482.0,
      "step": 43515
    },
    {
      "entropy": 5.661256551742554,
      "epoch": 3.9477503628447024,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003517071849436629,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2200719103217125,
      "num_tokens": 90313696.0,
      "step": 43520
    },
    {
      "entropy": 5.665653133392334,
      "epoch": 3.9482039187227866,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035167676613877366,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2145412966609001,
      "num_tokens": 90324564.0,
      "step": 43525
    },
    {
      "entropy": 5.635372877120972,
      "epoch": 3.948657474600871,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003516463457481894,
      "loss": 4.793,
      "mean_token_accuracy": 0.23202825635671614,
      "num_tokens": 90334440.0,
      "step": 43530
    },
    {
      "entropy": 5.7076925277709964,
      "epoch": 3.9491110304789547,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003516159237725392,
      "loss": 4.9294,
      "mean_token_accuracy": 0.22238634526729584,
      "num_tokens": 90344029.0,
      "step": 43535
    },
    {
      "entropy": 5.718100929260254,
      "epoch": 3.949564586357039,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003515855002124521,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2175098955631256,
      "num_tokens": 90354628.0,
      "step": 43540
    },
    {
      "entropy": 5.672612094879151,
      "epoch": 3.9500181422351233,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003515550750685573,
      "loss": 4.8083,
      "mean_token_accuracy": 0.2323197677731514,
      "num_tokens": 90365091.0,
      "step": 43545
    },
    {
      "entropy": 5.668745040893555,
      "epoch": 3.9504716981132075,
      "grad_norm": 1.25,
      "learning_rate": 0.0003515246483414842,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21860220283269882,
      "num_tokens": 90374197.0,
      "step": 43550
    },
    {
      "entropy": 5.6756429195404055,
      "epoch": 3.950925253991292,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035149422003186176,
      "loss": 4.9331,
      "mean_token_accuracy": 0.22352002710103988,
      "num_tokens": 90384686.0,
      "step": 43555
    },
    {
      "entropy": 5.725683355331421,
      "epoch": 3.951378809869376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035146379014031943,
      "loss": 4.9734,
      "mean_token_accuracy": 0.22019983381032943,
      "num_tokens": 90396329.0,
      "step": 43560
    },
    {
      "entropy": 5.722560596466065,
      "epoch": 3.9518323657474603,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003514333586674864,
      "loss": 4.9615,
      "mean_token_accuracy": 0.2203424468636513,
      "num_tokens": 90408957.0,
      "step": 43565
    },
    {
      "entropy": 5.620771789550782,
      "epoch": 3.952285921625544,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000351402925613992,
      "loss": 4.8629,
      "mean_token_accuracy": 0.21978174299001693,
      "num_tokens": 90420194.0,
      "step": 43570
    },
    {
      "entropy": 5.629699087142944,
      "epoch": 3.9527394775036284,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003513724909804656,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22595855444669724,
      "num_tokens": 90430881.0,
      "step": 43575
    },
    {
      "entropy": 5.629531669616699,
      "epoch": 3.9531930333817127,
      "grad_norm": 1.375,
      "learning_rate": 0.00035134205476753674,
      "loss": 4.8884,
      "mean_token_accuracy": 0.21324550360441208,
      "num_tokens": 90440713.0,
      "step": 43580
    },
    {
      "entropy": 5.670223999023437,
      "epoch": 3.953646589259797,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003513116169758347,
      "loss": 4.9597,
      "mean_token_accuracy": 0.21602357625961305,
      "num_tokens": 90451274.0,
      "step": 43585
    },
    {
      "entropy": 5.744808530807495,
      "epoch": 3.9541001451378808,
      "grad_norm": 1.25,
      "learning_rate": 0.0003512811776059889,
      "loss": 4.9865,
      "mean_token_accuracy": 0.2171168878674507,
      "num_tokens": 90461503.0,
      "step": 43590
    },
    {
      "entropy": 5.665510654449463,
      "epoch": 3.954553701015965,
      "grad_norm": 1.34375,
      "learning_rate": 0.000351250736658629,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23298123180866243,
      "num_tokens": 90471178.0,
      "step": 43595
    },
    {
      "entropy": 5.632181882858276,
      "epoch": 3.9550072568940493,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003512202941343844,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22214230448007583,
      "num_tokens": 90481573.0,
      "step": 43600
    },
    {
      "entropy": 5.688739395141601,
      "epoch": 3.9554608127721336,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003511898500338847,
      "loss": 4.9617,
      "mean_token_accuracy": 0.21614936739206314,
      "num_tokens": 90491853.0,
      "step": 43605
    },
    {
      "entropy": 5.676394081115722,
      "epoch": 3.955914368650218,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003511594043577594,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21660227179527283,
      "num_tokens": 90502958.0,
      "step": 43610
    },
    {
      "entropy": 5.619556093215943,
      "epoch": 3.956367924528302,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035112895710663833,
      "loss": 4.8807,
      "mean_token_accuracy": 0.22116641998291015,
      "num_tokens": 90512555.0,
      "step": 43615
    },
    {
      "entropy": 5.6964256286621096,
      "epoch": 3.956821480406386,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035109850828115105,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2234891787171364,
      "num_tokens": 90522722.0,
      "step": 43620
    },
    {
      "entropy": 5.661873483657837,
      "epoch": 3.95727503628447,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035106805788192707,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2224666103720665,
      "num_tokens": 90533182.0,
      "step": 43625
    },
    {
      "entropy": 5.701808261871338,
      "epoch": 3.9577285921625545,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003510376059095963,
      "loss": 4.9659,
      "mean_token_accuracy": 0.2112626001238823,
      "num_tokens": 90543495.0,
      "step": 43630
    },
    {
      "entropy": 5.626921081542969,
      "epoch": 3.9581821480406387,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003510071523647886,
      "loss": 4.8225,
      "mean_token_accuracy": 0.22718972116708755,
      "num_tokens": 90553680.0,
      "step": 43635
    },
    {
      "entropy": 5.630295085906982,
      "epoch": 3.9586357039187225,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003509766972481336,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22706011831760406,
      "num_tokens": 90563513.0,
      "step": 43640
    },
    {
      "entropy": 5.763727140426636,
      "epoch": 3.959089259796807,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000350946240560261,
      "loss": 4.9522,
      "mean_token_accuracy": 0.22226848006248473,
      "num_tokens": 90573337.0,
      "step": 43645
    },
    {
      "entropy": 5.641994285583496,
      "epoch": 3.959542815674891,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003509157823018009,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22536929994821547,
      "num_tokens": 90582979.0,
      "step": 43650
    },
    {
      "entropy": 5.731965208053589,
      "epoch": 3.9599963715529753,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035088532247338306,
      "loss": 4.9933,
      "mean_token_accuracy": 0.21136672645807267,
      "num_tokens": 90593552.0,
      "step": 43655
    },
    {
      "entropy": 5.705495691299438,
      "epoch": 3.9604499274310596,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035085486107563733,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2216441199183464,
      "num_tokens": 90603813.0,
      "step": 43660
    },
    {
      "entropy": 5.7087073802948,
      "epoch": 3.960903483309144,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035082439810919385,
      "loss": 4.8547,
      "mean_token_accuracy": 0.23074931204319,
      "num_tokens": 90613885.0,
      "step": 43665
    },
    {
      "entropy": 5.677077341079712,
      "epoch": 3.9613570391872277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003507939335746824,
      "loss": 4.8061,
      "mean_token_accuracy": 0.2214651510119438,
      "num_tokens": 90624665.0,
      "step": 43670
    },
    {
      "entropy": 5.646004390716553,
      "epoch": 3.961810595065312,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003507634674727332,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22370633184909822,
      "num_tokens": 90635496.0,
      "step": 43675
    },
    {
      "entropy": 5.697833728790283,
      "epoch": 3.9622641509433962,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003507329998039762,
      "loss": 4.8774,
      "mean_token_accuracy": 0.2209860160946846,
      "num_tokens": 90645502.0,
      "step": 43680
    },
    {
      "entropy": 5.580631494522095,
      "epoch": 3.9627177068214805,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035070253056904133,
      "loss": 4.7918,
      "mean_token_accuracy": 0.23052534013986586,
      "num_tokens": 90654724.0,
      "step": 43685
    },
    {
      "entropy": 5.622163105010986,
      "epoch": 3.9631712626995643,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003506720597685589,
      "loss": 4.8671,
      "mean_token_accuracy": 0.227325701713562,
      "num_tokens": 90665280.0,
      "step": 43690
    },
    {
      "entropy": 5.665242767333984,
      "epoch": 3.9636248185776486,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035064158740315894,
      "loss": 4.9233,
      "mean_token_accuracy": 0.221587111055851,
      "num_tokens": 90675804.0,
      "step": 43695
    },
    {
      "entropy": 5.6679527759552,
      "epoch": 3.964078374455733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035061111347347174,
      "loss": 4.8509,
      "mean_token_accuracy": 0.2242763251066208,
      "num_tokens": 90686636.0,
      "step": 43700
    },
    {
      "entropy": 5.640151596069336,
      "epoch": 3.964531930333817,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035058063798012746,
      "loss": 4.885,
      "mean_token_accuracy": 0.22426643520593642,
      "num_tokens": 90697666.0,
      "step": 43705
    },
    {
      "entropy": 5.727765130996704,
      "epoch": 3.9649854862119014,
      "grad_norm": 1.828125,
      "learning_rate": 0.0003505501609237563,
      "loss": 4.9771,
      "mean_token_accuracy": 0.21919141411781312,
      "num_tokens": 90708111.0,
      "step": 43710
    },
    {
      "entropy": 5.671432590484619,
      "epoch": 3.9654390420899857,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035051968230498853,
      "loss": 4.847,
      "mean_token_accuracy": 0.22903651744127274,
      "num_tokens": 90718711.0,
      "step": 43715
    },
    {
      "entropy": 5.616025638580322,
      "epoch": 3.96589259796807,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035048920212445454,
      "loss": 4.8329,
      "mean_token_accuracy": 0.231365367770195,
      "num_tokens": 90729585.0,
      "step": 43720
    },
    {
      "entropy": 5.5750213146209715,
      "epoch": 3.9663461538461537,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035045872038278467,
      "loss": 4.7708,
      "mean_token_accuracy": 0.23267118334770204,
      "num_tokens": 90739654.0,
      "step": 43725
    },
    {
      "entropy": 5.6397035121917725,
      "epoch": 3.966799709724238,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003504282370806091,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22622293680906297,
      "num_tokens": 90749456.0,
      "step": 43730
    },
    {
      "entropy": 5.626939821243286,
      "epoch": 3.9672532656023223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035039775221855845,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22617058604955673,
      "num_tokens": 90760051.0,
      "step": 43735
    },
    {
      "entropy": 5.60809326171875,
      "epoch": 3.9677068214804065,
      "grad_norm": 1.21875,
      "learning_rate": 0.000350367265797263,
      "loss": 4.826,
      "mean_token_accuracy": 0.22533978819847106,
      "num_tokens": 90770452.0,
      "step": 43740
    },
    {
      "entropy": 5.655235671997071,
      "epoch": 3.9681603773584904,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035033677781735336,
      "loss": 4.9061,
      "mean_token_accuracy": 0.22192625999450682,
      "num_tokens": 90780150.0,
      "step": 43745
    },
    {
      "entropy": 5.652624654769897,
      "epoch": 3.9686139332365746,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003503062882794599,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23167807906866072,
      "num_tokens": 90790338.0,
      "step": 43750
    },
    {
      "entropy": 5.6471888542175295,
      "epoch": 3.969067489114659,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035027579718421327,
      "loss": 4.87,
      "mean_token_accuracy": 0.22395725697278976,
      "num_tokens": 90800195.0,
      "step": 43755
    },
    {
      "entropy": 5.678124904632568,
      "epoch": 3.969521044992743,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003502453045322439,
      "loss": 4.9564,
      "mean_token_accuracy": 0.21838459521532058,
      "num_tokens": 90810872.0,
      "step": 43760
    },
    {
      "entropy": 5.674132204055786,
      "epoch": 3.9699746008708274,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035021481032418254,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22320411801338197,
      "num_tokens": 90821018.0,
      "step": 43765
    },
    {
      "entropy": 5.6932634830474855,
      "epoch": 3.9704281567489117,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003501843145606597,
      "loss": 4.8997,
      "mean_token_accuracy": 0.21695664674043655,
      "num_tokens": 90830274.0,
      "step": 43770
    },
    {
      "entropy": 5.631274795532226,
      "epoch": 3.9708817126269955,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000350153817242306,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23322552293539048,
      "num_tokens": 90839926.0,
      "step": 43775
    },
    {
      "entropy": 5.576013135910034,
      "epoch": 3.97133526850508,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003501233183697523,
      "loss": 4.7865,
      "mean_token_accuracy": 0.22268013060092925,
      "num_tokens": 90850263.0,
      "step": 43780
    },
    {
      "entropy": 5.658271551132202,
      "epoch": 3.971788824383164,
      "grad_norm": 1.25,
      "learning_rate": 0.00035009281794362926,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21503333896398544,
      "num_tokens": 90861747.0,
      "step": 43785
    },
    {
      "entropy": 5.680157852172852,
      "epoch": 3.9722423802612483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035006231596456755,
      "loss": 4.9774,
      "mean_token_accuracy": 0.2119024872779846,
      "num_tokens": 90872186.0,
      "step": 43790
    },
    {
      "entropy": 5.69822244644165,
      "epoch": 3.972695936139332,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035003181243319797,
      "loss": 4.8622,
      "mean_token_accuracy": 0.22887516766786575,
      "num_tokens": 90881088.0,
      "step": 43795
    },
    {
      "entropy": 5.593884086608886,
      "epoch": 3.9731494920174164,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003500013073501514,
      "loss": 4.844,
      "mean_token_accuracy": 0.22143300473690034,
      "num_tokens": 90890589.0,
      "step": 43800
    },
    {
      "entropy": 5.641145515441894,
      "epoch": 3.9736030478955007,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003499708007160588,
      "loss": 4.931,
      "mean_token_accuracy": 0.2198511466383934,
      "num_tokens": 90901692.0,
      "step": 43805
    },
    {
      "entropy": 5.7143457412719725,
      "epoch": 3.974056603773585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003499402925315509,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2204224780201912,
      "num_tokens": 90911811.0,
      "step": 43810
    },
    {
      "entropy": 5.69916877746582,
      "epoch": 3.974510159651669,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034990978279725846,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2313077464699745,
      "num_tokens": 90922323.0,
      "step": 43815
    },
    {
      "entropy": 5.784437036514282,
      "epoch": 3.9749637155297535,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034987927151381283,
      "loss": 5.0496,
      "mean_token_accuracy": 0.2066980391740799,
      "num_tokens": 90932834.0,
      "step": 43820
    },
    {
      "entropy": 5.698661184310913,
      "epoch": 3.9754172714078373,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034984875868184466,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22844714522361756,
      "num_tokens": 90943616.0,
      "step": 43825
    },
    {
      "entropy": 5.697509860992431,
      "epoch": 3.9758708272859216,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034981824430198517,
      "loss": 4.9746,
      "mean_token_accuracy": 0.21706928163766862,
      "num_tokens": 90954468.0,
      "step": 43830
    },
    {
      "entropy": 5.677630853652954,
      "epoch": 3.976324383164006,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034978772837486536,
      "loss": 4.8233,
      "mean_token_accuracy": 0.22597233206033707,
      "num_tokens": 90965109.0,
      "step": 43835
    },
    {
      "entropy": 5.6912651538848875,
      "epoch": 3.97677793904209,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034975721090111625,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22189621776342391,
      "num_tokens": 90974789.0,
      "step": 43840
    },
    {
      "entropy": 5.668655872344971,
      "epoch": 3.977231494920174,
      "grad_norm": 1.25,
      "learning_rate": 0.0003497266918813689,
      "loss": 4.9755,
      "mean_token_accuracy": 0.2170195236802101,
      "num_tokens": 90985154.0,
      "step": 43845
    },
    {
      "entropy": 5.664903926849365,
      "epoch": 3.977685050798258,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003496961713162545,
      "loss": 4.9255,
      "mean_token_accuracy": 0.22212358862161635,
      "num_tokens": 90996007.0,
      "step": 43850
    },
    {
      "entropy": 5.705722808837891,
      "epoch": 3.9781386066763424,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003496656492064043,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2135677695274353,
      "num_tokens": 91006253.0,
      "step": 43855
    },
    {
      "entropy": 5.736240339279175,
      "epoch": 3.9785921625544267,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003496351255524494,
      "loss": 5.0193,
      "mean_token_accuracy": 0.21396485716104507,
      "num_tokens": 91017169.0,
      "step": 43860
    },
    {
      "entropy": 5.667433547973633,
      "epoch": 3.979045718432511,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034960460035502117,
      "loss": 4.8738,
      "mean_token_accuracy": 0.21900187134742738,
      "num_tokens": 91026684.0,
      "step": 43865
    },
    {
      "entropy": 5.659739971160889,
      "epoch": 3.9794992743105952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003495740736147508,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2249004766345024,
      "num_tokens": 91037462.0,
      "step": 43870
    },
    {
      "entropy": 5.682798624038696,
      "epoch": 3.9799528301886795,
      "grad_norm": 1.125,
      "learning_rate": 0.00034954354533226953,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2138870894908905,
      "num_tokens": 91049418.0,
      "step": 43875
    },
    {
      "entropy": 5.625389623641968,
      "epoch": 3.9804063860667633,
      "grad_norm": 1.25,
      "learning_rate": 0.0003495130155082087,
      "loss": 4.834,
      "mean_token_accuracy": 0.22384393513202666,
      "num_tokens": 91061013.0,
      "step": 43880
    },
    {
      "entropy": 5.659660053253174,
      "epoch": 3.9808599419448476,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003494824841431998,
      "loss": 4.8886,
      "mean_token_accuracy": 0.22629764080047607,
      "num_tokens": 91070974.0,
      "step": 43885
    },
    {
      "entropy": 5.687782430648804,
      "epoch": 3.981313497822932,
      "grad_norm": 1.25,
      "learning_rate": 0.0003494519512378741,
      "loss": 4.9505,
      "mean_token_accuracy": 0.2256431683897972,
      "num_tokens": 91082681.0,
      "step": 43890
    },
    {
      "entropy": 5.665985488891602,
      "epoch": 3.9817670537010157,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034942141679286307,
      "loss": 4.8668,
      "mean_token_accuracy": 0.23574488162994384,
      "num_tokens": 91093552.0,
      "step": 43895
    },
    {
      "entropy": 5.687515687942505,
      "epoch": 3.9822206095791,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003493908808087982,
      "loss": 4.9261,
      "mean_token_accuracy": 0.2167578235268593,
      "num_tokens": 91105470.0,
      "step": 43900
    },
    {
      "entropy": 5.668826484680176,
      "epoch": 3.9826741654571842,
      "grad_norm": 1.140625,
      "learning_rate": 0.000349360343286311,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22682953923940657,
      "num_tokens": 91115822.0,
      "step": 43905
    },
    {
      "entropy": 5.635017395019531,
      "epoch": 3.9831277213352685,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034932980422603283,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23062579184770585,
      "num_tokens": 91125722.0,
      "step": 43910
    },
    {
      "entropy": 5.78600378036499,
      "epoch": 3.9835812772133528,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003492992636285955,
      "loss": 5.034,
      "mean_token_accuracy": 0.2230457440018654,
      "num_tokens": 91135783.0,
      "step": 43915
    },
    {
      "entropy": 5.683391761779785,
      "epoch": 3.984034833091437,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003492687214946304,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22641801983118057,
      "num_tokens": 91146151.0,
      "step": 43920
    },
    {
      "entropy": 5.671380138397216,
      "epoch": 3.9844883889695213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034923817782476927,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21648330688476564,
      "num_tokens": 91156711.0,
      "step": 43925
    },
    {
      "entropy": 5.644983911514283,
      "epoch": 3.984941944847605,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003492076326196436,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21649480909109114,
      "num_tokens": 91166601.0,
      "step": 43930
    },
    {
      "entropy": 5.730920743942261,
      "epoch": 3.9853955007256894,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034917708587988523,
      "loss": 4.9541,
      "mean_token_accuracy": 0.2093385711312294,
      "num_tokens": 91176556.0,
      "step": 43935
    },
    {
      "entropy": 5.655743741989136,
      "epoch": 3.9858490566037736,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034914653760612584,
      "loss": 4.864,
      "mean_token_accuracy": 0.22366306483745574,
      "num_tokens": 91186499.0,
      "step": 43940
    },
    {
      "entropy": 5.643047571182251,
      "epoch": 3.986302612481858,
      "grad_norm": 1.25,
      "learning_rate": 0.0003491159877989971,
      "loss": 4.8642,
      "mean_token_accuracy": 0.21299372166395186,
      "num_tokens": 91197088.0,
      "step": 43945
    },
    {
      "entropy": 5.645539665222168,
      "epoch": 3.9867561683599417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000349085436459131,
      "loss": 4.9088,
      "mean_token_accuracy": 0.22560027092695237,
      "num_tokens": 91207188.0,
      "step": 43950
    },
    {
      "entropy": 5.643137979507446,
      "epoch": 3.987209724238026,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034905488358715907,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22218265235424042,
      "num_tokens": 91217349.0,
      "step": 43955
    },
    {
      "entropy": 5.699260997772217,
      "epoch": 3.9876632801161103,
      "grad_norm": 1.25,
      "learning_rate": 0.0003490243291837133,
      "loss": 4.9307,
      "mean_token_accuracy": 0.22137006670236586,
      "num_tokens": 91227775.0,
      "step": 43960
    },
    {
      "entropy": 5.659594202041626,
      "epoch": 3.9881168359941945,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034899377324942556,
      "loss": 4.7973,
      "mean_token_accuracy": 0.232569183409214,
      "num_tokens": 91238315.0,
      "step": 43965
    },
    {
      "entropy": 5.673978233337403,
      "epoch": 3.988570391872279,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034896321578492767,
      "loss": 4.9987,
      "mean_token_accuracy": 0.21239070743322372,
      "num_tokens": 91249178.0,
      "step": 43970
    },
    {
      "entropy": 5.634776592254639,
      "epoch": 3.989023947750363,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003489326567908516,
      "loss": 4.8647,
      "mean_token_accuracy": 0.22213993966579437,
      "num_tokens": 91257911.0,
      "step": 43975
    },
    {
      "entropy": 5.66892147064209,
      "epoch": 3.989477503628447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034890209626782943,
      "loss": 4.9018,
      "mean_token_accuracy": 0.22171754539012908,
      "num_tokens": 91267491.0,
      "step": 43980
    },
    {
      "entropy": 5.662269067764282,
      "epoch": 3.989931059506531,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003488715342164931,
      "loss": 4.8892,
      "mean_token_accuracy": 0.22416286021471024,
      "num_tokens": 91278292.0,
      "step": 43985
    },
    {
      "entropy": 5.711904764175415,
      "epoch": 3.9903846153846154,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00034884097063747453,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21531758159399034,
      "num_tokens": 91287871.0,
      "step": 43990
    },
    {
      "entropy": 5.62492790222168,
      "epoch": 3.9908381712626997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003488104055314059,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2305402785539627,
      "num_tokens": 91297201.0,
      "step": 43995
    },
    {
      "entropy": 5.584606409072876,
      "epoch": 3.9912917271407835,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003487798388989193,
      "loss": 4.7844,
      "mean_token_accuracy": 0.23055097311735154,
      "num_tokens": 91306449.0,
      "step": 44000
    },
    {
      "entropy": 5.57442045211792,
      "epoch": 3.9917452830188678,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034874927074064684,
      "loss": 4.7969,
      "mean_token_accuracy": 0.22127773016691207,
      "num_tokens": 91317516.0,
      "step": 44005
    },
    {
      "entropy": 5.664388370513916,
      "epoch": 3.992198838896952,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003487187010572206,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21966809183359146,
      "num_tokens": 91326850.0,
      "step": 44010
    },
    {
      "entropy": 5.669976854324341,
      "epoch": 3.9926523947750363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003486881298492729,
      "loss": 4.9068,
      "mean_token_accuracy": 0.22310503870248793,
      "num_tokens": 91338068.0,
      "step": 44015
    },
    {
      "entropy": 5.746202421188355,
      "epoch": 3.9931059506531206,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003486575571174359,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22514592111110687,
      "num_tokens": 91349627.0,
      "step": 44020
    },
    {
      "entropy": 5.686382913589478,
      "epoch": 3.993559506531205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034862698286234185,
      "loss": 4.9649,
      "mean_token_accuracy": 0.21305015087127685,
      "num_tokens": 91361229.0,
      "step": 44025
    },
    {
      "entropy": 5.679896306991577,
      "epoch": 3.9940130624092887,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003485964070846229,
      "loss": 4.9855,
      "mean_token_accuracy": 0.2128543496131897,
      "num_tokens": 91372348.0,
      "step": 44030
    },
    {
      "entropy": 5.627528333663941,
      "epoch": 3.994466618287373,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003485658297849117,
      "loss": 4.8377,
      "mean_token_accuracy": 0.22429872751235963,
      "num_tokens": 91383028.0,
      "step": 44035
    },
    {
      "entropy": 5.688759613037109,
      "epoch": 3.994920174165457,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034853525096384034,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2189745455980301,
      "num_tokens": 91393263.0,
      "step": 44040
    },
    {
      "entropy": 5.724278926849365,
      "epoch": 3.9953737300435415,
      "grad_norm": 1.375,
      "learning_rate": 0.0003485046706220411,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22322145402431487,
      "num_tokens": 91403881.0,
      "step": 44045
    },
    {
      "entropy": 5.705347490310669,
      "epoch": 3.9958272859216253,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003484740887601467,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21835314631462097,
      "num_tokens": 91413723.0,
      "step": 44050
    },
    {
      "entropy": 5.694570159912109,
      "epoch": 3.9962808417997095,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034844350537878937,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22168629765510559,
      "num_tokens": 91424168.0,
      "step": 44055
    },
    {
      "entropy": 5.757430505752564,
      "epoch": 3.996734397677794,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034841292047860164,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21228594034910203,
      "num_tokens": 91433133.0,
      "step": 44060
    },
    {
      "entropy": 5.692979335784912,
      "epoch": 3.997187953555878,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000348382334060216,
      "loss": 5.0022,
      "mean_token_accuracy": 0.20999932438135147,
      "num_tokens": 91444521.0,
      "step": 44065
    },
    {
      "entropy": 5.597850227355957,
      "epoch": 3.9976415094339623,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000348351746124265,
      "loss": 4.8755,
      "mean_token_accuracy": 0.22766720056533812,
      "num_tokens": 91454870.0,
      "step": 44070
    },
    {
      "entropy": 5.679509258270263,
      "epoch": 3.9980950653120466,
      "grad_norm": 1.25,
      "learning_rate": 0.0003483211566713812,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22879617810249328,
      "num_tokens": 91465472.0,
      "step": 44075
    },
    {
      "entropy": 5.7269110679626465,
      "epoch": 3.998548621190131,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003482905657021971,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2221173882484436,
      "num_tokens": 91475467.0,
      "step": 44080
    },
    {
      "entropy": 5.728068685531616,
      "epoch": 3.9990021770682147,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003482599732173456,
      "loss": 4.9216,
      "mean_token_accuracy": 0.22487736046314238,
      "num_tokens": 91485615.0,
      "step": 44085
    },
    {
      "entropy": 5.6725341796875,
      "epoch": 3.999455732946299,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003482293792174591,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22877063304185868,
      "num_tokens": 91495790.0,
      "step": 44090
    },
    {
      "entropy": 5.665362310409546,
      "epoch": 3.9999092888243832,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003481987837031703,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21953017264604568,
      "num_tokens": 91507164.0,
      "step": 44095
    },
    {
      "entropy": 5.62367000579834,
      "epoch": 4.000362844702467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003481681866751121,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23152355700731278,
      "num_tokens": 91516732.0,
      "step": 44100
    },
    {
      "entropy": 5.609641122817993,
      "epoch": 4.000816400580551,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034813758813391716,
      "loss": 4.7843,
      "mean_token_accuracy": 0.2339000418782234,
      "num_tokens": 91527073.0,
      "step": 44105
    },
    {
      "entropy": 5.639034366607666,
      "epoch": 4.001269956458636,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003481069880802183,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2266480579972267,
      "num_tokens": 91538314.0,
      "step": 44110
    },
    {
      "entropy": 5.668161392211914,
      "epoch": 4.00172351233672,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034807638651464815,
      "loss": 4.8761,
      "mean_token_accuracy": 0.2249499797821045,
      "num_tokens": 91549159.0,
      "step": 44115
    },
    {
      "entropy": 5.660141658782959,
      "epoch": 4.002177068214804,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003480457834378398,
      "loss": 4.8665,
      "mean_token_accuracy": 0.22389106452465057,
      "num_tokens": 91560901.0,
      "step": 44120
    },
    {
      "entropy": 5.65052752494812,
      "epoch": 4.002630624092888,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034801517885042597,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22240670621395112,
      "num_tokens": 91570797.0,
      "step": 44125
    },
    {
      "entropy": 5.754239320755005,
      "epoch": 4.003084179970973,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003479845727530397,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21522759944200515,
      "num_tokens": 91580971.0,
      "step": 44130
    },
    {
      "entropy": 5.6647007942199705,
      "epoch": 4.003537735849057,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034795396514631386,
      "loss": 4.8728,
      "mean_token_accuracy": 0.23024609982967376,
      "num_tokens": 91590834.0,
      "step": 44135
    },
    {
      "entropy": 5.658629846572876,
      "epoch": 4.003991291727141,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034792335603088135,
      "loss": 4.8981,
      "mean_token_accuracy": 0.21986954659223557,
      "num_tokens": 91602099.0,
      "step": 44140
    },
    {
      "entropy": 5.632980442047119,
      "epoch": 4.004444847605225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034789274540737535,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23817332834005356,
      "num_tokens": 91612269.0,
      "step": 44145
    },
    {
      "entropy": 5.715988397598267,
      "epoch": 4.004898403483309,
      "grad_norm": 1.25,
      "learning_rate": 0.00034786213327642874,
      "loss": 4.8617,
      "mean_token_accuracy": 0.22462040036916733,
      "num_tokens": 91623644.0,
      "step": 44150
    },
    {
      "entropy": 5.682221460342407,
      "epoch": 4.005351959361393,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003478315196386747,
      "loss": 4.835,
      "mean_token_accuracy": 0.2328446999192238,
      "num_tokens": 91633563.0,
      "step": 44155
    },
    {
      "entropy": 5.629967498779297,
      "epoch": 4.005805515239477,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003478009044947463,
      "loss": 4.8112,
      "mean_token_accuracy": 0.2278311640024185,
      "num_tokens": 91644648.0,
      "step": 44160
    },
    {
      "entropy": 5.57977819442749,
      "epoch": 4.006259071117562,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003477702878452766,
      "loss": 4.6937,
      "mean_token_accuracy": 0.2477676525712013,
      "num_tokens": 91655366.0,
      "step": 44165
    },
    {
      "entropy": 5.607641983032226,
      "epoch": 4.006712626995646,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003477396696908988,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2301895424723625,
      "num_tokens": 91665474.0,
      "step": 44170
    },
    {
      "entropy": 5.668443250656128,
      "epoch": 4.00716618287373,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034770905003224606,
      "loss": 4.8534,
      "mean_token_accuracy": 0.21977997571229935,
      "num_tokens": 91675291.0,
      "step": 44175
    },
    {
      "entropy": 5.586932802200318,
      "epoch": 4.007619738751814,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003476784288699517,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23321352750062943,
      "num_tokens": 91686472.0,
      "step": 44180
    },
    {
      "entropy": 5.684142255783081,
      "epoch": 4.008073294629899,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034764780620464886,
      "loss": 4.8905,
      "mean_token_accuracy": 0.22337005585432052,
      "num_tokens": 91696347.0,
      "step": 44185
    },
    {
      "entropy": 5.63988242149353,
      "epoch": 4.008526850507983,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003476171820369709,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2208269640803337,
      "num_tokens": 91706504.0,
      "step": 44190
    },
    {
      "entropy": 5.6283097743988035,
      "epoch": 4.008980406386066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034758655636755103,
      "loss": 4.7959,
      "mean_token_accuracy": 0.2407778799533844,
      "num_tokens": 91716559.0,
      "step": 44195
    },
    {
      "entropy": 5.686963653564453,
      "epoch": 4.009433962264151,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034755592919702275,
      "loss": 4.8476,
      "mean_token_accuracy": 0.22040775269269944,
      "num_tokens": 91726312.0,
      "step": 44200
    },
    {
      "entropy": 5.61539945602417,
      "epoch": 4.009887518142235,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034752530052601944,
      "loss": 4.7939,
      "mean_token_accuracy": 0.2281578853726387,
      "num_tokens": 91736882.0,
      "step": 44205
    },
    {
      "entropy": 5.643258762359619,
      "epoch": 4.010341074020319,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034749467035517437,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22000845968723298,
      "num_tokens": 91746905.0,
      "step": 44210
    },
    {
      "entropy": 5.6665655136108395,
      "epoch": 4.010794629898403,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034746403868512094,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22082189470529556,
      "num_tokens": 91757609.0,
      "step": 44215
    },
    {
      "entropy": 5.690264177322388,
      "epoch": 4.011248185776488,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034743340551649285,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23034935891628266,
      "num_tokens": 91768017.0,
      "step": 44220
    },
    {
      "entropy": 5.5815088748931885,
      "epoch": 4.011701741654572,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034740277084992346,
      "loss": 4.7273,
      "mean_token_accuracy": 0.23843913972377778,
      "num_tokens": 91777820.0,
      "step": 44225
    },
    {
      "entropy": 5.593440008163452,
      "epoch": 4.012155297532656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034737213468604626,
      "loss": 4.8411,
      "mean_token_accuracy": 0.21970656961202623,
      "num_tokens": 91789011.0,
      "step": 44230
    },
    {
      "entropy": 5.677101802825928,
      "epoch": 4.0126088534107405,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003473414970254949,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22159491628408431,
      "num_tokens": 91799165.0,
      "step": 44235
    },
    {
      "entropy": 5.646427488327026,
      "epoch": 4.013062409288825,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034731085786890306,
      "loss": 4.8722,
      "mean_token_accuracy": 0.227496874332428,
      "num_tokens": 91809602.0,
      "step": 44240
    },
    {
      "entropy": 5.690667772293091,
      "epoch": 4.013515965166909,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034728021721690407,
      "loss": 4.9247,
      "mean_token_accuracy": 0.221859672665596,
      "num_tokens": 91820092.0,
      "step": 44245
    },
    {
      "entropy": 5.6781834125518795,
      "epoch": 4.013969521044992,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003472495750701318,
      "loss": 4.8442,
      "mean_token_accuracy": 0.2285470813512802,
      "num_tokens": 91829730.0,
      "step": 44250
    },
    {
      "entropy": 5.674529409408569,
      "epoch": 4.014423076923077,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034721893142922,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22150059938430786,
      "num_tokens": 91840697.0,
      "step": 44255
    },
    {
      "entropy": 5.661016130447388,
      "epoch": 4.014876632801161,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003471882862948023,
      "loss": 4.8697,
      "mean_token_accuracy": 0.22147967517375947,
      "num_tokens": 91852145.0,
      "step": 44260
    },
    {
      "entropy": 5.680552291870117,
      "epoch": 4.015330188679245,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034715763966751233,
      "loss": 4.8282,
      "mean_token_accuracy": 0.22101673036813735,
      "num_tokens": 91861940.0,
      "step": 44265
    },
    {
      "entropy": 5.726807165145874,
      "epoch": 4.0157837445573294,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034712699154798403,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22533690333366393,
      "num_tokens": 91872179.0,
      "step": 44270
    },
    {
      "entropy": 5.68064169883728,
      "epoch": 4.016237300435414,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034709634193685126,
      "loss": 4.8029,
      "mean_token_accuracy": 0.22966966181993484,
      "num_tokens": 91882051.0,
      "step": 44275
    },
    {
      "entropy": 5.63207516670227,
      "epoch": 4.016690856313498,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003470656908347476,
      "loss": 4.8685,
      "mean_token_accuracy": 0.22183433622121812,
      "num_tokens": 91892672.0,
      "step": 44280
    },
    {
      "entropy": 5.590099620819092,
      "epoch": 4.017144412191582,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003470350382423072,
      "loss": 4.7558,
      "mean_token_accuracy": 0.23339740931987762,
      "num_tokens": 91903418.0,
      "step": 44285
    },
    {
      "entropy": 5.630136013031006,
      "epoch": 4.0175979680696665,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003470043841601639,
      "loss": 4.8398,
      "mean_token_accuracy": 0.23290229886770247,
      "num_tokens": 91913076.0,
      "step": 44290
    },
    {
      "entropy": 5.61550121307373,
      "epoch": 4.018051523947751,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034697372858895147,
      "loss": 4.8284,
      "mean_token_accuracy": 0.23611175268888474,
      "num_tokens": 91923464.0,
      "step": 44295
    },
    {
      "entropy": 5.697745132446289,
      "epoch": 4.018505079825834,
      "grad_norm": 1.328125,
      "learning_rate": 0.000346943071529304,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22109244018793106,
      "num_tokens": 91934399.0,
      "step": 44300
    },
    {
      "entropy": 5.742668533325196,
      "epoch": 4.018958635703918,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034691241298185553,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22678199261426926,
      "num_tokens": 91945330.0,
      "step": 44305
    },
    {
      "entropy": 5.696223402023316,
      "epoch": 4.019412191582003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034688175294724004,
      "loss": 4.8893,
      "mean_token_accuracy": 0.21869195699691774,
      "num_tokens": 91956379.0,
      "step": 44310
    },
    {
      "entropy": 5.552106237411499,
      "epoch": 4.019865747460087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034685109142609154,
      "loss": 4.7101,
      "mean_token_accuracy": 0.2322893977165222,
      "num_tokens": 91965969.0,
      "step": 44315
    },
    {
      "entropy": 5.719584178924561,
      "epoch": 4.020319303338171,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003468204284190442,
      "loss": 4.9806,
      "mean_token_accuracy": 0.2136175036430359,
      "num_tokens": 91977005.0,
      "step": 44320
    },
    {
      "entropy": 5.689169931411743,
      "epoch": 4.0207728592162555,
      "grad_norm": 1.25,
      "learning_rate": 0.00034678976392673206,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22458529472351074,
      "num_tokens": 91988735.0,
      "step": 44325
    },
    {
      "entropy": 5.688523006439209,
      "epoch": 4.02122641509434,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034675909794978926,
      "loss": 4.8546,
      "mean_token_accuracy": 0.22859515845775605,
      "num_tokens": 91998743.0,
      "step": 44330
    },
    {
      "entropy": 5.681196832656861,
      "epoch": 4.021679970972424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034672843048885017,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22578403055667878,
      "num_tokens": 92009465.0,
      "step": 44335
    },
    {
      "entropy": 5.601125860214234,
      "epoch": 4.022133526850508,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034669776154454875,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23315410763025285,
      "num_tokens": 92020818.0,
      "step": 44340
    },
    {
      "entropy": 5.672318077087402,
      "epoch": 4.0225870827285926,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034666709111751945,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22158267945051194,
      "num_tokens": 92031528.0,
      "step": 44345
    },
    {
      "entropy": 5.7074450016021725,
      "epoch": 4.023040638606676,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003466364192083963,
      "loss": 4.9113,
      "mean_token_accuracy": 0.22390609830617905,
      "num_tokens": 92041856.0,
      "step": 44350
    },
    {
      "entropy": 5.695340156555176,
      "epoch": 4.02349419448476,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034660574581781386,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23131307661533357,
      "num_tokens": 92052117.0,
      "step": 44355
    },
    {
      "entropy": 5.691923713684082,
      "epoch": 4.0239477503628445,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003465750709464063,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22301280349493027,
      "num_tokens": 92062405.0,
      "step": 44360
    },
    {
      "entropy": 5.674297904968261,
      "epoch": 4.024401306240929,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000346544394594808,
      "loss": 4.8915,
      "mean_token_accuracy": 0.22742346674203873,
      "num_tokens": 92073269.0,
      "step": 44365
    },
    {
      "entropy": 5.643460416793824,
      "epoch": 4.024854862119013,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034651371676365343,
      "loss": 4.7971,
      "mean_token_accuracy": 0.23453737646341324,
      "num_tokens": 92083645.0,
      "step": 44370
    },
    {
      "entropy": 5.756990957260132,
      "epoch": 4.025308417997097,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034648303745357705,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22420720756053925,
      "num_tokens": 92093696.0,
      "step": 44375
    },
    {
      "entropy": 5.662577962875366,
      "epoch": 4.0257619738751815,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003464523566652131,
      "loss": 4.874,
      "mean_token_accuracy": 0.22283466905355453,
      "num_tokens": 92103217.0,
      "step": 44380
    },
    {
      "entropy": 5.607229948043823,
      "epoch": 4.026215529753266,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034642167439919634,
      "loss": 4.7877,
      "mean_token_accuracy": 0.22635677754878997,
      "num_tokens": 92112934.0,
      "step": 44385
    },
    {
      "entropy": 5.612427234649658,
      "epoch": 4.02666908563135,
      "grad_norm": 1.25,
      "learning_rate": 0.0003463909906561611,
      "loss": 4.8537,
      "mean_token_accuracy": 0.22943898290395737,
      "num_tokens": 92123066.0,
      "step": 44390
    },
    {
      "entropy": 5.683171558380127,
      "epoch": 4.027122641509434,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003463603054367419,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22709472328424454,
      "num_tokens": 92133911.0,
      "step": 44395
    },
    {
      "entropy": 5.608241939544678,
      "epoch": 4.027576197387518,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003463296187415735,
      "loss": 4.7845,
      "mean_token_accuracy": 0.23208570182323457,
      "num_tokens": 92143941.0,
      "step": 44400
    },
    {
      "entropy": 5.624104022979736,
      "epoch": 4.028029753265602,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003462989305712904,
      "loss": 4.8016,
      "mean_token_accuracy": 0.2274986431002617,
      "num_tokens": 92153934.0,
      "step": 44405
    },
    {
      "entropy": 5.612405300140381,
      "epoch": 4.028483309143686,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003462682409265272,
      "loss": 4.7815,
      "mean_token_accuracy": 0.2356450840830803,
      "num_tokens": 92163642.0,
      "step": 44410
    },
    {
      "entropy": 5.63348970413208,
      "epoch": 4.0289368650217705,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003462375498079186,
      "loss": 4.8621,
      "mean_token_accuracy": 0.22458656132221222,
      "num_tokens": 92174034.0,
      "step": 44415
    },
    {
      "entropy": 5.683361673355103,
      "epoch": 4.029390420899855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034620685721609937,
      "loss": 4.8513,
      "mean_token_accuracy": 0.22842602729797362,
      "num_tokens": 92183714.0,
      "step": 44420
    },
    {
      "entropy": 5.680710887908935,
      "epoch": 4.029843976777939,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034617616315170413,
      "loss": 4.9158,
      "mean_token_accuracy": 0.22030563801527023,
      "num_tokens": 92195291.0,
      "step": 44425
    },
    {
      "entropy": 5.663345384597778,
      "epoch": 4.030297532656023,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003461454676153678,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2296669900417328,
      "num_tokens": 92206159.0,
      "step": 44430
    },
    {
      "entropy": 5.681922769546508,
      "epoch": 4.030751088534108,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000346114770607725,
      "loss": 4.894,
      "mean_token_accuracy": 0.2254640281200409,
      "num_tokens": 92216928.0,
      "step": 44435
    },
    {
      "entropy": 5.590759754180908,
      "epoch": 4.031204644412192,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003460840721294106,
      "loss": 4.8076,
      "mean_token_accuracy": 0.22976668179035187,
      "num_tokens": 92227029.0,
      "step": 44440
    },
    {
      "entropy": 5.658042335510254,
      "epoch": 4.031658200290276,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034605337218105955,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22761137336492537,
      "num_tokens": 92237310.0,
      "step": 44445
    },
    {
      "entropy": 5.631627368927002,
      "epoch": 4.03211175616836,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003460226707633066,
      "loss": 4.8298,
      "mean_token_accuracy": 0.22943200170993805,
      "num_tokens": 92247841.0,
      "step": 44450
    },
    {
      "entropy": 5.670345163345337,
      "epoch": 4.032565312046444,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034599196787678667,
      "loss": 4.8547,
      "mean_token_accuracy": 0.22401938289403917,
      "num_tokens": 92258513.0,
      "step": 44455
    },
    {
      "entropy": 5.633939743041992,
      "epoch": 4.033018867924528,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003459612635221349,
      "loss": 4.8636,
      "mean_token_accuracy": 0.2253674179315567,
      "num_tokens": 92268659.0,
      "step": 44460
    },
    {
      "entropy": 5.634753799438476,
      "epoch": 4.033472423802612,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034593055769998595,
      "loss": 4.8572,
      "mean_token_accuracy": 0.22036066949367522,
      "num_tokens": 92279810.0,
      "step": 44465
    },
    {
      "entropy": 5.664070272445679,
      "epoch": 4.0339259796806965,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003458998504109751,
      "loss": 4.8533,
      "mean_token_accuracy": 0.22927334308624267,
      "num_tokens": 92289625.0,
      "step": 44470
    },
    {
      "entropy": 5.6108012199401855,
      "epoch": 4.034379535558781,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003458691416557372,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22815120667219163,
      "num_tokens": 92300669.0,
      "step": 44475
    },
    {
      "entropy": 5.656001996994019,
      "epoch": 4.034833091436865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034583843143490744,
      "loss": 4.8804,
      "mean_token_accuracy": 0.23028798550367355,
      "num_tokens": 92312160.0,
      "step": 44480
    },
    {
      "entropy": 5.708610582351684,
      "epoch": 4.035286647314949,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003458077197491208,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2161591589450836,
      "num_tokens": 92322677.0,
      "step": 44485
    },
    {
      "entropy": 5.641248512268066,
      "epoch": 4.035740203193034,
      "grad_norm": 1.25,
      "learning_rate": 0.0003457770065990126,
      "loss": 4.782,
      "mean_token_accuracy": 0.23111440986394882,
      "num_tokens": 92333148.0,
      "step": 44490
    },
    {
      "entropy": 5.6804444789886475,
      "epoch": 4.036193759071118,
      "grad_norm": 1.375,
      "learning_rate": 0.0003457462919852178,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22654978483915328,
      "num_tokens": 92344189.0,
      "step": 44495
    },
    {
      "entropy": 5.6912617683410645,
      "epoch": 4.036647314949202,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003457155759083717,
      "loss": 4.8436,
      "mean_token_accuracy": 0.23271348476409912,
      "num_tokens": 92356136.0,
      "step": 44500
    },
    {
      "entropy": 5.616119813919068,
      "epoch": 4.0371008708272855,
      "grad_norm": 1.25,
      "learning_rate": 0.0003456848583691094,
      "loss": 4.7463,
      "mean_token_accuracy": 0.2305932894349098,
      "num_tokens": 92366393.0,
      "step": 44505
    },
    {
      "entropy": 5.638161134719849,
      "epoch": 4.03755442670537,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003456541393680662,
      "loss": 4.7879,
      "mean_token_accuracy": 0.2310309961438179,
      "num_tokens": 92376819.0,
      "step": 44510
    },
    {
      "entropy": 5.642312240600586,
      "epoch": 4.038007982583454,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034562341890587747,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22103004455566405,
      "num_tokens": 92386704.0,
      "step": 44515
    },
    {
      "entropy": 5.65023398399353,
      "epoch": 4.038461538461538,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003455926969831784,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21993288546800613,
      "num_tokens": 92397412.0,
      "step": 44520
    },
    {
      "entropy": 5.725669908523559,
      "epoch": 4.038915094339623,
      "grad_norm": 1.25,
      "learning_rate": 0.00034556197360060436,
      "loss": 4.9439,
      "mean_token_accuracy": 0.22121664732694626,
      "num_tokens": 92408641.0,
      "step": 44525
    },
    {
      "entropy": 5.649687671661377,
      "epoch": 4.039368650217707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003455312487587908,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2286076918244362,
      "num_tokens": 92419341.0,
      "step": 44530
    },
    {
      "entropy": 5.7088817119598385,
      "epoch": 4.039822206095791,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003455005224583729,
      "loss": 4.8755,
      "mean_token_accuracy": 0.22246959805488586,
      "num_tokens": 92429297.0,
      "step": 44535
    },
    {
      "entropy": 5.739550065994263,
      "epoch": 4.040275761973875,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034546979469998637,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22285029739141465,
      "num_tokens": 92439851.0,
      "step": 44540
    },
    {
      "entropy": 5.590083408355713,
      "epoch": 4.04072931785196,
      "grad_norm": 1.25,
      "learning_rate": 0.00034543906548426646,
      "loss": 4.7981,
      "mean_token_accuracy": 0.23707572221755982,
      "num_tokens": 92451096.0,
      "step": 44545
    },
    {
      "entropy": 5.644199371337891,
      "epoch": 4.041182873730044,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034540833481184867,
      "loss": 4.8384,
      "mean_token_accuracy": 0.22562169134616852,
      "num_tokens": 92460516.0,
      "step": 44550
    },
    {
      "entropy": 5.734266376495361,
      "epoch": 4.041636429608127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034537760268336866,
      "loss": 4.9131,
      "mean_token_accuracy": 0.22014155089855195,
      "num_tokens": 92471429.0,
      "step": 44555
    },
    {
      "entropy": 5.672460412979126,
      "epoch": 4.042089985486212,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003453468690994619,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22560233771800994,
      "num_tokens": 92482047.0,
      "step": 44560
    },
    {
      "entropy": 5.609123849868775,
      "epoch": 4.042543541364296,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034531613406076387,
      "loss": 4.8033,
      "mean_token_accuracy": 0.23258839398622513,
      "num_tokens": 92492206.0,
      "step": 44565
    },
    {
      "entropy": 5.61236572265625,
      "epoch": 4.04299709724238,
      "grad_norm": 1.25,
      "learning_rate": 0.0003452853975679103,
      "loss": 4.7637,
      "mean_token_accuracy": 0.2292274311184883,
      "num_tokens": 92501979.0,
      "step": 44570
    },
    {
      "entropy": 5.765503072738648,
      "epoch": 4.043450653120464,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034525465962153683,
      "loss": 4.9214,
      "mean_token_accuracy": 0.22355792969465255,
      "num_tokens": 92512388.0,
      "step": 44575
    },
    {
      "entropy": 5.713842105865479,
      "epoch": 4.043904208998549,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034522392022227915,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21787252724170686,
      "num_tokens": 92522805.0,
      "step": 44580
    },
    {
      "entropy": 5.6409131526947025,
      "epoch": 4.044357764876633,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034519317937077283,
      "loss": 4.852,
      "mean_token_accuracy": 0.21904472261667252,
      "num_tokens": 92532957.0,
      "step": 44585
    },
    {
      "entropy": 5.6419586658477785,
      "epoch": 4.044811320754717,
      "grad_norm": 1.375,
      "learning_rate": 0.0003451624370676537,
      "loss": 4.8694,
      "mean_token_accuracy": 0.2216425746679306,
      "num_tokens": 92543451.0,
      "step": 44590
    },
    {
      "entropy": 5.759320878982544,
      "epoch": 4.045264876632801,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034513169331355754,
      "loss": 4.8583,
      "mean_token_accuracy": 0.2199556663632393,
      "num_tokens": 92553856.0,
      "step": 44595
    },
    {
      "entropy": 5.624948883056641,
      "epoch": 4.045718432510886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034510094810911995,
      "loss": 4.7908,
      "mean_token_accuracy": 0.23299646228551865,
      "num_tokens": 92563816.0,
      "step": 44600
    },
    {
      "entropy": 5.5846916198730465,
      "epoch": 4.04617198838897,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034507020145497705,
      "loss": 4.8645,
      "mean_token_accuracy": 0.23018748313188553,
      "num_tokens": 92574062.0,
      "step": 44605
    },
    {
      "entropy": 5.595385599136352,
      "epoch": 4.046625544267053,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034503945335176444,
      "loss": 4.7991,
      "mean_token_accuracy": 0.2357534497976303,
      "num_tokens": 92583935.0,
      "step": 44610
    },
    {
      "entropy": 5.708527374267578,
      "epoch": 4.047079100145138,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034500870380011807,
      "loss": 4.8707,
      "mean_token_accuracy": 0.23073073029518126,
      "num_tokens": 92594037.0,
      "step": 44615
    },
    {
      "entropy": 5.632271385192871,
      "epoch": 4.047532656023222,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034497795280067386,
      "loss": 4.8277,
      "mean_token_accuracy": 0.22318391799926757,
      "num_tokens": 92603336.0,
      "step": 44620
    },
    {
      "entropy": 5.587202072143555,
      "epoch": 4.047986211901306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034494720035406777,
      "loss": 4.7643,
      "mean_token_accuracy": 0.23421523869037628,
      "num_tokens": 92614210.0,
      "step": 44625
    },
    {
      "entropy": 5.698281288146973,
      "epoch": 4.04843976777939,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003449164464609358,
      "loss": 4.8767,
      "mean_token_accuracy": 0.2246902123093605,
      "num_tokens": 92624036.0,
      "step": 44630
    },
    {
      "entropy": 5.680367660522461,
      "epoch": 4.048893323657475,
      "grad_norm": 1.25,
      "learning_rate": 0.0003448856911219139,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2162610486149788,
      "num_tokens": 92634588.0,
      "step": 44635
    },
    {
      "entropy": 5.655884265899658,
      "epoch": 4.049346879535559,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034485493433763804,
      "loss": 4.8166,
      "mean_token_accuracy": 0.22789756804704667,
      "num_tokens": 92645027.0,
      "step": 44640
    },
    {
      "entropy": 5.792963933944702,
      "epoch": 4.049800435413643,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034482417610874437,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2181426152586937,
      "num_tokens": 92656669.0,
      "step": 44645
    },
    {
      "entropy": 5.679650592803955,
      "epoch": 4.0502539912917275,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003447934164358689,
      "loss": 4.9285,
      "mean_token_accuracy": 0.2216082990169525,
      "num_tokens": 92666558.0,
      "step": 44650
    },
    {
      "entropy": 5.637512874603272,
      "epoch": 4.050707547169812,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034476265531964786,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22565652132034303,
      "num_tokens": 92677384.0,
      "step": 44655
    },
    {
      "entropy": 5.614072370529175,
      "epoch": 4.051161103047895,
      "grad_norm": 1.609375,
      "learning_rate": 0.0003447318927607173,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2319969803094864,
      "num_tokens": 92687619.0,
      "step": 44660
    },
    {
      "entropy": 5.599501895904541,
      "epoch": 4.051614658925979,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003447011287597134,
      "loss": 4.749,
      "mean_token_accuracy": 0.23312059342861174,
      "num_tokens": 92697784.0,
      "step": 44665
    },
    {
      "entropy": 5.669119167327881,
      "epoch": 4.052068214804064,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003446703633172724,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21804513037204742,
      "num_tokens": 92709255.0,
      "step": 44670
    },
    {
      "entropy": 5.627715635299682,
      "epoch": 4.052521770682148,
      "grad_norm": 1.375,
      "learning_rate": 0.0003446395964340305,
      "loss": 4.803,
      "mean_token_accuracy": 0.23277949839830397,
      "num_tokens": 92720714.0,
      "step": 44675
    },
    {
      "entropy": 5.649722528457642,
      "epoch": 4.052975326560232,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034460882811062417,
      "loss": 4.8121,
      "mean_token_accuracy": 0.2265733599662781,
      "num_tokens": 92730783.0,
      "step": 44680
    },
    {
      "entropy": 5.697773218154907,
      "epoch": 4.0534288824383164,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034457805834768937,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2200439915060997,
      "num_tokens": 92741943.0,
      "step": 44685
    },
    {
      "entropy": 5.627641344070435,
      "epoch": 4.053882438316401,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003445472871458627,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2277112364768982,
      "num_tokens": 92753510.0,
      "step": 44690
    },
    {
      "entropy": 5.629557943344116,
      "epoch": 4.054335994194485,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003445165145057803,
      "loss": 4.8052,
      "mean_token_accuracy": 0.23348532915115355,
      "num_tokens": 92763655.0,
      "step": 44695
    },
    {
      "entropy": 5.691075134277344,
      "epoch": 4.054789550072569,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034448574042807866,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22064407616853715,
      "num_tokens": 92773732.0,
      "step": 44700
    },
    {
      "entropy": 5.714195108413696,
      "epoch": 4.0552431059506535,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003444549649133943,
      "loss": 4.8145,
      "mean_token_accuracy": 0.22589239180088044,
      "num_tokens": 92784822.0,
      "step": 44705
    },
    {
      "entropy": 5.735225486755371,
      "epoch": 4.055696661828737,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003444241879623636,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21155412644147872,
      "num_tokens": 92795270.0,
      "step": 44710
    },
    {
      "entropy": 5.657004308700562,
      "epoch": 4.056150217706821,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003443934095756229,
      "loss": 4.8315,
      "mean_token_accuracy": 0.22736946195364,
      "num_tokens": 92806578.0,
      "step": 44715
    },
    {
      "entropy": 5.684420204162597,
      "epoch": 4.056603773584905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003443626297538088,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2264147639274597,
      "num_tokens": 92815932.0,
      "step": 44720
    },
    {
      "entropy": 5.637942171096801,
      "epoch": 4.05705732946299,
      "grad_norm": 1.640625,
      "learning_rate": 0.0003443318484975579,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2338358834385872,
      "num_tokens": 92825536.0,
      "step": 44725
    },
    {
      "entropy": 5.65257158279419,
      "epoch": 4.057510885341074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003443010658075066,
      "loss": 4.8116,
      "mean_token_accuracy": 0.2279555842280388,
      "num_tokens": 92836451.0,
      "step": 44730
    },
    {
      "entropy": 5.706453847885132,
      "epoch": 4.057964441219158,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003442702816842917,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2217036709189415,
      "num_tokens": 92846923.0,
      "step": 44735
    },
    {
      "entropy": 5.642614412307739,
      "epoch": 4.0584179970972425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034423949612854974,
      "loss": 4.7935,
      "mean_token_accuracy": 0.22996752411127092,
      "num_tokens": 92857549.0,
      "step": 44740
    },
    {
      "entropy": 5.683959913253784,
      "epoch": 4.058871552975327,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003442087091409173,
      "loss": 4.9242,
      "mean_token_accuracy": 0.22380694895982742,
      "num_tokens": 92867059.0,
      "step": 44745
    },
    {
      "entropy": 5.647807455062866,
      "epoch": 4.059325108853411,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003441779207220311,
      "loss": 4.8657,
      "mean_token_accuracy": 0.21780441403388978,
      "num_tokens": 92878138.0,
      "step": 44750
    },
    {
      "entropy": 5.673136806488037,
      "epoch": 4.059778664731495,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003441471308725279,
      "loss": 4.879,
      "mean_token_accuracy": 0.2276317670941353,
      "num_tokens": 92888931.0,
      "step": 44755
    },
    {
      "entropy": 5.6840739250183105,
      "epoch": 4.060232220609579,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034411633959304444,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21783434301614762,
      "num_tokens": 92899718.0,
      "step": 44760
    },
    {
      "entropy": 5.683983373641968,
      "epoch": 4.060685776487663,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034408554688421743,
      "loss": 4.835,
      "mean_token_accuracy": 0.23158248215913774,
      "num_tokens": 92910381.0,
      "step": 44765
    },
    {
      "entropy": 5.638430881500244,
      "epoch": 4.061139332365747,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003440547527466837,
      "loss": 4.8237,
      "mean_token_accuracy": 0.22981788069009781,
      "num_tokens": 92919905.0,
      "step": 44770
    },
    {
      "entropy": 5.616773843765259,
      "epoch": 4.0615928882438315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034402395718108016,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23332347571849824,
      "num_tokens": 92930210.0,
      "step": 44775
    },
    {
      "entropy": 5.663282632827759,
      "epoch": 4.062046444121916,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034399316018804356,
      "loss": 4.8339,
      "mean_token_accuracy": 0.229477259516716,
      "num_tokens": 92940832.0,
      "step": 44780
    },
    {
      "entropy": 5.728418207168579,
      "epoch": 4.0625,
      "grad_norm": 1.375,
      "learning_rate": 0.0003439623617682107,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2175066888332367,
      "num_tokens": 92951349.0,
      "step": 44785
    },
    {
      "entropy": 5.656303358078003,
      "epoch": 4.062953555878084,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003439315619222188,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2270846888422966,
      "num_tokens": 92961642.0,
      "step": 44790
    },
    {
      "entropy": 5.642505407333374,
      "epoch": 4.0634071117561685,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003439007606507046,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2269844591617584,
      "num_tokens": 92971061.0,
      "step": 44795
    },
    {
      "entropy": 5.721338081359863,
      "epoch": 4.063860667634253,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003438699579543051,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21646183878183364,
      "num_tokens": 92981732.0,
      "step": 44800
    },
    {
      "entropy": 5.720691967010498,
      "epoch": 4.064314223512337,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034383915383365726,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21890389025211335,
      "num_tokens": 92991570.0,
      "step": 44805
    },
    {
      "entropy": 5.688021612167359,
      "epoch": 4.064767779390421,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003438083482893982,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2271180510520935,
      "num_tokens": 93001623.0,
      "step": 44810
    },
    {
      "entropy": 5.700196599960327,
      "epoch": 4.065221335268505,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034377754132216496,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22592390924692154,
      "num_tokens": 93012167.0,
      "step": 44815
    },
    {
      "entropy": 5.633208990097046,
      "epoch": 4.065674891146589,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003437467329325947,
      "loss": 4.8133,
      "mean_token_accuracy": 0.2338806062936783,
      "num_tokens": 93022090.0,
      "step": 44820
    },
    {
      "entropy": 5.645627832412719,
      "epoch": 4.066128447024673,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003437159231213244,
      "loss": 4.7339,
      "mean_token_accuracy": 0.23746891617774962,
      "num_tokens": 93030917.0,
      "step": 44825
    },
    {
      "entropy": 5.655427265167236,
      "epoch": 4.0665820029027575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003436851118889914,
      "loss": 4.8315,
      "mean_token_accuracy": 0.22182028144598007,
      "num_tokens": 93040113.0,
      "step": 44830
    },
    {
      "entropy": 5.664164066314697,
      "epoch": 4.067035558780842,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003436542992362326,
      "loss": 4.8728,
      "mean_token_accuracy": 0.230484339594841,
      "num_tokens": 93050053.0,
      "step": 44835
    },
    {
      "entropy": 5.749315977096558,
      "epoch": 4.067489114658926,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003436234851636856,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21561993509531022,
      "num_tokens": 93060581.0,
      "step": 44840
    },
    {
      "entropy": 5.711426877975464,
      "epoch": 4.06794267053701,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00034359266967198734,
      "loss": 4.9189,
      "mean_token_accuracy": 0.22212810218334197,
      "num_tokens": 93070158.0,
      "step": 44845
    },
    {
      "entropy": 5.590299463272094,
      "epoch": 4.068396226415095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003435618527617752,
      "loss": 4.8155,
      "mean_token_accuracy": 0.22881777286529542,
      "num_tokens": 93080804.0,
      "step": 44850
    },
    {
      "entropy": 5.698328971862793,
      "epoch": 4.068849782293179,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003435310344336865,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22443929463624954,
      "num_tokens": 93091346.0,
      "step": 44855
    },
    {
      "entropy": 5.620715570449829,
      "epoch": 4.069303338171263,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003435002146883585,
      "loss": 4.8175,
      "mean_token_accuracy": 0.23096765279769899,
      "num_tokens": 93100931.0,
      "step": 44860
    },
    {
      "entropy": 5.64532585144043,
      "epoch": 4.0697568940493465,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003434693935264286,
      "loss": 4.833,
      "mean_token_accuracy": 0.2314332216978073,
      "num_tokens": 93110181.0,
      "step": 44865
    },
    {
      "entropy": 5.699645280838013,
      "epoch": 4.070210449927431,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003434385709485343,
      "loss": 4.836,
      "mean_token_accuracy": 0.22298609912395478,
      "num_tokens": 93120140.0,
      "step": 44870
    },
    {
      "entropy": 5.721970176696777,
      "epoch": 4.070664005805515,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003434077469553128,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21836236864328384,
      "num_tokens": 93130853.0,
      "step": 44875
    },
    {
      "entropy": 5.672637224197388,
      "epoch": 4.071117561683599,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003433769215474017,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22929693311452864,
      "num_tokens": 93141430.0,
      "step": 44880
    },
    {
      "entropy": 5.7178362846374515,
      "epoch": 4.0715711175616836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034334609472543845,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21823622286319733,
      "num_tokens": 93152270.0,
      "step": 44885
    },
    {
      "entropy": 5.706017780303955,
      "epoch": 4.072024673439768,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003433152664900605,
      "loss": 4.9122,
      "mean_token_accuracy": 0.22154334485530852,
      "num_tokens": 93162850.0,
      "step": 44890
    },
    {
      "entropy": 5.653424978256226,
      "epoch": 4.072478229317852,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003432844368419055,
      "loss": 4.7452,
      "mean_token_accuracy": 0.23382721096277237,
      "num_tokens": 93172265.0,
      "step": 44895
    },
    {
      "entropy": 5.63842363357544,
      "epoch": 4.072931785195936,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003432536057816108,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23524775505065917,
      "num_tokens": 93182756.0,
      "step": 44900
    },
    {
      "entropy": 5.645920705795288,
      "epoch": 4.073385341074021,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034322277330981426,
      "loss": 4.8735,
      "mean_token_accuracy": 0.21888897716999053,
      "num_tokens": 93192710.0,
      "step": 44905
    },
    {
      "entropy": 5.644369411468506,
      "epoch": 4.073838896952105,
      "grad_norm": 1.375,
      "learning_rate": 0.00034319193942715343,
      "loss": 4.8283,
      "mean_token_accuracy": 0.22990211397409438,
      "num_tokens": 93202531.0,
      "step": 44910
    },
    {
      "entropy": 5.7184381008148195,
      "epoch": 4.074292452830188,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034316110413426586,
      "loss": 4.9421,
      "mean_token_accuracy": 0.22301193028688432,
      "num_tokens": 93212208.0,
      "step": 44915
    },
    {
      "entropy": 5.77139983177185,
      "epoch": 4.0747460087082725,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003431302674317892,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21703479886054994,
      "num_tokens": 93222932.0,
      "step": 44920
    },
    {
      "entropy": 5.664581775665283,
      "epoch": 4.075199564586357,
      "grad_norm": 1.375,
      "learning_rate": 0.00034309942932036136,
      "loss": 4.7562,
      "mean_token_accuracy": 0.2378673493862152,
      "num_tokens": 93232049.0,
      "step": 44925
    },
    {
      "entropy": 5.649416732788086,
      "epoch": 4.075653120464441,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034306858980061994,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22571207582950592,
      "num_tokens": 93244230.0,
      "step": 44930
    },
    {
      "entropy": 5.655362701416015,
      "epoch": 4.076106676342525,
      "grad_norm": 1.375,
      "learning_rate": 0.00034303774887320274,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22493933886289597,
      "num_tokens": 93254159.0,
      "step": 44935
    },
    {
      "entropy": 5.669178581237793,
      "epoch": 4.07656023222061,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034300690653874753,
      "loss": 4.7838,
      "mean_token_accuracy": 0.2320196285843849,
      "num_tokens": 93263742.0,
      "step": 44940
    },
    {
      "entropy": 5.664962720870972,
      "epoch": 4.077013788098694,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034297606279789215,
      "loss": 4.9552,
      "mean_token_accuracy": 0.2174091726541519,
      "num_tokens": 93274446.0,
      "step": 44945
    },
    {
      "entropy": 5.674808597564697,
      "epoch": 4.077467343976778,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003429452176512745,
      "loss": 4.9222,
      "mean_token_accuracy": 0.22100184857845306,
      "num_tokens": 93284102.0,
      "step": 44950
    },
    {
      "entropy": 5.673649883270263,
      "epoch": 4.077920899854862,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034291437109953244,
      "loss": 4.7899,
      "mean_token_accuracy": 0.23807226568460466,
      "num_tokens": 93295096.0,
      "step": 44955
    },
    {
      "entropy": 5.630706644058227,
      "epoch": 4.078374455732947,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003428835231433039,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2250870794057846,
      "num_tokens": 93305377.0,
      "step": 44960
    },
    {
      "entropy": 5.732788848876953,
      "epoch": 4.07882801161103,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034285267378322674,
      "loss": 4.913,
      "mean_token_accuracy": 0.22364114671945573,
      "num_tokens": 93315859.0,
      "step": 44965
    },
    {
      "entropy": 5.652154445648193,
      "epoch": 4.079281567489114,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034282182301993903,
      "loss": 4.818,
      "mean_token_accuracy": 0.22866998463869095,
      "num_tokens": 93326519.0,
      "step": 44970
    },
    {
      "entropy": 5.677565002441407,
      "epoch": 4.079735123367199,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003427909708540787,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22327892780303954,
      "num_tokens": 93337055.0,
      "step": 44975
    },
    {
      "entropy": 5.691798877716065,
      "epoch": 4.080188679245283,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034276011728628393,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21740656942129136,
      "num_tokens": 93346565.0,
      "step": 44980
    },
    {
      "entropy": 5.654440116882324,
      "epoch": 4.080642235123367,
      "grad_norm": 1.375,
      "learning_rate": 0.00034272926231719255,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2255293011665344,
      "num_tokens": 93356366.0,
      "step": 44985
    },
    {
      "entropy": 5.677212429046631,
      "epoch": 4.081095791001451,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003426984059474428,
      "loss": 4.868,
      "mean_token_accuracy": 0.21957886815071107,
      "num_tokens": 93367317.0,
      "step": 44990
    },
    {
      "entropy": 5.6544046878814695,
      "epoch": 4.081549346879536,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003426675481776728,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22476110458374024,
      "num_tokens": 93379601.0,
      "step": 44995
    },
    {
      "entropy": 5.6626802444458,
      "epoch": 4.08200290275762,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034263668900852065,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2374037727713585,
      "num_tokens": 93389740.0,
      "step": 45000
    },
    {
      "epoch": 4.08200290275762,
      "eval_entropy": 5.415978300841146,
      "eval_loss": 4.963794708251953,
      "eval_mean_token_accuracy": 0.22841118181722736,
      "eval_num_tokens": 93389740.0,
      "eval_runtime": 20.7334,
      "eval_samples_per_second": 1705.463,
      "eval_steps_per_second": 213.183,
      "step": 45000
    },
    {
      "entropy": 5.76001386642456,
      "epoch": 4.082456458635704,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003426058284406244,
      "loss": 4.9755,
      "mean_token_accuracy": 0.21899300515651704,
      "num_tokens": 93400232.0,
      "step": 45005
    },
    {
      "entropy": 5.722826671600342,
      "epoch": 4.082910014513788,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003425749664746226,
      "loss": 4.9031,
      "mean_token_accuracy": 0.22981218695640565,
      "num_tokens": 93410662.0,
      "step": 45010
    },
    {
      "entropy": 5.659625625610351,
      "epoch": 4.083363570391873,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003425441031111532,
      "loss": 4.7516,
      "mean_token_accuracy": 0.23781661242246627,
      "num_tokens": 93420328.0,
      "step": 45015
    },
    {
      "entropy": 5.688152742385864,
      "epoch": 4.083817126269956,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034251323835085453,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21787895560264586,
      "num_tokens": 93430878.0,
      "step": 45020
    },
    {
      "entropy": 5.739926385879516,
      "epoch": 4.08427068214804,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003424823721943649,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2200404554605484,
      "num_tokens": 93441784.0,
      "step": 45025
    },
    {
      "entropy": 5.645525598526001,
      "epoch": 4.084724238026125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003424515046423226,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22602445632219315,
      "num_tokens": 93453058.0,
      "step": 45030
    },
    {
      "entropy": 5.701102352142334,
      "epoch": 4.085177793904209,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003424206356953659,
      "loss": 4.9042,
      "mean_token_accuracy": 0.22581003457307816,
      "num_tokens": 93463089.0,
      "step": 45035
    },
    {
      "entropy": 5.6917809009552,
      "epoch": 4.085631349782293,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034238976535413333,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2289608433842659,
      "num_tokens": 93473315.0,
      "step": 45040
    },
    {
      "entropy": 5.7374776840209964,
      "epoch": 4.086084905660377,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003423588936192633,
      "loss": 4.9614,
      "mean_token_accuracy": 0.21516617387533188,
      "num_tokens": 93485113.0,
      "step": 45045
    },
    {
      "entropy": 5.622793197631836,
      "epoch": 4.086538461538462,
      "grad_norm": 1.375,
      "learning_rate": 0.0003423280204913941,
      "loss": 4.8137,
      "mean_token_accuracy": 0.23042899817228318,
      "num_tokens": 93494269.0,
      "step": 45050
    },
    {
      "entropy": 5.606999731063842,
      "epoch": 4.086992017416546,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003422971459711644,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2321794241666794,
      "num_tokens": 93505052.0,
      "step": 45055
    },
    {
      "entropy": 5.6597411155700685,
      "epoch": 4.08744557329463,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034226627005921244,
      "loss": 4.8458,
      "mean_token_accuracy": 0.22276150435209274,
      "num_tokens": 93514986.0,
      "step": 45060
    },
    {
      "entropy": 5.695908737182617,
      "epoch": 4.0878991291727145,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000342235392756177,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22508749216794968,
      "num_tokens": 93525286.0,
      "step": 45065
    },
    {
      "entropy": 5.703288841247558,
      "epoch": 4.088352685050798,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034220451406269636,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21571753174066544,
      "num_tokens": 93536034.0,
      "step": 45070
    },
    {
      "entropy": 5.598156547546386,
      "epoch": 4.088806240928882,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034217363397940937,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2273579865694046,
      "num_tokens": 93546283.0,
      "step": 45075
    },
    {
      "entropy": 5.59804368019104,
      "epoch": 4.089259796806966,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034214275250695446,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22409984171390535,
      "num_tokens": 93555811.0,
      "step": 45080
    },
    {
      "entropy": 5.71259822845459,
      "epoch": 4.089713352685051,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003421118696459703,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21810808777809143,
      "num_tokens": 93567154.0,
      "step": 45085
    },
    {
      "entropy": 5.6417242050170895,
      "epoch": 4.090166908563135,
      "grad_norm": 1.25,
      "learning_rate": 0.00034208098539709566,
      "loss": 4.8301,
      "mean_token_accuracy": 0.2349393218755722,
      "num_tokens": 93577654.0,
      "step": 45090
    },
    {
      "entropy": 5.649134731292724,
      "epoch": 4.090620464441219,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034205009976096906,
      "loss": 4.8677,
      "mean_token_accuracy": 0.22095982879400253,
      "num_tokens": 93588712.0,
      "step": 45095
    },
    {
      "entropy": 5.67066798210144,
      "epoch": 4.0910740203193035,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003420192127382293,
      "loss": 4.9059,
      "mean_token_accuracy": 0.2203037515282631,
      "num_tokens": 93599186.0,
      "step": 45100
    },
    {
      "entropy": 5.653915023803711,
      "epoch": 4.091527576197388,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034198832432951517,
      "loss": 4.7702,
      "mean_token_accuracy": 0.23333630859851837,
      "num_tokens": 93609439.0,
      "step": 45105
    },
    {
      "entropy": 5.677346897125244,
      "epoch": 4.091981132075472,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003419574345354654,
      "loss": 4.8529,
      "mean_token_accuracy": 0.2196532353758812,
      "num_tokens": 93619903.0,
      "step": 45110
    },
    {
      "entropy": 5.653753089904785,
      "epoch": 4.092434687953556,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034192654335671893,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23201809376478194,
      "num_tokens": 93629895.0,
      "step": 45115
    },
    {
      "entropy": 5.660627555847168,
      "epoch": 4.09288824383164,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034189565079391436,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21982201486825942,
      "num_tokens": 93640677.0,
      "step": 45120
    },
    {
      "entropy": 5.692655467987061,
      "epoch": 4.093341799709724,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003418647568476907,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2249619647860527,
      "num_tokens": 93651541.0,
      "step": 45125
    },
    {
      "entropy": 5.728185319900513,
      "epoch": 4.093795355587808,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003418338615186869,
      "loss": 4.9076,
      "mean_token_accuracy": 0.21813030689954757,
      "num_tokens": 93661901.0,
      "step": 45130
    },
    {
      "entropy": 5.682808351516724,
      "epoch": 4.094248911465892,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034180296480754174,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2221786230802536,
      "num_tokens": 93672307.0,
      "step": 45135
    },
    {
      "entropy": 5.664837074279785,
      "epoch": 4.094702467343977,
      "grad_norm": 1.375,
      "learning_rate": 0.0003417720667148943,
      "loss": 4.8216,
      "mean_token_accuracy": 0.215610633790493,
      "num_tokens": 93683802.0,
      "step": 45140
    },
    {
      "entropy": 5.738400554656982,
      "epoch": 4.095156023222061,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003417411672413835,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21194143146276473,
      "num_tokens": 93693467.0,
      "step": 45145
    },
    {
      "entropy": 5.668906784057617,
      "epoch": 4.095609579100145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034171026638764833,
      "loss": 4.8425,
      "mean_token_accuracy": 0.2330695942044258,
      "num_tokens": 93703858.0,
      "step": 45150
    },
    {
      "entropy": 5.656031799316406,
      "epoch": 4.0960631349782295,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034167936415432785,
      "loss": 4.853,
      "mean_token_accuracy": 0.23564156293869018,
      "num_tokens": 93714457.0,
      "step": 45155
    },
    {
      "entropy": 5.665589141845703,
      "epoch": 4.096516690856314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003416484605420611,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22803847938776017,
      "num_tokens": 93725009.0,
      "step": 45160
    },
    {
      "entropy": 5.7171369075775145,
      "epoch": 4.096970246734398,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034161755555148725,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2169937163591385,
      "num_tokens": 93736357.0,
      "step": 45165
    },
    {
      "entropy": 5.641406297683716,
      "epoch": 4.097423802612482,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034158664918324536,
      "loss": 4.8185,
      "mean_token_accuracy": 0.22552237808704376,
      "num_tokens": 93746711.0,
      "step": 45170
    },
    {
      "entropy": 5.7034728050231935,
      "epoch": 4.097877358490566,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034155574143797455,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2179391413927078,
      "num_tokens": 93757732.0,
      "step": 45175
    },
    {
      "entropy": 5.66833610534668,
      "epoch": 4.09833091436865,
      "grad_norm": 1.375,
      "learning_rate": 0.00034152483231631413,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2288547307252884,
      "num_tokens": 93767048.0,
      "step": 45180
    },
    {
      "entropy": 5.824919176101685,
      "epoch": 4.098784470246734,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003414939218189032,
      "loss": 5.0682,
      "mean_token_accuracy": 0.21120611429214478,
      "num_tokens": 93778283.0,
      "step": 45185
    },
    {
      "entropy": 5.7421308040618895,
      "epoch": 4.0992380261248185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000341463009946381,
      "loss": 4.9733,
      "mean_token_accuracy": 0.21656214892864228,
      "num_tokens": 93788776.0,
      "step": 45190
    },
    {
      "entropy": 5.700044250488281,
      "epoch": 4.099691582002903,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034143209669938685,
      "loss": 4.883,
      "mean_token_accuracy": 0.21803147941827775,
      "num_tokens": 93798695.0,
      "step": 45195
    },
    {
      "entropy": 5.7308789730072025,
      "epoch": 4.100145137880987,
      "grad_norm": 1.25,
      "learning_rate": 0.00034140118207856,
      "loss": 4.9693,
      "mean_token_accuracy": 0.22296274751424788,
      "num_tokens": 93808993.0,
      "step": 45200
    },
    {
      "entropy": 5.725953722000122,
      "epoch": 4.100598693759071,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034137026608453973,
      "loss": 4.9766,
      "mean_token_accuracy": 0.21785318851470947,
      "num_tokens": 93818361.0,
      "step": 45205
    },
    {
      "entropy": 5.70720272064209,
      "epoch": 4.1010522496371555,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003413393487179655,
      "loss": 4.911,
      "mean_token_accuracy": 0.23547320812940598,
      "num_tokens": 93828988.0,
      "step": 45210
    },
    {
      "entropy": 5.791929578781128,
      "epoch": 4.10150580551524,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003413084299794767,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2215809315443039,
      "num_tokens": 93839350.0,
      "step": 45215
    },
    {
      "entropy": 5.703439235687256,
      "epoch": 4.101959361393324,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034127750986971255,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22897980511188507,
      "num_tokens": 93849939.0,
      "step": 45220
    },
    {
      "entropy": 5.670909023284912,
      "epoch": 4.1024129172714074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003412465883893126,
      "loss": 4.8965,
      "mean_token_accuracy": 0.22460082769393921,
      "num_tokens": 93861278.0,
      "step": 45225
    },
    {
      "entropy": 5.717130088806153,
      "epoch": 4.102866473149492,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003412156655389164,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22266440689563752,
      "num_tokens": 93871693.0,
      "step": 45230
    },
    {
      "entropy": 5.671990060806275,
      "epoch": 4.103320029027576,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003411847413191633,
      "loss": 4.8832,
      "mean_token_accuracy": 0.2262684002518654,
      "num_tokens": 93883098.0,
      "step": 45235
    },
    {
      "entropy": 5.643094491958618,
      "epoch": 4.10377358490566,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034115381573069297,
      "loss": 4.7936,
      "mean_token_accuracy": 0.22806403338909148,
      "num_tokens": 93893347.0,
      "step": 45240
    },
    {
      "entropy": 5.738433074951172,
      "epoch": 4.1042271407837445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034112288877414485,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21406890004873275,
      "num_tokens": 93903964.0,
      "step": 45245
    },
    {
      "entropy": 5.693863868713379,
      "epoch": 4.104680696661829,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003410919604501585,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2248979777097702,
      "num_tokens": 93914526.0,
      "step": 45250
    },
    {
      "entropy": 5.755918216705322,
      "epoch": 4.105134252539913,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003410610307593735,
      "loss": 4.9721,
      "mean_token_accuracy": 0.2249869227409363,
      "num_tokens": 93924430.0,
      "step": 45255
    },
    {
      "entropy": 5.716763544082641,
      "epoch": 4.105587808417997,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003410300997024297,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22883870154619218,
      "num_tokens": 93935150.0,
      "step": 45260
    },
    {
      "entropy": 5.633342456817627,
      "epoch": 4.106041364296082,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003409991672799665,
      "loss": 4.8137,
      "mean_token_accuracy": 0.23588439375162124,
      "num_tokens": 93944496.0,
      "step": 45265
    },
    {
      "entropy": 5.684167146682739,
      "epoch": 4.106494920174166,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034096823349262376,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2192351371049881,
      "num_tokens": 93955745.0,
      "step": 45270
    },
    {
      "entropy": 5.613743114471435,
      "epoch": 4.106948476052249,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003409372983410411,
      "loss": 4.7918,
      "mean_token_accuracy": 0.2272347167134285,
      "num_tokens": 93965930.0,
      "step": 45275
    },
    {
      "entropy": 5.643866682052613,
      "epoch": 4.1074020319303335,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003409063618258583,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2328326880931854,
      "num_tokens": 93976435.0,
      "step": 45280
    },
    {
      "entropy": 5.545793581008911,
      "epoch": 4.107855587808418,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034087542394771524,
      "loss": 4.7311,
      "mean_token_accuracy": 0.23665671050548553,
      "num_tokens": 93987294.0,
      "step": 45285
    },
    {
      "entropy": 5.694854021072388,
      "epoch": 4.108309143686502,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034084448470725147,
      "loss": 4.8691,
      "mean_token_accuracy": 0.2259235665202141,
      "num_tokens": 93997177.0,
      "step": 45290
    },
    {
      "entropy": 5.752830266952515,
      "epoch": 4.108762699564586,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003408135441051071,
      "loss": 4.9529,
      "mean_token_accuracy": 0.21901662945747374,
      "num_tokens": 94007663.0,
      "step": 45295
    },
    {
      "entropy": 5.677536821365356,
      "epoch": 4.1092162554426706,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003407826021419218,
      "loss": 4.7978,
      "mean_token_accuracy": 0.22582892924547196,
      "num_tokens": 94016899.0,
      "step": 45300
    },
    {
      "entropy": 5.67814884185791,
      "epoch": 4.109669811320755,
      "grad_norm": 1.375,
      "learning_rate": 0.0003407516588183356,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21812294274568558,
      "num_tokens": 94027696.0,
      "step": 45305
    },
    {
      "entropy": 5.646271181106568,
      "epoch": 4.110123367198839,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003407207141349882,
      "loss": 4.8761,
      "mean_token_accuracy": 0.22462472915649415,
      "num_tokens": 94038075.0,
      "step": 45310
    },
    {
      "entropy": 5.685809135437012,
      "epoch": 4.110576923076923,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003406897680925199,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21939867585897446,
      "num_tokens": 94049082.0,
      "step": 45315
    },
    {
      "entropy": 5.69429235458374,
      "epoch": 4.111030478955008,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034065882069157023,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22121205031871796,
      "num_tokens": 94060367.0,
      "step": 45320
    },
    {
      "entropy": 5.667037630081177,
      "epoch": 4.111484034833092,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034062787193277954,
      "loss": 4.8327,
      "mean_token_accuracy": 0.2274642691016197,
      "num_tokens": 94069415.0,
      "step": 45325
    },
    {
      "entropy": 5.722782325744629,
      "epoch": 4.111937590711175,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034059692181678775,
      "loss": 4.91,
      "mean_token_accuracy": 0.21705766022205353,
      "num_tokens": 94079047.0,
      "step": 45330
    },
    {
      "entropy": 5.658153343200683,
      "epoch": 4.1123911465892595,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034056597034423484,
      "loss": 4.8259,
      "mean_token_accuracy": 0.2258522093296051,
      "num_tokens": 94089631.0,
      "step": 45335
    },
    {
      "entropy": 5.637837314605713,
      "epoch": 4.112844702467344,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034053501751576097,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22640466541051865,
      "num_tokens": 94099265.0,
      "step": 45340
    },
    {
      "entropy": 5.603198003768921,
      "epoch": 4.113298258345428,
      "grad_norm": 1.25,
      "learning_rate": 0.0003405040633320063,
      "loss": 4.8645,
      "mean_token_accuracy": 0.21789870262145997,
      "num_tokens": 94109673.0,
      "step": 45345
    },
    {
      "entropy": 5.643795442581177,
      "epoch": 4.113751814223512,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003404731077936108,
      "loss": 4.7719,
      "mean_token_accuracy": 0.2338087722659111,
      "num_tokens": 94120004.0,
      "step": 45350
    },
    {
      "entropy": 5.639846086502075,
      "epoch": 4.114205370101597,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003404421509012148,
      "loss": 4.8578,
      "mean_token_accuracy": 0.2245064362883568,
      "num_tokens": 94129935.0,
      "step": 45355
    },
    {
      "entropy": 5.678585243225098,
      "epoch": 4.114658925979681,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034041119265545847,
      "loss": 4.902,
      "mean_token_accuracy": 0.22027744352817535,
      "num_tokens": 94139035.0,
      "step": 45360
    },
    {
      "entropy": 5.70267424583435,
      "epoch": 4.115112481857765,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034038023305698196,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22503584176301955,
      "num_tokens": 94149281.0,
      "step": 45365
    },
    {
      "entropy": 5.658447790145874,
      "epoch": 4.115566037735849,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003403492721064256,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22493139803409576,
      "num_tokens": 94160455.0,
      "step": 45370
    },
    {
      "entropy": 5.691318845748901,
      "epoch": 4.116019593613934,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034031830980442955,
      "loss": 4.898,
      "mean_token_accuracy": 0.21858693212270736,
      "num_tokens": 94170334.0,
      "step": 45375
    },
    {
      "entropy": 5.650800514221191,
      "epoch": 4.116473149492017,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034028734615163436,
      "loss": 4.8206,
      "mean_token_accuracy": 0.22225605845451354,
      "num_tokens": 94181055.0,
      "step": 45380
    },
    {
      "entropy": 5.6686070442199705,
      "epoch": 4.116926705370101,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003402563811486802,
      "loss": 4.8235,
      "mean_token_accuracy": 0.2321927949786186,
      "num_tokens": 94191182.0,
      "step": 45385
    },
    {
      "entropy": 5.696384477615356,
      "epoch": 4.117380261248186,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003402254147962073,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22072302550077438,
      "num_tokens": 94201229.0,
      "step": 45390
    },
    {
      "entropy": 5.623414182662964,
      "epoch": 4.11783381712627,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003401944470948563,
      "loss": 4.8389,
      "mean_token_accuracy": 0.2286711409687996,
      "num_tokens": 94210956.0,
      "step": 45395
    },
    {
      "entropy": 5.66583833694458,
      "epoch": 4.118287373004354,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034016347804526753,
      "loss": 4.9,
      "mean_token_accuracy": 0.22061336040496826,
      "num_tokens": 94220740.0,
      "step": 45400
    },
    {
      "entropy": 5.616199493408203,
      "epoch": 4.118740928882438,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034013250764808133,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24380480200052262,
      "num_tokens": 94231497.0,
      "step": 45405
    },
    {
      "entropy": 5.67497501373291,
      "epoch": 4.119194484760523,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003401015359039384,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22862010300159455,
      "num_tokens": 94240492.0,
      "step": 45410
    },
    {
      "entropy": 5.625301551818848,
      "epoch": 4.119648040638607,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003400705628134791,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23075455278158188,
      "num_tokens": 94251586.0,
      "step": 45415
    },
    {
      "entropy": 5.657022476196289,
      "epoch": 4.120101596516691,
      "grad_norm": 1.34375,
      "learning_rate": 0.000340039588377344,
      "loss": 4.8579,
      "mean_token_accuracy": 0.23169631212949754,
      "num_tokens": 94261635.0,
      "step": 45420
    },
    {
      "entropy": 5.682074499130249,
      "epoch": 4.120555152394775,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003400086125961735,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21863821297883987,
      "num_tokens": 94269881.0,
      "step": 45425
    },
    {
      "entropy": 5.6766022682189945,
      "epoch": 4.121008708272859,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033997763547060847,
      "loss": 4.8213,
      "mean_token_accuracy": 0.22783233374357223,
      "num_tokens": 94279412.0,
      "step": 45430
    },
    {
      "entropy": 5.671648263931274,
      "epoch": 4.121462264150943,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003399466570012893,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22102775871753694,
      "num_tokens": 94290157.0,
      "step": 45435
    },
    {
      "entropy": 5.70859112739563,
      "epoch": 4.121915820029027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033991567718885674,
      "loss": 4.9034,
      "mean_token_accuracy": 0.22150469273328782,
      "num_tokens": 94299693.0,
      "step": 45440
    },
    {
      "entropy": 5.781943368911743,
      "epoch": 4.122369375907112,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003398846960339515,
      "loss": 4.9868,
      "mean_token_accuracy": 0.2201607570052147,
      "num_tokens": 94310245.0,
      "step": 45445
    },
    {
      "entropy": 5.668787813186645,
      "epoch": 4.122822931785196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003398537135372141,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2243713304400444,
      "num_tokens": 94321190.0,
      "step": 45450
    },
    {
      "entropy": 5.724693059921265,
      "epoch": 4.12327648766328,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003398227296992854,
      "loss": 4.9206,
      "mean_token_accuracy": 0.22508213967084884,
      "num_tokens": 94330273.0,
      "step": 45455
    },
    {
      "entropy": 5.747559309005737,
      "epoch": 4.123730043541364,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033979174452080616,
      "loss": 4.9667,
      "mean_token_accuracy": 0.21284834146499634,
      "num_tokens": 94341097.0,
      "step": 45460
    },
    {
      "entropy": 5.687551498413086,
      "epoch": 4.124183599419449,
      "grad_norm": 1.25,
      "learning_rate": 0.0003397607580024171,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22462472319602966,
      "num_tokens": 94352329.0,
      "step": 45465
    },
    {
      "entropy": 5.658494567871093,
      "epoch": 4.124637155297533,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003397297701447591,
      "loss": 4.817,
      "mean_token_accuracy": 0.22459701001644133,
      "num_tokens": 94362551.0,
      "step": 45470
    },
    {
      "entropy": 5.657405662536621,
      "epoch": 4.125090711175617,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000339698780948473,
      "loss": 4.9259,
      "mean_token_accuracy": 0.22020042985677718,
      "num_tokens": 94372405.0,
      "step": 45475
    },
    {
      "entropy": 5.713830089569091,
      "epoch": 4.125544267053701,
      "grad_norm": 1.25,
      "learning_rate": 0.0003396677904141995,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22904844731092452,
      "num_tokens": 94382347.0,
      "step": 45480
    },
    {
      "entropy": 5.610005855560303,
      "epoch": 4.125997822931785,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033963679854257965,
      "loss": 4.8929,
      "mean_token_accuracy": 0.22397020161151887,
      "num_tokens": 94393428.0,
      "step": 45485
    },
    {
      "entropy": 5.680051374435425,
      "epoch": 4.126451378809869,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003396058053342544,
      "loss": 4.887,
      "mean_token_accuracy": 0.23005108237266542,
      "num_tokens": 94403941.0,
      "step": 45490
    },
    {
      "entropy": 5.692630815505981,
      "epoch": 4.126904934687953,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033957481078986465,
      "loss": 4.8692,
      "mean_token_accuracy": 0.2214859664440155,
      "num_tokens": 94414206.0,
      "step": 45495
    },
    {
      "entropy": 5.717976760864258,
      "epoch": 4.127358490566038,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033954381491005126,
      "loss": 4.8471,
      "mean_token_accuracy": 0.23197571486234664,
      "num_tokens": 94424481.0,
      "step": 45500
    },
    {
      "entropy": 5.6614902973175045,
      "epoch": 4.127812046444122,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003395128176954554,
      "loss": 4.8317,
      "mean_token_accuracy": 0.22894318401813507,
      "num_tokens": 94434714.0,
      "step": 45505
    },
    {
      "entropy": 5.7151110649108885,
      "epoch": 4.128265602322206,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000339481819146718,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21963382065296172,
      "num_tokens": 94445267.0,
      "step": 45510
    },
    {
      "entropy": 5.707913446426391,
      "epoch": 4.1287191582002905,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033945081926448013,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22955146729946135,
      "num_tokens": 94455724.0,
      "step": 45515
    },
    {
      "entropy": 5.705229091644287,
      "epoch": 4.129172714078375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033941981804938294,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2216176763176918,
      "num_tokens": 94466655.0,
      "step": 45520
    },
    {
      "entropy": 5.615235090255737,
      "epoch": 4.129626269956459,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033938881550206744,
      "loss": 4.815,
      "mean_token_accuracy": 0.22803997248411179,
      "num_tokens": 94477699.0,
      "step": 45525
    },
    {
      "entropy": 5.661374616622925,
      "epoch": 4.130079825834542,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033935781162317485,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22512202560901642,
      "num_tokens": 94487639.0,
      "step": 45530
    },
    {
      "entropy": 5.801564645767212,
      "epoch": 4.130533381712627,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033932680641334635,
      "loss": 5.0035,
      "mean_token_accuracy": 0.21225543767213823,
      "num_tokens": 94498687.0,
      "step": 45535
    },
    {
      "entropy": 5.704364824295044,
      "epoch": 4.130986937590711,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003392957998732231,
      "loss": 4.8468,
      "mean_token_accuracy": 0.23066934794187546,
      "num_tokens": 94508706.0,
      "step": 45540
    },
    {
      "entropy": 5.618652629852295,
      "epoch": 4.131440493468795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033926479200344626,
      "loss": 4.7776,
      "mean_token_accuracy": 0.23333349972963333,
      "num_tokens": 94517246.0,
      "step": 45545
    },
    {
      "entropy": 5.613811016082764,
      "epoch": 4.131894049346879,
      "grad_norm": 1.25,
      "learning_rate": 0.0003392337828046572,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23112025260925292,
      "num_tokens": 94527956.0,
      "step": 45550
    },
    {
      "entropy": 5.714540481567383,
      "epoch": 4.132347605224964,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003392027722774971,
      "loss": 4.935,
      "mean_token_accuracy": 0.21406676471233368,
      "num_tokens": 94539101.0,
      "step": 45555
    },
    {
      "entropy": 5.665484237670898,
      "epoch": 4.132801161103048,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033917176042260735,
      "loss": 4.8096,
      "mean_token_accuracy": 0.2254219651222229,
      "num_tokens": 94549597.0,
      "step": 45560
    },
    {
      "entropy": 5.725054740905762,
      "epoch": 4.133254716981132,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003391407472406292,
      "loss": 4.94,
      "mean_token_accuracy": 0.22266182601451873,
      "num_tokens": 94559911.0,
      "step": 45565
    },
    {
      "entropy": 5.667895317077637,
      "epoch": 4.1337082728592165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033910973273220416,
      "loss": 4.8054,
      "mean_token_accuracy": 0.22711390107870102,
      "num_tokens": 94569396.0,
      "step": 45570
    },
    {
      "entropy": 5.663702917098999,
      "epoch": 4.134161828737301,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033907871689797337,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2295814648270607,
      "num_tokens": 94579817.0,
      "step": 45575
    },
    {
      "entropy": 5.598019647598266,
      "epoch": 4.134615384615385,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003390476997385785,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22663536965847014,
      "num_tokens": 94590398.0,
      "step": 45580
    },
    {
      "entropy": 5.62575478553772,
      "epoch": 4.135068940493468,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003390166812546608,
      "loss": 4.8398,
      "mean_token_accuracy": 0.2249739184975624,
      "num_tokens": 94600528.0,
      "step": 45585
    },
    {
      "entropy": 5.671099233627319,
      "epoch": 4.135522496371553,
      "grad_norm": 1.25,
      "learning_rate": 0.00033898566144686184,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22154261469841002,
      "num_tokens": 94611107.0,
      "step": 45590
    },
    {
      "entropy": 5.73763747215271,
      "epoch": 4.135976052249637,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003389546403158231,
      "loss": 4.9428,
      "mean_token_accuracy": 0.22034264653921126,
      "num_tokens": 94620932.0,
      "step": 45595
    },
    {
      "entropy": 5.735232877731323,
      "epoch": 4.136429608127721,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033892361786218604,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2195568263530731,
      "num_tokens": 94630476.0,
      "step": 45600
    },
    {
      "entropy": 5.634406661987304,
      "epoch": 4.1368831640058055,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003388925940865924,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22063252180814744,
      "num_tokens": 94641487.0,
      "step": 45605
    },
    {
      "entropy": 5.703968477249146,
      "epoch": 4.13733671988389,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033886156898968353,
      "loss": 4.7808,
      "mean_token_accuracy": 0.23053215742111205,
      "num_tokens": 94651668.0,
      "step": 45610
    },
    {
      "entropy": 5.633303499221801,
      "epoch": 4.137790275761974,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033883054257210116,
      "loss": 4.8849,
      "mean_token_accuracy": 0.21612417250871657,
      "num_tokens": 94661218.0,
      "step": 45615
    },
    {
      "entropy": 5.623491001129151,
      "epoch": 4.138243831640058,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000338799514834487,
      "loss": 4.8252,
      "mean_token_accuracy": 0.22095274031162263,
      "num_tokens": 94672412.0,
      "step": 45620
    },
    {
      "entropy": 5.685295724868775,
      "epoch": 4.1386973875181425,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003387684857774824,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21403622925281524,
      "num_tokens": 94683102.0,
      "step": 45625
    },
    {
      "entropy": 5.672245645523072,
      "epoch": 4.139150943396227,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003387374554017294,
      "loss": 4.8613,
      "mean_token_accuracy": 0.23045235872268677,
      "num_tokens": 94692869.0,
      "step": 45630
    },
    {
      "entropy": 5.678238487243652,
      "epoch": 4.13960449927431,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033870642370786966,
      "loss": 4.8205,
      "mean_token_accuracy": 0.23011452853679656,
      "num_tokens": 94702437.0,
      "step": 45635
    },
    {
      "entropy": 5.655371904373169,
      "epoch": 4.1400580551523944,
      "grad_norm": 1.25,
      "learning_rate": 0.00033867539069654476,
      "loss": 4.9,
      "mean_token_accuracy": 0.2237161949276924,
      "num_tokens": 94713130.0,
      "step": 45640
    },
    {
      "entropy": 5.750271606445312,
      "epoch": 4.140511611030479,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033864435636839647,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21517866253852844,
      "num_tokens": 94723251.0,
      "step": 45645
    },
    {
      "entropy": 5.701170539855957,
      "epoch": 4.140965166908563,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003386133207240668,
      "loss": 4.8755,
      "mean_token_accuracy": 0.23121597915887832,
      "num_tokens": 94733805.0,
      "step": 45650
    },
    {
      "entropy": 5.692129898071289,
      "epoch": 4.141418722786647,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003385822837641974,
      "loss": 4.813,
      "mean_token_accuracy": 0.22503754049539565,
      "num_tokens": 94743654.0,
      "step": 45655
    },
    {
      "entropy": 5.6265904903411865,
      "epoch": 4.1418722786647315,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033855124548943025,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2254452720284462,
      "num_tokens": 94754635.0,
      "step": 45660
    },
    {
      "entropy": 5.691283750534057,
      "epoch": 4.142325834542816,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00033852020590040705,
      "loss": 4.8992,
      "mean_token_accuracy": 0.22603995949029923,
      "num_tokens": 94765371.0,
      "step": 45665
    },
    {
      "entropy": 5.653284215927124,
      "epoch": 4.1427793904209,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003384891649977698,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23326751291751863,
      "num_tokens": 94775780.0,
      "step": 45670
    },
    {
      "entropy": 5.6912096500396725,
      "epoch": 4.143232946298984,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033845812278216045,
      "loss": 4.9015,
      "mean_token_accuracy": 0.22231681942939757,
      "num_tokens": 94785345.0,
      "step": 45675
    },
    {
      "entropy": 5.641998767852783,
      "epoch": 4.143686502177069,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033842707925422095,
      "loss": 4.8237,
      "mean_token_accuracy": 0.22652334868907928,
      "num_tokens": 94796059.0,
      "step": 45680
    },
    {
      "entropy": 5.645629262924194,
      "epoch": 4.144140058055152,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033839603441459333,
      "loss": 4.845,
      "mean_token_accuracy": 0.22868782579898833,
      "num_tokens": 94805867.0,
      "step": 45685
    },
    {
      "entropy": 5.629306125640869,
      "epoch": 4.144593613933236,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003383649882639195,
      "loss": 4.854,
      "mean_token_accuracy": 0.22390590608119965,
      "num_tokens": 94817172.0,
      "step": 45690
    },
    {
      "entropy": 5.676585006713867,
      "epoch": 4.1450471698113205,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003383339408028416,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22290365248918534,
      "num_tokens": 94828748.0,
      "step": 45695
    },
    {
      "entropy": 5.641875982284546,
      "epoch": 4.145500725689405,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003383028920320017,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22626406103372573,
      "num_tokens": 94838704.0,
      "step": 45700
    },
    {
      "entropy": 5.683857583999634,
      "epoch": 4.145954281567489,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033827184195204174,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2174295336008072,
      "num_tokens": 94849555.0,
      "step": 45705
    },
    {
      "entropy": 5.719347190856934,
      "epoch": 4.146407837445573,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003382407905636041,
      "loss": 4.9598,
      "mean_token_accuracy": 0.21727394759655,
      "num_tokens": 94862238.0,
      "step": 45710
    },
    {
      "entropy": 5.565261745452881,
      "epoch": 4.1468613933236576,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003382097378673308,
      "loss": 4.7433,
      "mean_token_accuracy": 0.24257777333259584,
      "num_tokens": 94872960.0,
      "step": 45715
    },
    {
      "entropy": 5.603706741333008,
      "epoch": 4.147314949201742,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000338178683863864,
      "loss": 4.8378,
      "mean_token_accuracy": 0.23169205635786055,
      "num_tokens": 94883612.0,
      "step": 45720
    },
    {
      "entropy": 5.629396867752075,
      "epoch": 4.147768505079826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003381476285538458,
      "loss": 4.808,
      "mean_token_accuracy": 0.23540480583906173,
      "num_tokens": 94893176.0,
      "step": 45725
    },
    {
      "entropy": 5.664601755142212,
      "epoch": 4.14822206095791,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003381165719379187,
      "loss": 4.8227,
      "mean_token_accuracy": 0.2265253707766533,
      "num_tokens": 94903860.0,
      "step": 45730
    },
    {
      "entropy": 5.648194360733032,
      "epoch": 4.148675616835995,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003380855140167247,
      "loss": 4.8075,
      "mean_token_accuracy": 0.22897300720214844,
      "num_tokens": 94915681.0,
      "step": 45735
    },
    {
      "entropy": 5.643002367019653,
      "epoch": 4.149129172714078,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003380544547909061,
      "loss": 4.8389,
      "mean_token_accuracy": 0.2265712082386017,
      "num_tokens": 94927062.0,
      "step": 45740
    },
    {
      "entropy": 5.661525106430053,
      "epoch": 4.149582728592162,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003380233942611055,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22940066307783127,
      "num_tokens": 94938838.0,
      "step": 45745
    },
    {
      "entropy": 5.602386140823365,
      "epoch": 4.1500362844702465,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000337992332427965,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22933536618947983,
      "num_tokens": 94949678.0,
      "step": 45750
    },
    {
      "entropy": 5.668763065338135,
      "epoch": 4.150489840348331,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000337961269292127,
      "loss": 4.9132,
      "mean_token_accuracy": 0.22446864545345308,
      "num_tokens": 94960314.0,
      "step": 45755
    },
    {
      "entropy": 5.675050973892212,
      "epoch": 4.150943396226415,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033793020485423385,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22197601497173308,
      "num_tokens": 94969831.0,
      "step": 45760
    },
    {
      "entropy": 5.681710052490234,
      "epoch": 4.151396952104499,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033789913911492814,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2222761780023575,
      "num_tokens": 94980046.0,
      "step": 45765
    },
    {
      "entropy": 5.628504085540771,
      "epoch": 4.151850507982584,
      "grad_norm": 1.25,
      "learning_rate": 0.0003378680720748522,
      "loss": 4.875,
      "mean_token_accuracy": 0.2255491778254509,
      "num_tokens": 94990850.0,
      "step": 45770
    },
    {
      "entropy": 5.710575342178345,
      "epoch": 4.152304063860668,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003378370037346484,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21948318481445311,
      "num_tokens": 95001658.0,
      "step": 45775
    },
    {
      "entropy": 5.677415180206299,
      "epoch": 4.152757619738752,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033780593409495936,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22447699755430223,
      "num_tokens": 95011912.0,
      "step": 45780
    },
    {
      "entropy": 5.670226049423218,
      "epoch": 4.153211175616836,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003377748631564276,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22418673783540727,
      "num_tokens": 95021299.0,
      "step": 45785
    },
    {
      "entropy": 5.652838563919067,
      "epoch": 4.15366473149492,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033774379091969583,
      "loss": 4.8478,
      "mean_token_accuracy": 0.23108469396829606,
      "num_tokens": 95031520.0,
      "step": 45790
    },
    {
      "entropy": 5.599844932556152,
      "epoch": 4.154118287373004,
      "grad_norm": 1.25,
      "learning_rate": 0.00033771271738540634,
      "loss": 4.7757,
      "mean_token_accuracy": 0.2340543806552887,
      "num_tokens": 95042127.0,
      "step": 45795
    },
    {
      "entropy": 5.659517574310303,
      "epoch": 4.154571843251088,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033768164255420184,
      "loss": 4.8485,
      "mean_token_accuracy": 0.22394797950983047,
      "num_tokens": 95052181.0,
      "step": 45800
    },
    {
      "entropy": 5.745415258407593,
      "epoch": 4.155025399129173,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033765056642672507,
      "loss": 4.8994,
      "mean_token_accuracy": 0.223867204785347,
      "num_tokens": 95062583.0,
      "step": 45805
    },
    {
      "entropy": 5.710824298858642,
      "epoch": 4.155478955007257,
      "grad_norm": 1.25,
      "learning_rate": 0.00033761948900361857,
      "loss": 4.8752,
      "mean_token_accuracy": 0.2287884086370468,
      "num_tokens": 95073250.0,
      "step": 45810
    },
    {
      "entropy": 5.700158548355103,
      "epoch": 4.155932510885341,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003375884102855251,
      "loss": 4.8609,
      "mean_token_accuracy": 0.22304863929748536,
      "num_tokens": 95083544.0,
      "step": 45815
    },
    {
      "entropy": 5.620473098754883,
      "epoch": 4.156386066763425,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033755733027308746,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22599489241838455,
      "num_tokens": 95093933.0,
      "step": 45820
    },
    {
      "entropy": 5.649737691879272,
      "epoch": 4.15683962264151,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003375262489669481,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22793227732181548,
      "num_tokens": 95104468.0,
      "step": 45825
    },
    {
      "entropy": 5.74627103805542,
      "epoch": 4.157293178519594,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033749516636775014,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22400702089071273,
      "num_tokens": 95114285.0,
      "step": 45830
    },
    {
      "entropy": 5.732165002822876,
      "epoch": 4.157746734397678,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033746408247613616,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2212749734520912,
      "num_tokens": 95123984.0,
      "step": 45835
    },
    {
      "entropy": 5.716455554962158,
      "epoch": 4.1582002902757615,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000337432997292749,
      "loss": 4.9235,
      "mean_token_accuracy": 0.22392043173313142,
      "num_tokens": 95135466.0,
      "step": 45840
    },
    {
      "entropy": 5.637009811401367,
      "epoch": 4.158653846153846,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003374019108182315,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23243832141160964,
      "num_tokens": 95145223.0,
      "step": 45845
    },
    {
      "entropy": 5.736983871459961,
      "epoch": 4.15910740203193,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003373708230532267,
      "loss": 4.8978,
      "mean_token_accuracy": 0.2218468353152275,
      "num_tokens": 95154916.0,
      "step": 45850
    },
    {
      "entropy": 5.657283926010132,
      "epoch": 4.159560957910014,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033733973399837735,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2262278601527214,
      "num_tokens": 95164717.0,
      "step": 45855
    },
    {
      "entropy": 5.738411140441895,
      "epoch": 4.160014513788099,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003373086436543263,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21771416515111924,
      "num_tokens": 95175544.0,
      "step": 45860
    },
    {
      "entropy": 5.658340501785278,
      "epoch": 4.160468069666183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003372775520217168,
      "loss": 4.9229,
      "mean_token_accuracy": 0.22239864319562913,
      "num_tokens": 95185997.0,
      "step": 45865
    },
    {
      "entropy": 5.632626390457153,
      "epoch": 4.160921625544267,
      "grad_norm": 1.25,
      "learning_rate": 0.0003372464591011916,
      "loss": 4.861,
      "mean_token_accuracy": 0.2191475123167038,
      "num_tokens": 95196967.0,
      "step": 45870
    },
    {
      "entropy": 5.62531361579895,
      "epoch": 4.161375181422351,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033721536489339374,
      "loss": 4.8122,
      "mean_token_accuracy": 0.2264287441968918,
      "num_tokens": 95206788.0,
      "step": 45875
    },
    {
      "entropy": 5.739284181594849,
      "epoch": 4.161828737300436,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033718426939896627,
      "loss": 4.896,
      "mean_token_accuracy": 0.22622128427028657,
      "num_tokens": 95217806.0,
      "step": 45880
    },
    {
      "entropy": 5.6841668605804445,
      "epoch": 4.16228229317852,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033715317261855235,
      "loss": 4.8142,
      "mean_token_accuracy": 0.23467088490724564,
      "num_tokens": 95228311.0,
      "step": 45885
    },
    {
      "entropy": 5.61879529953003,
      "epoch": 4.162735849056604,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033712207455279485,
      "loss": 4.8274,
      "mean_token_accuracy": 0.22531543672084808,
      "num_tokens": 95238872.0,
      "step": 45890
    },
    {
      "entropy": 5.568365049362183,
      "epoch": 4.163189404934688,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003370909752023371,
      "loss": 4.7998,
      "mean_token_accuracy": 0.23678479492664337,
      "num_tokens": 95248575.0,
      "step": 45895
    },
    {
      "entropy": 5.654595994949341,
      "epoch": 4.163642960812772,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003370598745678221,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2235287234187126,
      "num_tokens": 95259323.0,
      "step": 45900
    },
    {
      "entropy": 5.697019910812378,
      "epoch": 4.164096516690856,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003370287726498932,
      "loss": 4.8403,
      "mean_token_accuracy": 0.22598668783903123,
      "num_tokens": 95270018.0,
      "step": 45905
    },
    {
      "entropy": 5.6743532657623295,
      "epoch": 4.16455007256894,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033699766944919325,
      "loss": 4.8166,
      "mean_token_accuracy": 0.23115460723638534,
      "num_tokens": 95279783.0,
      "step": 45910
    },
    {
      "entropy": 5.6850687026977536,
      "epoch": 4.165003628447025,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033696656496636586,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22663683146238328,
      "num_tokens": 95291200.0,
      "step": 45915
    },
    {
      "entropy": 5.70444016456604,
      "epoch": 4.165457184325109,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033693545920205416,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2164785459637642,
      "num_tokens": 95301641.0,
      "step": 45920
    },
    {
      "entropy": 5.6081986904144285,
      "epoch": 4.165910740203193,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033690435215690125,
      "loss": 4.7917,
      "mean_token_accuracy": 0.22975433021783828,
      "num_tokens": 95312636.0,
      "step": 45925
    },
    {
      "entropy": 5.619971752166748,
      "epoch": 4.1663642960812775,
      "grad_norm": 1.375,
      "learning_rate": 0.0003368732438315506,
      "loss": 4.8197,
      "mean_token_accuracy": 0.23392310440540315,
      "num_tokens": 95322168.0,
      "step": 45930
    },
    {
      "entropy": 5.613124370574951,
      "epoch": 4.166817851959362,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033684213422664554,
      "loss": 4.804,
      "mean_token_accuracy": 0.2308719962835312,
      "num_tokens": 95333162.0,
      "step": 45935
    },
    {
      "entropy": 5.610457944869995,
      "epoch": 4.167271407837446,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003368110233428294,
      "loss": 4.8391,
      "mean_token_accuracy": 0.22784648090600967,
      "num_tokens": 95343806.0,
      "step": 45940
    },
    {
      "entropy": 5.655176162719727,
      "epoch": 4.167724963715529,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033677991118074554,
      "loss": 4.767,
      "mean_token_accuracy": 0.23597723990678787,
      "num_tokens": 95353906.0,
      "step": 45945
    },
    {
      "entropy": 5.7291559219360355,
      "epoch": 4.168178519593614,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003367487977410373,
      "loss": 4.9639,
      "mean_token_accuracy": 0.21363867223262786,
      "num_tokens": 95363801.0,
      "step": 45950
    },
    {
      "entropy": 5.703226852416992,
      "epoch": 4.168632075471698,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033671768302434823,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22326904237270356,
      "num_tokens": 95375012.0,
      "step": 45955
    },
    {
      "entropy": 5.72789511680603,
      "epoch": 4.169085631349782,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033668656703132174,
      "loss": 4.8908,
      "mean_token_accuracy": 0.23269332200288773,
      "num_tokens": 95386429.0,
      "step": 45960
    },
    {
      "entropy": 5.634442090988159,
      "epoch": 4.169539187227866,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003366554497626014,
      "loss": 4.8657,
      "mean_token_accuracy": 0.22063346654176713,
      "num_tokens": 95396534.0,
      "step": 45965
    },
    {
      "entropy": 5.65145354270935,
      "epoch": 4.169992743105951,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033662433121883057,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22448988556861876,
      "num_tokens": 95406621.0,
      "step": 45970
    },
    {
      "entropy": 5.681789922714233,
      "epoch": 4.170446298984035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003365932114006528,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22154108583927154,
      "num_tokens": 95417587.0,
      "step": 45975
    },
    {
      "entropy": 5.605659246444702,
      "epoch": 4.170899854862119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033656209030871184,
      "loss": 4.7498,
      "mean_token_accuracy": 0.23565778732299805,
      "num_tokens": 95428288.0,
      "step": 45980
    },
    {
      "entropy": 5.643091487884521,
      "epoch": 4.1713534107402035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033653096794365115,
      "loss": 4.8341,
      "mean_token_accuracy": 0.2264254242181778,
      "num_tokens": 95438093.0,
      "step": 45985
    },
    {
      "entropy": 5.583966159820557,
      "epoch": 4.171806966618288,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033649984430611437,
      "loss": 4.7763,
      "mean_token_accuracy": 0.23059134930372238,
      "num_tokens": 95448397.0,
      "step": 45990
    },
    {
      "entropy": 5.745728158950806,
      "epoch": 4.172260522496371,
      "grad_norm": 1.25,
      "learning_rate": 0.00033646871939674506,
      "loss": 5.0002,
      "mean_token_accuracy": 0.21399605572223662,
      "num_tokens": 95459533.0,
      "step": 45995
    },
    {
      "entropy": 5.6715991497039795,
      "epoch": 4.172714078374455,
      "grad_norm": 1.328125,
      "learning_rate": 0.000336437593216187,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2279481530189514,
      "num_tokens": 95469507.0,
      "step": 46000
    },
    {
      "entropy": 5.7735827445983885,
      "epoch": 4.17316763425254,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003364064657650839,
      "loss": 4.9342,
      "mean_token_accuracy": 0.21621053814888,
      "num_tokens": 95479742.0,
      "step": 46005
    },
    {
      "entropy": 5.619962692260742,
      "epoch": 4.173621190130624,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033637533704407936,
      "loss": 4.7918,
      "mean_token_accuracy": 0.23762890696525574,
      "num_tokens": 95489751.0,
      "step": 46010
    },
    {
      "entropy": 5.716387367248535,
      "epoch": 4.174074746008708,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003363442070538173,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21937399208545685,
      "num_tokens": 95500580.0,
      "step": 46015
    },
    {
      "entropy": 5.741354894638062,
      "epoch": 4.1745283018867925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033631307579494137,
      "loss": 4.9458,
      "mean_token_accuracy": 0.22390652000904082,
      "num_tokens": 95511018.0,
      "step": 46020
    },
    {
      "entropy": 5.729398632049561,
      "epoch": 4.174981857764877,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033628194326809543,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22720793783664703,
      "num_tokens": 95521220.0,
      "step": 46025
    },
    {
      "entropy": 5.660093545913696,
      "epoch": 4.175435413642961,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033625080947392324,
      "loss": 4.854,
      "mean_token_accuracy": 0.22366981506347655,
      "num_tokens": 95531426.0,
      "step": 46030
    },
    {
      "entropy": 5.709395503997802,
      "epoch": 4.175888969521045,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003362196744130687,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2330963507294655,
      "num_tokens": 95542698.0,
      "step": 46035
    },
    {
      "entropy": 5.709615802764892,
      "epoch": 4.1763425253991295,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033618853808617573,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23700927942991257,
      "num_tokens": 95553545.0,
      "step": 46040
    },
    {
      "entropy": 5.744338417053223,
      "epoch": 4.176796081277214,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003361574004938882,
      "loss": 4.9058,
      "mean_token_accuracy": 0.2242059513926506,
      "num_tokens": 95564452.0,
      "step": 46045
    },
    {
      "entropy": 5.6580479621887205,
      "epoch": 4.177249637155297,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033612626163685,
      "loss": 4.8874,
      "mean_token_accuracy": 0.233695150911808,
      "num_tokens": 95574756.0,
      "step": 46050
    },
    {
      "entropy": 5.74639253616333,
      "epoch": 4.1777031930333814,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033609512151570517,
      "loss": 4.9673,
      "mean_token_accuracy": 0.21630845963954926,
      "num_tokens": 95585147.0,
      "step": 46055
    },
    {
      "entropy": 5.698428583145142,
      "epoch": 4.178156748911466,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033606398013109765,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2246038571000099,
      "num_tokens": 95596025.0,
      "step": 46060
    },
    {
      "entropy": 5.687728500366211,
      "epoch": 4.17861030478955,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003360328374836715,
      "loss": 4.8706,
      "mean_token_accuracy": 0.22005927115678786,
      "num_tokens": 95607191.0,
      "step": 46065
    },
    {
      "entropy": 5.698869609832764,
      "epoch": 4.179063860667634,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003360016935740706,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22165920585393906,
      "num_tokens": 95617256.0,
      "step": 46070
    },
    {
      "entropy": 5.710460042953491,
      "epoch": 4.1795174165457185,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033597054840293927,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22560470551252365,
      "num_tokens": 95627638.0,
      "step": 46075
    },
    {
      "entropy": 5.676480531692505,
      "epoch": 4.179970972423803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003359394019709214,
      "loss": 4.8217,
      "mean_token_accuracy": 0.22575396448373794,
      "num_tokens": 95637662.0,
      "step": 46080
    },
    {
      "entropy": 5.677263307571411,
      "epoch": 4.180424528301887,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033590825427866117,
      "loss": 4.8396,
      "mean_token_accuracy": 0.22200418263673782,
      "num_tokens": 95647505.0,
      "step": 46085
    },
    {
      "entropy": 5.648023796081543,
      "epoch": 4.180878084179971,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003358771053268027,
      "loss": 4.7962,
      "mean_token_accuracy": 0.21993348151445388,
      "num_tokens": 95657810.0,
      "step": 46090
    },
    {
      "entropy": 5.631044530868531,
      "epoch": 4.181331640058056,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033584595511599024,
      "loss": 4.8107,
      "mean_token_accuracy": 0.22664395570755005,
      "num_tokens": 95667838.0,
      "step": 46095
    },
    {
      "entropy": 5.69027624130249,
      "epoch": 4.181785195936139,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003358148036468679,
      "loss": 4.852,
      "mean_token_accuracy": 0.2323600947856903,
      "num_tokens": 95678497.0,
      "step": 46100
    },
    {
      "entropy": 5.630828619003296,
      "epoch": 4.182238751814223,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033578365092007986,
      "loss": 4.7747,
      "mean_token_accuracy": 0.22508976608514786,
      "num_tokens": 95688656.0,
      "step": 46105
    },
    {
      "entropy": 5.706089019775391,
      "epoch": 4.1826923076923075,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003357524969362705,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2185644045472145,
      "num_tokens": 95699156.0,
      "step": 46110
    },
    {
      "entropy": 5.709915685653686,
      "epoch": 4.183145863570392,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033572134169608395,
      "loss": 4.8804,
      "mean_token_accuracy": 0.23125759065151213,
      "num_tokens": 95709680.0,
      "step": 46115
    },
    {
      "entropy": 5.748576498031616,
      "epoch": 4.183599419448476,
      "grad_norm": 1.375,
      "learning_rate": 0.00033569018520016465,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22202230393886566,
      "num_tokens": 95719142.0,
      "step": 46120
    },
    {
      "entropy": 5.683540487289429,
      "epoch": 4.18405297532656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003356590274491568,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21833272874355317,
      "num_tokens": 95729415.0,
      "step": 46125
    },
    {
      "entropy": 5.6179084300994875,
      "epoch": 4.1845065312046446,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003356278684437049,
      "loss": 4.8696,
      "mean_token_accuracy": 0.2224570944905281,
      "num_tokens": 95739775.0,
      "step": 46130
    },
    {
      "entropy": 5.711679363250733,
      "epoch": 4.184960087082729,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033559670818445315,
      "loss": 4.9363,
      "mean_token_accuracy": 0.22121995240449904,
      "num_tokens": 95751540.0,
      "step": 46135
    },
    {
      "entropy": 5.797618579864502,
      "epoch": 4.185413642960813,
      "grad_norm": 1.359375,
      "learning_rate": 0.000335565546672046,
      "loss": 5.0424,
      "mean_token_accuracy": 0.2182249128818512,
      "num_tokens": 95761885.0,
      "step": 46140
    },
    {
      "entropy": 5.724494981765747,
      "epoch": 4.185867198838897,
      "grad_norm": 1.21875,
      "learning_rate": 0.000335534383907128,
      "loss": 4.8802,
      "mean_token_accuracy": 0.2263200581073761,
      "num_tokens": 95772284.0,
      "step": 46145
    },
    {
      "entropy": 5.6371794700622555,
      "epoch": 4.186320754716981,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003355032198903434,
      "loss": 4.8135,
      "mean_token_accuracy": 0.22623331099748611,
      "num_tokens": 95783373.0,
      "step": 46150
    },
    {
      "entropy": 5.583518314361572,
      "epoch": 4.186774310595065,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033547205462233694,
      "loss": 4.7584,
      "mean_token_accuracy": 0.23460909575223923,
      "num_tokens": 95793226.0,
      "step": 46155
    },
    {
      "entropy": 5.550963544845581,
      "epoch": 4.187227866473149,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033544088810375287,
      "loss": 4.7725,
      "mean_token_accuracy": 0.22641163170337678,
      "num_tokens": 95803559.0,
      "step": 46160
    },
    {
      "entropy": 5.637325096130371,
      "epoch": 4.1876814223512335,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003354097203352358,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2176286295056343,
      "num_tokens": 95814620.0,
      "step": 46165
    },
    {
      "entropy": 5.6432798385620115,
      "epoch": 4.188134978229318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003353785513174305,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2286339744925499,
      "num_tokens": 95825024.0,
      "step": 46170
    },
    {
      "entropy": 5.701894378662109,
      "epoch": 4.188588534107402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033534738105098124,
      "loss": 4.9055,
      "mean_token_accuracy": 0.22606592178344725,
      "num_tokens": 95837373.0,
      "step": 46175
    },
    {
      "entropy": 5.687940168380737,
      "epoch": 4.189042089985486,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033531620953653276,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22362521588802337,
      "num_tokens": 95846962.0,
      "step": 46180
    },
    {
      "entropy": 5.679882669448853,
      "epoch": 4.189495645863571,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003352850367747297,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22488296329975127,
      "num_tokens": 95857492.0,
      "step": 46185
    },
    {
      "entropy": 5.64149661064148,
      "epoch": 4.189949201741655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003352538627662168,
      "loss": 4.8141,
      "mean_token_accuracy": 0.2241443157196045,
      "num_tokens": 95867903.0,
      "step": 46190
    },
    {
      "entropy": 5.761031198501587,
      "epoch": 4.190402757619739,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003352226875116386,
      "loss": 4.9822,
      "mean_token_accuracy": 0.2197064623236656,
      "num_tokens": 95878720.0,
      "step": 46195
    },
    {
      "entropy": 5.6713916778564455,
      "epoch": 4.1908563134978225,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003351915110116399,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22056437879800797,
      "num_tokens": 95888406.0,
      "step": 46200
    },
    {
      "entropy": 5.655045127868652,
      "epoch": 4.191309869375907,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033516033326686537,
      "loss": 4.8866,
      "mean_token_accuracy": 0.21591058224439622,
      "num_tokens": 95898431.0,
      "step": 46205
    },
    {
      "entropy": 5.7061375141143795,
      "epoch": 4.191763425253991,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033512915427795975,
      "loss": 4.8522,
      "mean_token_accuracy": 0.23053670078516006,
      "num_tokens": 95909089.0,
      "step": 46210
    },
    {
      "entropy": 5.650598764419556,
      "epoch": 4.192216981132075,
      "grad_norm": 1.328125,
      "learning_rate": 0.000335097974045568,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22529788613319396,
      "num_tokens": 95919787.0,
      "step": 46215
    },
    {
      "entropy": 5.683562994003296,
      "epoch": 4.19267053701016,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033506679257033485,
      "loss": 4.9176,
      "mean_token_accuracy": 0.22522805631160736,
      "num_tokens": 95930620.0,
      "step": 46220
    },
    {
      "entropy": 5.679514217376709,
      "epoch": 4.193124092888244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003350356098529051,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2304411843419075,
      "num_tokens": 95940258.0,
      "step": 46225
    },
    {
      "entropy": 5.696340322494507,
      "epoch": 4.193577648766328,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003350044258939236,
      "loss": 4.9326,
      "mean_token_accuracy": 0.22667087465524674,
      "num_tokens": 95950062.0,
      "step": 46230
    },
    {
      "entropy": 5.652636623382568,
      "epoch": 4.194031204644412,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033497324069403536,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2318427234888077,
      "num_tokens": 95960658.0,
      "step": 46235
    },
    {
      "entropy": 5.701156949996948,
      "epoch": 4.194484760522497,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003349420542538851,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23048902302980423,
      "num_tokens": 95970497.0,
      "step": 46240
    },
    {
      "entropy": 5.696426820755005,
      "epoch": 4.194938316400581,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033491086657411797,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21871609836816788,
      "num_tokens": 95980405.0,
      "step": 46245
    },
    {
      "entropy": 5.74472074508667,
      "epoch": 4.195391872278664,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003348796776553788,
      "loss": 4.9636,
      "mean_token_accuracy": 0.2173682987689972,
      "num_tokens": 95991111.0,
      "step": 46250
    },
    {
      "entropy": 5.7374592304229735,
      "epoch": 4.1958454281567485,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003348484874983127,
      "loss": 4.8964,
      "mean_token_accuracy": 0.22624988704919816,
      "num_tokens": 96001962.0,
      "step": 46255
    },
    {
      "entropy": 5.670695638656616,
      "epoch": 4.196298984034833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003348172961035646,
      "loss": 4.8389,
      "mean_token_accuracy": 0.2258397713303566,
      "num_tokens": 96011774.0,
      "step": 46260
    },
    {
      "entropy": 5.6405387878417965,
      "epoch": 4.196752539912917,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003347861034717795,
      "loss": 4.8836,
      "mean_token_accuracy": 0.22053113281726838,
      "num_tokens": 96021915.0,
      "step": 46265
    },
    {
      "entropy": 5.614178800582886,
      "epoch": 4.197206095791001,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033475490960360266,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22341588735580445,
      "num_tokens": 96032851.0,
      "step": 46270
    },
    {
      "entropy": 5.628676700592041,
      "epoch": 4.197659651669086,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000334723714499679,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23020168095827104,
      "num_tokens": 96042630.0,
      "step": 46275
    },
    {
      "entropy": 5.656966257095337,
      "epoch": 4.19811320754717,
      "grad_norm": 1.484375,
      "learning_rate": 0.00033469251816065375,
      "loss": 4.7998,
      "mean_token_accuracy": 0.23116175085306168,
      "num_tokens": 96052154.0,
      "step": 46280
    },
    {
      "entropy": 5.709416818618775,
      "epoch": 4.198566763425254,
      "grad_norm": 1.421875,
      "learning_rate": 0.000334661320587172,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22897300273180007,
      "num_tokens": 96062301.0,
      "step": 46285
    },
    {
      "entropy": 5.721240282058716,
      "epoch": 4.199020319303338,
      "grad_norm": 1.328125,
      "learning_rate": 0.000334630121779879,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2206738695502281,
      "num_tokens": 96072792.0,
      "step": 46290
    },
    {
      "entropy": 5.7347647190094,
      "epoch": 4.199473875181423,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033459892173941973,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21921680271625518,
      "num_tokens": 96082610.0,
      "step": 46295
    },
    {
      "entropy": 5.764332437515259,
      "epoch": 4.199927431059507,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003345677204664397,
      "loss": 4.9504,
      "mean_token_accuracy": 0.22139354050159454,
      "num_tokens": 96093095.0,
      "step": 46300
    },
    {
      "entropy": 5.653356266021729,
      "epoch": 4.20038098693759,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033453651796158397,
      "loss": 4.8444,
      "mean_token_accuracy": 0.23290420025587083,
      "num_tokens": 96103218.0,
      "step": 46305
    },
    {
      "entropy": 5.63716983795166,
      "epoch": 4.200834542815675,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033450531422549786,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22758259177207946,
      "num_tokens": 96113937.0,
      "step": 46310
    },
    {
      "entropy": 5.778512382507325,
      "epoch": 4.201288098693759,
      "grad_norm": 1.59375,
      "learning_rate": 0.00033447410925882673,
      "loss": 5.0159,
      "mean_token_accuracy": 0.221172334253788,
      "num_tokens": 96124378.0,
      "step": 46315
    },
    {
      "entropy": 5.617116403579712,
      "epoch": 4.201741654571843,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003344429030622159,
      "loss": 4.7553,
      "mean_token_accuracy": 0.2324612870812416,
      "num_tokens": 96133657.0,
      "step": 46320
    },
    {
      "entropy": 5.634434843063355,
      "epoch": 4.202195210449927,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033441169563631055,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2298687994480133,
      "num_tokens": 96143669.0,
      "step": 46325
    },
    {
      "entropy": 5.6544252872467045,
      "epoch": 4.202648766328012,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033438048698175625,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22578764855861663,
      "num_tokens": 96154782.0,
      "step": 46330
    },
    {
      "entropy": 5.678867053985596,
      "epoch": 4.203102322206096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003343492770991984,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22567752599716187,
      "num_tokens": 96165128.0,
      "step": 46335
    },
    {
      "entropy": 5.663485908508301,
      "epoch": 4.20355587808418,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003343180659892824,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22269556373357774,
      "num_tokens": 96176909.0,
      "step": 46340
    },
    {
      "entropy": 5.676246547698975,
      "epoch": 4.2040094339622645,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003342868536526536,
      "loss": 4.8365,
      "mean_token_accuracy": 0.23048171252012253,
      "num_tokens": 96188287.0,
      "step": 46345
    },
    {
      "entropy": 5.603690671920776,
      "epoch": 4.204462989840349,
      "grad_norm": 1.375,
      "learning_rate": 0.00033425564008995764,
      "loss": 4.8345,
      "mean_token_accuracy": 0.22889405637979507,
      "num_tokens": 96198852.0,
      "step": 46350
    },
    {
      "entropy": 5.646171998977661,
      "epoch": 4.204916545718432,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003342244253018399,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2231426492333412,
      "num_tokens": 96208111.0,
      "step": 46355
    },
    {
      "entropy": 5.674980974197387,
      "epoch": 4.205370101596516,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003341932092889459,
      "loss": 4.8409,
      "mean_token_accuracy": 0.2309754431247711,
      "num_tokens": 96218925.0,
      "step": 46360
    },
    {
      "entropy": 5.69445948600769,
      "epoch": 4.205823657474601,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003341619920519213,
      "loss": 4.9098,
      "mean_token_accuracy": 0.2319643661379814,
      "num_tokens": 96228573.0,
      "step": 46365
    },
    {
      "entropy": 5.694979572296143,
      "epoch": 4.206277213352685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033413077359141166,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22321068197488786,
      "num_tokens": 96239109.0,
      "step": 46370
    },
    {
      "entropy": 5.694328022003174,
      "epoch": 4.206730769230769,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003340995539080625,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21522919684648514,
      "num_tokens": 96249286.0,
      "step": 46375
    },
    {
      "entropy": 5.620876121520996,
      "epoch": 4.207184325108853,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033406833300251947,
      "loss": 4.8452,
      "mean_token_accuracy": 0.22553667575120925,
      "num_tokens": 96259515.0,
      "step": 46380
    },
    {
      "entropy": 5.640081262588501,
      "epoch": 4.207637880986938,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003340371108754284,
      "loss": 4.7976,
      "mean_token_accuracy": 0.22967431098222732,
      "num_tokens": 96269136.0,
      "step": 46385
    },
    {
      "entropy": 5.70844874382019,
      "epoch": 4.208091436865022,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003340058875274347,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22910353690385818,
      "num_tokens": 96279792.0,
      "step": 46390
    },
    {
      "entropy": 5.66708517074585,
      "epoch": 4.208544992743106,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003339746629591842,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2255984514951706,
      "num_tokens": 96289606.0,
      "step": 46395
    },
    {
      "entropy": 5.626698398590088,
      "epoch": 4.2089985486211905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033394343717132275,
      "loss": 4.9153,
      "mean_token_accuracy": 0.21923031955957412,
      "num_tokens": 96299903.0,
      "step": 46400
    },
    {
      "entropy": 5.752191781997681,
      "epoch": 4.209452104499274,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033391221016449595,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21440760046243668,
      "num_tokens": 96310179.0,
      "step": 46405
    },
    {
      "entropy": 5.725928544998169,
      "epoch": 4.209905660377358,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033388098193934965,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2212996780872345,
      "num_tokens": 96320948.0,
      "step": 46410
    },
    {
      "entropy": 5.627597093582153,
      "epoch": 4.210359216255442,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003338497524965296,
      "loss": 4.8302,
      "mean_token_accuracy": 0.22239601463079453,
      "num_tokens": 96332046.0,
      "step": 46415
    },
    {
      "entropy": 5.6298182010650635,
      "epoch": 4.210812772133527,
      "grad_norm": 1.375,
      "learning_rate": 0.00033381852183668166,
      "loss": 4.7983,
      "mean_token_accuracy": 0.23208268731832504,
      "num_tokens": 96342327.0,
      "step": 46420
    },
    {
      "entropy": 5.641676425933838,
      "epoch": 4.211266328011611,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003337872899604517,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21738712042570113,
      "num_tokens": 96352048.0,
      "step": 46425
    },
    {
      "entropy": 5.651961708068848,
      "epoch": 4.211719883889695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033375605686848566,
      "loss": 4.7863,
      "mean_token_accuracy": 0.2290177971124649,
      "num_tokens": 96362917.0,
      "step": 46430
    },
    {
      "entropy": 5.661551189422608,
      "epoch": 4.2121734397677795,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003337248225614293,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2274763897061348,
      "num_tokens": 96373261.0,
      "step": 46435
    },
    {
      "entropy": 5.675552082061768,
      "epoch": 4.212626995645864,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003336935870399288,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21607214212417603,
      "num_tokens": 96384279.0,
      "step": 46440
    },
    {
      "entropy": 5.742441177368164,
      "epoch": 4.213080551523948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003336623503046299,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22264773398637772,
      "num_tokens": 96394071.0,
      "step": 46445
    },
    {
      "entropy": 5.658328247070313,
      "epoch": 4.213534107402032,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033363111235617854,
      "loss": 4.8699,
      "mean_token_accuracy": 0.2297147184610367,
      "num_tokens": 96404233.0,
      "step": 46450
    },
    {
      "entropy": 5.6747157096862795,
      "epoch": 4.2139876632801165,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033359987319522095,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22754011899232865,
      "num_tokens": 96415388.0,
      "step": 46455
    },
    {
      "entropy": 5.732484245300293,
      "epoch": 4.2144412191582,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033356863282240295,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22248462736606597,
      "num_tokens": 96425744.0,
      "step": 46460
    },
    {
      "entropy": 5.695106267929077,
      "epoch": 4.214894775036284,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003335373912383707,
      "loss": 4.862,
      "mean_token_accuracy": 0.2353838324546814,
      "num_tokens": 96436340.0,
      "step": 46465
    },
    {
      "entropy": 5.747853851318359,
      "epoch": 4.2153483309143684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003335061484437704,
      "loss": 4.975,
      "mean_token_accuracy": 0.21487181186676024,
      "num_tokens": 96447384.0,
      "step": 46470
    },
    {
      "entropy": 5.647762203216553,
      "epoch": 4.215801886792453,
      "grad_norm": 1.296875,
      "learning_rate": 0.000333474904439248,
      "loss": 4.8439,
      "mean_token_accuracy": 0.23008871972560882,
      "num_tokens": 96456938.0,
      "step": 46475
    },
    {
      "entropy": 5.6212774276733395,
      "epoch": 4.216255442670537,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003334436592254496,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22531364113092422,
      "num_tokens": 96467343.0,
      "step": 46480
    },
    {
      "entropy": 5.652690887451172,
      "epoch": 4.216708998548621,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033341241280302153,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22551099061965943,
      "num_tokens": 96477443.0,
      "step": 46485
    },
    {
      "entropy": 5.660575437545776,
      "epoch": 4.2171625544267055,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033338116517260985,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21605146676301956,
      "num_tokens": 96487711.0,
      "step": 46490
    },
    {
      "entropy": 5.76071195602417,
      "epoch": 4.21761611030479,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003333499163348608,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2178516075015068,
      "num_tokens": 96497839.0,
      "step": 46495
    },
    {
      "entropy": 5.6869344234466555,
      "epoch": 4.218069666182874,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033331866629042055,
      "loss": 4.9409,
      "mean_token_accuracy": 0.2209418907761574,
      "num_tokens": 96508237.0,
      "step": 46500
    },
    {
      "entropy": 5.625384140014648,
      "epoch": 4.218523222060958,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003332874150399355,
      "loss": 4.7795,
      "mean_token_accuracy": 0.22704704105854034,
      "num_tokens": 96518877.0,
      "step": 46505
    },
    {
      "entropy": 5.648231363296508,
      "epoch": 4.218976777939042,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003332561625840519,
      "loss": 4.8399,
      "mean_token_accuracy": 0.22353625744581224,
      "num_tokens": 96530734.0,
      "step": 46510
    },
    {
      "entropy": 5.732207155227661,
      "epoch": 4.219430333817126,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003332249089234159,
      "loss": 4.9431,
      "mean_token_accuracy": 0.22388130128383638,
      "num_tokens": 96540983.0,
      "step": 46515
    },
    {
      "entropy": 5.735537195205689,
      "epoch": 4.21988388969521,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033319365405867394,
      "loss": 4.952,
      "mean_token_accuracy": 0.22156168520450592,
      "num_tokens": 96551866.0,
      "step": 46520
    },
    {
      "entropy": 5.681655502319336,
      "epoch": 4.2203374455732945,
      "grad_norm": 1.375,
      "learning_rate": 0.00033316239799047247,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2233220160007477,
      "num_tokens": 96562515.0,
      "step": 46525
    },
    {
      "entropy": 5.630547428131104,
      "epoch": 4.220791001451379,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033313114071945764,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22379629909992219,
      "num_tokens": 96573266.0,
      "step": 46530
    },
    {
      "entropy": 5.650281715393066,
      "epoch": 4.221244557329463,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003330998822462762,
      "loss": 4.8226,
      "mean_token_accuracy": 0.23221779763698577,
      "num_tokens": 96583070.0,
      "step": 46535
    },
    {
      "entropy": 5.644886445999146,
      "epoch": 4.221698113207547,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033306862257157427,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23255024403333663,
      "num_tokens": 96593582.0,
      "step": 46540
    },
    {
      "entropy": 5.6363588809967045,
      "epoch": 4.2221516690856316,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003330373616959984,
      "loss": 4.8488,
      "mean_token_accuracy": 0.23394003361463547,
      "num_tokens": 96603233.0,
      "step": 46545
    },
    {
      "entropy": 5.6279326438903805,
      "epoch": 4.222605224963716,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033300609962019523,
      "loss": 4.8285,
      "mean_token_accuracy": 0.22774043083190917,
      "num_tokens": 96614361.0,
      "step": 46550
    },
    {
      "entropy": 5.650464200973511,
      "epoch": 4.2230587808418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000332974836344811,
      "loss": 4.8964,
      "mean_token_accuracy": 0.22408775985240936,
      "num_tokens": 96624161.0,
      "step": 46555
    },
    {
      "entropy": 5.7006810188293455,
      "epoch": 4.2235123367198835,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033294357187049246,
      "loss": 4.9005,
      "mean_token_accuracy": 0.22136003971099855,
      "num_tokens": 96634329.0,
      "step": 46560
    },
    {
      "entropy": 5.7343731880187985,
      "epoch": 4.223965892597968,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033291230619788604,
      "loss": 4.9374,
      "mean_token_accuracy": 0.22745105028152465,
      "num_tokens": 96645575.0,
      "step": 46565
    },
    {
      "entropy": 5.710940170288086,
      "epoch": 4.224419448476052,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033288103932763846,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21736026257276536,
      "num_tokens": 96656224.0,
      "step": 46570
    },
    {
      "entropy": 5.716402959823609,
      "epoch": 4.224873004354136,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003328497712603962,
      "loss": 4.8552,
      "mean_token_accuracy": 0.23283409625291823,
      "num_tokens": 96665950.0,
      "step": 46575
    },
    {
      "entropy": 5.611243534088135,
      "epoch": 4.2253265602322205,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033281850199680586,
      "loss": 4.7493,
      "mean_token_accuracy": 0.23403723090887069,
      "num_tokens": 96676916.0,
      "step": 46580
    },
    {
      "entropy": 5.680910301208496,
      "epoch": 4.225780116110305,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003327872315375142,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21581359058618546,
      "num_tokens": 96687289.0,
      "step": 46585
    },
    {
      "entropy": 5.694920206069947,
      "epoch": 4.226233671988389,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003327559598831679,
      "loss": 4.8936,
      "mean_token_accuracy": 0.22803395241498947,
      "num_tokens": 96698149.0,
      "step": 46590
    },
    {
      "entropy": 5.655440187454223,
      "epoch": 4.226687227866473,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033272468703441363,
      "loss": 4.863,
      "mean_token_accuracy": 0.2261739954352379,
      "num_tokens": 96709466.0,
      "step": 46595
    },
    {
      "entropy": 5.692148208618164,
      "epoch": 4.227140783744558,
      "grad_norm": 1.3125,
      "learning_rate": 0.000332693412991898,
      "loss": 4.9163,
      "mean_token_accuracy": 0.22609069049358368,
      "num_tokens": 96720179.0,
      "step": 46600
    },
    {
      "entropy": 5.730257368087768,
      "epoch": 4.227594339622642,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003326621377562681,
      "loss": 4.921,
      "mean_token_accuracy": 0.21624848544597625,
      "num_tokens": 96730783.0,
      "step": 46605
    },
    {
      "entropy": 5.7256816864013675,
      "epoch": 4.228047895500726,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033263086132817035,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2196589559316635,
      "num_tokens": 96740903.0,
      "step": 46610
    },
    {
      "entropy": 5.677406930923462,
      "epoch": 4.2285014513788095,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033259958370825165,
      "loss": 4.8592,
      "mean_token_accuracy": 0.2334198921918869,
      "num_tokens": 96750600.0,
      "step": 46615
    },
    {
      "entropy": 5.657588529586792,
      "epoch": 4.228955007256894,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033256830489715905,
      "loss": 4.8106,
      "mean_token_accuracy": 0.22996080666780472,
      "num_tokens": 96760550.0,
      "step": 46620
    },
    {
      "entropy": 5.690332317352295,
      "epoch": 4.229408563134978,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003325370248955392,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2272874429821968,
      "num_tokens": 96770784.0,
      "step": 46625
    },
    {
      "entropy": 5.643504905700683,
      "epoch": 4.229862119013062,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003325057437040389,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23126513957977296,
      "num_tokens": 96780572.0,
      "step": 46630
    },
    {
      "entropy": 5.642734098434448,
      "epoch": 4.230315674891147,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003324744613233052,
      "loss": 4.7987,
      "mean_token_accuracy": 0.22768006026744841,
      "num_tokens": 96790504.0,
      "step": 46635
    },
    {
      "entropy": 5.6875114917755125,
      "epoch": 4.230769230769231,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033244317775398506,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22599518299102783,
      "num_tokens": 96800856.0,
      "step": 46640
    },
    {
      "entropy": 5.641908311843872,
      "epoch": 4.231222786647315,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003324118929967253,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22116409987211227,
      "num_tokens": 96811293.0,
      "step": 46645
    },
    {
      "entropy": 5.566002941131591,
      "epoch": 4.231676342525399,
      "grad_norm": 1.265625,
      "learning_rate": 0.000332380607052173,
      "loss": 4.7864,
      "mean_token_accuracy": 0.2353133425116539,
      "num_tokens": 96821844.0,
      "step": 46650
    },
    {
      "entropy": 5.67760157585144,
      "epoch": 4.232129898403484,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003323493199209751,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21882238537073134,
      "num_tokens": 96833438.0,
      "step": 46655
    },
    {
      "entropy": 5.632921171188355,
      "epoch": 4.232583454281568,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003323180316037787,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22766333669424058,
      "num_tokens": 96843086.0,
      "step": 46660
    },
    {
      "entropy": 5.688339757919311,
      "epoch": 4.233037010159651,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003322867421012308,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22545151561498641,
      "num_tokens": 96853064.0,
      "step": 46665
    },
    {
      "entropy": 5.76229338645935,
      "epoch": 4.2334905660377355,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003322554514139784,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2213694468140602,
      "num_tokens": 96864213.0,
      "step": 46670
    },
    {
      "entropy": 5.76208724975586,
      "epoch": 4.23394412191582,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033222415954266873,
      "loss": 5.0068,
      "mean_token_accuracy": 0.21810267120599747,
      "num_tokens": 96875344.0,
      "step": 46675
    },
    {
      "entropy": 5.745862627029419,
      "epoch": 4.234397677793904,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033219286648794886,
      "loss": 4.9453,
      "mean_token_accuracy": 0.22638072073459625,
      "num_tokens": 96885985.0,
      "step": 46680
    },
    {
      "entropy": 5.624972105026245,
      "epoch": 4.234851233671988,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003321615722504659,
      "loss": 4.7784,
      "mean_token_accuracy": 0.23417041301727295,
      "num_tokens": 96897512.0,
      "step": 46685
    },
    {
      "entropy": 5.655225849151611,
      "epoch": 4.235304789550073,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003321302768308671,
      "loss": 4.946,
      "mean_token_accuracy": 0.2111332595348358,
      "num_tokens": 96907630.0,
      "step": 46690
    },
    {
      "entropy": 5.615852022171021,
      "epoch": 4.235758345428157,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033209898022979964,
      "loss": 4.7925,
      "mean_token_accuracy": 0.23420303910970688,
      "num_tokens": 96917400.0,
      "step": 46695
    },
    {
      "entropy": 5.77334303855896,
      "epoch": 4.236211901306241,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003320676824479106,
      "loss": 4.981,
      "mean_token_accuracy": 0.22056981325149536,
      "num_tokens": 96927700.0,
      "step": 46700
    },
    {
      "entropy": 5.718997192382813,
      "epoch": 4.236665457184325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003320363834858474,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22380704581737518,
      "num_tokens": 96938925.0,
      "step": 46705
    },
    {
      "entropy": 5.634799528121948,
      "epoch": 4.23711901306241,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003320050833442573,
      "loss": 4.8056,
      "mean_token_accuracy": 0.2273781418800354,
      "num_tokens": 96948956.0,
      "step": 46710
    },
    {
      "entropy": 5.637960147857666,
      "epoch": 4.237572568940493,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003319737820237875,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22205977588891984,
      "num_tokens": 96958565.0,
      "step": 46715
    },
    {
      "entropy": 5.636889505386352,
      "epoch": 4.238026124818577,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033194247952508544,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22467081993818283,
      "num_tokens": 96968261.0,
      "step": 46720
    },
    {
      "entropy": 5.707002592086792,
      "epoch": 4.238479680696662,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033191117584879836,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2208097219467163,
      "num_tokens": 96978621.0,
      "step": 46725
    },
    {
      "entropy": 5.661589193344116,
      "epoch": 4.238933236574746,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003318798709955736,
      "loss": 4.8378,
      "mean_token_accuracy": 0.22556596994400024,
      "num_tokens": 96989081.0,
      "step": 46730
    },
    {
      "entropy": 5.76829948425293,
      "epoch": 4.23938679245283,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003318485649660586,
      "loss": 4.9932,
      "mean_token_accuracy": 0.21397027522325515,
      "num_tokens": 97000518.0,
      "step": 46735
    },
    {
      "entropy": 5.673490381240844,
      "epoch": 4.239840348330914,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033181725776090093,
      "loss": 4.8675,
      "mean_token_accuracy": 0.2255713611841202,
      "num_tokens": 97012038.0,
      "step": 46740
    },
    {
      "entropy": 5.68008508682251,
      "epoch": 4.240293904208999,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033178594938074787,
      "loss": 4.8475,
      "mean_token_accuracy": 0.22247387170791627,
      "num_tokens": 97022256.0,
      "step": 46745
    },
    {
      "entropy": 5.691211891174317,
      "epoch": 4.240747460087083,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003317546398262468,
      "loss": 4.9559,
      "mean_token_accuracy": 0.22166019529104233,
      "num_tokens": 97033063.0,
      "step": 46750
    },
    {
      "entropy": 5.657290744781494,
      "epoch": 4.241201015965167,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003317233290980454,
      "loss": 4.9041,
      "mean_token_accuracy": 0.22303344905376435,
      "num_tokens": 97043711.0,
      "step": 46755
    },
    {
      "entropy": 5.696803903579712,
      "epoch": 4.2416545718432515,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033169201719679105,
      "loss": 4.8268,
      "mean_token_accuracy": 0.2341694012284279,
      "num_tokens": 97053582.0,
      "step": 46760
    },
    {
      "entropy": 5.728871393203735,
      "epoch": 4.242108127721336,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033166070412313133,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2251521423459053,
      "num_tokens": 97064535.0,
      "step": 46765
    },
    {
      "entropy": 5.660004806518555,
      "epoch": 4.242561683599419,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003316293898777138,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2206759497523308,
      "num_tokens": 97074295.0,
      "step": 46770
    },
    {
      "entropy": 5.647121143341065,
      "epoch": 4.243015239477503,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033159807446118614,
      "loss": 4.8876,
      "mean_token_accuracy": 0.2193903759121895,
      "num_tokens": 97084781.0,
      "step": 46775
    },
    {
      "entropy": 5.75445065498352,
      "epoch": 4.243468795355588,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003315667578741958,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2187938004732132,
      "num_tokens": 97094848.0,
      "step": 46780
    },
    {
      "entropy": 5.716567182540894,
      "epoch": 4.243922351233672,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003315354401173905,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2293837547302246,
      "num_tokens": 97104747.0,
      "step": 46785
    },
    {
      "entropy": 5.633616733551025,
      "epoch": 4.244375907111756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003315041211914179,
      "loss": 4.8211,
      "mean_token_accuracy": 0.22519677132368088,
      "num_tokens": 97115599.0,
      "step": 46790
    },
    {
      "entropy": 5.657943773269653,
      "epoch": 4.24482946298984,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003314728010969257,
      "loss": 4.9711,
      "mean_token_accuracy": 0.22055939733982086,
      "num_tokens": 97127492.0,
      "step": 46795
    },
    {
      "entropy": 5.715530729293823,
      "epoch": 4.245283018867925,
      "grad_norm": 1.375,
      "learning_rate": 0.00033144147983456154,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21817528158426286,
      "num_tokens": 97138336.0,
      "step": 46800
    },
    {
      "entropy": 5.676424264907837,
      "epoch": 4.245736574746009,
      "grad_norm": 1.375,
      "learning_rate": 0.0003314101574049732,
      "loss": 4.8414,
      "mean_token_accuracy": 0.2285957530140877,
      "num_tokens": 97149196.0,
      "step": 46805
    },
    {
      "entropy": 5.719177150726319,
      "epoch": 4.246190130624093,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033137883380880847,
      "loss": 4.9312,
      "mean_token_accuracy": 0.22124050855636596,
      "num_tokens": 97159622.0,
      "step": 46810
    },
    {
      "entropy": 5.732594442367554,
      "epoch": 4.2466436865021775,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000331347509046715,
      "loss": 4.97,
      "mean_token_accuracy": 0.21201612800359726,
      "num_tokens": 97170149.0,
      "step": 46815
    },
    {
      "entropy": 5.739738893508911,
      "epoch": 4.247097242380261,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003313161831193408,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2203262284398079,
      "num_tokens": 97180409.0,
      "step": 46820
    },
    {
      "entropy": 5.701783609390259,
      "epoch": 4.247550798258345,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033128485602733354,
      "loss": 4.9122,
      "mean_token_accuracy": 0.22342330813407899,
      "num_tokens": 97190899.0,
      "step": 46825
    },
    {
      "entropy": 5.721866035461426,
      "epoch": 4.248004354136429,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003312535277713411,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22733433693647384,
      "num_tokens": 97200759.0,
      "step": 46830
    },
    {
      "entropy": 5.685166883468628,
      "epoch": 4.248457910014514,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033122219835201135,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2319476678967476,
      "num_tokens": 97211476.0,
      "step": 46835
    },
    {
      "entropy": 5.7116588115692135,
      "epoch": 4.248911465892598,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003311908677699923,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21850743591785432,
      "num_tokens": 97222374.0,
      "step": 46840
    },
    {
      "entropy": 5.695628547668457,
      "epoch": 4.249365021770682,
      "grad_norm": 1.375,
      "learning_rate": 0.0003311595360259317,
      "loss": 4.8984,
      "mean_token_accuracy": 0.22292670607566833,
      "num_tokens": 97233079.0,
      "step": 46845
    },
    {
      "entropy": 5.598395824432373,
      "epoch": 4.2498185776487665,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033112820312047756,
      "loss": 4.7892,
      "mean_token_accuracy": 0.23457126766443254,
      "num_tokens": 97243796.0,
      "step": 46850
    },
    {
      "entropy": 5.626037979125977,
      "epoch": 4.250272133526851,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033109686905427794,
      "loss": 4.7605,
      "mean_token_accuracy": 0.23499299436807633,
      "num_tokens": 97253341.0,
      "step": 46855
    },
    {
      "entropy": 5.733515787124634,
      "epoch": 4.250725689404935,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033106553382798075,
      "loss": 4.942,
      "mean_token_accuracy": 0.21816972643136978,
      "num_tokens": 97264244.0,
      "step": 46860
    },
    {
      "entropy": 5.645431280136108,
      "epoch": 4.251179245283019,
      "grad_norm": 1.375,
      "learning_rate": 0.00033103419744223413,
      "loss": 4.8458,
      "mean_token_accuracy": 0.22767426669597626,
      "num_tokens": 97274239.0,
      "step": 46865
    },
    {
      "entropy": 5.613644599914551,
      "epoch": 4.251632801161103,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003310028598976859,
      "loss": 4.7696,
      "mean_token_accuracy": 0.2356210932135582,
      "num_tokens": 97284709.0,
      "step": 46870
    },
    {
      "entropy": 5.620578193664551,
      "epoch": 4.252086357039187,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033097152119498425,
      "loss": 4.799,
      "mean_token_accuracy": 0.22977874726057052,
      "num_tokens": 97295262.0,
      "step": 46875
    },
    {
      "entropy": 5.672809267044068,
      "epoch": 4.252539912917271,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003309401813347774,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2187335103750229,
      "num_tokens": 97306143.0,
      "step": 46880
    },
    {
      "entropy": 5.704710197448731,
      "epoch": 4.2529934687953554,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003309088403177132,
      "loss": 4.8117,
      "mean_token_accuracy": 0.23305461257696153,
      "num_tokens": 97316248.0,
      "step": 46885
    },
    {
      "entropy": 5.750253343582154,
      "epoch": 4.25344702467344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033087749814444,
      "loss": 4.9737,
      "mean_token_accuracy": 0.22110236883163453,
      "num_tokens": 97327704.0,
      "step": 46890
    },
    {
      "entropy": 5.744761657714844,
      "epoch": 4.253900580551524,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033084615481560585,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21289312839508057,
      "num_tokens": 97337817.0,
      "step": 46895
    },
    {
      "entropy": 5.731951761245727,
      "epoch": 4.254354136429608,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033081481033185903,
      "loss": 4.9563,
      "mean_token_accuracy": 0.22055644243955613,
      "num_tokens": 97348315.0,
      "step": 46900
    },
    {
      "entropy": 5.659780740737915,
      "epoch": 4.2548076923076925,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003307834646938477,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22482480108737946,
      "num_tokens": 97358842.0,
      "step": 46905
    },
    {
      "entropy": 5.7696269989013675,
      "epoch": 4.255261248185777,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003307521179022201,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21794855296611787,
      "num_tokens": 97369247.0,
      "step": 46910
    },
    {
      "entropy": 5.68321213722229,
      "epoch": 4.255714804063861,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003307207699576246,
      "loss": 4.8448,
      "mean_token_accuracy": 0.22952199280261992,
      "num_tokens": 97379668.0,
      "step": 46915
    },
    {
      "entropy": 5.663760709762573,
      "epoch": 4.256168359941945,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033068942086070924,
      "loss": 4.8478,
      "mean_token_accuracy": 0.2255804255604744,
      "num_tokens": 97389131.0,
      "step": 46920
    },
    {
      "entropy": 5.730261564254761,
      "epoch": 4.256621915820029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033065807061212254,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2156697317957878,
      "num_tokens": 97400583.0,
      "step": 46925
    },
    {
      "entropy": 5.7204207420349125,
      "epoch": 4.257075471698113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003306267192125128,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2233791321516037,
      "num_tokens": 97410828.0,
      "step": 46930
    },
    {
      "entropy": 5.717368221282959,
      "epoch": 4.257529027576197,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003305953666625283,
      "loss": 4.9233,
      "mean_token_accuracy": 0.2185799479484558,
      "num_tokens": 97421190.0,
      "step": 46935
    },
    {
      "entropy": 5.66599850654602,
      "epoch": 4.2579825834542815,
      "grad_norm": 1.25,
      "learning_rate": 0.00033056401296281733,
      "loss": 4.9459,
      "mean_token_accuracy": 0.21589289009571075,
      "num_tokens": 97432257.0,
      "step": 46940
    },
    {
      "entropy": 5.674975824356079,
      "epoch": 4.258436139332366,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003305326581140286,
      "loss": 4.9287,
      "mean_token_accuracy": 0.22064141631126405,
      "num_tokens": 97442272.0,
      "step": 46945
    },
    {
      "entropy": 5.667259407043457,
      "epoch": 4.25888969521045,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033050130211681017,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22867862433195113,
      "num_tokens": 97451462.0,
      "step": 46950
    },
    {
      "entropy": 5.651415634155273,
      "epoch": 4.259343251088534,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033046994497181077,
      "loss": 4.7692,
      "mean_token_accuracy": 0.22875206917524338,
      "num_tokens": 97461579.0,
      "step": 46955
    },
    {
      "entropy": 5.762288522720337,
      "epoch": 4.2597968069666186,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033043858667967875,
      "loss": 4.9638,
      "mean_token_accuracy": 0.2201904386281967,
      "num_tokens": 97472892.0,
      "step": 46960
    },
    {
      "entropy": 5.644103956222534,
      "epoch": 4.260250362844703,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033040722724106257,
      "loss": 4.8386,
      "mean_token_accuracy": 0.22571358531713487,
      "num_tokens": 97484278.0,
      "step": 46965
    },
    {
      "entropy": 5.659617280960083,
      "epoch": 4.260703918722786,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033037586665661094,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2219989240169525,
      "num_tokens": 97494127.0,
      "step": 46970
    },
    {
      "entropy": 5.656411266326904,
      "epoch": 4.2611574746008705,
      "grad_norm": 1.25,
      "learning_rate": 0.0003303445049269722,
      "loss": 4.8025,
      "mean_token_accuracy": 0.23088260889053344,
      "num_tokens": 97504213.0,
      "step": 46975
    },
    {
      "entropy": 5.74541540145874,
      "epoch": 4.261611030478955,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033031314205279503,
      "loss": 4.9458,
      "mean_token_accuracy": 0.22348013669252395,
      "num_tokens": 97515242.0,
      "step": 46980
    },
    {
      "entropy": 5.757241582870483,
      "epoch": 4.262064586357039,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033028177803472793,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2247408375144005,
      "num_tokens": 97525956.0,
      "step": 46985
    },
    {
      "entropy": 5.6540446281433105,
      "epoch": 4.262518142235123,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003302504128734196,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23293780237436296,
      "num_tokens": 97535973.0,
      "step": 46990
    },
    {
      "entropy": 5.630394220352173,
      "epoch": 4.2629716981132075,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033021904656951867,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22867006212472915,
      "num_tokens": 97547580.0,
      "step": 46995
    },
    {
      "entropy": 5.666845893859863,
      "epoch": 4.263425253991292,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033018767912367376,
      "loss": 4.8476,
      "mean_token_accuracy": 0.22088335007429122,
      "num_tokens": 97558050.0,
      "step": 47000
    },
    {
      "entropy": 5.710594606399536,
      "epoch": 4.263878809869376,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033015631053653357,
      "loss": 4.8217,
      "mean_token_accuracy": 0.2318890631198883,
      "num_tokens": 97567997.0,
      "step": 47005
    },
    {
      "entropy": 5.675081968307495,
      "epoch": 4.26433236574746,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003301249408087469,
      "loss": 4.8451,
      "mean_token_accuracy": 0.2296413630247116,
      "num_tokens": 97577220.0,
      "step": 47010
    },
    {
      "entropy": 5.6979320526123045,
      "epoch": 4.264785921625545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033009356994096237,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2285613849759102,
      "num_tokens": 97588421.0,
      "step": 47015
    },
    {
      "entropy": 5.6226012229919435,
      "epoch": 4.265239477503629,
      "grad_norm": 1.25,
      "learning_rate": 0.00033006219793382867,
      "loss": 4.793,
      "mean_token_accuracy": 0.23029449880123137,
      "num_tokens": 97598624.0,
      "step": 47020
    },
    {
      "entropy": 5.691341543197632,
      "epoch": 4.265693033381712,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003300308247879948,
      "loss": 4.8814,
      "mean_token_accuracy": 0.2207088738679886,
      "num_tokens": 97609128.0,
      "step": 47025
    },
    {
      "entropy": 5.6923095703125,
      "epoch": 4.2661465892597965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003299994505041094,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22919802665710448,
      "num_tokens": 97619549.0,
      "step": 47030
    },
    {
      "entropy": 5.6800089359283445,
      "epoch": 4.266600145137881,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032996807508282134,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22280146628618241,
      "num_tokens": 97630184.0,
      "step": 47035
    },
    {
      "entropy": 5.660915565490723,
      "epoch": 4.267053701015965,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032993669852477954,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22722068428993225,
      "num_tokens": 97640569.0,
      "step": 47040
    },
    {
      "entropy": 5.659248495101929,
      "epoch": 4.267507256894049,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003299053208306328,
      "loss": 4.8312,
      "mean_token_accuracy": 0.2284248188138008,
      "num_tokens": 97651006.0,
      "step": 47045
    },
    {
      "entropy": 5.658032178878784,
      "epoch": 4.267960812772134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003298739420010299,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22692545056343078,
      "num_tokens": 97661567.0,
      "step": 47050
    },
    {
      "entropy": 5.650281095504761,
      "epoch": 4.268414368650218,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003298425620366201,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22889568358659745,
      "num_tokens": 97673058.0,
      "step": 47055
    },
    {
      "entropy": 5.781052684783935,
      "epoch": 4.268867924528302,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003298111809380521,
      "loss": 5.0038,
      "mean_token_accuracy": 0.21487070620059967,
      "num_tokens": 97684509.0,
      "step": 47060
    },
    {
      "entropy": 5.764691877365112,
      "epoch": 4.269321480406386,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032977979870597477,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2179829940199852,
      "num_tokens": 97693954.0,
      "step": 47065
    },
    {
      "entropy": 5.699521827697754,
      "epoch": 4.269775036284471,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003297484153410374,
      "loss": 4.8238,
      "mean_token_accuracy": 0.2289520174264908,
      "num_tokens": 97704345.0,
      "step": 47070
    },
    {
      "entropy": 5.611645889282227,
      "epoch": 4.270228592162554,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032971703084388885,
      "loss": 4.7461,
      "mean_token_accuracy": 0.2410870298743248,
      "num_tokens": 97714857.0,
      "step": 47075
    },
    {
      "entropy": 5.671855115890503,
      "epoch": 4.270682148040638,
      "grad_norm": 1.234375,
      "learning_rate": 0.000329685645215178,
      "loss": 4.892,
      "mean_token_accuracy": 0.2232367977499962,
      "num_tokens": 97725921.0,
      "step": 47080
    },
    {
      "entropy": 5.6664213180542,
      "epoch": 4.2711357039187225,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032965425845555413,
      "loss": 4.8581,
      "mean_token_accuracy": 0.2381445974111557,
      "num_tokens": 97735361.0,
      "step": 47085
    },
    {
      "entropy": 5.665806865692138,
      "epoch": 4.271589259796807,
      "grad_norm": 1.25,
      "learning_rate": 0.00032962287056566633,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2219262659549713,
      "num_tokens": 97745915.0,
      "step": 47090
    },
    {
      "entropy": 5.698756551742553,
      "epoch": 4.272042815674891,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003295914815461637,
      "loss": 4.855,
      "mean_token_accuracy": 0.23077422380447388,
      "num_tokens": 97755663.0,
      "step": 47095
    },
    {
      "entropy": 5.738223314285278,
      "epoch": 4.272496371552975,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032956009139769514,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21545568406581878,
      "num_tokens": 97766300.0,
      "step": 47100
    },
    {
      "entropy": 5.7091155529022215,
      "epoch": 4.27294992743106,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032952870012091013,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2232059806585312,
      "num_tokens": 97776584.0,
      "step": 47105
    },
    {
      "entropy": 5.651587915420532,
      "epoch": 4.273403483309144,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032949730771645767,
      "loss": 4.8243,
      "mean_token_accuracy": 0.22760859429836272,
      "num_tokens": 97786230.0,
      "step": 47110
    },
    {
      "entropy": 5.718541860580444,
      "epoch": 4.273857039187228,
      "grad_norm": 1.3125,
      "learning_rate": 0.000329465914184987,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21965525448322296,
      "num_tokens": 97796671.0,
      "step": 47115
    },
    {
      "entropy": 5.640603303909302,
      "epoch": 4.274310595065312,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003294345195271473,
      "loss": 4.8567,
      "mean_token_accuracy": 0.22892607748508453,
      "num_tokens": 97807569.0,
      "step": 47120
    },
    {
      "entropy": 5.742707204818726,
      "epoch": 4.274764150943396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032940312374358794,
      "loss": 5.0014,
      "mean_token_accuracy": 0.21365204751491546,
      "num_tokens": 97818680.0,
      "step": 47125
    },
    {
      "entropy": 5.7019205570220945,
      "epoch": 4.27521770682148,
      "grad_norm": 1.421875,
      "learning_rate": 0.000329371726834958,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2238304927945137,
      "num_tokens": 97828955.0,
      "step": 47130
    },
    {
      "entropy": 5.687495470046997,
      "epoch": 4.275671262699564,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000329340328801907,
      "loss": 4.874,
      "mean_token_accuracy": 0.23057658076286316,
      "num_tokens": 97839000.0,
      "step": 47135
    },
    {
      "entropy": 5.657153940200805,
      "epoch": 4.276124818577649,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003293089296450841,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2144649088382721,
      "num_tokens": 97850140.0,
      "step": 47140
    },
    {
      "entropy": 5.654101753234864,
      "epoch": 4.276578374455733,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003292775293651388,
      "loss": 4.7942,
      "mean_token_accuracy": 0.2294245958328247,
      "num_tokens": 97860887.0,
      "step": 47145
    },
    {
      "entropy": 5.698821640014648,
      "epoch": 4.277031930333817,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003292461279627203,
      "loss": 4.8504,
      "mean_token_accuracy": 0.22793372720479965,
      "num_tokens": 97871190.0,
      "step": 47150
    },
    {
      "entropy": 5.663764429092407,
      "epoch": 4.277485486211901,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032921472543847804,
      "loss": 4.8041,
      "mean_token_accuracy": 0.2328104481101036,
      "num_tokens": 97881479.0,
      "step": 47155
    },
    {
      "entropy": 5.631919717788696,
      "epoch": 4.277939042089986,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003291833217930614,
      "loss": 4.878,
      "mean_token_accuracy": 0.22875399589538575,
      "num_tokens": 97892340.0,
      "step": 47160
    },
    {
      "entropy": 5.676173973083496,
      "epoch": 4.27839259796807,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032915191702711995,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22004313319921492,
      "num_tokens": 97902882.0,
      "step": 47165
    },
    {
      "entropy": 5.627889823913574,
      "epoch": 4.278846153846154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032912051114130297,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23036182224750518,
      "num_tokens": 97913187.0,
      "step": 47170
    },
    {
      "entropy": 5.624239444732666,
      "epoch": 4.2792997097242385,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00032908910413626,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23218010663986205,
      "num_tokens": 97922434.0,
      "step": 47175
    },
    {
      "entropy": 5.548560905456543,
      "epoch": 4.279753265602322,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003290576960126407,
      "loss": 4.7395,
      "mean_token_accuracy": 0.23581626266241074,
      "num_tokens": 97933020.0,
      "step": 47180
    },
    {
      "entropy": 5.6319740295410154,
      "epoch": 4.280206821480406,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032902628677109434,
      "loss": 4.8443,
      "mean_token_accuracy": 0.2259765788912773,
      "num_tokens": 97944103.0,
      "step": 47185
    },
    {
      "entropy": 5.709651136398316,
      "epoch": 4.28066037735849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003289948764122707,
      "loss": 4.8565,
      "mean_token_accuracy": 0.23097035884857178,
      "num_tokens": 97953692.0,
      "step": 47190
    },
    {
      "entropy": 5.72479887008667,
      "epoch": 4.281113933236575,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003289634649368193,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22914205342531205,
      "num_tokens": 97963592.0,
      "step": 47195
    },
    {
      "entropy": 5.751935958862305,
      "epoch": 4.281567489114659,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003289320523453895,
      "loss": 5.0502,
      "mean_token_accuracy": 0.21587749272584916,
      "num_tokens": 97974092.0,
      "step": 47200
    },
    {
      "entropy": 5.6376060962677,
      "epoch": 4.282021044992743,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032890063863863126,
      "loss": 4.7625,
      "mean_token_accuracy": 0.23120422214269637,
      "num_tokens": 97984175.0,
      "step": 47205
    },
    {
      "entropy": 5.636980056762695,
      "epoch": 4.282474600870827,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003288692238171941,
      "loss": 4.8874,
      "mean_token_accuracy": 0.23617047369480132,
      "num_tokens": 97994993.0,
      "step": 47210
    },
    {
      "entropy": 5.6741691589355465,
      "epoch": 4.282928156748912,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032883780788172756,
      "loss": 4.9002,
      "mean_token_accuracy": 0.218203404545784,
      "num_tokens": 98005892.0,
      "step": 47215
    },
    {
      "entropy": 5.623246669769287,
      "epoch": 4.283381712626996,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003288063908328814,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23167288154363633,
      "num_tokens": 98016726.0,
      "step": 47220
    },
    {
      "entropy": 5.725933265686035,
      "epoch": 4.28383526850508,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003287749726713055,
      "loss": 4.9054,
      "mean_token_accuracy": 0.22654304951429366,
      "num_tokens": 98028500.0,
      "step": 47225
    },
    {
      "entropy": 5.752040052413941,
      "epoch": 4.284288824383164,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003287435533976494,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21943883150815963,
      "num_tokens": 98039304.0,
      "step": 47230
    },
    {
      "entropy": 5.695298337936402,
      "epoch": 4.284742380261248,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003287121330125629,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22220477908849717,
      "num_tokens": 98049448.0,
      "step": 47235
    },
    {
      "entropy": 5.701738929748535,
      "epoch": 4.285195936139332,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032868071151669585,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2332577019929886,
      "num_tokens": 98059532.0,
      "step": 47240
    },
    {
      "entropy": 5.61998610496521,
      "epoch": 4.285649492017416,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000328649288910698,
      "loss": 4.7823,
      "mean_token_accuracy": 0.22975722551345826,
      "num_tokens": 98069385.0,
      "step": 47245
    },
    {
      "entropy": 5.694291448593139,
      "epoch": 4.286103047895501,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003286178651952192,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2296536386013031,
      "num_tokens": 98079354.0,
      "step": 47250
    },
    {
      "entropy": 5.711437082290649,
      "epoch": 4.286556603773585,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032858644037090924,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21793979853391648,
      "num_tokens": 98090864.0,
      "step": 47255
    },
    {
      "entropy": 5.66371111869812,
      "epoch": 4.287010159651669,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003285550144384181,
      "loss": 4.8395,
      "mean_token_accuracy": 0.22818129807710646,
      "num_tokens": 98101515.0,
      "step": 47260
    },
    {
      "entropy": 5.702941465377807,
      "epoch": 4.2874637155297535,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032852358739839555,
      "loss": 4.8582,
      "mean_token_accuracy": 0.22469830214977266,
      "num_tokens": 98111786.0,
      "step": 47265
    },
    {
      "entropy": 5.683735179901123,
      "epoch": 4.287917271407838,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032849215925149155,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23230572789907455,
      "num_tokens": 98121732.0,
      "step": 47270
    },
    {
      "entropy": 5.724025249481201,
      "epoch": 4.288370827285922,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003284607299983561,
      "loss": 4.9923,
      "mean_token_accuracy": 0.21983799934387208,
      "num_tokens": 98131417.0,
      "step": 47275
    },
    {
      "entropy": 5.7351508140563965,
      "epoch": 4.288824383164005,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032842929963963915,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2256147414445877,
      "num_tokens": 98141664.0,
      "step": 47280
    },
    {
      "entropy": 5.666127300262451,
      "epoch": 4.28927793904209,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032839786817599065,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22550419121980667,
      "num_tokens": 98150959.0,
      "step": 47285
    },
    {
      "entropy": 5.682355499267578,
      "epoch": 4.289731494920174,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003283664356080606,
      "loss": 4.966,
      "mean_token_accuracy": 0.2208641618490219,
      "num_tokens": 98161957.0,
      "step": 47290
    },
    {
      "entropy": 5.632346057891846,
      "epoch": 4.290185050798258,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032833500193649916,
      "loss": 4.8001,
      "mean_token_accuracy": 0.22397990375757218,
      "num_tokens": 98172053.0,
      "step": 47295
    },
    {
      "entropy": 5.67668514251709,
      "epoch": 4.2906386066763424,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032830356716195624,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2245911180973053,
      "num_tokens": 98182074.0,
      "step": 47300
    },
    {
      "entropy": 5.725739431381226,
      "epoch": 4.291092162554427,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003282721312850819,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2233437180519104,
      "num_tokens": 98192591.0,
      "step": 47305
    },
    {
      "entropy": 5.674335956573486,
      "epoch": 4.291545718432511,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003282406943065264,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23457320034503937,
      "num_tokens": 98203426.0,
      "step": 47310
    },
    {
      "entropy": 5.637891721725464,
      "epoch": 4.291999274310595,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032820925622693977,
      "loss": 4.8485,
      "mean_token_accuracy": 0.22813299000263215,
      "num_tokens": 98214342.0,
      "step": 47315
    },
    {
      "entropy": 5.680193567276001,
      "epoch": 4.2924528301886795,
      "grad_norm": 1.25,
      "learning_rate": 0.00032817781704697204,
      "loss": 4.8473,
      "mean_token_accuracy": 0.22287966310977936,
      "num_tokens": 98224155.0,
      "step": 47320
    },
    {
      "entropy": 5.663097906112671,
      "epoch": 4.292906386066764,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003281463767672736,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23103382289409638,
      "num_tokens": 98233613.0,
      "step": 47325
    },
    {
      "entropy": 5.658789014816284,
      "epoch": 4.293359941944848,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003281149353884946,
      "loss": 4.8485,
      "mean_token_accuracy": 0.2316870391368866,
      "num_tokens": 98243117.0,
      "step": 47330
    },
    {
      "entropy": 5.656659364700317,
      "epoch": 4.293813497822931,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003280834929112851,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22910133004188538,
      "num_tokens": 98253619.0,
      "step": 47335
    },
    {
      "entropy": 5.654382658004761,
      "epoch": 4.294267053701016,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003280520493362955,
      "loss": 4.8647,
      "mean_token_accuracy": 0.22564778476953506,
      "num_tokens": 98264221.0,
      "step": 47340
    },
    {
      "entropy": 5.663712453842163,
      "epoch": 4.2947206095791,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003280206046641759,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2270164504647255,
      "num_tokens": 98275530.0,
      "step": 47345
    },
    {
      "entropy": 5.76324462890625,
      "epoch": 4.295174165457184,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032798915889557687,
      "loss": 4.9546,
      "mean_token_accuracy": 0.2223445236682892,
      "num_tokens": 98285937.0,
      "step": 47350
    },
    {
      "entropy": 5.722216606140137,
      "epoch": 4.2956277213352685,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003279577120311483,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22277185022830964,
      "num_tokens": 98295690.0,
      "step": 47355
    },
    {
      "entropy": 5.6390468120574955,
      "epoch": 4.296081277213353,
      "grad_norm": 1.40625,
      "learning_rate": 0.000327926264071541,
      "loss": 4.8573,
      "mean_token_accuracy": 0.2283207058906555,
      "num_tokens": 98305073.0,
      "step": 47360
    },
    {
      "entropy": 5.696509647369385,
      "epoch": 4.296534833091437,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003278948150174049,
      "loss": 4.9459,
      "mean_token_accuracy": 0.2187576100230217,
      "num_tokens": 98316008.0,
      "step": 47365
    },
    {
      "entropy": 5.651239633560181,
      "epoch": 4.296988388969521,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003278633648693906,
      "loss": 4.8309,
      "mean_token_accuracy": 0.22334285080432892,
      "num_tokens": 98327032.0,
      "step": 47370
    },
    {
      "entropy": 5.710759210586548,
      "epoch": 4.2974419448476056,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003278319136281484,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22898487001657486,
      "num_tokens": 98336085.0,
      "step": 47375
    },
    {
      "entropy": 5.698973941802978,
      "epoch": 4.297895500725689,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032780046129432885,
      "loss": 4.881,
      "mean_token_accuracy": 0.23336286693811417,
      "num_tokens": 98346180.0,
      "step": 47380
    },
    {
      "entropy": 5.6831762313842775,
      "epoch": 4.298349056603773,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032776900786858214,
      "loss": 4.8081,
      "mean_token_accuracy": 0.22490776032209397,
      "num_tokens": 98356001.0,
      "step": 47385
    },
    {
      "entropy": 5.735995531082153,
      "epoch": 4.2988026124818575,
      "grad_norm": 1.359375,
      "learning_rate": 0.000327737553351559,
      "loss": 4.9531,
      "mean_token_accuracy": 0.22149443626403809,
      "num_tokens": 98366744.0,
      "step": 47390
    },
    {
      "entropy": 5.636945629119873,
      "epoch": 4.299256168359942,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032770609774390985,
      "loss": 4.8184,
      "mean_token_accuracy": 0.2351287066936493,
      "num_tokens": 98377350.0,
      "step": 47395
    },
    {
      "entropy": 5.64010705947876,
      "epoch": 4.299709724238026,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003276746410462852,
      "loss": 4.849,
      "mean_token_accuracy": 0.23031098395586014,
      "num_tokens": 98387926.0,
      "step": 47400
    },
    {
      "entropy": 5.757110738754273,
      "epoch": 4.30016328011611,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032764318325933546,
      "loss": 4.9338,
      "mean_token_accuracy": 0.22473542392253876,
      "num_tokens": 98397117.0,
      "step": 47405
    },
    {
      "entropy": 5.752656745910644,
      "epoch": 4.3006168359941945,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003276117243837113,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22306031882762908,
      "num_tokens": 98407983.0,
      "step": 47410
    },
    {
      "entropy": 5.693604421615601,
      "epoch": 4.301070391872279,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003275802644200633,
      "loss": 4.908,
      "mean_token_accuracy": 0.2198258236050606,
      "num_tokens": 98419940.0,
      "step": 47415
    },
    {
      "entropy": 5.627506589889526,
      "epoch": 4.301523947750363,
      "grad_norm": 1.453125,
      "learning_rate": 0.000327548803369042,
      "loss": 4.8164,
      "mean_token_accuracy": 0.23481301814317704,
      "num_tokens": 98430487.0,
      "step": 47420
    },
    {
      "entropy": 5.664304542541504,
      "epoch": 4.301977503628447,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032751734123129814,
      "loss": 4.8511,
      "mean_token_accuracy": 0.231725250184536,
      "num_tokens": 98440231.0,
      "step": 47425
    },
    {
      "entropy": 5.7077515602111815,
      "epoch": 4.302431059506532,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003274858780074822,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2230115592479706,
      "num_tokens": 98449741.0,
      "step": 47430
    },
    {
      "entropy": 5.6739737033844,
      "epoch": 4.302884615384615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032745441369824485,
      "loss": 4.8016,
      "mean_token_accuracy": 0.2390468329191208,
      "num_tokens": 98461068.0,
      "step": 47435
    },
    {
      "entropy": 5.683884954452514,
      "epoch": 4.303338171262699,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032742294830423696,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2266627088189125,
      "num_tokens": 98472585.0,
      "step": 47440
    },
    {
      "entropy": 5.716053915023804,
      "epoch": 4.3037917271407835,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003273914818261092,
      "loss": 4.901,
      "mean_token_accuracy": 0.22745440453290938,
      "num_tokens": 98482874.0,
      "step": 47445
    },
    {
      "entropy": 5.701595592498779,
      "epoch": 4.304245283018868,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032736001426451216,
      "loss": 4.8576,
      "mean_token_accuracy": 0.22890810072422027,
      "num_tokens": 98492995.0,
      "step": 47450
    },
    {
      "entropy": 5.7431660175323485,
      "epoch": 4.304698838896952,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003273285456200966,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22010931819677354,
      "num_tokens": 98503965.0,
      "step": 47455
    },
    {
      "entropy": 5.687218046188354,
      "epoch": 4.305152394775036,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003272970758935135,
      "loss": 4.884,
      "mean_token_accuracy": 0.21982184201478958,
      "num_tokens": 98515584.0,
      "step": 47460
    },
    {
      "entropy": 5.712123966217041,
      "epoch": 4.305605950653121,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032726560508541356,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22632251381874086,
      "num_tokens": 98527134.0,
      "step": 47465
    },
    {
      "entropy": 5.72504072189331,
      "epoch": 4.306059506531205,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032723413319644756,
      "loss": 4.9028,
      "mean_token_accuracy": 0.22476772964000702,
      "num_tokens": 98537189.0,
      "step": 47470
    },
    {
      "entropy": 5.704430150985718,
      "epoch": 4.306513062409289,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003272026602272663,
      "loss": 4.8732,
      "mean_token_accuracy": 0.22671725898981093,
      "num_tokens": 98547803.0,
      "step": 47475
    },
    {
      "entropy": 5.656298732757568,
      "epoch": 4.306966618287373,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032717118617852086,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22278762012720107,
      "num_tokens": 98558332.0,
      "step": 47480
    },
    {
      "entropy": 5.723431205749511,
      "epoch": 4.307420174165458,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003271397110508619,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21766755729913712,
      "num_tokens": 98569046.0,
      "step": 47485
    },
    {
      "entropy": 5.6893257141113285,
      "epoch": 4.307873730043541,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003271082348449405,
      "loss": 4.8737,
      "mean_token_accuracy": 0.2260136842727661,
      "num_tokens": 98579358.0,
      "step": 47490
    },
    {
      "entropy": 5.681375789642334,
      "epoch": 4.308327285921625,
      "grad_norm": 1.25,
      "learning_rate": 0.0003270767575614074,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2232773020863533,
      "num_tokens": 98589224.0,
      "step": 47495
    },
    {
      "entropy": 5.725817108154297,
      "epoch": 4.3087808417997095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003270452792009138,
      "loss": 4.9483,
      "mean_token_accuracy": 0.2256453052163124,
      "num_tokens": 98598954.0,
      "step": 47500
    },
    {
      "entropy": 5.65271897315979,
      "epoch": 4.309234397677794,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032701379976411043,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2240918129682541,
      "num_tokens": 98608813.0,
      "step": 47505
    },
    {
      "entropy": 5.645434856414795,
      "epoch": 4.309687953555878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003269823192516486,
      "loss": 4.8496,
      "mean_token_accuracy": 0.2281813606619835,
      "num_tokens": 98618746.0,
      "step": 47510
    },
    {
      "entropy": 5.6741170406341555,
      "epoch": 4.310141509433962,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032695083766417913,
      "loss": 4.8491,
      "mean_token_accuracy": 0.23122671842575074,
      "num_tokens": 98628804.0,
      "step": 47515
    },
    {
      "entropy": 5.700046586990356,
      "epoch": 4.310595065312047,
      "grad_norm": 1.328125,
      "learning_rate": 0.000326919355002353,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22894047796726227,
      "num_tokens": 98640056.0,
      "step": 47520
    },
    {
      "entropy": 5.660945177078247,
      "epoch": 4.311048621190131,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032688787126682143,
      "loss": 4.8241,
      "mean_token_accuracy": 0.23013663440942764,
      "num_tokens": 98651042.0,
      "step": 47525
    },
    {
      "entropy": 5.677336072921753,
      "epoch": 4.311502177068215,
      "grad_norm": 1.25,
      "learning_rate": 0.0003268563864582354,
      "loss": 4.843,
      "mean_token_accuracy": 0.22753780335187912,
      "num_tokens": 98661405.0,
      "step": 47530
    },
    {
      "entropy": 5.718926525115966,
      "epoch": 4.3119557329462985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032682490057724616,
      "loss": 4.9407,
      "mean_token_accuracy": 0.22202659100294114,
      "num_tokens": 98672323.0,
      "step": 47535
    },
    {
      "entropy": 5.735607481002807,
      "epoch": 4.312409288824383,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0003267934136245047,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22524276077747346,
      "num_tokens": 98681431.0,
      "step": 47540
    },
    {
      "entropy": 5.571204090118409,
      "epoch": 4.312862844702467,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003267619256006623,
      "loss": 4.7991,
      "mean_token_accuracy": 0.24031769037246703,
      "num_tokens": 98692345.0,
      "step": 47545
    },
    {
      "entropy": 5.657980918884277,
      "epoch": 4.313316400580551,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003267304365063701,
      "loss": 4.9325,
      "mean_token_accuracy": 0.22305618226528168,
      "num_tokens": 98703425.0,
      "step": 47550
    },
    {
      "entropy": 5.708440589904785,
      "epoch": 4.313769956458636,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003266989463422793,
      "loss": 4.8825,
      "mean_token_accuracy": 0.22638604789972305,
      "num_tokens": 98714193.0,
      "step": 47555
    },
    {
      "entropy": 5.749283075332642,
      "epoch": 4.31422351233672,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032666745510904105,
      "loss": 4.8635,
      "mean_token_accuracy": 0.23029360175132751,
      "num_tokens": 98723550.0,
      "step": 47560
    },
    {
      "entropy": 5.64726881980896,
      "epoch": 4.314677068214804,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032663596280730673,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22294517308473588,
      "num_tokens": 98734075.0,
      "step": 47565
    },
    {
      "entropy": 5.657389545440674,
      "epoch": 4.315130624092888,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032660446943772755,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22976389676332473,
      "num_tokens": 98745447.0,
      "step": 47570
    },
    {
      "entropy": 5.707361030578613,
      "epoch": 4.315584179970973,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003265729750009547,
      "loss": 4.9439,
      "mean_token_accuracy": 0.22419597804546357,
      "num_tokens": 98756169.0,
      "step": 47575
    },
    {
      "entropy": 5.703860235214234,
      "epoch": 4.316037735849057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003265414794976397,
      "loss": 4.9223,
      "mean_token_accuracy": 0.22036930471658706,
      "num_tokens": 98767112.0,
      "step": 47580
    },
    {
      "entropy": 5.740336894989014,
      "epoch": 4.316491291727141,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003265099829284337,
      "loss": 4.9329,
      "mean_token_accuracy": 0.22451334446668625,
      "num_tokens": 98777533.0,
      "step": 47585
    },
    {
      "entropy": 5.7530943870544435,
      "epoch": 4.316944847605225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003264784852939881,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2183598831295967,
      "num_tokens": 98788101.0,
      "step": 47590
    },
    {
      "entropy": 5.773582029342651,
      "epoch": 4.317398403483309,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032644698659495437,
      "loss": 4.9171,
      "mean_token_accuracy": 0.22269751280546188,
      "num_tokens": 98799438.0,
      "step": 47595
    },
    {
      "entropy": 5.65052433013916,
      "epoch": 4.317851959361393,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032641548683198384,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2212439477443695,
      "num_tokens": 98810325.0,
      "step": 47600
    },
    {
      "entropy": 5.616851425170898,
      "epoch": 4.318305515239477,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003263839860057279,
      "loss": 4.8387,
      "mean_token_accuracy": 0.22813035249710084,
      "num_tokens": 98821897.0,
      "step": 47605
    },
    {
      "entropy": 5.665419340133667,
      "epoch": 4.318759071117562,
      "grad_norm": 1.390625,
      "learning_rate": 0.000326352484116838,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22758509069681168,
      "num_tokens": 98833338.0,
      "step": 47610
    },
    {
      "entropy": 5.723418760299682,
      "epoch": 4.319212626995646,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032632098116596576,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22723012864589692,
      "num_tokens": 98844308.0,
      "step": 47615
    },
    {
      "entropy": 5.703376817703247,
      "epoch": 4.31966618287373,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032628947715376247,
      "loss": 4.8804,
      "mean_token_accuracy": 0.23022021502256393,
      "num_tokens": 98854491.0,
      "step": 47620
    },
    {
      "entropy": 5.687318134307861,
      "epoch": 4.320119738751814,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003262579720808797,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21656761616468428,
      "num_tokens": 98864702.0,
      "step": 47625
    },
    {
      "entropy": 5.664725589752197,
      "epoch": 4.320573294629899,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032622646594796905,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22775158882141114,
      "num_tokens": 98874682.0,
      "step": 47630
    },
    {
      "entropy": 5.724459409713745,
      "epoch": 4.321026850507983,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003261949587556821,
      "loss": 4.9735,
      "mean_token_accuracy": 0.21476083844900132,
      "num_tokens": 98885130.0,
      "step": 47635
    },
    {
      "entropy": 5.675551033020019,
      "epoch": 4.321480406386067,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032616345050467016,
      "loss": 4.886,
      "mean_token_accuracy": 0.22595305889844894,
      "num_tokens": 98896398.0,
      "step": 47640
    },
    {
      "entropy": 5.672812175750733,
      "epoch": 4.321933962264151,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003261319411955852,
      "loss": 4.8487,
      "mean_token_accuracy": 0.23337628692388535,
      "num_tokens": 98906571.0,
      "step": 47645
    },
    {
      "entropy": 5.651860952377319,
      "epoch": 4.322387518142235,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003261004308290786,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2294571042060852,
      "num_tokens": 98916666.0,
      "step": 47650
    },
    {
      "entropy": 5.719958114624023,
      "epoch": 4.322841074020319,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003260689194058021,
      "loss": 4.8714,
      "mean_token_accuracy": 0.21983617842197417,
      "num_tokens": 98926599.0,
      "step": 47655
    },
    {
      "entropy": 5.635683155059814,
      "epoch": 4.323294629898403,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032603740692640733,
      "loss": 4.812,
      "mean_token_accuracy": 0.22517455071210862,
      "num_tokens": 98936512.0,
      "step": 47660
    },
    {
      "entropy": 5.709617042541504,
      "epoch": 4.323748185776488,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003260058933915459,
      "loss": 4.9934,
      "mean_token_accuracy": 0.2098014920949936,
      "num_tokens": 98946604.0,
      "step": 47665
    },
    {
      "entropy": 5.752908849716187,
      "epoch": 4.324201741654572,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003259743788018697,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2181127443909645,
      "num_tokens": 98956257.0,
      "step": 47670
    },
    {
      "entropy": 5.7294378757476805,
      "epoch": 4.324655297532656,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032594286315803026,
      "loss": 4.932,
      "mean_token_accuracy": 0.2204469084739685,
      "num_tokens": 98966346.0,
      "step": 47675
    },
    {
      "entropy": 5.674243783950805,
      "epoch": 4.3251088534107405,
      "grad_norm": 1.25,
      "learning_rate": 0.0003259113464606795,
      "loss": 4.8305,
      "mean_token_accuracy": 0.22845391780138016,
      "num_tokens": 98977410.0,
      "step": 47680
    },
    {
      "entropy": 5.723272466659546,
      "epoch": 4.325562409288825,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003258798287104691,
      "loss": 4.9252,
      "mean_token_accuracy": 0.22434550374746323,
      "num_tokens": 98987488.0,
      "step": 47685
    },
    {
      "entropy": 5.661989068984985,
      "epoch": 4.326015965166908,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032584830990805094,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23168001770973207,
      "num_tokens": 98996792.0,
      "step": 47690
    },
    {
      "entropy": 5.649671030044556,
      "epoch": 4.326469521044992,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032581679005407667,
      "loss": 4.8039,
      "mean_token_accuracy": 0.22948745042085647,
      "num_tokens": 99007425.0,
      "step": 47695
    },
    {
      "entropy": 5.649986696243286,
      "epoch": 4.326923076923077,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003257852691491983,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2244899556040764,
      "num_tokens": 99018275.0,
      "step": 47700
    },
    {
      "entropy": 5.712569665908814,
      "epoch": 4.327376632801161,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032575374719406755,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22437282502651215,
      "num_tokens": 99028853.0,
      "step": 47705
    },
    {
      "entropy": 5.714393901824951,
      "epoch": 4.327830188679245,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032572222418933644,
      "loss": 4.8995,
      "mean_token_accuracy": 0.21886498630046844,
      "num_tokens": 99039423.0,
      "step": 47710
    },
    {
      "entropy": 5.701907396316528,
      "epoch": 4.3282837445573294,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003256907001356568,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21604047417640687,
      "num_tokens": 99050041.0,
      "step": 47715
    },
    {
      "entropy": 5.687766027450562,
      "epoch": 4.328737300435414,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003256591750336806,
      "loss": 4.9162,
      "mean_token_accuracy": 0.22190199047327042,
      "num_tokens": 99062050.0,
      "step": 47720
    },
    {
      "entropy": 5.602911329269409,
      "epoch": 4.329190856313498,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003256276488840596,
      "loss": 4.783,
      "mean_token_accuracy": 0.22968636453151703,
      "num_tokens": 99072337.0,
      "step": 47725
    },
    {
      "entropy": 5.582111358642578,
      "epoch": 4.329644412191582,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032559612168744613,
      "loss": 4.7989,
      "mean_token_accuracy": 0.22836442440748214,
      "num_tokens": 99082468.0,
      "step": 47730
    },
    {
      "entropy": 5.666590833663941,
      "epoch": 4.3300979680696665,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003255645934444918,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22091216295957566,
      "num_tokens": 99091757.0,
      "step": 47735
    },
    {
      "entropy": 5.728024005889893,
      "epoch": 4.330551523947751,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003255330641558489,
      "loss": 4.9445,
      "mean_token_accuracy": 0.2189284458756447,
      "num_tokens": 99103192.0,
      "step": 47740
    },
    {
      "entropy": 5.626647424697876,
      "epoch": 4.331005079825834,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003255015338221693,
      "loss": 4.7799,
      "mean_token_accuracy": 0.2293917328119278,
      "num_tokens": 99112660.0,
      "step": 47745
    },
    {
      "entropy": 5.738275337219238,
      "epoch": 4.331458635703918,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032547000244410515,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2241392746567726,
      "num_tokens": 99123867.0,
      "step": 47750
    },
    {
      "entropy": 5.692056322097779,
      "epoch": 4.331912191582003,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032543847002230845,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22272821962833406,
      "num_tokens": 99134445.0,
      "step": 47755
    },
    {
      "entropy": 5.657172155380249,
      "epoch": 4.332365747460087,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003254069365574313,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22691401988267898,
      "num_tokens": 99143483.0,
      "step": 47760
    },
    {
      "entropy": 5.6528760433197025,
      "epoch": 4.332819303338171,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003253754020501259,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23200729936361314,
      "num_tokens": 99154514.0,
      "step": 47765
    },
    {
      "entropy": 5.661369800567627,
      "epoch": 4.3332728592162555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032534386650104444,
      "loss": 4.9212,
      "mean_token_accuracy": 0.2137492299079895,
      "num_tokens": 99165969.0,
      "step": 47770
    },
    {
      "entropy": 5.593125247955323,
      "epoch": 4.33372641509434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003253123299108388,
      "loss": 4.7876,
      "mean_token_accuracy": 0.23204695880413057,
      "num_tokens": 99176595.0,
      "step": 47775
    },
    {
      "entropy": 5.700669193267823,
      "epoch": 4.334179970972424,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003252807922801615,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22432172447443008,
      "num_tokens": 99186916.0,
      "step": 47780
    },
    {
      "entropy": 5.731165838241577,
      "epoch": 4.334633526850508,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003252492536096646,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2133137121796608,
      "num_tokens": 99197892.0,
      "step": 47785
    },
    {
      "entropy": 5.743923377990723,
      "epoch": 4.3350870827285926,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003252177139000003,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22680503576993943,
      "num_tokens": 99208373.0,
      "step": 47790
    },
    {
      "entropy": 5.707012367248535,
      "epoch": 4.335540638606676,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003251861731518209,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22761063128709794,
      "num_tokens": 99218788.0,
      "step": 47795
    },
    {
      "entropy": 5.770860385894776,
      "epoch": 4.33599419448476,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032515463136577864,
      "loss": 4.9438,
      "mean_token_accuracy": 0.22563234716653824,
      "num_tokens": 99228653.0,
      "step": 47800
    },
    {
      "entropy": 5.671164798736572,
      "epoch": 4.3364477503628445,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032512308854252577,
      "loss": 4.804,
      "mean_token_accuracy": 0.23209425061941147,
      "num_tokens": 99239925.0,
      "step": 47805
    },
    {
      "entropy": 5.632341098785401,
      "epoch": 4.336901306240929,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003250915446827147,
      "loss": 4.7608,
      "mean_token_accuracy": 0.23194602727890015,
      "num_tokens": 99250091.0,
      "step": 47810
    },
    {
      "entropy": 5.648854684829712,
      "epoch": 4.337354862119013,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003250599997869977,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21233980655670165,
      "num_tokens": 99260799.0,
      "step": 47815
    },
    {
      "entropy": 5.630479717254639,
      "epoch": 4.337808417997097,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003250284538560272,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23726825714111327,
      "num_tokens": 99270338.0,
      "step": 47820
    },
    {
      "entropy": 5.674000692367554,
      "epoch": 4.3382619738751815,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003249969068904554,
      "loss": 4.8625,
      "mean_token_accuracy": 0.2235625520348549,
      "num_tokens": 99280419.0,
      "step": 47825
    },
    {
      "entropy": 5.737470293045044,
      "epoch": 4.338715529753266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032496535889093496,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2193329319357872,
      "num_tokens": 99291590.0,
      "step": 47830
    },
    {
      "entropy": 5.705243444442749,
      "epoch": 4.33916908563135,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032493380985811814,
      "loss": 4.9189,
      "mean_token_accuracy": 0.22330475896596907,
      "num_tokens": 99301987.0,
      "step": 47835
    },
    {
      "entropy": 5.733073616027832,
      "epoch": 4.339622641509434,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032490225979265734,
      "loss": 4.9118,
      "mean_token_accuracy": 0.219923534989357,
      "num_tokens": 99312231.0,
      "step": 47840
    },
    {
      "entropy": 5.704366683959961,
      "epoch": 4.340076197387518,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032487070869520517,
      "loss": 4.83,
      "mean_token_accuracy": 0.22698791325092316,
      "num_tokens": 99322784.0,
      "step": 47845
    },
    {
      "entropy": 5.729310941696167,
      "epoch": 4.340529753265602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000324839156566414,
      "loss": 4.9327,
      "mean_token_accuracy": 0.22054585665464402,
      "num_tokens": 99333989.0,
      "step": 47850
    },
    {
      "entropy": 5.633415174484253,
      "epoch": 4.340983309143686,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003248076034069364,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22615868598222733,
      "num_tokens": 99345034.0,
      "step": 47855
    },
    {
      "entropy": 5.639495372772217,
      "epoch": 4.3414368650217705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003247760492174248,
      "loss": 4.8084,
      "mean_token_accuracy": 0.22888606190681457,
      "num_tokens": 99356062.0,
      "step": 47860
    },
    {
      "entropy": 5.6990961074829105,
      "epoch": 4.341890420899855,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003247444939985319,
      "loss": 4.8434,
      "mean_token_accuracy": 0.23153524100780487,
      "num_tokens": 99365315.0,
      "step": 47865
    },
    {
      "entropy": 5.78768482208252,
      "epoch": 4.342343976777939,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032471293775091013,
      "loss": 4.8902,
      "mean_token_accuracy": 0.23138579428195954,
      "num_tokens": 99374897.0,
      "step": 47870
    },
    {
      "entropy": 5.655116558074951,
      "epoch": 4.342797532656023,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032468138047521216,
      "loss": 4.8303,
      "mean_token_accuracy": 0.2294129639863968,
      "num_tokens": 99383624.0,
      "step": 47875
    },
    {
      "entropy": 5.636184978485107,
      "epoch": 4.343251088534108,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003246498221720906,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22156410813331603,
      "num_tokens": 99393791.0,
      "step": 47880
    },
    {
      "entropy": 5.67339301109314,
      "epoch": 4.343704644412192,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000324618262842198,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2275726854801178,
      "num_tokens": 99403687.0,
      "step": 47885
    },
    {
      "entropy": 5.726601314544678,
      "epoch": 4.344158200290276,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032458670248618714,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22387943118810655,
      "num_tokens": 99413411.0,
      "step": 47890
    },
    {
      "entropy": 5.738475036621094,
      "epoch": 4.34461175616836,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003245551411047106,
      "loss": 4.9219,
      "mean_token_accuracy": 0.22198912352323533,
      "num_tokens": 99425563.0,
      "step": 47895
    },
    {
      "entropy": 5.657603549957275,
      "epoch": 4.345065312046444,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032452357869842117,
      "loss": 4.8543,
      "mean_token_accuracy": 0.23643344938755034,
      "num_tokens": 99436297.0,
      "step": 47900
    },
    {
      "entropy": 5.69480676651001,
      "epoch": 4.345518867924528,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003244920152679715,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22540836930274963,
      "num_tokens": 99445745.0,
      "step": 47905
    },
    {
      "entropy": 5.641480398178101,
      "epoch": 4.345972423802612,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032446045081401426,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2231060892343521,
      "num_tokens": 99456915.0,
      "step": 47910
    },
    {
      "entropy": 5.746298313140869,
      "epoch": 4.3464259796806965,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003244288853372024,
      "loss": 4.8887,
      "mean_token_accuracy": 0.21986974477767945,
      "num_tokens": 99466778.0,
      "step": 47915
    },
    {
      "entropy": 5.721002054214478,
      "epoch": 4.346879535558781,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003243973188381885,
      "loss": 4.8496,
      "mean_token_accuracy": 0.22375288605690002,
      "num_tokens": 99475984.0,
      "step": 47920
    },
    {
      "entropy": 5.611246776580811,
      "epoch": 4.347333091436865,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032436575131762553,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2224792718887329,
      "num_tokens": 99486238.0,
      "step": 47925
    },
    {
      "entropy": 5.780407857894898,
      "epoch": 4.347786647314949,
      "grad_norm": 1.53125,
      "learning_rate": 0.00032433418277616623,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22479342371225358,
      "num_tokens": 99497806.0,
      "step": 47930
    },
    {
      "entropy": 5.687092351913452,
      "epoch": 4.348240203193034,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003243026132144635,
      "loss": 4.8745,
      "mean_token_accuracy": 0.21963958740234374,
      "num_tokens": 99507214.0,
      "step": 47935
    },
    {
      "entropy": 5.733822011947632,
      "epoch": 4.348693759071118,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003242710426331701,
      "loss": 4.9186,
      "mean_token_accuracy": 0.2217227354645729,
      "num_tokens": 99517025.0,
      "step": 47940
    },
    {
      "entropy": 5.739341449737549,
      "epoch": 4.349147314949202,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032423947103293893,
      "loss": 4.914,
      "mean_token_accuracy": 0.2255868822336197,
      "num_tokens": 99526972.0,
      "step": 47945
    },
    {
      "entropy": 5.721365308761596,
      "epoch": 4.3496008708272855,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032420789841442306,
      "loss": 4.9098,
      "mean_token_accuracy": 0.21642527878284454,
      "num_tokens": 99538583.0,
      "step": 47950
    },
    {
      "entropy": 5.613011074066162,
      "epoch": 4.35005442670537,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032417632477827526,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22948300391435622,
      "num_tokens": 99549178.0,
      "step": 47955
    },
    {
      "entropy": 5.689499616622925,
      "epoch": 4.350507982583454,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032414475012514847,
      "loss": 4.852,
      "mean_token_accuracy": 0.23075646460056304,
      "num_tokens": 99558491.0,
      "step": 47960
    },
    {
      "entropy": 5.684060573577881,
      "epoch": 4.350961538461538,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003241131744556958,
      "loss": 4.918,
      "mean_token_accuracy": 0.21956392973661423,
      "num_tokens": 99568711.0,
      "step": 47965
    },
    {
      "entropy": 5.658336591720581,
      "epoch": 4.351415094339623,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003240815977705702,
      "loss": 4.882,
      "mean_token_accuracy": 0.22137337177991867,
      "num_tokens": 99578329.0,
      "step": 47970
    },
    {
      "entropy": 5.638819742202759,
      "epoch": 4.351868650217707,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032405002007042453,
      "loss": 4.7342,
      "mean_token_accuracy": 0.2339604079723358,
      "num_tokens": 99588646.0,
      "step": 47975
    },
    {
      "entropy": 5.726035690307617,
      "epoch": 4.352322206095791,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032401844135591213,
      "loss": 4.9057,
      "mean_token_accuracy": 0.22094153463840485,
      "num_tokens": 99598780.0,
      "step": 47980
    },
    {
      "entropy": 5.709539604187012,
      "epoch": 4.352775761973875,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003239868616276857,
      "loss": 4.8767,
      "mean_token_accuracy": 0.22480917423963548,
      "num_tokens": 99609528.0,
      "step": 47985
    },
    {
      "entropy": 5.613860988616944,
      "epoch": 4.35322931785196,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003239552808863986,
      "loss": 4.7971,
      "mean_token_accuracy": 0.22886189371347426,
      "num_tokens": 99619928.0,
      "step": 47990
    },
    {
      "entropy": 5.605508708953858,
      "epoch": 4.353682873730044,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032392369913270376,
      "loss": 4.7713,
      "mean_token_accuracy": 0.2326163098216057,
      "num_tokens": 99630135.0,
      "step": 47995
    },
    {
      "entropy": 5.713461303710938,
      "epoch": 4.354136429608127,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003238921163672544,
      "loss": 4.9879,
      "mean_token_accuracy": 0.21180067509412764,
      "num_tokens": 99641557.0,
      "step": 48000
    },
    {
      "epoch": 4.354136429608127,
      "eval_entropy": 5.389565399976877,
      "eval_loss": 4.96226692199707,
      "eval_mean_token_accuracy": 0.22942079486687797,
      "eval_num_tokens": 99641557.0,
      "eval_runtime": 20.6495,
      "eval_samples_per_second": 1712.393,
      "eval_steps_per_second": 214.049,
      "step": 48000
    },
    {
      "entropy": 5.706479501724243,
      "epoch": 4.354589985486212,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003238605325907035,
      "loss": 4.8761,
      "mean_token_accuracy": 0.22449006140232086,
      "num_tokens": 99651244.0,
      "step": 48005
    },
    {
      "entropy": 5.718190383911133,
      "epoch": 4.355043541364296,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032382894780370446,
      "loss": 4.8414,
      "mean_token_accuracy": 0.22278761267662048,
      "num_tokens": 99662317.0,
      "step": 48010
    },
    {
      "entropy": 5.656660795211792,
      "epoch": 4.35549709724238,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032379736200691036,
      "loss": 4.8153,
      "mean_token_accuracy": 0.22842392921447754,
      "num_tokens": 99671624.0,
      "step": 48015
    },
    {
      "entropy": 5.7166101932525635,
      "epoch": 4.355950653120464,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032376577520097427,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21680358052253723,
      "num_tokens": 99682076.0,
      "step": 48020
    },
    {
      "entropy": 5.6633082866668705,
      "epoch": 4.356404208998549,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032373418738654953,
      "loss": 4.861,
      "mean_token_accuracy": 0.2296249523758888,
      "num_tokens": 99692076.0,
      "step": 48025
    },
    {
      "entropy": 5.679394054412842,
      "epoch": 4.356857764876633,
      "grad_norm": 1.375,
      "learning_rate": 0.0003237025985642895,
      "loss": 4.843,
      "mean_token_accuracy": 0.22733064442873002,
      "num_tokens": 99701455.0,
      "step": 48030
    },
    {
      "entropy": 5.713113117218017,
      "epoch": 4.357311320754717,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032367100873484725,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22275718599557875,
      "num_tokens": 99712034.0,
      "step": 48035
    },
    {
      "entropy": 5.67870078086853,
      "epoch": 4.357764876632801,
      "grad_norm": 1.234375,
      "learning_rate": 0.000323639417898876,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2203628733754158,
      "num_tokens": 99722965.0,
      "step": 48040
    },
    {
      "entropy": 5.578767347335815,
      "epoch": 4.358218432510886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003236078260570293,
      "loss": 4.7943,
      "mean_token_accuracy": 0.22731985449790953,
      "num_tokens": 99732847.0,
      "step": 48045
    },
    {
      "entropy": 5.638531684875488,
      "epoch": 4.35867198838897,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003235762332099604,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23807456344366074,
      "num_tokens": 99743639.0,
      "step": 48050
    },
    {
      "entropy": 5.704068803787232,
      "epoch": 4.359125544267053,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003235446393583227,
      "loss": 4.7924,
      "mean_token_accuracy": 0.2365147739648819,
      "num_tokens": 99753374.0,
      "step": 48055
    },
    {
      "entropy": 5.654393100738526,
      "epoch": 4.359579100145138,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003235130445027693,
      "loss": 4.8309,
      "mean_token_accuracy": 0.227109132707119,
      "num_tokens": 99763361.0,
      "step": 48060
    },
    {
      "entropy": 5.722085189819336,
      "epoch": 4.360032656023222,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032348144864395383,
      "loss": 4.9377,
      "mean_token_accuracy": 0.22294931560754777,
      "num_tokens": 99773813.0,
      "step": 48065
    },
    {
      "entropy": 5.702458047866822,
      "epoch": 4.360486211901306,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003234498517825297,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22262655943632126,
      "num_tokens": 99784812.0,
      "step": 48070
    },
    {
      "entropy": 5.679151201248169,
      "epoch": 4.36093976777939,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032341825391915026,
      "loss": 4.854,
      "mean_token_accuracy": 0.22705024778842925,
      "num_tokens": 99795197.0,
      "step": 48075
    },
    {
      "entropy": 5.598389673233032,
      "epoch": 4.361393323657475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000323386655054469,
      "loss": 4.8918,
      "mean_token_accuracy": 0.22666011452674867,
      "num_tokens": 99805883.0,
      "step": 48080
    },
    {
      "entropy": 5.6966533184051515,
      "epoch": 4.361846879535559,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003233550551891394,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2259916767477989,
      "num_tokens": 99816940.0,
      "step": 48085
    },
    {
      "entropy": 5.692325305938721,
      "epoch": 4.362300435413643,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032332345432381493,
      "loss": 4.8591,
      "mean_token_accuracy": 0.22401827871799468,
      "num_tokens": 99827926.0,
      "step": 48090
    },
    {
      "entropy": 5.622637939453125,
      "epoch": 4.3627539912917275,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003232918524591491,
      "loss": 4.8168,
      "mean_token_accuracy": 0.23099080324172974,
      "num_tokens": 99837900.0,
      "step": 48095
    },
    {
      "entropy": 5.715040254592895,
      "epoch": 4.363207547169811,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003232602495957955,
      "loss": 4.919,
      "mean_token_accuracy": 0.22066500335931777,
      "num_tokens": 99848655.0,
      "step": 48100
    },
    {
      "entropy": 5.6466954231262205,
      "epoch": 4.363661103047895,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032322864573440757,
      "loss": 4.8326,
      "mean_token_accuracy": 0.2276299074292183,
      "num_tokens": 99859715.0,
      "step": 48105
    },
    {
      "entropy": 5.686735820770264,
      "epoch": 4.364114658925979,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003231970408756389,
      "loss": 4.8161,
      "mean_token_accuracy": 0.2358422875404358,
      "num_tokens": 99870175.0,
      "step": 48110
    },
    {
      "entropy": 5.749360513687134,
      "epoch": 4.364568214804064,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003231654350201432,
      "loss": 4.9045,
      "mean_token_accuracy": 0.2202591672539711,
      "num_tokens": 99879794.0,
      "step": 48115
    },
    {
      "entropy": 5.698179101943969,
      "epoch": 4.365021770682148,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003231338281685741,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2304138123989105,
      "num_tokens": 99889418.0,
      "step": 48120
    },
    {
      "entropy": 5.723806953430175,
      "epoch": 4.365475326560232,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003231022203215851,
      "loss": 4.9019,
      "mean_token_accuracy": 0.22079143226146697,
      "num_tokens": 99898741.0,
      "step": 48125
    },
    {
      "entropy": 5.63426775932312,
      "epoch": 4.3659288824383164,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032307061147982994,
      "loss": 4.8274,
      "mean_token_accuracy": 0.22336673736572266,
      "num_tokens": 99909167.0,
      "step": 48130
    },
    {
      "entropy": 5.6447359085083,
      "epoch": 4.366382438316401,
      "grad_norm": 1.375,
      "learning_rate": 0.0003230390016439623,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22817673683166503,
      "num_tokens": 99918786.0,
      "step": 48135
    },
    {
      "entropy": 5.650050115585327,
      "epoch": 4.366835994194485,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003230073908146359,
      "loss": 4.9199,
      "mean_token_accuracy": 0.2175696849822998,
      "num_tokens": 99928758.0,
      "step": 48140
    },
    {
      "entropy": 5.667374515533448,
      "epoch": 4.367289550072569,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003229757789925044,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22104677855968474,
      "num_tokens": 99939721.0,
      "step": 48145
    },
    {
      "entropy": 5.653213119506836,
      "epoch": 4.3677431059506535,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003229441661782215,
      "loss": 4.8041,
      "mean_token_accuracy": 0.22633409947156907,
      "num_tokens": 99949308.0,
      "step": 48150
    },
    {
      "entropy": 5.67603120803833,
      "epoch": 4.368196661828737,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032291255237244104,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22738104909658433,
      "num_tokens": 99960284.0,
      "step": 48155
    },
    {
      "entropy": 5.633839750289917,
      "epoch": 4.368650217706821,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032288093757581684,
      "loss": 4.7881,
      "mean_token_accuracy": 0.22759477645158768,
      "num_tokens": 99970701.0,
      "step": 48160
    },
    {
      "entropy": 5.71251950263977,
      "epoch": 4.369103773584905,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003228493217890026,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22423505038022995,
      "num_tokens": 99981362.0,
      "step": 48165
    },
    {
      "entropy": 5.689724969863891,
      "epoch": 4.36955732946299,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003228177050126523,
      "loss": 4.8708,
      "mean_token_accuracy": 0.23222921341657637,
      "num_tokens": 99991383.0,
      "step": 48170
    },
    {
      "entropy": 5.657776975631714,
      "epoch": 4.370010885341074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032278608724741956,
      "loss": 4.869,
      "mean_token_accuracy": 0.22141768485307695,
      "num_tokens": 100001176.0,
      "step": 48175
    },
    {
      "entropy": 5.644967794418335,
      "epoch": 4.370464441219158,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003227544684939584,
      "loss": 4.8254,
      "mean_token_accuracy": 0.2374468520283699,
      "num_tokens": 100011467.0,
      "step": 48180
    },
    {
      "entropy": 5.672161102294922,
      "epoch": 4.3709179970972425,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003227228487529226,
      "loss": 4.8496,
      "mean_token_accuracy": 0.22693271189928055,
      "num_tokens": 100020976.0,
      "step": 48185
    },
    {
      "entropy": 5.631677961349487,
      "epoch": 4.371371552975327,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003226912280249662,
      "loss": 4.8467,
      "mean_token_accuracy": 0.2285946175456047,
      "num_tokens": 100032250.0,
      "step": 48190
    },
    {
      "entropy": 5.718546628952026,
      "epoch": 4.371825108853411,
      "grad_norm": 1.28125,
      "learning_rate": 0.000322659606310743,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22566657662391662,
      "num_tokens": 100043383.0,
      "step": 48195
    },
    {
      "entropy": 5.662776184082031,
      "epoch": 4.372278664731495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032262798361090705,
      "loss": 4.7833,
      "mean_token_accuracy": 0.2329989418387413,
      "num_tokens": 100053396.0,
      "step": 48200
    },
    {
      "entropy": 5.65639443397522,
      "epoch": 4.37273222060958,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003225963599261123,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22864966839551926,
      "num_tokens": 100064159.0,
      "step": 48205
    },
    {
      "entropy": 5.634120464324951,
      "epoch": 4.373185776487663,
      "grad_norm": 1.375,
      "learning_rate": 0.00032256473525701257,
      "loss": 4.816,
      "mean_token_accuracy": 0.23759115487337112,
      "num_tokens": 100074722.0,
      "step": 48210
    },
    {
      "entropy": 5.63558669090271,
      "epoch": 4.373639332365747,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032253310960426215,
      "loss": 4.8344,
      "mean_token_accuracy": 0.23577409088611603,
      "num_tokens": 100085067.0,
      "step": 48215
    },
    {
      "entropy": 5.691960096359253,
      "epoch": 4.3740928882438315,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032250148296851484,
      "loss": 4.8931,
      "mean_token_accuracy": 0.22797372937202454,
      "num_tokens": 100095806.0,
      "step": 48220
    },
    {
      "entropy": 5.758006954193116,
      "epoch": 4.374546444121916,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003224698553504247,
      "loss": 4.9494,
      "mean_token_accuracy": 0.2184480458498001,
      "num_tokens": 100106069.0,
      "step": 48225
    },
    {
      "entropy": 5.759176301956177,
      "epoch": 4.375,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032243822675064585,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21750841289758682,
      "num_tokens": 100117136.0,
      "step": 48230
    },
    {
      "entropy": 5.697377872467041,
      "epoch": 4.375453555878084,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003224065971698324,
      "loss": 4.8863,
      "mean_token_accuracy": 0.20956885516643525,
      "num_tokens": 100126757.0,
      "step": 48235
    },
    {
      "entropy": 5.8024101734161375,
      "epoch": 4.3759071117561685,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003223749666086385,
      "loss": 4.9943,
      "mean_token_accuracy": 0.22345636487007142,
      "num_tokens": 100138431.0,
      "step": 48240
    },
    {
      "entropy": 5.719567060470581,
      "epoch": 4.376360667634253,
      "grad_norm": 1.25,
      "learning_rate": 0.0003223433350677182,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22122462838888168,
      "num_tokens": 100148919.0,
      "step": 48245
    },
    {
      "entropy": 5.764693832397461,
      "epoch": 4.376814223512337,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003223117025477256,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2194630831480026,
      "num_tokens": 100159175.0,
      "step": 48250
    },
    {
      "entropy": 5.695107793807983,
      "epoch": 4.37726777939042,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003222800690493151,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22275409251451492,
      "num_tokens": 100169451.0,
      "step": 48255
    },
    {
      "entropy": 5.705862712860108,
      "epoch": 4.377721335268505,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003222484345731405,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2258662074804306,
      "num_tokens": 100180077.0,
      "step": 48260
    },
    {
      "entropy": 5.687623977661133,
      "epoch": 4.378174891146589,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032221679911985634,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2260860487818718,
      "num_tokens": 100190175.0,
      "step": 48265
    },
    {
      "entropy": 5.6554161548614506,
      "epoch": 4.378628447024673,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032218516269011676,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23445080518722533,
      "num_tokens": 100200285.0,
      "step": 48270
    },
    {
      "entropy": 5.557846784591675,
      "epoch": 4.3790820029027575,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032215352528457597,
      "loss": 4.6988,
      "mean_token_accuracy": 0.23558185547590255,
      "num_tokens": 100210685.0,
      "step": 48275
    },
    {
      "entropy": 5.641123247146607,
      "epoch": 4.379535558780842,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032212188690388825,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2335732564330101,
      "num_tokens": 100220486.0,
      "step": 48280
    },
    {
      "entropy": 5.644009685516357,
      "epoch": 4.379989114658926,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032209024754870796,
      "loss": 4.7871,
      "mean_token_accuracy": 0.22393751442432402,
      "num_tokens": 100229007.0,
      "step": 48285
    },
    {
      "entropy": 5.650397920608521,
      "epoch": 4.38044267053701,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003220586072196893,
      "loss": 4.8115,
      "mean_token_accuracy": 0.22585691809654235,
      "num_tokens": 100240363.0,
      "step": 48290
    },
    {
      "entropy": 5.60833511352539,
      "epoch": 4.380896226415095,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032202696591748667,
      "loss": 4.8185,
      "mean_token_accuracy": 0.23000694960355758,
      "num_tokens": 100251193.0,
      "step": 48295
    },
    {
      "entropy": 5.662957572937012,
      "epoch": 4.381349782293179,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032199532364275436,
      "loss": 4.8455,
      "mean_token_accuracy": 0.2279832273721695,
      "num_tokens": 100261248.0,
      "step": 48300
    },
    {
      "entropy": 5.670694875717163,
      "epoch": 4.381803338171263,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003219636803961468,
      "loss": 4.8412,
      "mean_token_accuracy": 0.22127762138843537,
      "num_tokens": 100271174.0,
      "step": 48305
    },
    {
      "entropy": 5.69003701210022,
      "epoch": 4.3822568940493465,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032193203617831835,
      "loss": 4.9055,
      "mean_token_accuracy": 0.22600615322589873,
      "num_tokens": 100282428.0,
      "step": 48310
    },
    {
      "entropy": 5.706641292572021,
      "epoch": 4.382710449927431,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032190039098992345,
      "loss": 4.9048,
      "mean_token_accuracy": 0.221664160490036,
      "num_tokens": 100292559.0,
      "step": 48315
    },
    {
      "entropy": 5.647461557388306,
      "epoch": 4.383164005805515,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003218687448316166,
      "loss": 4.7829,
      "mean_token_accuracy": 0.23104466646909713,
      "num_tokens": 100303956.0,
      "step": 48320
    },
    {
      "entropy": 5.624965572357178,
      "epoch": 4.383617561683599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032183709770405204,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2263238474726677,
      "num_tokens": 100315281.0,
      "step": 48325
    },
    {
      "entropy": 5.694290256500244,
      "epoch": 4.3840711175616836,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003218054496078844,
      "loss": 4.895,
      "mean_token_accuracy": 0.22328037172555923,
      "num_tokens": 100326761.0,
      "step": 48330
    },
    {
      "entropy": 5.705591201782227,
      "epoch": 4.384524673439768,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032177380054376813,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22472509890794753,
      "num_tokens": 100337219.0,
      "step": 48335
    },
    {
      "entropy": 5.70663423538208,
      "epoch": 4.384978229317852,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032174215051235776,
      "loss": 4.8563,
      "mean_token_accuracy": 0.23288750350475312,
      "num_tokens": 100347135.0,
      "step": 48340
    },
    {
      "entropy": 5.582832622528076,
      "epoch": 4.385431785195936,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032171049951430775,
      "loss": 4.7247,
      "mean_token_accuracy": 0.2378337025642395,
      "num_tokens": 100357476.0,
      "step": 48345
    },
    {
      "entropy": 5.671077299118042,
      "epoch": 4.385885341074021,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032167884755027285,
      "loss": 4.825,
      "mean_token_accuracy": 0.22346246838569642,
      "num_tokens": 100367860.0,
      "step": 48350
    },
    {
      "entropy": 5.693797159194946,
      "epoch": 4.386338896952105,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003216471946209074,
      "loss": 4.8321,
      "mean_token_accuracy": 0.22731355726718902,
      "num_tokens": 100378613.0,
      "step": 48355
    },
    {
      "entropy": 5.677419567108155,
      "epoch": 4.386792452830189,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000321615540726866,
      "loss": 4.8891,
      "mean_token_accuracy": 0.22598228454589844,
      "num_tokens": 100389499.0,
      "step": 48360
    },
    {
      "entropy": 5.716799688339234,
      "epoch": 4.3872460087082725,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032158388586880346,
      "loss": 4.9056,
      "mean_token_accuracy": 0.2266518771648407,
      "num_tokens": 100400538.0,
      "step": 48365
    },
    {
      "entropy": 5.673984336853027,
      "epoch": 4.387699564586357,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032155223004737423,
      "loss": 4.9052,
      "mean_token_accuracy": 0.2332645207643509,
      "num_tokens": 100411528.0,
      "step": 48370
    },
    {
      "entropy": 5.684253263473511,
      "epoch": 4.388153120464441,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032152057326323305,
      "loss": 4.8668,
      "mean_token_accuracy": 0.21794661283493041,
      "num_tokens": 100421349.0,
      "step": 48375
    },
    {
      "entropy": 5.6323692321777346,
      "epoch": 4.388606676342525,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032148891551703445,
      "loss": 4.795,
      "mean_token_accuracy": 0.23052536100149154,
      "num_tokens": 100431132.0,
      "step": 48380
    },
    {
      "entropy": 5.6525288105010985,
      "epoch": 4.38906023222061,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003214572568094334,
      "loss": 4.8717,
      "mean_token_accuracy": 0.22933972924947738,
      "num_tokens": 100441436.0,
      "step": 48385
    },
    {
      "entropy": 5.6054487228393555,
      "epoch": 4.389513788098694,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003214255971410843,
      "loss": 4.734,
      "mean_token_accuracy": 0.2320557177066803,
      "num_tokens": 100451711.0,
      "step": 48390
    },
    {
      "entropy": 5.777924108505249,
      "epoch": 4.389967343976778,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000321393936512642,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21756030917167662,
      "num_tokens": 100461177.0,
      "step": 48395
    },
    {
      "entropy": 5.642672348022461,
      "epoch": 4.390420899854862,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032136227492476145,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22615667581558227,
      "num_tokens": 100471391.0,
      "step": 48400
    },
    {
      "entropy": 5.722684383392334,
      "epoch": 4.390874455732947,
      "grad_norm": 1.25,
      "learning_rate": 0.0003213306123780971,
      "loss": 4.8468,
      "mean_token_accuracy": 0.22554196715354918,
      "num_tokens": 100480747.0,
      "step": 48405
    },
    {
      "entropy": 5.685224342346191,
      "epoch": 4.39132801161103,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032129894887330396,
      "loss": 4.8829,
      "mean_token_accuracy": 0.23193800151348115,
      "num_tokens": 100491451.0,
      "step": 48410
    },
    {
      "entropy": 5.7391808986663815,
      "epoch": 4.391781567489114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003212672844110367,
      "loss": 4.8971,
      "mean_token_accuracy": 0.225168676674366,
      "num_tokens": 100501801.0,
      "step": 48415
    },
    {
      "entropy": 5.663092184066772,
      "epoch": 4.392235123367199,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032123561899195026,
      "loss": 4.8742,
      "mean_token_accuracy": 0.23342036902904512,
      "num_tokens": 100511857.0,
      "step": 48420
    },
    {
      "entropy": 5.728339099884034,
      "epoch": 4.392688679245283,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003212039526166994,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22407063245773315,
      "num_tokens": 100523654.0,
      "step": 48425
    },
    {
      "entropy": 5.761644649505615,
      "epoch": 4.393142235123367,
      "grad_norm": 1.375,
      "learning_rate": 0.00032117228528593907,
      "loss": 4.9425,
      "mean_token_accuracy": 0.22090758979320527,
      "num_tokens": 100533030.0,
      "step": 48430
    },
    {
      "entropy": 5.694897890090942,
      "epoch": 4.393595791001451,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000321140617000324,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21959954500198364,
      "num_tokens": 100543414.0,
      "step": 48435
    },
    {
      "entropy": 5.666410732269287,
      "epoch": 4.394049346879536,
      "grad_norm": 1.375,
      "learning_rate": 0.00032110894776050936,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22590952813625337,
      "num_tokens": 100553303.0,
      "step": 48440
    },
    {
      "entropy": 5.620059442520142,
      "epoch": 4.39450290275762,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032107727756714984,
      "loss": 4.8357,
      "mean_token_accuracy": 0.23085491508245468,
      "num_tokens": 100563030.0,
      "step": 48445
    },
    {
      "entropy": 5.722546339035034,
      "epoch": 4.394956458635704,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003210456064209005,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21863256245851517,
      "num_tokens": 100573216.0,
      "step": 48450
    },
    {
      "entropy": 5.625833559036255,
      "epoch": 4.395410014513788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032101393432241635,
      "loss": 4.7924,
      "mean_token_accuracy": 0.2247015878558159,
      "num_tokens": 100582646.0,
      "step": 48455
    },
    {
      "entropy": 5.6770120620727536,
      "epoch": 4.395863570391873,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032098226127235215,
      "loss": 4.9038,
      "mean_token_accuracy": 0.21946977227926254,
      "num_tokens": 100593689.0,
      "step": 48460
    },
    {
      "entropy": 5.660766077041626,
      "epoch": 4.396317126269956,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003209505872713632,
      "loss": 4.9448,
      "mean_token_accuracy": 0.2165007099509239,
      "num_tokens": 100604721.0,
      "step": 48465
    },
    {
      "entropy": 5.710666751861572,
      "epoch": 4.39677068214804,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032091891232010434,
      "loss": 4.9241,
      "mean_token_accuracy": 0.22488236874341966,
      "num_tokens": 100614684.0,
      "step": 48470
    },
    {
      "entropy": 5.684283828735351,
      "epoch": 4.397224238026125,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032088723641923066,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22374772876501084,
      "num_tokens": 100624434.0,
      "step": 48475
    },
    {
      "entropy": 5.770476388931274,
      "epoch": 4.397677793904209,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032085555956939726,
      "loss": 5.0031,
      "mean_token_accuracy": 0.21274882853031157,
      "num_tokens": 100635656.0,
      "step": 48480
    },
    {
      "entropy": 5.5909655570983885,
      "epoch": 4.398131349782293,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003208238817712593,
      "loss": 4.741,
      "mean_token_accuracy": 0.23832974582910538,
      "num_tokens": 100645749.0,
      "step": 48485
    },
    {
      "entropy": 5.64471321105957,
      "epoch": 4.398584905660377,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003207922030254717,
      "loss": 4.8392,
      "mean_token_accuracy": 0.2285049721598625,
      "num_tokens": 100655814.0,
      "step": 48490
    },
    {
      "entropy": 5.648770046234131,
      "epoch": 4.399038461538462,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032076052333268964,
      "loss": 4.8513,
      "mean_token_accuracy": 0.22562450915575027,
      "num_tokens": 100665579.0,
      "step": 48495
    },
    {
      "entropy": 5.775399398803711,
      "epoch": 4.399492017416546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003207288426935683,
      "loss": 5.0279,
      "mean_token_accuracy": 0.21439120471477507,
      "num_tokens": 100676421.0,
      "step": 48500
    },
    {
      "entropy": 5.701840353012085,
      "epoch": 4.39994557329463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003206971611087629,
      "loss": 4.8308,
      "mean_token_accuracy": 0.2352561891078949,
      "num_tokens": 100686689.0,
      "step": 48505
    },
    {
      "entropy": 5.796198940277099,
      "epoch": 4.4003991291727145,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003206654785789285,
      "loss": 4.9529,
      "mean_token_accuracy": 0.2228643476963043,
      "num_tokens": 100697324.0,
      "step": 48510
    },
    {
      "entropy": 5.633426141738892,
      "epoch": 4.400852685050798,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032063379510472036,
      "loss": 4.7961,
      "mean_token_accuracy": 0.23130947351455688,
      "num_tokens": 100706003.0,
      "step": 48515
    },
    {
      "entropy": 5.693969774246216,
      "epoch": 4.401306240928882,
      "grad_norm": 1.25,
      "learning_rate": 0.0003206021106867938,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2329317092895508,
      "num_tokens": 100716755.0,
      "step": 48520
    },
    {
      "entropy": 5.643677806854248,
      "epoch": 4.401759796806966,
      "grad_norm": 1.25,
      "learning_rate": 0.0003205704253258039,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22047752290964126,
      "num_tokens": 100727846.0,
      "step": 48525
    },
    {
      "entropy": 5.690610694885254,
      "epoch": 4.402213352685051,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000320538739022406,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22514562010765077,
      "num_tokens": 100738293.0,
      "step": 48530
    },
    {
      "entropy": 5.6880041599273685,
      "epoch": 4.402666908563135,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003205070517772554,
      "loss": 4.8086,
      "mean_token_accuracy": 0.22784607112407684,
      "num_tokens": 100748031.0,
      "step": 48535
    },
    {
      "entropy": 5.6400306701660154,
      "epoch": 4.403120464441219,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003204753635910073,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23215733021497725,
      "num_tokens": 100758825.0,
      "step": 48540
    },
    {
      "entropy": 5.694318151473999,
      "epoch": 4.4035740203193035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032044367446431714,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2309201255440712,
      "num_tokens": 100768226.0,
      "step": 48545
    },
    {
      "entropy": 5.64250226020813,
      "epoch": 4.404027576197388,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032041198439784025,
      "loss": 4.8029,
      "mean_token_accuracy": 0.22965584546327591,
      "num_tokens": 100778125.0,
      "step": 48550
    },
    {
      "entropy": 5.685123634338379,
      "epoch": 4.404481132075472,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003203802933922319,
      "loss": 4.8525,
      "mean_token_accuracy": 0.23229414224624634,
      "num_tokens": 100788670.0,
      "step": 48555
    },
    {
      "entropy": 5.681270980834961,
      "epoch": 4.404934687953556,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032034860144814756,
      "loss": 4.8937,
      "mean_token_accuracy": 0.224245522916317,
      "num_tokens": 100800077.0,
      "step": 48560
    },
    {
      "entropy": 5.672792577743531,
      "epoch": 4.40538824383164,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003203169085662425,
      "loss": 4.8559,
      "mean_token_accuracy": 0.22554195523262024,
      "num_tokens": 100809432.0,
      "step": 48565
    },
    {
      "entropy": 5.6720270156860355,
      "epoch": 4.405841799709724,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003202852147471723,
      "loss": 4.8381,
      "mean_token_accuracy": 0.2361274167895317,
      "num_tokens": 100819675.0,
      "step": 48570
    },
    {
      "entropy": 5.713675832748413,
      "epoch": 4.406295355587808,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032025351999159237,
      "loss": 4.871,
      "mean_token_accuracy": 0.2277238115668297,
      "num_tokens": 100829779.0,
      "step": 48575
    },
    {
      "entropy": 5.68006157875061,
      "epoch": 4.406748911465892,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032022182430015795,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21911661177873612,
      "num_tokens": 100839840.0,
      "step": 48580
    },
    {
      "entropy": 5.661422967910767,
      "epoch": 4.407202467343977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003201901276735248,
      "loss": 4.8974,
      "mean_token_accuracy": 0.22679625004529952,
      "num_tokens": 100849495.0,
      "step": 48585
    },
    {
      "entropy": 5.653108692169189,
      "epoch": 4.407656023222061,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032015843011234826,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22436591386795043,
      "num_tokens": 100860214.0,
      "step": 48590
    },
    {
      "entropy": 5.788898229598999,
      "epoch": 4.408109579100145,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032012673161728396,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2092674434185028,
      "num_tokens": 100870773.0,
      "step": 48595
    },
    {
      "entropy": 5.741449356079102,
      "epoch": 4.4085631349782295,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032009503218898716,
      "loss": 4.8794,
      "mean_token_accuracy": 0.2359814926981926,
      "num_tokens": 100880654.0,
      "step": 48600
    },
    {
      "entropy": 5.615416145324707,
      "epoch": 4.409016690856314,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003200633318281138,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22389124184846879,
      "num_tokens": 100891217.0,
      "step": 48605
    },
    {
      "entropy": 5.650545978546143,
      "epoch": 4.409470246734398,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032003163053531913,
      "loss": 4.7797,
      "mean_token_accuracy": 0.22917840629816055,
      "num_tokens": 100901424.0,
      "step": 48610
    },
    {
      "entropy": 5.729248905181885,
      "epoch": 4.409923802612482,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031999992831125893,
      "loss": 4.959,
      "mean_token_accuracy": 0.2269819498062134,
      "num_tokens": 100912069.0,
      "step": 48615
    },
    {
      "entropy": 5.676454067230225,
      "epoch": 4.410377358490566,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003199682251565887,
      "loss": 4.8529,
      "mean_token_accuracy": 0.21943309158086777,
      "num_tokens": 100922750.0,
      "step": 48620
    },
    {
      "entropy": 5.708251810073852,
      "epoch": 4.41083091436865,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031993652107196416,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2261740446090698,
      "num_tokens": 100933052.0,
      "step": 48625
    },
    {
      "entropy": 5.690996074676514,
      "epoch": 4.411284470246734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031990481605804084,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2292378231883049,
      "num_tokens": 100944519.0,
      "step": 48630
    },
    {
      "entropy": 5.654050254821778,
      "epoch": 4.4117380261248185,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031987311011547455,
      "loss": 4.9076,
      "mean_token_accuracy": 0.21871311962604523,
      "num_tokens": 100954901.0,
      "step": 48635
    },
    {
      "entropy": 5.667327117919922,
      "epoch": 4.412191582002903,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003198414032449209,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2249605268239975,
      "num_tokens": 100964931.0,
      "step": 48640
    },
    {
      "entropy": 5.706310033798218,
      "epoch": 4.412645137880987,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003198096954470356,
      "loss": 4.8564,
      "mean_token_accuracy": 0.23334578573703765,
      "num_tokens": 100974877.0,
      "step": 48645
    },
    {
      "entropy": 5.740013790130615,
      "epoch": 4.413098693759071,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031977798672247444,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2288014695048332,
      "num_tokens": 100986122.0,
      "step": 48650
    },
    {
      "entropy": 5.731851816177368,
      "epoch": 4.4135522496371555,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031974627707189303,
      "loss": 4.8648,
      "mean_token_accuracy": 0.22133907079696655,
      "num_tokens": 100996489.0,
      "step": 48655
    },
    {
      "entropy": 5.677108573913574,
      "epoch": 4.41400580551524,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031971456649594723,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22869823276996612,
      "num_tokens": 101006453.0,
      "step": 48660
    },
    {
      "entropy": 5.74294695854187,
      "epoch": 4.414459361393324,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003196828549952927,
      "loss": 4.9231,
      "mean_token_accuracy": 0.22457430213689805,
      "num_tokens": 101015609.0,
      "step": 48665
    },
    {
      "entropy": 5.718839263916015,
      "epoch": 4.4149129172714074,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003196511425705855,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22479408979415894,
      "num_tokens": 101027406.0,
      "step": 48670
    },
    {
      "entropy": 5.662295913696289,
      "epoch": 4.415366473149492,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003196194292224812,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21948989778757094,
      "num_tokens": 101038255.0,
      "step": 48675
    },
    {
      "entropy": 5.662010908126831,
      "epoch": 4.415820029027576,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031958771495163574,
      "loss": 4.8685,
      "mean_token_accuracy": 0.223648564517498,
      "num_tokens": 101048857.0,
      "step": 48680
    },
    {
      "entropy": 5.731916379928589,
      "epoch": 4.41627358490566,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000319555999758705,
      "loss": 4.9418,
      "mean_token_accuracy": 0.22665096670389176,
      "num_tokens": 101058640.0,
      "step": 48685
    },
    {
      "entropy": 5.7373395442962645,
      "epoch": 4.4167271407837445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031952428364434484,
      "loss": 5.0226,
      "mean_token_accuracy": 0.21954674124717713,
      "num_tokens": 101069537.0,
      "step": 48690
    },
    {
      "entropy": 5.723458576202392,
      "epoch": 4.417180696661829,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031949256660921104,
      "loss": 4.9018,
      "mean_token_accuracy": 0.21509345173835753,
      "num_tokens": 101080202.0,
      "step": 48695
    },
    {
      "entropy": 5.691107749938965,
      "epoch": 4.417634252539913,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031946084865395975,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22690445333719253,
      "num_tokens": 101089500.0,
      "step": 48700
    },
    {
      "entropy": 5.767582893371582,
      "epoch": 4.418087808417997,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031942912977924674,
      "loss": 4.9848,
      "mean_token_accuracy": 0.21876843720674516,
      "num_tokens": 101099858.0,
      "step": 48705
    },
    {
      "entropy": 5.7089385986328125,
      "epoch": 4.418541364296082,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031939740998572806,
      "loss": 4.9437,
      "mean_token_accuracy": 0.22004939913749694,
      "num_tokens": 101110834.0,
      "step": 48710
    },
    {
      "entropy": 5.728222799301148,
      "epoch": 4.418994920174166,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003193656892740595,
      "loss": 4.9229,
      "mean_token_accuracy": 0.22487525641918182,
      "num_tokens": 101121003.0,
      "step": 48715
    },
    {
      "entropy": 5.732079124450683,
      "epoch": 4.419448476052249,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003193339676448973,
      "loss": 4.9181,
      "mean_token_accuracy": 0.2207103982567787,
      "num_tokens": 101131925.0,
      "step": 48720
    },
    {
      "entropy": 5.676371431350708,
      "epoch": 4.4199020319303335,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003193022450988973,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23044830560684204,
      "num_tokens": 101143130.0,
      "step": 48725
    },
    {
      "entropy": 5.66291356086731,
      "epoch": 4.420355587808418,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003192705216367156,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2273336187005043,
      "num_tokens": 101155524.0,
      "step": 48730
    },
    {
      "entropy": 5.6316310405731205,
      "epoch": 4.420809143686502,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031923879725900817,
      "loss": 4.8672,
      "mean_token_accuracy": 0.23235570639371872,
      "num_tokens": 101164815.0,
      "step": 48735
    },
    {
      "entropy": 5.693700456619263,
      "epoch": 4.421262699564586,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031920707196643125,
      "loss": 4.976,
      "mean_token_accuracy": 0.22281210422515868,
      "num_tokens": 101174910.0,
      "step": 48740
    },
    {
      "entropy": 5.6253915309906,
      "epoch": 4.4217162554426706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003191753457596407,
      "loss": 4.737,
      "mean_token_accuracy": 0.2347467452287674,
      "num_tokens": 101185698.0,
      "step": 48745
    },
    {
      "entropy": 5.669201231002807,
      "epoch": 4.422169811320755,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031914361863929276,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22552573382854463,
      "num_tokens": 101197001.0,
      "step": 48750
    },
    {
      "entropy": 5.650038242340088,
      "epoch": 4.422623367198839,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003191118906060436,
      "loss": 4.8341,
      "mean_token_accuracy": 0.22572482824325563,
      "num_tokens": 101206705.0,
      "step": 48755
    },
    {
      "entropy": 5.692517518997192,
      "epoch": 4.423076923076923,
      "grad_norm": 1.25,
      "learning_rate": 0.00031908016166054933,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22394585758447647,
      "num_tokens": 101216484.0,
      "step": 48760
    },
    {
      "entropy": 5.681727647781372,
      "epoch": 4.423530478955008,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003190484318034659,
      "loss": 4.8816,
      "mean_token_accuracy": 0.22456322461366654,
      "num_tokens": 101227169.0,
      "step": 48765
    },
    {
      "entropy": 5.711324214935303,
      "epoch": 4.423984034833092,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003190167010354499,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22762604355812072,
      "num_tokens": 101236956.0,
      "step": 48770
    },
    {
      "entropy": 5.693299102783203,
      "epoch": 4.424437590711175,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003189849693571572,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22204425930976868,
      "num_tokens": 101248681.0,
      "step": 48775
    },
    {
      "entropy": 5.716499614715576,
      "epoch": 4.4248911465892595,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003189532367692441,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22168020904064178,
      "num_tokens": 101258375.0,
      "step": 48780
    },
    {
      "entropy": 5.650789499282837,
      "epoch": 4.425344702467344,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031892150327236693,
      "loss": 4.8559,
      "mean_token_accuracy": 0.23254477083683014,
      "num_tokens": 101268513.0,
      "step": 48785
    },
    {
      "entropy": 5.708165311813355,
      "epoch": 4.425798258345428,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003188897688671818,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21464011520147325,
      "num_tokens": 101278998.0,
      "step": 48790
    },
    {
      "entropy": 5.755346775054932,
      "epoch": 4.426251814223512,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031885803355434516,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2270659476518631,
      "num_tokens": 101289090.0,
      "step": 48795
    },
    {
      "entropy": 5.705514669418335,
      "epoch": 4.426705370101597,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003188262973345131,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22452063411474227,
      "num_tokens": 101299697.0,
      "step": 48800
    },
    {
      "entropy": 5.70561146736145,
      "epoch": 4.427158925979681,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031879456020834213,
      "loss": 4.9414,
      "mean_token_accuracy": 0.2148512214422226,
      "num_tokens": 101310470.0,
      "step": 48805
    },
    {
      "entropy": 5.712173748016357,
      "epoch": 4.427612481857765,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031876282217648845,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22452938705682754,
      "num_tokens": 101321864.0,
      "step": 48810
    },
    {
      "entropy": 5.678779077529907,
      "epoch": 4.428066037735849,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031873108323960845,
      "loss": 4.8569,
      "mean_token_accuracy": 0.23217401057481765,
      "num_tokens": 101331302.0,
      "step": 48815
    },
    {
      "entropy": 5.690540599822998,
      "epoch": 4.428519593613933,
      "grad_norm": 1.25,
      "learning_rate": 0.00031869934339835856,
      "loss": 4.9301,
      "mean_token_accuracy": 0.22089801281690596,
      "num_tokens": 101341387.0,
      "step": 48820
    },
    {
      "entropy": 5.693362903594971,
      "epoch": 4.428973149492017,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031866760265339514,
      "loss": 4.8265,
      "mean_token_accuracy": 0.22790978997945785,
      "num_tokens": 101351235.0,
      "step": 48825
    },
    {
      "entropy": 5.6901990413665775,
      "epoch": 4.429426705370101,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031863586100537453,
      "loss": 4.9003,
      "mean_token_accuracy": 0.2206323504447937,
      "num_tokens": 101361289.0,
      "step": 48830
    },
    {
      "entropy": 5.653173971176147,
      "epoch": 4.429880261248186,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003186041184549531,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22461865991353988,
      "num_tokens": 101371666.0,
      "step": 48835
    },
    {
      "entropy": 5.674940013885498,
      "epoch": 4.43033381712627,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003185723750027875,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2297504261136055,
      "num_tokens": 101381195.0,
      "step": 48840
    },
    {
      "entropy": 5.750525379180909,
      "epoch": 4.430787373004354,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031854063064953403,
      "loss": 4.9374,
      "mean_token_accuracy": 0.2271145761013031,
      "num_tokens": 101392478.0,
      "step": 48845
    },
    {
      "entropy": 5.728920316696167,
      "epoch": 4.431240928882438,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003185088853958492,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21685962677001952,
      "num_tokens": 101405005.0,
      "step": 48850
    },
    {
      "entropy": 5.681728315353394,
      "epoch": 4.431694484760523,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031847713924238965,
      "loss": 4.888,
      "mean_token_accuracy": 0.21742024570703505,
      "num_tokens": 101416588.0,
      "step": 48855
    },
    {
      "entropy": 5.681982803344726,
      "epoch": 4.432148040638607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031844539218981165,
      "loss": 4.9011,
      "mean_token_accuracy": 0.2201018065214157,
      "num_tokens": 101426784.0,
      "step": 48860
    },
    {
      "entropy": 5.6550849914550785,
      "epoch": 4.432601596516691,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003184136442387719,
      "loss": 4.8262,
      "mean_token_accuracy": 0.2256109222769737,
      "num_tokens": 101437919.0,
      "step": 48865
    },
    {
      "entropy": 5.684061098098755,
      "epoch": 4.433055152394775,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031838189538992693,
      "loss": 4.9167,
      "mean_token_accuracy": 0.22476656436920167,
      "num_tokens": 101448654.0,
      "step": 48870
    },
    {
      "entropy": 5.707367944717407,
      "epoch": 4.433508708272859,
      "grad_norm": 1.25,
      "learning_rate": 0.0003183501456439334,
      "loss": 4.8144,
      "mean_token_accuracy": 0.22878600358963014,
      "num_tokens": 101458648.0,
      "step": 48875
    },
    {
      "entropy": 5.648741769790649,
      "epoch": 4.433962264150943,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031831839500144767,
      "loss": 4.8397,
      "mean_token_accuracy": 0.22134959995746611,
      "num_tokens": 101469339.0,
      "step": 48880
    },
    {
      "entropy": 5.700949382781983,
      "epoch": 4.434415820029027,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00031828664346312657,
      "loss": 4.874,
      "mean_token_accuracy": 0.2291574776172638,
      "num_tokens": 101479452.0,
      "step": 48885
    },
    {
      "entropy": 5.669437837600708,
      "epoch": 4.434869375907112,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003182548910296267,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22313859611749648,
      "num_tokens": 101489147.0,
      "step": 48890
    },
    {
      "entropy": 5.635786485671997,
      "epoch": 4.435322931785196,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003182231377016046,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23186724185943602,
      "num_tokens": 101498958.0,
      "step": 48895
    },
    {
      "entropy": 5.731043720245362,
      "epoch": 4.43577648766328,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000318191383479717,
      "loss": 4.9564,
      "mean_token_accuracy": 0.22086626440286636,
      "num_tokens": 101509197.0,
      "step": 48900
    },
    {
      "entropy": 5.745031929016113,
      "epoch": 4.436230043541364,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003181596283646205,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22487464398145676,
      "num_tokens": 101520018.0,
      "step": 48905
    },
    {
      "entropy": 5.71612811088562,
      "epoch": 4.436683599419449,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031812787235697206,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22282404005527495,
      "num_tokens": 101530755.0,
      "step": 48910
    },
    {
      "entropy": 5.652226114273072,
      "epoch": 4.437137155297533,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031809611545742804,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21327615827322005,
      "num_tokens": 101541028.0,
      "step": 48915
    },
    {
      "entropy": 5.690514755249024,
      "epoch": 4.437590711175617,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003180643576666455,
      "loss": 4.843,
      "mean_token_accuracy": 0.23039474338293076,
      "num_tokens": 101550826.0,
      "step": 48920
    },
    {
      "entropy": 5.734674787521362,
      "epoch": 4.4380442670537015,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003180325989852811,
      "loss": 4.9009,
      "mean_token_accuracy": 0.22476475238800048,
      "num_tokens": 101560657.0,
      "step": 48925
    },
    {
      "entropy": 5.658048820495606,
      "epoch": 4.438497822931785,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003180008394139914,
      "loss": 4.861,
      "mean_token_accuracy": 0.22852977514266967,
      "num_tokens": 101570504.0,
      "step": 48930
    },
    {
      "entropy": 5.725140476226807,
      "epoch": 4.438951378809869,
      "grad_norm": 1.375,
      "learning_rate": 0.00031796907895343356,
      "loss": 4.9918,
      "mean_token_accuracy": 0.21601316928863526,
      "num_tokens": 101581729.0,
      "step": 48935
    },
    {
      "entropy": 5.739936494827271,
      "epoch": 4.439404934687953,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031793731760426414,
      "loss": 4.9642,
      "mean_token_accuracy": 0.22021788656711577,
      "num_tokens": 101592649.0,
      "step": 48940
    },
    {
      "entropy": 5.793572950363159,
      "epoch": 4.439858490566038,
      "grad_norm": 1.375,
      "learning_rate": 0.00031790555536714013,
      "loss": 5.0072,
      "mean_token_accuracy": 0.22201725840568542,
      "num_tokens": 101603824.0,
      "step": 48945
    },
    {
      "entropy": 5.715573024749756,
      "epoch": 4.440312046444122,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003178737922427182,
      "loss": 4.9179,
      "mean_token_accuracy": 0.22401766180992128,
      "num_tokens": 101615659.0,
      "step": 48950
    },
    {
      "entropy": 5.784754037857056,
      "epoch": 4.440765602322206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003178420282316554,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21614880412817,
      "num_tokens": 101625937.0,
      "step": 48955
    },
    {
      "entropy": 5.692641067504883,
      "epoch": 4.4412191582002905,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003178102633346085,
      "loss": 4.853,
      "mean_token_accuracy": 0.23233896791934966,
      "num_tokens": 101637352.0,
      "step": 48960
    },
    {
      "entropy": 5.689813709259033,
      "epoch": 4.441672714078375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003177784975522344,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23415756821632386,
      "num_tokens": 101647939.0,
      "step": 48965
    },
    {
      "entropy": 5.7498492240905765,
      "epoch": 4.442126269956459,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031774673088519014,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2220427706837654,
      "num_tokens": 101658200.0,
      "step": 48970
    },
    {
      "entropy": 5.7066789150238035,
      "epoch": 4.442579825834542,
      "grad_norm": 1.375,
      "learning_rate": 0.0003177149633341326,
      "loss": 4.8824,
      "mean_token_accuracy": 0.22449959218502044,
      "num_tokens": 101668624.0,
      "step": 48975
    },
    {
      "entropy": 5.777685832977295,
      "epoch": 4.443033381712627,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003176831948997187,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2125133365392685,
      "num_tokens": 101678939.0,
      "step": 48980
    },
    {
      "entropy": 5.718263578414917,
      "epoch": 4.443486937590711,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003176514255826054,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22839555889368057,
      "num_tokens": 101689365.0,
      "step": 48985
    },
    {
      "entropy": 5.688775873184204,
      "epoch": 4.443940493468795,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003176196553834498,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22006253749132157,
      "num_tokens": 101699429.0,
      "step": 48990
    },
    {
      "entropy": 5.681458330154419,
      "epoch": 4.444394049346879,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031758788430290893,
      "loss": 4.9245,
      "mean_token_accuracy": 0.2205912947654724,
      "num_tokens": 101709452.0,
      "step": 48995
    },
    {
      "entropy": 5.648981332778931,
      "epoch": 4.444847605224964,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031755611234163966,
      "loss": 4.83,
      "mean_token_accuracy": 0.22939110845327376,
      "num_tokens": 101719783.0,
      "step": 49000
    },
    {
      "entropy": 5.66691780090332,
      "epoch": 4.445301161103048,
      "grad_norm": 1.375,
      "learning_rate": 0.0003175243395002992,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2324678495526314,
      "num_tokens": 101729100.0,
      "step": 49005
    },
    {
      "entropy": 5.689895868301392,
      "epoch": 4.445754716981132,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031749256577954453,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22761654406785964,
      "num_tokens": 101738711.0,
      "step": 49010
    },
    {
      "entropy": 5.660838651657104,
      "epoch": 4.4462082728592165,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003174607911800328,
      "loss": 4.865,
      "mean_token_accuracy": 0.22449371069669724,
      "num_tokens": 101749624.0,
      "step": 49015
    },
    {
      "entropy": 5.667410612106323,
      "epoch": 4.446661828737301,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031742901570242107,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22528573870658875,
      "num_tokens": 101760951.0,
      "step": 49020
    },
    {
      "entropy": 5.675141859054565,
      "epoch": 4.447115384615385,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031739723934736655,
      "loss": 4.8142,
      "mean_token_accuracy": 0.23112359195947646,
      "num_tokens": 101770773.0,
      "step": 49025
    },
    {
      "entropy": 5.709124946594239,
      "epoch": 4.447568940493468,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003173654621155262,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21943217664957046,
      "num_tokens": 101781566.0,
      "step": 49030
    },
    {
      "entropy": 5.672196102142334,
      "epoch": 4.448022496371553,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031733368400755735,
      "loss": 4.8892,
      "mean_token_accuracy": 0.22893578559160233,
      "num_tokens": 101792268.0,
      "step": 49035
    },
    {
      "entropy": 5.732755851745606,
      "epoch": 4.448476052249637,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031730190502411715,
      "loss": 4.9321,
      "mean_token_accuracy": 0.22195803970098496,
      "num_tokens": 101802500.0,
      "step": 49040
    },
    {
      "entropy": 5.725253009796143,
      "epoch": 4.448929608127721,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003172701251658628,
      "loss": 4.8515,
      "mean_token_accuracy": 0.22088802307844163,
      "num_tokens": 101813851.0,
      "step": 49045
    },
    {
      "entropy": 5.799509906768799,
      "epoch": 4.4493831640058055,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031723834443345134,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21947232782840728,
      "num_tokens": 101824774.0,
      "step": 49050
    },
    {
      "entropy": 5.666823959350586,
      "epoch": 4.44983671988389,
      "grad_norm": 1.625,
      "learning_rate": 0.00031720656282754027,
      "loss": 4.7699,
      "mean_token_accuracy": 0.24122696220874787,
      "num_tokens": 101835975.0,
      "step": 49055
    },
    {
      "entropy": 5.634105014801025,
      "epoch": 4.450290275761974,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031717478034878676,
      "loss": 4.8278,
      "mean_token_accuracy": 0.23018525093793868,
      "num_tokens": 101846764.0,
      "step": 49060
    },
    {
      "entropy": 5.656715393066406,
      "epoch": 4.450743831640058,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031714299699784796,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22538535594940184,
      "num_tokens": 101857968.0,
      "step": 49065
    },
    {
      "entropy": 5.690466451644897,
      "epoch": 4.4511973875181425,
      "grad_norm": 1.375,
      "learning_rate": 0.0003171112127753813,
      "loss": 4.8807,
      "mean_token_accuracy": 0.2320683240890503,
      "num_tokens": 101868484.0,
      "step": 49070
    },
    {
      "entropy": 5.729061841964722,
      "epoch": 4.451650943396227,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003170794276820441,
      "loss": 4.8718,
      "mean_token_accuracy": 0.22664163261651993,
      "num_tokens": 101877969.0,
      "step": 49075
    },
    {
      "entropy": 5.646499967575073,
      "epoch": 4.452104499274311,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031704764171849344,
      "loss": 4.8567,
      "mean_token_accuracy": 0.23062149584293365,
      "num_tokens": 101888635.0,
      "step": 49080
    },
    {
      "entropy": 5.67963809967041,
      "epoch": 4.4525580551523944,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031701585488538695,
      "loss": 4.8238,
      "mean_token_accuracy": 0.22866480499505998,
      "num_tokens": 101898340.0,
      "step": 49085
    },
    {
      "entropy": 5.7443970203399655,
      "epoch": 4.453011611030479,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003169840671833819,
      "loss": 4.9422,
      "mean_token_accuracy": 0.21564559042453765,
      "num_tokens": 101908993.0,
      "step": 49090
    },
    {
      "entropy": 5.635062789916992,
      "epoch": 4.453465166908563,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003169522786131355,
      "loss": 4.8139,
      "mean_token_accuracy": 0.22952888011932374,
      "num_tokens": 101919675.0,
      "step": 49095
    },
    {
      "entropy": 5.748192214965821,
      "epoch": 4.453918722786647,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031692048917530535,
      "loss": 4.9532,
      "mean_token_accuracy": 0.22083310037851334,
      "num_tokens": 101929949.0,
      "step": 49100
    },
    {
      "entropy": 5.676390695571899,
      "epoch": 4.4543722786647315,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031688869887054887,
      "loss": 4.7532,
      "mean_token_accuracy": 0.24162102043628692,
      "num_tokens": 101940025.0,
      "step": 49105
    },
    {
      "entropy": 5.6652747631073,
      "epoch": 4.454825834542816,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031685690769952336,
      "loss": 4.811,
      "mean_token_accuracy": 0.23445447981357576,
      "num_tokens": 101950344.0,
      "step": 49110
    },
    {
      "entropy": 5.736001205444336,
      "epoch": 4.4552793904209,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003168251156628863,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22559948414564132,
      "num_tokens": 101961405.0,
      "step": 49115
    },
    {
      "entropy": 5.638861989974975,
      "epoch": 4.455732946298984,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031679332276129523,
      "loss": 4.7744,
      "mean_token_accuracy": 0.23217560201883317,
      "num_tokens": 101972164.0,
      "step": 49120
    },
    {
      "entropy": 5.66761965751648,
      "epoch": 4.456186502177069,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031676152899540763,
      "loss": 4.853,
      "mean_token_accuracy": 0.22048295438289642,
      "num_tokens": 101981573.0,
      "step": 49125
    },
    {
      "entropy": 5.6621520042419435,
      "epoch": 4.456640058055152,
      "grad_norm": 1.390625,
      "learning_rate": 0.000316729734365881,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22522128373384476,
      "num_tokens": 101991051.0,
      "step": 49130
    },
    {
      "entropy": 5.680356121063232,
      "epoch": 4.457093613933236,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003166979388733727,
      "loss": 4.8498,
      "mean_token_accuracy": 0.23260704427957535,
      "num_tokens": 102000003.0,
      "step": 49135
    },
    {
      "entropy": 5.717011260986328,
      "epoch": 4.4575471698113205,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003166661425185405,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21650610417127608,
      "num_tokens": 102010489.0,
      "step": 49140
    },
    {
      "entropy": 5.688056993484497,
      "epoch": 4.458000725689405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031663434530204183,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21538309156894683,
      "num_tokens": 102020486.0,
      "step": 49145
    },
    {
      "entropy": 5.712220907211304,
      "epoch": 4.458454281567489,
      "grad_norm": 1.25,
      "learning_rate": 0.0003166025472245342,
      "loss": 4.8637,
      "mean_token_accuracy": 0.2231267809867859,
      "num_tokens": 102030098.0,
      "step": 49150
    },
    {
      "entropy": 5.7407286167144775,
      "epoch": 4.458907837445573,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003165707482866754,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22466369420289994,
      "num_tokens": 102040850.0,
      "step": 49155
    },
    {
      "entropy": 5.65698356628418,
      "epoch": 4.4593613933236576,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031653894848912296,
      "loss": 4.7942,
      "mean_token_accuracy": 0.22890523225069045,
      "num_tokens": 102051056.0,
      "step": 49160
    },
    {
      "entropy": 5.71702470779419,
      "epoch": 4.459814949201742,
      "grad_norm": 1.25,
      "learning_rate": 0.0003165071478325344,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21775954365730285,
      "num_tokens": 102061727.0,
      "step": 49165
    },
    {
      "entropy": 5.69984917640686,
      "epoch": 4.460268505079826,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003164753463175675,
      "loss": 4.92,
      "mean_token_accuracy": 0.22378232479095458,
      "num_tokens": 102071501.0,
      "step": 49170
    },
    {
      "entropy": 5.7138221740722654,
      "epoch": 4.46072206095791,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031644354394487977,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23841129690408708,
      "num_tokens": 102082511.0,
      "step": 49175
    },
    {
      "entropy": 5.747981405258178,
      "epoch": 4.461175616835995,
      "grad_norm": 1.375,
      "learning_rate": 0.00031641174071512905,
      "loss": 4.8746,
      "mean_token_accuracy": 0.23152710348367692,
      "num_tokens": 102093010.0,
      "step": 49180
    },
    {
      "entropy": 5.653470849990844,
      "epoch": 4.461629172714078,
      "grad_norm": 1.46875,
      "learning_rate": 0.000316379936628973,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22837564945220948,
      "num_tokens": 102102691.0,
      "step": 49185
    },
    {
      "entropy": 5.667874526977539,
      "epoch": 4.462082728592162,
      "grad_norm": 1.375,
      "learning_rate": 0.00031634813168706933,
      "loss": 4.8293,
      "mean_token_accuracy": 0.22891950458288193,
      "num_tokens": 102112726.0,
      "step": 49190
    },
    {
      "entropy": 5.670906257629395,
      "epoch": 4.4625362844702465,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031631632589007575,
      "loss": 4.8756,
      "mean_token_accuracy": 0.23098576664924622,
      "num_tokens": 102122879.0,
      "step": 49195
    },
    {
      "entropy": 5.701549339294433,
      "epoch": 4.462989840348331,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031628451923864997,
      "loss": 4.8896,
      "mean_token_accuracy": 0.22618578672409057,
      "num_tokens": 102133485.0,
      "step": 49200
    },
    {
      "entropy": 5.737494611740113,
      "epoch": 4.463443396226415,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031625271173344984,
      "loss": 4.974,
      "mean_token_accuracy": 0.2252035468816757,
      "num_tokens": 102143439.0,
      "step": 49205
    },
    {
      "entropy": 5.6540158748626705,
      "epoch": 4.463896952104499,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031622090337513314,
      "loss": 4.8294,
      "mean_token_accuracy": 0.2306519240140915,
      "num_tokens": 102153808.0,
      "step": 49210
    },
    {
      "entropy": 5.768898582458496,
      "epoch": 4.464350507982584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031618909416435766,
      "loss": 4.952,
      "mean_token_accuracy": 0.2167825371026993,
      "num_tokens": 102163375.0,
      "step": 49215
    },
    {
      "entropy": 5.714747095108033,
      "epoch": 4.464804063860668,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003161572841017811,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22629363536834718,
      "num_tokens": 102174402.0,
      "step": 49220
    },
    {
      "entropy": 5.6153707027435305,
      "epoch": 4.465257619738752,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031612547318806154,
      "loss": 4.7736,
      "mean_token_accuracy": 0.23196899592876435,
      "num_tokens": 102184446.0,
      "step": 49225
    },
    {
      "entropy": 5.666836738586426,
      "epoch": 4.465711175616836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031609366142385675,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22365206629037857,
      "num_tokens": 102194398.0,
      "step": 49230
    },
    {
      "entropy": 5.630029535293579,
      "epoch": 4.46616473149492,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003160618488098245,
      "loss": 4.8185,
      "mean_token_accuracy": 0.23420944213867187,
      "num_tokens": 102205953.0,
      "step": 49235
    },
    {
      "entropy": 5.657793760299683,
      "epoch": 4.466618287373004,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031603003534662273,
      "loss": 4.8285,
      "mean_token_accuracy": 0.23655975013971328,
      "num_tokens": 102215815.0,
      "step": 49240
    },
    {
      "entropy": 5.708395385742188,
      "epoch": 4.467071843251088,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031599822103490946,
      "loss": 4.8285,
      "mean_token_accuracy": 0.2270008310675621,
      "num_tokens": 102225703.0,
      "step": 49245
    },
    {
      "entropy": 5.753926992416382,
      "epoch": 4.467525399129173,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003159664058753425,
      "loss": 5.0086,
      "mean_token_accuracy": 0.21056965589523316,
      "num_tokens": 102236627.0,
      "step": 49250
    },
    {
      "entropy": 5.671736145019532,
      "epoch": 4.467978955007257,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003159345898685797,
      "loss": 4.8922,
      "mean_token_accuracy": 0.23127419650554656,
      "num_tokens": 102246539.0,
      "step": 49255
    },
    {
      "entropy": 5.696890163421631,
      "epoch": 4.468432510885341,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003159027730152793,
      "loss": 4.89,
      "mean_token_accuracy": 0.22175315618515015,
      "num_tokens": 102256734.0,
      "step": 49260
    },
    {
      "entropy": 5.738433790206909,
      "epoch": 4.468886066763425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003158709553160991,
      "loss": 5.0018,
      "mean_token_accuracy": 0.21315175741910936,
      "num_tokens": 102267469.0,
      "step": 49265
    },
    {
      "entropy": 5.735507822036743,
      "epoch": 4.46933962264151,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031583913677169705,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22604428976774216,
      "num_tokens": 102278435.0,
      "step": 49270
    },
    {
      "entropy": 5.7637616157531735,
      "epoch": 4.469793178519594,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031580731738273135,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22151069790124894,
      "num_tokens": 102289171.0,
      "step": 49275
    },
    {
      "entropy": 5.676257562637329,
      "epoch": 4.470246734397678,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003157754971498599,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23366160392761232,
      "num_tokens": 102299254.0,
      "step": 49280
    },
    {
      "entropy": 5.71058874130249,
      "epoch": 4.4707002902757615,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003157436760737408,
      "loss": 4.905,
      "mean_token_accuracy": 0.22193235903978348,
      "num_tokens": 102308471.0,
      "step": 49285
    },
    {
      "entropy": 5.728515386581421,
      "epoch": 4.471153846153846,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000315711854155032,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22010568231344224,
      "num_tokens": 102319005.0,
      "step": 49290
    },
    {
      "entropy": 5.716075706481933,
      "epoch": 4.47160740203193,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031568003139439185,
      "loss": 4.8683,
      "mean_token_accuracy": 0.23048002719879152,
      "num_tokens": 102329568.0,
      "step": 49295
    },
    {
      "entropy": 5.803271389007568,
      "epoch": 4.472060957910014,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003156482077924781,
      "loss": 4.9906,
      "mean_token_accuracy": 0.21854897886514663,
      "num_tokens": 102340810.0,
      "step": 49300
    },
    {
      "entropy": 5.658351421356201,
      "epoch": 4.472514513788099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003156163833499491,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22166965901851654,
      "num_tokens": 102351305.0,
      "step": 49305
    },
    {
      "entropy": 5.716815996170044,
      "epoch": 4.472968069666183,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031558455806746306,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22055397033691407,
      "num_tokens": 102362057.0,
      "step": 49310
    },
    {
      "entropy": 5.770926380157471,
      "epoch": 4.473421625544267,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003155527319456779,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21886435449123381,
      "num_tokens": 102372612.0,
      "step": 49315
    },
    {
      "entropy": 5.6805384159088135,
      "epoch": 4.473875181422351,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031552090498525195,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23052327185869217,
      "num_tokens": 102383337.0,
      "step": 49320
    },
    {
      "entropy": 5.712318038940429,
      "epoch": 4.474328737300436,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003154890771868433,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22391318380832673,
      "num_tokens": 102394196.0,
      "step": 49325
    },
    {
      "entropy": 5.757188129425049,
      "epoch": 4.47478229317852,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031545724855111025,
      "loss": 4.9139,
      "mean_token_accuracy": 0.2271278902888298,
      "num_tokens": 102404677.0,
      "step": 49330
    },
    {
      "entropy": 5.700198078155518,
      "epoch": 4.475235849056604,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031542541907871095,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23824246376752853,
      "num_tokens": 102413445.0,
      "step": 49335
    },
    {
      "entropy": 5.662498474121094,
      "epoch": 4.475689404934688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003153935887703038,
      "loss": 4.8723,
      "mean_token_accuracy": 0.23155819475650788,
      "num_tokens": 102424527.0,
      "step": 49340
    },
    {
      "entropy": 5.731548738479614,
      "epoch": 4.476142960812772,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003153617576265469,
      "loss": 4.856,
      "mean_token_accuracy": 0.219210746884346,
      "num_tokens": 102434940.0,
      "step": 49345
    },
    {
      "entropy": 5.77527756690979,
      "epoch": 4.476596516690856,
      "grad_norm": 1.5625,
      "learning_rate": 0.00031532992564809847,
      "loss": 4.9535,
      "mean_token_accuracy": 0.22771701067686081,
      "num_tokens": 102446678.0,
      "step": 49350
    },
    {
      "entropy": 5.642897701263427,
      "epoch": 4.47705007256894,
      "grad_norm": 1.1875,
      "learning_rate": 0.000315298092835617,
      "loss": 4.848,
      "mean_token_accuracy": 0.23006006181240082,
      "num_tokens": 102457255.0,
      "step": 49355
    },
    {
      "entropy": 5.72709641456604,
      "epoch": 4.477503628447025,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031526625918976066,
      "loss": 5.0112,
      "mean_token_accuracy": 0.22288602888584136,
      "num_tokens": 102468385.0,
      "step": 49360
    },
    {
      "entropy": 5.693874788284302,
      "epoch": 4.477957184325109,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031523442471118783,
      "loss": 4.8398,
      "mean_token_accuracy": 0.2228361561894417,
      "num_tokens": 102478204.0,
      "step": 49365
    },
    {
      "entropy": 5.757846832275391,
      "epoch": 4.478410740203193,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031520258940055684,
      "loss": 4.8993,
      "mean_token_accuracy": 0.22610832303762435,
      "num_tokens": 102488492.0,
      "step": 49370
    },
    {
      "entropy": 5.68573317527771,
      "epoch": 4.4788642960812775,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031517075325852606,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22581779211759567,
      "num_tokens": 102500309.0,
      "step": 49375
    },
    {
      "entropy": 5.6710327625274655,
      "epoch": 4.479317851959362,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003151389162857539,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22687178552150727,
      "num_tokens": 102511196.0,
      "step": 49380
    },
    {
      "entropy": 5.686135864257812,
      "epoch": 4.479771407837446,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031510707848289864,
      "loss": 4.8476,
      "mean_token_accuracy": 0.23003244400024414,
      "num_tokens": 102522100.0,
      "step": 49385
    },
    {
      "entropy": 5.677272987365723,
      "epoch": 4.480224963715529,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031507523985061886,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23701518923044204,
      "num_tokens": 102531677.0,
      "step": 49390
    },
    {
      "entropy": 5.637654590606689,
      "epoch": 4.480678519593614,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031504340038957294,
      "loss": 4.7621,
      "mean_token_accuracy": 0.2367106020450592,
      "num_tokens": 102541854.0,
      "step": 49395
    },
    {
      "entropy": 5.6900773525238035,
      "epoch": 4.481132075471698,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003150115601004192,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22702303826808928,
      "num_tokens": 102551187.0,
      "step": 49400
    },
    {
      "entropy": 5.72037901878357,
      "epoch": 4.481585631349782,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003149797189838162,
      "loss": 4.8993,
      "mean_token_accuracy": 0.22085327357053758,
      "num_tokens": 102562568.0,
      "step": 49405
    },
    {
      "entropy": 5.698154163360596,
      "epoch": 4.482039187227866,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031494787704042243,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21815808564424516,
      "num_tokens": 102573170.0,
      "step": 49410
    },
    {
      "entropy": 5.630895709991455,
      "epoch": 4.482492743105951,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031491603427089644,
      "loss": 4.7365,
      "mean_token_accuracy": 0.2358236461877823,
      "num_tokens": 102583950.0,
      "step": 49415
    },
    {
      "entropy": 5.685453128814697,
      "epoch": 4.482946298984035,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031488419067589665,
      "loss": 4.865,
      "mean_token_accuracy": 0.22736094295978546,
      "num_tokens": 102594370.0,
      "step": 49420
    },
    {
      "entropy": 5.709194660186768,
      "epoch": 4.483399854862119,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003148523462560816,
      "loss": 4.8437,
      "mean_token_accuracy": 0.23229633420705795,
      "num_tokens": 102604837.0,
      "step": 49425
    },
    {
      "entropy": 5.704143762588501,
      "epoch": 4.4838534107402035,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003148205010121099,
      "loss": 4.9291,
      "mean_token_accuracy": 0.22321943044662476,
      "num_tokens": 102616170.0,
      "step": 49430
    },
    {
      "entropy": 5.683032274246216,
      "epoch": 4.484306966618288,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003147886549446401,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22771499305963516,
      "num_tokens": 102627283.0,
      "step": 49435
    },
    {
      "entropy": 5.737686061859131,
      "epoch": 4.484760522496371,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003147568080543308,
      "loss": 4.9166,
      "mean_token_accuracy": 0.21566525399684905,
      "num_tokens": 102637087.0,
      "step": 49440
    },
    {
      "entropy": 5.706229543685913,
      "epoch": 4.485214078374455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003147249603418406,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22491077929735184,
      "num_tokens": 102646818.0,
      "step": 49445
    },
    {
      "entropy": 5.6755835056304935,
      "epoch": 4.48566763425254,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003146931118078281,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23492573350667953,
      "num_tokens": 102656940.0,
      "step": 49450
    },
    {
      "entropy": 5.699328851699829,
      "epoch": 4.486121190130624,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003146612624529518,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22663700133562087,
      "num_tokens": 102666841.0,
      "step": 49455
    },
    {
      "entropy": 5.7099823474884035,
      "epoch": 4.486574746008708,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031462941227787056,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2235623851418495,
      "num_tokens": 102676821.0,
      "step": 49460
    },
    {
      "entropy": 5.785326910018921,
      "epoch": 4.4870283018867925,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003145975612832431,
      "loss": 4.9734,
      "mean_token_accuracy": 0.22331641912460326,
      "num_tokens": 102687340.0,
      "step": 49465
    },
    {
      "entropy": 5.687460994720459,
      "epoch": 4.487481857764877,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003145657094697278,
      "loss": 4.884,
      "mean_token_accuracy": 0.2247382953763008,
      "num_tokens": 102698320.0,
      "step": 49470
    },
    {
      "entropy": 5.747256851196289,
      "epoch": 4.487935413642961,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003145338568379836,
      "loss": 4.9179,
      "mean_token_accuracy": 0.22527152746915818,
      "num_tokens": 102708685.0,
      "step": 49475
    },
    {
      "entropy": 5.677592802047729,
      "epoch": 4.488388969521045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003145020033886692,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22277004271745682,
      "num_tokens": 102719680.0,
      "step": 49480
    },
    {
      "entropy": 5.631991958618164,
      "epoch": 4.4888425253991295,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031447014912244325,
      "loss": 4.8273,
      "mean_token_accuracy": 0.23073019683361054,
      "num_tokens": 102731156.0,
      "step": 49485
    },
    {
      "entropy": 5.684357976913452,
      "epoch": 4.489296081277214,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003144382940399646,
      "loss": 4.807,
      "mean_token_accuracy": 0.233288936316967,
      "num_tokens": 102740790.0,
      "step": 49490
    },
    {
      "entropy": 5.670986127853394,
      "epoch": 4.489749637155297,
      "grad_norm": 1.40625,
      "learning_rate": 0.000314406438141892,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22721543163061142,
      "num_tokens": 102750672.0,
      "step": 49495
    },
    {
      "entropy": 5.633837747573852,
      "epoch": 4.4902031930333814,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003143745814288842,
      "loss": 4.9074,
      "mean_token_accuracy": 0.2187287300825119,
      "num_tokens": 102760364.0,
      "step": 49500
    },
    {
      "entropy": 5.7826902866363525,
      "epoch": 4.490656748911466,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003143427239016001,
      "loss": 4.9536,
      "mean_token_accuracy": 0.22175161987543107,
      "num_tokens": 102770415.0,
      "step": 49505
    },
    {
      "entropy": 5.7089149951934814,
      "epoch": 4.49111030478955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003143108655606983,
      "loss": 4.8541,
      "mean_token_accuracy": 0.23046558499336242,
      "num_tokens": 102781203.0,
      "step": 49510
    },
    {
      "entropy": 5.690456581115723,
      "epoch": 4.491563860667634,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031427900640683783,
      "loss": 4.8318,
      "mean_token_accuracy": 0.227487613260746,
      "num_tokens": 102791071.0,
      "step": 49515
    },
    {
      "entropy": 5.725067853927612,
      "epoch": 4.4920174165457185,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031424714644067754,
      "loss": 4.9477,
      "mean_token_accuracy": 0.21808468401432038,
      "num_tokens": 102801755.0,
      "step": 49520
    },
    {
      "entropy": 5.7322381973266605,
      "epoch": 4.492470972423803,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031421528566287623,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22299295663833618,
      "num_tokens": 102812947.0,
      "step": 49525
    },
    {
      "entropy": 5.7126991748809814,
      "epoch": 4.492924528301887,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003141834240740928,
      "loss": 4.874,
      "mean_token_accuracy": 0.2265447661280632,
      "num_tokens": 102822917.0,
      "step": 49530
    },
    {
      "entropy": 5.712336206436158,
      "epoch": 4.493378084179971,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003141515616749862,
      "loss": 4.873,
      "mean_token_accuracy": 0.22555318772792815,
      "num_tokens": 102833102.0,
      "step": 49535
    },
    {
      "entropy": 5.749135971069336,
      "epoch": 4.493831640058055,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003141196984662153,
      "loss": 4.9064,
      "mean_token_accuracy": 0.22949060201644897,
      "num_tokens": 102843888.0,
      "step": 49540
    },
    {
      "entropy": 5.727388048171997,
      "epoch": 4.494285195936139,
      "grad_norm": 1.25,
      "learning_rate": 0.0003140878344484391,
      "loss": 4.9318,
      "mean_token_accuracy": 0.22388771772384644,
      "num_tokens": 102854417.0,
      "step": 49545
    },
    {
      "entropy": 5.668524885177613,
      "epoch": 4.494738751814223,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003140559696223165,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2322499707341194,
      "num_tokens": 102865206.0,
      "step": 49550
    },
    {
      "entropy": 5.750192594528198,
      "epoch": 4.4951923076923075,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003140241039885065,
      "loss": 4.8486,
      "mean_token_accuracy": 0.2295156314969063,
      "num_tokens": 102874967.0,
      "step": 49555
    },
    {
      "entropy": 5.660081291198731,
      "epoch": 4.495645863570392,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003139922375476681,
      "loss": 4.8323,
      "mean_token_accuracy": 0.22603032290935515,
      "num_tokens": 102884808.0,
      "step": 49560
    },
    {
      "entropy": 5.637185764312744,
      "epoch": 4.496099419448476,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003139603703004603,
      "loss": 4.783,
      "mean_token_accuracy": 0.23519599437713623,
      "num_tokens": 102895206.0,
      "step": 49565
    },
    {
      "entropy": 5.755924081802368,
      "epoch": 4.49655297532656,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003139285022475421,
      "loss": 4.9258,
      "mean_token_accuracy": 0.2242951825261116,
      "num_tokens": 102906658.0,
      "step": 49570
    },
    {
      "entropy": 5.677575492858887,
      "epoch": 4.4970065312046446,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031389663338957254,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2230847030878067,
      "num_tokens": 102917071.0,
      "step": 49575
    },
    {
      "entropy": 5.652672958374024,
      "epoch": 4.497460087082729,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031386476372721067,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22483044266700744,
      "num_tokens": 102927280.0,
      "step": 49580
    },
    {
      "entropy": 5.708342742919922,
      "epoch": 4.497913642960813,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003138328932611156,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21776343435049056,
      "num_tokens": 102937670.0,
      "step": 49585
    },
    {
      "entropy": 5.703975629806519,
      "epoch": 4.498367198838897,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031380102199194646,
      "loss": 4.8957,
      "mean_token_accuracy": 0.22328962683677672,
      "num_tokens": 102947874.0,
      "step": 49590
    },
    {
      "entropy": 5.734690237045288,
      "epoch": 4.498820754716981,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003137691499203622,
      "loss": 4.9035,
      "mean_token_accuracy": 0.22053515017032624,
      "num_tokens": 102959311.0,
      "step": 49595
    },
    {
      "entropy": 5.669805955886841,
      "epoch": 4.499274310595065,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031373727704702217,
      "loss": 4.83,
      "mean_token_accuracy": 0.22890401184558867,
      "num_tokens": 102969866.0,
      "step": 49600
    },
    {
      "entropy": 5.6739161014556885,
      "epoch": 4.499727866473149,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031370540337258523,
      "loss": 4.8077,
      "mean_token_accuracy": 0.23393996953964233,
      "num_tokens": 102979672.0,
      "step": 49605
    },
    {
      "entropy": 5.6604705333709715,
      "epoch": 4.5001814223512335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003136735288977108,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22458747029304504,
      "num_tokens": 102991050.0,
      "step": 49610
    },
    {
      "entropy": 5.687786340713501,
      "epoch": 4.500634978229318,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031364165362305783,
      "loss": 4.9185,
      "mean_token_accuracy": 0.22306693494319915,
      "num_tokens": 103001700.0,
      "step": 49615
    },
    {
      "entropy": 5.708021545410157,
      "epoch": 4.501088534107402,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003136097775492857,
      "loss": 4.917,
      "mean_token_accuracy": 0.22404645681381224,
      "num_tokens": 103013013.0,
      "step": 49620
    },
    {
      "entropy": 5.679308176040649,
      "epoch": 4.501542089985486,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003135779006770535,
      "loss": 4.8966,
      "mean_token_accuracy": 0.22345632016658784,
      "num_tokens": 103024110.0,
      "step": 49625
    },
    {
      "entropy": 5.763753890991211,
      "epoch": 4.501995645863571,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003135460230070205,
      "loss": 4.9102,
      "mean_token_accuracy": 0.21759968250989914,
      "num_tokens": 103034991.0,
      "step": 49630
    },
    {
      "entropy": 5.745695018768311,
      "epoch": 4.502449201741655,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031351414453984595,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22679725736379625,
      "num_tokens": 103045634.0,
      "step": 49635
    },
    {
      "entropy": 5.673786878585815,
      "epoch": 4.502902757619739,
      "grad_norm": 1.34375,
      "learning_rate": 0.000313482265276189,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22825949341058732,
      "num_tokens": 103054999.0,
      "step": 49640
    },
    {
      "entropy": 5.688971328735351,
      "epoch": 4.503356313497823,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031345038521670907,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21776329278945922,
      "num_tokens": 103065163.0,
      "step": 49645
    },
    {
      "entropy": 5.7502340316772464,
      "epoch": 4.503809869375907,
      "grad_norm": 1.375,
      "learning_rate": 0.0003134185043620654,
      "loss": 4.8624,
      "mean_token_accuracy": 0.23368613421916962,
      "num_tokens": 103075979.0,
      "step": 49650
    },
    {
      "entropy": 5.649557256698609,
      "epoch": 4.504263425253991,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003133866227129171,
      "loss": 4.7759,
      "mean_token_accuracy": 0.2281112030148506,
      "num_tokens": 103086124.0,
      "step": 49655
    },
    {
      "entropy": 5.641469478607178,
      "epoch": 4.504716981132075,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031335474026992377,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2246745929121971,
      "num_tokens": 103095330.0,
      "step": 49660
    },
    {
      "entropy": 5.682191610336304,
      "epoch": 4.50517053701016,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003133228570337447,
      "loss": 4.9105,
      "mean_token_accuracy": 0.22421835511922836,
      "num_tokens": 103106863.0,
      "step": 49665
    },
    {
      "entropy": 5.759317779541016,
      "epoch": 4.505624092888244,
      "grad_norm": 1.25,
      "learning_rate": 0.0003132909730050391,
      "loss": 4.9181,
      "mean_token_accuracy": 0.22401736974716185,
      "num_tokens": 103117256.0,
      "step": 49670
    },
    {
      "entropy": 5.731714582443237,
      "epoch": 4.506077648766328,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031325908818446643,
      "loss": 4.875,
      "mean_token_accuracy": 0.22776956409215926,
      "num_tokens": 103127601.0,
      "step": 49675
    },
    {
      "entropy": 5.703935384750366,
      "epoch": 4.506531204644412,
      "grad_norm": 1.5,
      "learning_rate": 0.00031322720257268607,
      "loss": 4.8228,
      "mean_token_accuracy": 0.22095379382371902,
      "num_tokens": 103136868.0,
      "step": 49680
    },
    {
      "entropy": 5.69203634262085,
      "epoch": 4.506984760522497,
      "grad_norm": 1.375,
      "learning_rate": 0.00031319531617035747,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22576983124017716,
      "num_tokens": 103146908.0,
      "step": 49685
    },
    {
      "entropy": 5.638894414901733,
      "epoch": 4.507438316400581,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031316342897813994,
      "loss": 4.8396,
      "mean_token_accuracy": 0.22508679181337357,
      "num_tokens": 103157244.0,
      "step": 49690
    },
    {
      "entropy": 5.638537740707397,
      "epoch": 4.507891872278664,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031313154099669294,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22400934100151063,
      "num_tokens": 103168502.0,
      "step": 49695
    },
    {
      "entropy": 5.6880530834198,
      "epoch": 4.5083454281567485,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003130996522266761,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22116229981184005,
      "num_tokens": 103179015.0,
      "step": 49700
    },
    {
      "entropy": 5.771174955368042,
      "epoch": 4.508798984034833,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003130677626687485,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2276221916079521,
      "num_tokens": 103189200.0,
      "step": 49705
    },
    {
      "entropy": 5.781101131439209,
      "epoch": 4.509252539912917,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003130358723235702,
      "loss": 5.0166,
      "mean_token_accuracy": 0.21430651396512984,
      "num_tokens": 103200683.0,
      "step": 49710
    },
    {
      "entropy": 5.712635803222656,
      "epoch": 4.509706095791001,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031300398119180013,
      "loss": 4.8979,
      "mean_token_accuracy": 0.23180828243494034,
      "num_tokens": 103210992.0,
      "step": 49715
    },
    {
      "entropy": 5.651263904571533,
      "epoch": 4.510159651669086,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003129720892740981,
      "loss": 4.8115,
      "mean_token_accuracy": 0.22605156004428864,
      "num_tokens": 103220783.0,
      "step": 49720
    },
    {
      "entropy": 5.642436981201172,
      "epoch": 4.51061320754717,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003129401965711236,
      "loss": 4.8396,
      "mean_token_accuracy": 0.23195305466651917,
      "num_tokens": 103231593.0,
      "step": 49725
    },
    {
      "entropy": 5.714063453674316,
      "epoch": 4.511066763425254,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003129083030835361,
      "loss": 4.9105,
      "mean_token_accuracy": 0.21947463899850844,
      "num_tokens": 103242416.0,
      "step": 49730
    },
    {
      "entropy": 5.674784517288208,
      "epoch": 4.511520319303338,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00031287640881199534,
      "loss": 4.8102,
      "mean_token_accuracy": 0.22916484922170638,
      "num_tokens": 103252535.0,
      "step": 49735
    },
    {
      "entropy": 5.7499898910522464,
      "epoch": 4.511973875181423,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003128445137571607,
      "loss": 4.9719,
      "mean_token_accuracy": 0.21584209948778152,
      "num_tokens": 103263039.0,
      "step": 49740
    },
    {
      "entropy": 5.713814687728882,
      "epoch": 4.512427431059507,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031281261791969196,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2248384341597557,
      "num_tokens": 103273304.0,
      "step": 49745
    },
    {
      "entropy": 5.74613733291626,
      "epoch": 4.51288098693759,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031278072130024856,
      "loss": 4.9651,
      "mean_token_accuracy": 0.22427776753902434,
      "num_tokens": 103283688.0,
      "step": 49750
    },
    {
      "entropy": 5.7757209777832035,
      "epoch": 4.513334542815675,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003127488238994902,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2219189316034317,
      "num_tokens": 103293572.0,
      "step": 49755
    },
    {
      "entropy": 5.732826089859008,
      "epoch": 4.513788098693759,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003127169257180766,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22303334921598433,
      "num_tokens": 103303474.0,
      "step": 49760
    },
    {
      "entropy": 5.79227499961853,
      "epoch": 4.514241654571843,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003126850267566674,
      "loss": 5.0135,
      "mean_token_accuracy": 0.21207593083381654,
      "num_tokens": 103314343.0,
      "step": 49765
    },
    {
      "entropy": 5.699454212188721,
      "epoch": 4.514695210449927,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003126531270159221,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2286601722240448,
      "num_tokens": 103323942.0,
      "step": 49770
    },
    {
      "entropy": 5.720363616943359,
      "epoch": 4.515148766328012,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003126212264965007,
      "loss": 4.8918,
      "mean_token_accuracy": 0.21938391625881196,
      "num_tokens": 103333785.0,
      "step": 49775
    },
    {
      "entropy": 5.627668762207032,
      "epoch": 4.515602322206096,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031258932519906266,
      "loss": 4.8011,
      "mean_token_accuracy": 0.2325948655605316,
      "num_tokens": 103344184.0,
      "step": 49780
    },
    {
      "entropy": 5.713441658020019,
      "epoch": 4.51605587808418,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031255742312426786,
      "loss": 4.94,
      "mean_token_accuracy": 0.22078813910484313,
      "num_tokens": 103355085.0,
      "step": 49785
    },
    {
      "entropy": 5.639127111434936,
      "epoch": 4.5165094339622645,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031252552027277595,
      "loss": 4.8252,
      "mean_token_accuracy": 0.22836390733718873,
      "num_tokens": 103365942.0,
      "step": 49790
    },
    {
      "entropy": 5.6651064395904545,
      "epoch": 4.516962989840349,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003124936166452467,
      "loss": 4.8043,
      "mean_token_accuracy": 0.23786956071853638,
      "num_tokens": 103375844.0,
      "step": 49795
    },
    {
      "entropy": 5.702456521987915,
      "epoch": 4.517416545718433,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003124617122423399,
      "loss": 4.9025,
      "mean_token_accuracy": 0.2322573482990265,
      "num_tokens": 103385871.0,
      "step": 49800
    },
    {
      "entropy": 5.661716794967651,
      "epoch": 4.517870101596516,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003124298070647154,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22489600926637648,
      "num_tokens": 103396888.0,
      "step": 49805
    },
    {
      "entropy": 5.68301477432251,
      "epoch": 4.518323657474601,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003123979011130329,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22625461518764495,
      "num_tokens": 103406913.0,
      "step": 49810
    },
    {
      "entropy": 5.684360218048096,
      "epoch": 4.518777213352685,
      "grad_norm": 1.25,
      "learning_rate": 0.0003123659943879523,
      "loss": 4.8186,
      "mean_token_accuracy": 0.23401186168193816,
      "num_tokens": 103416485.0,
      "step": 49815
    },
    {
      "entropy": 5.647102785110474,
      "epoch": 4.519230769230769,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003123340868901334,
      "loss": 4.7961,
      "mean_token_accuracy": 0.22966784685850145,
      "num_tokens": 103427476.0,
      "step": 49820
    },
    {
      "entropy": 5.650262403488159,
      "epoch": 4.519684325108853,
      "grad_norm": 1.375,
      "learning_rate": 0.00031230217862023604,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22766775041818618,
      "num_tokens": 103437500.0,
      "step": 49825
    },
    {
      "entropy": 5.681860876083374,
      "epoch": 4.520137880986938,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031227026957892013,
      "loss": 4.821,
      "mean_token_accuracy": 0.23005252182483674,
      "num_tokens": 103447646.0,
      "step": 49830
    },
    {
      "entropy": 5.683729219436645,
      "epoch": 4.520591436865022,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031223835976684555,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22656139880418777,
      "num_tokens": 103457698.0,
      "step": 49835
    },
    {
      "entropy": 5.741828155517578,
      "epoch": 4.521044992743106,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031220644918467224,
      "loss": 4.9198,
      "mean_token_accuracy": 0.23020777404308318,
      "num_tokens": 103467566.0,
      "step": 49840
    },
    {
      "entropy": 5.718018293380737,
      "epoch": 4.5214985486211905,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031217453783306007,
      "loss": 4.8902,
      "mean_token_accuracy": 0.22020101249217988,
      "num_tokens": 103477724.0,
      "step": 49845
    },
    {
      "entropy": 5.705986976623535,
      "epoch": 4.521952104499274,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003121426257126689,
      "loss": 4.8927,
      "mean_token_accuracy": 0.22343249022960662,
      "num_tokens": 103487864.0,
      "step": 49850
    },
    {
      "entropy": 5.648791170120239,
      "epoch": 4.522405660377358,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031211071282415885,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23350208401679992,
      "num_tokens": 103497229.0,
      "step": 49855
    },
    {
      "entropy": 5.646293449401855,
      "epoch": 4.522859216255442,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003120787991681898,
      "loss": 4.8021,
      "mean_token_accuracy": 0.23637476712465286,
      "num_tokens": 103506984.0,
      "step": 49860
    },
    {
      "entropy": 5.654769706726074,
      "epoch": 4.523312772133527,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003120468847454217,
      "loss": 4.8389,
      "mean_token_accuracy": 0.22460144460201265,
      "num_tokens": 103517287.0,
      "step": 49865
    },
    {
      "entropy": 5.663955926895142,
      "epoch": 4.523766328011611,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031201496955651466,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2247145101428032,
      "num_tokens": 103528575.0,
      "step": 49870
    },
    {
      "entropy": 5.740159034729004,
      "epoch": 4.524219883889695,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031198305360212854,
      "loss": 4.9817,
      "mean_token_accuracy": 0.21788249909877777,
      "num_tokens": 103538406.0,
      "step": 49875
    },
    {
      "entropy": 5.706613540649414,
      "epoch": 4.5246734397677795,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003119511368829235,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22481988221406937,
      "num_tokens": 103548738.0,
      "step": 49880
    },
    {
      "entropy": 5.652500629425049,
      "epoch": 4.525126995645864,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031191921939955947,
      "loss": 4.8175,
      "mean_token_accuracy": 0.22960299402475357,
      "num_tokens": 103558755.0,
      "step": 49885
    },
    {
      "entropy": 5.74186520576477,
      "epoch": 4.525580551523948,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003118873011526966,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2260322853922844,
      "num_tokens": 103569115.0,
      "step": 49890
    },
    {
      "entropy": 5.772779083251953,
      "epoch": 4.526034107402032,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003118553821429949,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22597311139106752,
      "num_tokens": 103580453.0,
      "step": 49895
    },
    {
      "entropy": 5.6915411949157715,
      "epoch": 4.5264876632801165,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003118234623711146,
      "loss": 4.9284,
      "mean_token_accuracy": 0.22523769587278367,
      "num_tokens": 103592995.0,
      "step": 49900
    },
    {
      "entropy": 5.634797382354736,
      "epoch": 4.5269412191582,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00031179154183771576,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23130361139774322,
      "num_tokens": 103603496.0,
      "step": 49905
    },
    {
      "entropy": 5.752328538894654,
      "epoch": 4.527394775036284,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003117596205434583,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22518282681703566,
      "num_tokens": 103615028.0,
      "step": 49910
    },
    {
      "entropy": 5.703857040405273,
      "epoch": 4.5278483309143684,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031172769848900256,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2244715839624405,
      "num_tokens": 103625430.0,
      "step": 49915
    },
    {
      "entropy": 5.7415163040161135,
      "epoch": 4.528301886792453,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003116957756750087,
      "loss": 4.9149,
      "mean_token_accuracy": 0.22314620316028594,
      "num_tokens": 103636409.0,
      "step": 49920
    },
    {
      "entropy": 5.702204370498658,
      "epoch": 4.528755442670537,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031166385210213684,
      "loss": 4.9327,
      "mean_token_accuracy": 0.23518973141908645,
      "num_tokens": 103646869.0,
      "step": 49925
    },
    {
      "entropy": 5.655885791778564,
      "epoch": 4.529208998548621,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031163192777104715,
      "loss": 4.7615,
      "mean_token_accuracy": 0.23136031329631807,
      "num_tokens": 103657098.0,
      "step": 49930
    },
    {
      "entropy": 5.7223052978515625,
      "epoch": 4.5296625544267055,
      "grad_norm": 1.375,
      "learning_rate": 0.0003116000026823999,
      "loss": 4.9427,
      "mean_token_accuracy": 0.22613623589277268,
      "num_tokens": 103667331.0,
      "step": 49935
    },
    {
      "entropy": 5.638739013671875,
      "epoch": 4.53011611030479,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003115680768368552,
      "loss": 4.8202,
      "mean_token_accuracy": 0.22776037752628325,
      "num_tokens": 103676755.0,
      "step": 49940
    },
    {
      "entropy": 5.651150560379028,
      "epoch": 4.530569666182874,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003115361502350734,
      "loss": 4.804,
      "mean_token_accuracy": 0.22751510739326478,
      "num_tokens": 103686334.0,
      "step": 49945
    },
    {
      "entropy": 5.677527523040771,
      "epoch": 4.531023222060957,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003115042228777148,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2246727928519249,
      "num_tokens": 103697127.0,
      "step": 49950
    },
    {
      "entropy": 5.635869264602661,
      "epoch": 4.531476777939043,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031147229476543935,
      "loss": 4.8412,
      "mean_token_accuracy": 0.2181378036737442,
      "num_tokens": 103707860.0,
      "step": 49955
    },
    {
      "entropy": 5.743034172058105,
      "epoch": 4.531930333817126,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031144036589890765,
      "loss": 4.9512,
      "mean_token_accuracy": 0.21722831428050995,
      "num_tokens": 103718816.0,
      "step": 49960
    },
    {
      "entropy": 5.649497938156128,
      "epoch": 4.53238388969521,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031140843627878003,
      "loss": 4.804,
      "mean_token_accuracy": 0.23024368584156035,
      "num_tokens": 103729203.0,
      "step": 49965
    },
    {
      "entropy": 5.75989089012146,
      "epoch": 4.5328374455732945,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031137650590571655,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2238464131951332,
      "num_tokens": 103740830.0,
      "step": 49970
    },
    {
      "entropy": 5.680646276473999,
      "epoch": 4.533291001451379,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003113445747803776,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22337727397680282,
      "num_tokens": 103752881.0,
      "step": 49975
    },
    {
      "entropy": 5.693352508544922,
      "epoch": 4.533744557329463,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003113126429034237,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22584230005741118,
      "num_tokens": 103762875.0,
      "step": 49980
    },
    {
      "entropy": 5.733663749694824,
      "epoch": 4.534198113207547,
      "grad_norm": 1.375,
      "learning_rate": 0.00031128071027551504,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22276131808757782,
      "num_tokens": 103772147.0,
      "step": 49985
    },
    {
      "entropy": 5.63450255393982,
      "epoch": 4.5346516690856316,
      "grad_norm": 1.3125,
      "learning_rate": 0.000311248776897312,
      "loss": 4.8218,
      "mean_token_accuracy": 0.2228258341550827,
      "num_tokens": 103782904.0,
      "step": 49990
    },
    {
      "entropy": 5.631860828399658,
      "epoch": 4.535105224963716,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031121684276947504,
      "loss": 4.8553,
      "mean_token_accuracy": 0.22851039171218873,
      "num_tokens": 103793202.0,
      "step": 49995
    },
    {
      "entropy": 5.810676193237304,
      "epoch": 4.5355587808418,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003111849078926646,
      "loss": 5.0637,
      "mean_token_accuracy": 0.2196551337838173,
      "num_tokens": 103805796.0,
      "step": 50000
    },
    {
      "entropy": 5.7590110301971436,
      "epoch": 4.5360123367198835,
      "grad_norm": 1.28125,
      "learning_rate": 0.000311152972267541,
      "loss": 4.9149,
      "mean_token_accuracy": 0.21807679086923598,
      "num_tokens": 103816622.0,
      "step": 50005
    },
    {
      "entropy": 5.703660202026367,
      "epoch": 4.536465892597968,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003111210358947647,
      "loss": 4.8066,
      "mean_token_accuracy": 0.22980775237083434,
      "num_tokens": 103826174.0,
      "step": 50010
    },
    {
      "entropy": 5.7522071361541744,
      "epoch": 4.536919448476052,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003110890987749962,
      "loss": 4.9428,
      "mean_token_accuracy": 0.22741507291793822,
      "num_tokens": 103836217.0,
      "step": 50015
    },
    {
      "entropy": 5.652793264389038,
      "epoch": 4.537373004354136,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003110571609088959,
      "loss": 4.7767,
      "mean_token_accuracy": 0.23268394768238068,
      "num_tokens": 103846661.0,
      "step": 50020
    },
    {
      "entropy": 5.697833061218262,
      "epoch": 4.5378265602322205,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003110252222971244,
      "loss": 4.932,
      "mean_token_accuracy": 0.22303052395582199,
      "num_tokens": 103857323.0,
      "step": 50025
    },
    {
      "entropy": 5.737350606918335,
      "epoch": 4.538280116110305,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003109932829403421,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2191044881939888,
      "num_tokens": 103867409.0,
      "step": 50030
    },
    {
      "entropy": 5.75800256729126,
      "epoch": 4.538733671988389,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003109613428392095,
      "loss": 4.899,
      "mean_token_accuracy": 0.22553782612085344,
      "num_tokens": 103877603.0,
      "step": 50035
    },
    {
      "entropy": 5.7097025394439695,
      "epoch": 4.539187227866473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031092940199438715,
      "loss": 4.9278,
      "mean_token_accuracy": 0.2236072301864624,
      "num_tokens": 103888236.0,
      "step": 50040
    },
    {
      "entropy": 5.6817139148712155,
      "epoch": 4.539640783744558,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003108974604065357,
      "loss": 4.8815,
      "mean_token_accuracy": 0.2291667252779007,
      "num_tokens": 103898474.0,
      "step": 50045
    },
    {
      "entropy": 5.664478445053101,
      "epoch": 4.540094339622642,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031086551807631553,
      "loss": 4.8634,
      "mean_token_accuracy": 0.23300401419401168,
      "num_tokens": 103909854.0,
      "step": 50050
    },
    {
      "entropy": 5.668638277053833,
      "epoch": 4.540547895500726,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031083357500438736,
      "loss": 4.876,
      "mean_token_accuracy": 0.2221869483590126,
      "num_tokens": 103919227.0,
      "step": 50055
    },
    {
      "entropy": 5.642814636230469,
      "epoch": 4.5410014513788095,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031080163119141167,
      "loss": 4.8324,
      "mean_token_accuracy": 0.23698274791240692,
      "num_tokens": 103929022.0,
      "step": 50060
    },
    {
      "entropy": 5.704234600067139,
      "epoch": 4.541455007256894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031076968663804917,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22372479140758514,
      "num_tokens": 103939441.0,
      "step": 50065
    },
    {
      "entropy": 5.666438913345337,
      "epoch": 4.541908563134978,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031073774134496047,
      "loss": 4.8345,
      "mean_token_accuracy": 0.222641783952713,
      "num_tokens": 103950312.0,
      "step": 50070
    },
    {
      "entropy": 5.696568584442138,
      "epoch": 4.542362119013062,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003107057953128061,
      "loss": 4.8911,
      "mean_token_accuracy": 0.2223531112074852,
      "num_tokens": 103960415.0,
      "step": 50075
    },
    {
      "entropy": 5.656058835983276,
      "epoch": 4.542815674891147,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003106738485422468,
      "loss": 4.8281,
      "mean_token_accuracy": 0.2276388868689537,
      "num_tokens": 103971183.0,
      "step": 50080
    },
    {
      "entropy": 5.769261455535888,
      "epoch": 4.543269230769231,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003106419010339433,
      "loss": 4.9629,
      "mean_token_accuracy": 0.2153464823961258,
      "num_tokens": 103982120.0,
      "step": 50085
    },
    {
      "entropy": 5.698762702941894,
      "epoch": 4.543722786647315,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031060995278855616,
      "loss": 4.8115,
      "mean_token_accuracy": 0.2342630997300148,
      "num_tokens": 103992116.0,
      "step": 50090
    },
    {
      "entropy": 5.6604996681213375,
      "epoch": 4.544176342525399,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031057800380674604,
      "loss": 4.8971,
      "mean_token_accuracy": 0.22539713382720947,
      "num_tokens": 104002646.0,
      "step": 50095
    },
    {
      "entropy": 5.711823415756226,
      "epoch": 4.544629898403484,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003105460540891738,
      "loss": 4.9498,
      "mean_token_accuracy": 0.21851164400577544,
      "num_tokens": 104013158.0,
      "step": 50100
    },
    {
      "entropy": 5.684101009368897,
      "epoch": 4.545083454281567,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003105141036365001,
      "loss": 4.8695,
      "mean_token_accuracy": 0.22324619442224503,
      "num_tokens": 104023485.0,
      "step": 50105
    },
    {
      "entropy": 5.729295349121093,
      "epoch": 4.545537010159651,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003104821524493857,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22723016142845154,
      "num_tokens": 104034588.0,
      "step": 50110
    },
    {
      "entropy": 5.743296337127686,
      "epoch": 4.5459905660377355,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003104502005284913,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2249019116163254,
      "num_tokens": 104045081.0,
      "step": 50115
    },
    {
      "entropy": 5.721516370773315,
      "epoch": 4.54644412191582,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003104182478744778,
      "loss": 4.842,
      "mean_token_accuracy": 0.2247536987066269,
      "num_tokens": 104056166.0,
      "step": 50120
    },
    {
      "entropy": 5.685320663452148,
      "epoch": 4.546897677793904,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003103862944880058,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2316370353102684,
      "num_tokens": 104066699.0,
      "step": 50125
    },
    {
      "entropy": 5.681048822402954,
      "epoch": 4.547351233671988,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003103543403697363,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22821453213691711,
      "num_tokens": 104077061.0,
      "step": 50130
    },
    {
      "entropy": 5.705513048171997,
      "epoch": 4.547804789550073,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003103223855203301,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21938000172376632,
      "num_tokens": 104087877.0,
      "step": 50135
    },
    {
      "entropy": 5.770837306976318,
      "epoch": 4.548258345428157,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031029042994044784,
      "loss": 4.9645,
      "mean_token_accuracy": 0.21957058161497117,
      "num_tokens": 104099483.0,
      "step": 50140
    },
    {
      "entropy": 5.693341636657715,
      "epoch": 4.548711901306241,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031025847363075056,
      "loss": 4.8409,
      "mean_token_accuracy": 0.22161760032176972,
      "num_tokens": 104109530.0,
      "step": 50145
    },
    {
      "entropy": 5.661199283599854,
      "epoch": 4.549165457184325,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031022651659189907,
      "loss": 4.8903,
      "mean_token_accuracy": 0.22210443168878555,
      "num_tokens": 104119610.0,
      "step": 50150
    },
    {
      "entropy": 5.7157995223999025,
      "epoch": 4.54961901306241,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003101945588245541,
      "loss": 4.936,
      "mean_token_accuracy": 0.21573855578899384,
      "num_tokens": 104130337.0,
      "step": 50155
    },
    {
      "entropy": 5.767745399475098,
      "epoch": 4.550072568940493,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003101626003293768,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21958310157060623,
      "num_tokens": 104139462.0,
      "step": 50160
    },
    {
      "entropy": 5.684769916534424,
      "epoch": 4.550526124818577,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003101306411070279,
      "loss": 4.7898,
      "mean_token_accuracy": 0.23763093054294587,
      "num_tokens": 104148640.0,
      "step": 50165
    },
    {
      "entropy": 5.686920022964477,
      "epoch": 4.550979680696662,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031009868115816847,
      "loss": 4.9021,
      "mean_token_accuracy": 0.23107923269271852,
      "num_tokens": 104159400.0,
      "step": 50170
    },
    {
      "entropy": 5.662400341033935,
      "epoch": 4.551433236574746,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031006672048345936,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23131699562072755,
      "num_tokens": 104170149.0,
      "step": 50175
    },
    {
      "entropy": 5.641582775115967,
      "epoch": 4.55188679245283,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003100347590835615,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23390598744153976,
      "num_tokens": 104180787.0,
      "step": 50180
    },
    {
      "entropy": 5.7260003089904785,
      "epoch": 4.552340348330914,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003100027969591359,
      "loss": 4.906,
      "mean_token_accuracy": 0.22534482628107072,
      "num_tokens": 104190961.0,
      "step": 50185
    },
    {
      "entropy": 5.676319360733032,
      "epoch": 4.552793904208999,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030997083411084356,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23051350861787795,
      "num_tokens": 104201886.0,
      "step": 50190
    },
    {
      "entropy": 5.736472225189209,
      "epoch": 4.553247460087083,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003099388705393454,
      "loss": 4.883,
      "mean_token_accuracy": 0.23167929649353028,
      "num_tokens": 104212834.0,
      "step": 50195
    },
    {
      "entropy": 5.690448141098022,
      "epoch": 4.553701015965167,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003099069062453025,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22715362310409545,
      "num_tokens": 104222384.0,
      "step": 50200
    },
    {
      "entropy": 5.645592594146729,
      "epoch": 4.5541545718432515,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030987494122937604,
      "loss": 4.8425,
      "mean_token_accuracy": 0.22933519780635833,
      "num_tokens": 104232971.0,
      "step": 50205
    },
    {
      "entropy": 5.712569761276245,
      "epoch": 4.554608127721336,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003098429754922267,
      "loss": 4.7986,
      "mean_token_accuracy": 0.22562622874975205,
      "num_tokens": 104245239.0,
      "step": 50210
    },
    {
      "entropy": 5.739643096923828,
      "epoch": 4.555061683599419,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030981100903451587,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2207686960697174,
      "num_tokens": 104256323.0,
      "step": 50215
    },
    {
      "entropy": 5.74214243888855,
      "epoch": 4.555515239477503,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003097790418569045,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22563227117061616,
      "num_tokens": 104267106.0,
      "step": 50220
    },
    {
      "entropy": 5.695002508163452,
      "epoch": 4.555968795355588,
      "grad_norm": 1.375,
      "learning_rate": 0.0003097470739600536,
      "loss": 4.8852,
      "mean_token_accuracy": 0.22683780044317245,
      "num_tokens": 104277336.0,
      "step": 50225
    },
    {
      "entropy": 5.69345989227295,
      "epoch": 4.556422351233672,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003097151053446244,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23037983626127242,
      "num_tokens": 104288809.0,
      "step": 50230
    },
    {
      "entropy": 5.6774908065795895,
      "epoch": 4.556875907111756,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003096831360112781,
      "loss": 4.8913,
      "mean_token_accuracy": 0.21912351101636887,
      "num_tokens": 104299434.0,
      "step": 50235
    },
    {
      "entropy": 5.628775358200073,
      "epoch": 4.55732946298984,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003096511659606756,
      "loss": 4.8173,
      "mean_token_accuracy": 0.22721707373857497,
      "num_tokens": 104309426.0,
      "step": 50240
    },
    {
      "entropy": 5.628371047973633,
      "epoch": 4.557783018867925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003096191951934781,
      "loss": 4.82,
      "mean_token_accuracy": 0.22716917991638183,
      "num_tokens": 104319530.0,
      "step": 50245
    },
    {
      "entropy": 5.668262958526611,
      "epoch": 4.558236574746009,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003095872237103468,
      "loss": 4.7954,
      "mean_token_accuracy": 0.23129143565893173,
      "num_tokens": 104329198.0,
      "step": 50250
    },
    {
      "entropy": 5.758710336685181,
      "epoch": 4.558690130624093,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030955525151194304,
      "loss": 4.9674,
      "mean_token_accuracy": 0.22428909689188004,
      "num_tokens": 104340481.0,
      "step": 50255
    },
    {
      "entropy": 5.757569026947022,
      "epoch": 4.559143686502177,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003095232785989277,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2262146443128586,
      "num_tokens": 104350382.0,
      "step": 50260
    },
    {
      "entropy": 5.685077428817749,
      "epoch": 4.559597242380261,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003094913049719623,
      "loss": 4.7815,
      "mean_token_accuracy": 0.2282022312283516,
      "num_tokens": 104360967.0,
      "step": 50265
    },
    {
      "entropy": 5.674678373336792,
      "epoch": 4.560050798258345,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030945933063170793,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22428934276103973,
      "num_tokens": 104371283.0,
      "step": 50270
    },
    {
      "entropy": 5.68880524635315,
      "epoch": 4.560504354136429,
      "grad_norm": 1.25,
      "learning_rate": 0.00030942735557882575,
      "loss": 4.8576,
      "mean_token_accuracy": 0.22517330199480057,
      "num_tokens": 104381545.0,
      "step": 50275
    },
    {
      "entropy": 5.760785913467407,
      "epoch": 4.560957910014514,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030939537981397716,
      "loss": 4.9741,
      "mean_token_accuracy": 0.21582166701555253,
      "num_tokens": 104392533.0,
      "step": 50280
    },
    {
      "entropy": 5.727894449234009,
      "epoch": 4.561411465892598,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030936340333782326,
      "loss": 4.9289,
      "mean_token_accuracy": 0.22064250707626343,
      "num_tokens": 104402230.0,
      "step": 50285
    },
    {
      "entropy": 5.724709701538086,
      "epoch": 4.561865021770682,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003093314261510254,
      "loss": 4.8553,
      "mean_token_accuracy": 0.22923006117343903,
      "num_tokens": 104412029.0,
      "step": 50290
    },
    {
      "entropy": 5.646293687820434,
      "epoch": 4.5623185776487665,
      "grad_norm": 1.390625,
      "learning_rate": 0.000309299448254245,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22563476115465164,
      "num_tokens": 104422107.0,
      "step": 50295
    },
    {
      "entropy": 5.712235307693481,
      "epoch": 4.562772133526851,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030926746964814324,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22536336183547973,
      "num_tokens": 104432073.0,
      "step": 50300
    },
    {
      "entropy": 5.68666820526123,
      "epoch": 4.563225689404935,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030923549033338147,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22781168967485427,
      "num_tokens": 104441998.0,
      "step": 50305
    },
    {
      "entropy": 5.595049524307251,
      "epoch": 4.563679245283019,
      "grad_norm": 1.3125,
      "learning_rate": 0.000309203510310621,
      "loss": 4.7468,
      "mean_token_accuracy": 0.23840146660804748,
      "num_tokens": 104451474.0,
      "step": 50310
    },
    {
      "entropy": 5.658726930618286,
      "epoch": 4.564132801161103,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003091715295805233,
      "loss": 4.7926,
      "mean_token_accuracy": 0.23647590428590776,
      "num_tokens": 104461351.0,
      "step": 50315
    },
    {
      "entropy": 5.722055435180664,
      "epoch": 4.564586357039187,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003091395481437496,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22799201756715776,
      "num_tokens": 104471013.0,
      "step": 50320
    },
    {
      "entropy": 5.726309108734131,
      "epoch": 4.565039912917271,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003091075660009614,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22307423651218414,
      "num_tokens": 104481863.0,
      "step": 50325
    },
    {
      "entropy": 5.716893196105957,
      "epoch": 4.5654934687953554,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030907558315282,
      "loss": 4.8784,
      "mean_token_accuracy": 0.21959338784217836,
      "num_tokens": 104492718.0,
      "step": 50330
    },
    {
      "entropy": 5.696963977813721,
      "epoch": 4.56594702467344,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030904359959998685,
      "loss": 4.8466,
      "mean_token_accuracy": 0.22843571454286576,
      "num_tokens": 104502837.0,
      "step": 50335
    },
    {
      "entropy": 5.718865394592285,
      "epoch": 4.566400580551524,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003090116153431234,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2282785639166832,
      "num_tokens": 104514202.0,
      "step": 50340
    },
    {
      "entropy": 5.727221155166626,
      "epoch": 4.566854136429608,
      "grad_norm": 1.515625,
      "learning_rate": 0.000308979630382891,
      "loss": 4.8889,
      "mean_token_accuracy": 0.22341345399618148,
      "num_tokens": 104525037.0,
      "step": 50345
    },
    {
      "entropy": 5.613678646087647,
      "epoch": 4.5673076923076925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030894764471995137,
      "loss": 4.8151,
      "mean_token_accuracy": 0.2274194359779358,
      "num_tokens": 104534883.0,
      "step": 50350
    },
    {
      "entropy": 5.646625232696533,
      "epoch": 4.567761248185777,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003089156583549657,
      "loss": 4.8588,
      "mean_token_accuracy": 0.23018560707569122,
      "num_tokens": 104544845.0,
      "step": 50355
    },
    {
      "entropy": 5.721243381500244,
      "epoch": 4.568214804063861,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003088836712885955,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22398525029420852,
      "num_tokens": 104555654.0,
      "step": 50360
    },
    {
      "entropy": 5.683901834487915,
      "epoch": 4.568668359941945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030885168352150243,
      "loss": 4.91,
      "mean_token_accuracy": 0.21804360747337342,
      "num_tokens": 104565015.0,
      "step": 50365
    },
    {
      "entropy": 5.607568788528442,
      "epoch": 4.569121915820029,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030881969505434795,
      "loss": 4.8305,
      "mean_token_accuracy": 0.22750867158174515,
      "num_tokens": 104575934.0,
      "step": 50370
    },
    {
      "entropy": 5.623124790191651,
      "epoch": 4.569575471698113,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030878770588779335,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23078940957784652,
      "num_tokens": 104585915.0,
      "step": 50375
    },
    {
      "entropy": 5.749627542495728,
      "epoch": 4.570029027576197,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030875571602250064,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22110754400491714,
      "num_tokens": 104595605.0,
      "step": 50380
    },
    {
      "entropy": 5.738490343093872,
      "epoch": 4.5704825834542815,
      "grad_norm": 1.21875,
      "learning_rate": 0.000308723725459131,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21640095561742784,
      "num_tokens": 104608251.0,
      "step": 50385
    },
    {
      "entropy": 5.736108827590942,
      "epoch": 4.570936139332366,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003086917341983461,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2344186782836914,
      "num_tokens": 104618586.0,
      "step": 50390
    },
    {
      "entropy": 5.6727721214294435,
      "epoch": 4.57138969521045,
      "grad_norm": 2.25,
      "learning_rate": 0.0003086597422408076,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22626133561134337,
      "num_tokens": 104629718.0,
      "step": 50395
    },
    {
      "entropy": 5.650493383407593,
      "epoch": 4.571843251088534,
      "grad_norm": 1.25,
      "learning_rate": 0.000308627749587177,
      "loss": 4.8379,
      "mean_token_accuracy": 0.22570976614952087,
      "num_tokens": 104639923.0,
      "step": 50400
    },
    {
      "entropy": 5.678637409210205,
      "epoch": 4.5722968069666186,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000308595756238116,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2239961802959442,
      "num_tokens": 104650890.0,
      "step": 50405
    },
    {
      "entropy": 5.6807976245880125,
      "epoch": 4.572750362844703,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003085637621942862,
      "loss": 4.8411,
      "mean_token_accuracy": 0.2321598008275032,
      "num_tokens": 104661231.0,
      "step": 50410
    },
    {
      "entropy": 5.668264627456665,
      "epoch": 4.573203918722786,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030853176745634925,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2267017975449562,
      "num_tokens": 104671641.0,
      "step": 50415
    },
    {
      "entropy": 5.680671501159668,
      "epoch": 4.5736574746008705,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003084997720249668,
      "loss": 4.8626,
      "mean_token_accuracy": 0.22951533943414687,
      "num_tokens": 104682310.0,
      "step": 50420
    },
    {
      "entropy": 5.751889896392822,
      "epoch": 4.574111030478955,
      "grad_norm": 1.375,
      "learning_rate": 0.00030846777590080057,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21732309311628342,
      "num_tokens": 104692414.0,
      "step": 50425
    },
    {
      "entropy": 5.679687643051148,
      "epoch": 4.574564586357039,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030843577908451215,
      "loss": 4.9262,
      "mean_token_accuracy": 0.221542689204216,
      "num_tokens": 104703617.0,
      "step": 50430
    },
    {
      "entropy": 5.62609224319458,
      "epoch": 4.575018142235123,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003084037815767633,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22741172164678575,
      "num_tokens": 104713494.0,
      "step": 50435
    },
    {
      "entropy": 5.706634044647217,
      "epoch": 4.5754716981132075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003083717833782158,
      "loss": 4.8297,
      "mean_token_accuracy": 0.2256772518157959,
      "num_tokens": 104723349.0,
      "step": 50440
    },
    {
      "entropy": 5.682857084274292,
      "epoch": 4.575925253991292,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003083397844895312,
      "loss": 4.7979,
      "mean_token_accuracy": 0.23713806569576262,
      "num_tokens": 104732903.0,
      "step": 50445
    },
    {
      "entropy": 5.6154913902282715,
      "epoch": 4.576378809869376,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003083077849113715,
      "loss": 4.8192,
      "mean_token_accuracy": 0.2314871519804001,
      "num_tokens": 104742533.0,
      "step": 50450
    },
    {
      "entropy": 5.62505784034729,
      "epoch": 4.57683236574746,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030827578464439823,
      "loss": 4.7589,
      "mean_token_accuracy": 0.2267906352877617,
      "num_tokens": 104752161.0,
      "step": 50455
    },
    {
      "entropy": 5.674819707870483,
      "epoch": 4.577285921625545,
      "grad_norm": 1.375,
      "learning_rate": 0.00030824378368927324,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2339810013771057,
      "num_tokens": 104762630.0,
      "step": 50460
    },
    {
      "entropy": 5.662899208068848,
      "epoch": 4.577739477503629,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030821178204665836,
      "loss": 4.8145,
      "mean_token_accuracy": 0.23296123445034028,
      "num_tokens": 104773529.0,
      "step": 50465
    },
    {
      "entropy": 5.660335826873779,
      "epoch": 4.578193033381712,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003081797797172154,
      "loss": 4.8005,
      "mean_token_accuracy": 0.2217215746641159,
      "num_tokens": 104783386.0,
      "step": 50470
    },
    {
      "entropy": 5.68530535697937,
      "epoch": 4.5786465892597965,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003081477767016061,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22392311245203017,
      "num_tokens": 104793254.0,
      "step": 50475
    },
    {
      "entropy": 5.677037239074707,
      "epoch": 4.579100145137881,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030811577300049226,
      "loss": 4.896,
      "mean_token_accuracy": 0.22788870483636856,
      "num_tokens": 104804345.0,
      "step": 50480
    },
    {
      "entropy": 5.666444158554077,
      "epoch": 4.579553701015965,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003080837686145359,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22501977682113647,
      "num_tokens": 104814891.0,
      "step": 50485
    },
    {
      "entropy": 5.652721738815307,
      "epoch": 4.580007256894049,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003080517635443987,
      "loss": 4.8217,
      "mean_token_accuracy": 0.22205309569835663,
      "num_tokens": 104825015.0,
      "step": 50490
    },
    {
      "entropy": 5.712806940078735,
      "epoch": 4.580460812772134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003080197577907427,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21813639551401137,
      "num_tokens": 104835735.0,
      "step": 50495
    },
    {
      "entropy": 5.678415536880493,
      "epoch": 4.580914368650218,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030798775135422966,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22211802005767822,
      "num_tokens": 104847484.0,
      "step": 50500
    },
    {
      "entropy": 5.697001361846924,
      "epoch": 4.581367924528302,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003079557442355215,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2252477526664734,
      "num_tokens": 104857783.0,
      "step": 50505
    },
    {
      "entropy": 5.6587347984313965,
      "epoch": 4.581821480406386,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003079237364352801,
      "loss": 4.8042,
      "mean_token_accuracy": 0.23977857530117036,
      "num_tokens": 104867498.0,
      "step": 50510
    },
    {
      "entropy": 5.718320560455322,
      "epoch": 4.58227503628447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003078917279541675,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2264880895614624,
      "num_tokens": 104876505.0,
      "step": 50515
    },
    {
      "entropy": 5.70214877128601,
      "epoch": 4.582728592162555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003078597187928456,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21839916408061982,
      "num_tokens": 104886932.0,
      "step": 50520
    },
    {
      "entropy": 5.723393344879151,
      "epoch": 4.583182148040638,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003078277089519764,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2327924907207489,
      "num_tokens": 104896733.0,
      "step": 50525
    },
    {
      "entropy": 5.707929468154907,
      "epoch": 4.5836357039187225,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030779569843222165,
      "loss": 4.9065,
      "mean_token_accuracy": 0.21330433785915376,
      "num_tokens": 104906846.0,
      "step": 50530
    },
    {
      "entropy": 5.583409023284912,
      "epoch": 4.584089259796807,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030776368723424365,
      "loss": 4.6935,
      "mean_token_accuracy": 0.24185246527194976,
      "num_tokens": 104916136.0,
      "step": 50535
    },
    {
      "entropy": 5.663533306121826,
      "epoch": 4.584542815674891,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003077316753587042,
      "loss": 4.8326,
      "mean_token_accuracy": 0.23333924561738967,
      "num_tokens": 104926145.0,
      "step": 50540
    },
    {
      "entropy": 5.726410245895385,
      "epoch": 4.584996371552975,
      "grad_norm": 1.5,
      "learning_rate": 0.00030769966280626526,
      "loss": 4.9456,
      "mean_token_accuracy": 0.22015639543533325,
      "num_tokens": 104935576.0,
      "step": 50545
    },
    {
      "entropy": 5.655355834960938,
      "epoch": 4.58544992743106,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030766764957758906,
      "loss": 4.7555,
      "mean_token_accuracy": 0.2364237368106842,
      "num_tokens": 104944846.0,
      "step": 50550
    },
    {
      "entropy": 5.671792221069336,
      "epoch": 4.585903483309144,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030763563567333754,
      "loss": 4.8176,
      "mean_token_accuracy": 0.23094723820686341,
      "num_tokens": 104954604.0,
      "step": 50555
    },
    {
      "entropy": 5.682136487960816,
      "epoch": 4.586357039187228,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030760362109417277,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22431804686784745,
      "num_tokens": 104964858.0,
      "step": 50560
    },
    {
      "entropy": 5.725125932693482,
      "epoch": 4.586810595065312,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003075716058407568,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22328484654426575,
      "num_tokens": 104974997.0,
      "step": 50565
    },
    {
      "entropy": 5.68674111366272,
      "epoch": 4.587264150943396,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030753958991375176,
      "loss": 4.831,
      "mean_token_accuracy": 0.22976437211036682,
      "num_tokens": 104984585.0,
      "step": 50570
    },
    {
      "entropy": 5.645398426055908,
      "epoch": 4.58771770682148,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030750757331381964,
      "loss": 4.8254,
      "mean_token_accuracy": 0.22918434590101242,
      "num_tokens": 104994843.0,
      "step": 50575
    },
    {
      "entropy": 5.670269584655761,
      "epoch": 4.588171262699564,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003074755560416226,
      "loss": 4.8968,
      "mean_token_accuracy": 0.2284742534160614,
      "num_tokens": 105006082.0,
      "step": 50580
    },
    {
      "entropy": 5.672900867462158,
      "epoch": 4.588624818577649,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030744353809782283,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22429999113082885,
      "num_tokens": 105016704.0,
      "step": 50585
    },
    {
      "entropy": 5.738408899307251,
      "epoch": 4.589078374455733,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030741151948308237,
      "loss": 4.94,
      "mean_token_accuracy": 0.22162844836711884,
      "num_tokens": 105027810.0,
      "step": 50590
    },
    {
      "entropy": 5.669487810134887,
      "epoch": 4.589531930333817,
      "grad_norm": 1.375,
      "learning_rate": 0.0003073795001980635,
      "loss": 4.8365,
      "mean_token_accuracy": 0.23173744827508927,
      "num_tokens": 105037488.0,
      "step": 50595
    },
    {
      "entropy": 5.705255126953125,
      "epoch": 4.589985486211901,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003073474802434282,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22812942117452623,
      "num_tokens": 105047461.0,
      "step": 50600
    },
    {
      "entropy": 5.719021463394165,
      "epoch": 4.590439042089986,
      "grad_norm": 1.34375,
      "learning_rate": 0.000307315459619839,
      "loss": 4.8989,
      "mean_token_accuracy": 0.2270221307873726,
      "num_tokens": 105058020.0,
      "step": 50605
    },
    {
      "entropy": 5.792120456695557,
      "epoch": 4.59089259796807,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003072834383279576,
      "loss": 4.917,
      "mean_token_accuracy": 0.21995415687561035,
      "num_tokens": 105069638.0,
      "step": 50610
    },
    {
      "entropy": 5.707047033309936,
      "epoch": 4.591346153846154,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003072514163684466,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2295903295278549,
      "num_tokens": 105078686.0,
      "step": 50615
    },
    {
      "entropy": 5.67106146812439,
      "epoch": 4.5917997097242385,
      "grad_norm": 1.25,
      "learning_rate": 0.0003072193937419681,
      "loss": 4.8794,
      "mean_token_accuracy": 0.22682609558105468,
      "num_tokens": 105089739.0,
      "step": 50620
    },
    {
      "entropy": 5.674495697021484,
      "epoch": 4.592253265602322,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030718737044918435,
      "loss": 4.8653,
      "mean_token_accuracy": 0.22388627529144287,
      "num_tokens": 105100673.0,
      "step": 50625
    },
    {
      "entropy": 5.691620540618897,
      "epoch": 4.592706821480406,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030715534649075745,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22991926819086075,
      "num_tokens": 105110750.0,
      "step": 50630
    },
    {
      "entropy": 5.616527509689331,
      "epoch": 4.59316037735849,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030712332186734985,
      "loss": 4.772,
      "mean_token_accuracy": 0.23851172775030136,
      "num_tokens": 105120809.0,
      "step": 50635
    },
    {
      "entropy": 5.698835134506226,
      "epoch": 4.593613933236575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003070912965796238,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22927975058555602,
      "num_tokens": 105131106.0,
      "step": 50640
    },
    {
      "entropy": 5.756080484390258,
      "epoch": 4.594067489114659,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030705927062824154,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22258019894361497,
      "num_tokens": 105140970.0,
      "step": 50645
    },
    {
      "entropy": 5.756182765960693,
      "epoch": 4.594521044992743,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003070272440138654,
      "loss": 4.885,
      "mean_token_accuracy": 0.21808476150035858,
      "num_tokens": 105152429.0,
      "step": 50650
    },
    {
      "entropy": 5.740137672424316,
      "epoch": 4.594974600870827,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003069952167371577,
      "loss": 4.947,
      "mean_token_accuracy": 0.22026603370904924,
      "num_tokens": 105163534.0,
      "step": 50655
    },
    {
      "entropy": 5.625470781326294,
      "epoch": 4.595428156748912,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030696318879878064,
      "loss": 4.8008,
      "mean_token_accuracy": 0.23092413544654847,
      "num_tokens": 105172798.0,
      "step": 50660
    },
    {
      "entropy": 5.736574935913086,
      "epoch": 4.595881712626996,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030693116019939686,
      "loss": 4.893,
      "mean_token_accuracy": 0.22852515429258347,
      "num_tokens": 105182576.0,
      "step": 50665
    },
    {
      "entropy": 5.693402528762817,
      "epoch": 4.596335268505079,
      "grad_norm": 1.59375,
      "learning_rate": 0.00030689913093966847,
      "loss": 4.8848,
      "mean_token_accuracy": 0.22636853605508805,
      "num_tokens": 105192070.0,
      "step": 50670
    },
    {
      "entropy": 5.709234571456909,
      "epoch": 4.5967888243831645,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000306867101020258,
      "loss": 4.8794,
      "mean_token_accuracy": 0.22921363711357118,
      "num_tokens": 105203151.0,
      "step": 50675
    },
    {
      "entropy": 5.727764225006103,
      "epoch": 4.597242380261248,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003068350704418277,
      "loss": 4.9279,
      "mean_token_accuracy": 0.2205573573708534,
      "num_tokens": 105213217.0,
      "step": 50680
    },
    {
      "entropy": 5.649460363388061,
      "epoch": 4.597695936139332,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030680303920504006,
      "loss": 4.7779,
      "mean_token_accuracy": 0.23605193346738815,
      "num_tokens": 105223450.0,
      "step": 50685
    },
    {
      "entropy": 5.746522045135498,
      "epoch": 4.598149492017416,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003067710073105574,
      "loss": 4.944,
      "mean_token_accuracy": 0.22442022264003753,
      "num_tokens": 105234705.0,
      "step": 50690
    },
    {
      "entropy": 5.815143823623657,
      "epoch": 4.598603047895501,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030673897475904223,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20582575500011444,
      "num_tokens": 105245042.0,
      "step": 50695
    },
    {
      "entropy": 5.741490268707276,
      "epoch": 4.599056603773585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030670694155115704,
      "loss": 4.8752,
      "mean_token_accuracy": 0.21678342819213867,
      "num_tokens": 105255830.0,
      "step": 50700
    },
    {
      "entropy": 5.660472106933594,
      "epoch": 4.599510159651669,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030667490768756416,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22672707140445708,
      "num_tokens": 105265844.0,
      "step": 50705
    },
    {
      "entropy": 5.637545442581176,
      "epoch": 4.5999637155297535,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003066428731689262,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22432522773742675,
      "num_tokens": 105276846.0,
      "step": 50710
    },
    {
      "entropy": 5.740696859359741,
      "epoch": 4.600417271407838,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030661083799590544,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22621923089027404,
      "num_tokens": 105286740.0,
      "step": 50715
    },
    {
      "entropy": 5.668060636520385,
      "epoch": 4.600870827285922,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030657880216916454,
      "loss": 4.8458,
      "mean_token_accuracy": 0.23194270581007004,
      "num_tokens": 105296710.0,
      "step": 50720
    },
    {
      "entropy": 5.704075002670288,
      "epoch": 4.601324383164005,
      "grad_norm": 1.34375,
      "learning_rate": 0.000306546765689366,
      "loss": 4.8998,
      "mean_token_accuracy": 0.23296107798814775,
      "num_tokens": 105307076.0,
      "step": 50725
    },
    {
      "entropy": 5.726117277145386,
      "epoch": 4.60177793904209,
      "grad_norm": 1.5,
      "learning_rate": 0.0003065147285571722,
      "loss": 4.8406,
      "mean_token_accuracy": 0.22680448442697526,
      "num_tokens": 105317486.0,
      "step": 50730
    },
    {
      "entropy": 5.673011827468872,
      "epoch": 4.602231494920174,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003064826907732459,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2269439622759819,
      "num_tokens": 105327771.0,
      "step": 50735
    },
    {
      "entropy": 5.684528589248657,
      "epoch": 4.602685050798258,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030645065233824955,
      "loss": 4.9133,
      "mean_token_accuracy": 0.22071736305952072,
      "num_tokens": 105339191.0,
      "step": 50740
    },
    {
      "entropy": 5.707553339004517,
      "epoch": 4.6031386066763424,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030641861325284557,
      "loss": 4.9084,
      "mean_token_accuracy": 0.22903488278388978,
      "num_tokens": 105349949.0,
      "step": 50745
    },
    {
      "entropy": 5.66290397644043,
      "epoch": 4.603592162554427,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030638657351769673,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21854560375213622,
      "num_tokens": 105360357.0,
      "step": 50750
    },
    {
      "entropy": 5.756484079360962,
      "epoch": 4.604045718432511,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030635453313346554,
      "loss": 4.8969,
      "mean_token_accuracy": 0.22640447318553925,
      "num_tokens": 105370765.0,
      "step": 50755
    },
    {
      "entropy": 5.727868509292603,
      "epoch": 4.604499274310595,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003063224921008146,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2243700623512268,
      "num_tokens": 105380797.0,
      "step": 50760
    },
    {
      "entropy": 5.715209341049194,
      "epoch": 4.6049528301886795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003062904504204066,
      "loss": 4.8661,
      "mean_token_accuracy": 0.22549018412828445,
      "num_tokens": 105390653.0,
      "step": 50765
    },
    {
      "entropy": 5.789652347564697,
      "epoch": 4.605406386066764,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030625840809290404,
      "loss": 5.0158,
      "mean_token_accuracy": 0.2199588105082512,
      "num_tokens": 105401308.0,
      "step": 50770
    },
    {
      "entropy": 5.680815315246582,
      "epoch": 4.605859941944848,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030622636511896977,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22145765125751496,
      "num_tokens": 105412412.0,
      "step": 50775
    },
    {
      "entropy": 5.773718595504761,
      "epoch": 4.606313497822931,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003061943214992662,
      "loss": 4.9477,
      "mean_token_accuracy": 0.21699339002370835,
      "num_tokens": 105424789.0,
      "step": 50780
    },
    {
      "entropy": 5.7120274066925045,
      "epoch": 4.606767053701016,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003061622772344561,
      "loss": 4.9631,
      "mean_token_accuracy": 0.22418549805879592,
      "num_tokens": 105434380.0,
      "step": 50785
    },
    {
      "entropy": 5.693018007278442,
      "epoch": 4.6072206095791,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003061302323252022,
      "loss": 4.8388,
      "mean_token_accuracy": 0.23313265442848205,
      "num_tokens": 105443889.0,
      "step": 50790
    },
    {
      "entropy": 5.731161975860596,
      "epoch": 4.607674165457184,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003060981867721671,
      "loss": 4.8659,
      "mean_token_accuracy": 0.2270519718527794,
      "num_tokens": 105454336.0,
      "step": 50795
    },
    {
      "entropy": 5.771022653579712,
      "epoch": 4.6081277213352685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030606614057601374,
      "loss": 4.9748,
      "mean_token_accuracy": 0.22002509385347366,
      "num_tokens": 105464601.0,
      "step": 50800
    },
    {
      "entropy": 5.756485843658448,
      "epoch": 4.608581277213353,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030603409373740456,
      "loss": 4.8887,
      "mean_token_accuracy": 0.21646765917539595,
      "num_tokens": 105475265.0,
      "step": 50805
    },
    {
      "entropy": 5.638398504257202,
      "epoch": 4.609034833091437,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030600204625700234,
      "loss": 4.7697,
      "mean_token_accuracy": 0.23697642236948013,
      "num_tokens": 105485491.0,
      "step": 50810
    },
    {
      "entropy": 5.655730056762695,
      "epoch": 4.609488388969521,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030596999813547007,
      "loss": 4.8398,
      "mean_token_accuracy": 0.22744027972221376,
      "num_tokens": 105495859.0,
      "step": 50815
    },
    {
      "entropy": 5.622395753860474,
      "epoch": 4.6099419448476056,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003059379493734702,
      "loss": 4.7977,
      "mean_token_accuracy": 0.2316683366894722,
      "num_tokens": 105506385.0,
      "step": 50820
    },
    {
      "entropy": 5.681451892852783,
      "epoch": 4.610395500725689,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030590589997166575,
      "loss": 4.7849,
      "mean_token_accuracy": 0.23272815495729446,
      "num_tokens": 105515829.0,
      "step": 50825
    },
    {
      "entropy": 5.663812112808228,
      "epoch": 4.610849056603773,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003058738499307193,
      "loss": 4.8054,
      "mean_token_accuracy": 0.23824314624071122,
      "num_tokens": 105525950.0,
      "step": 50830
    },
    {
      "entropy": 5.6618212223052975,
      "epoch": 4.6113026124818575,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030584179925129384,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2260940730571747,
      "num_tokens": 105536536.0,
      "step": 50835
    },
    {
      "entropy": 5.631273078918457,
      "epoch": 4.611756168359942,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030580974793405213,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23438195437192916,
      "num_tokens": 105546069.0,
      "step": 50840
    },
    {
      "entropy": 5.663151025772095,
      "epoch": 4.612209724238026,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030577769597965696,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22296520322561264,
      "num_tokens": 105556522.0,
      "step": 50845
    },
    {
      "entropy": 5.67576231956482,
      "epoch": 4.61266328011611,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003057456433887712,
      "loss": 4.7935,
      "mean_token_accuracy": 0.23023964762687682,
      "num_tokens": 105566431.0,
      "step": 50850
    },
    {
      "entropy": 5.676932716369629,
      "epoch": 4.6131168359941945,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030571359016205767,
      "loss": 4.8174,
      "mean_token_accuracy": 0.2361232101917267,
      "num_tokens": 105576270.0,
      "step": 50855
    },
    {
      "entropy": 5.678663682937622,
      "epoch": 4.613570391872279,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003056815363001793,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22400808334350586,
      "num_tokens": 105587144.0,
      "step": 50860
    },
    {
      "entropy": 5.6196653842926025,
      "epoch": 4.614023947750363,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030564948180379897,
      "loss": 4.851,
      "mean_token_accuracy": 0.2272627368569374,
      "num_tokens": 105596636.0,
      "step": 50865
    },
    {
      "entropy": 5.6914794921875,
      "epoch": 4.614477503628447,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003056174266735795,
      "loss": 4.8743,
      "mean_token_accuracy": 0.22701463252305984,
      "num_tokens": 105605825.0,
      "step": 50870
    },
    {
      "entropy": 5.771955060958862,
      "epoch": 4.614931059506532,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003055853709101839,
      "loss": 4.9533,
      "mean_token_accuracy": 0.2216496393084526,
      "num_tokens": 105615990.0,
      "step": 50875
    },
    {
      "entropy": 5.717739152908325,
      "epoch": 4.615384615384615,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030555331451427494,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22084965258836747,
      "num_tokens": 105626673.0,
      "step": 50880
    },
    {
      "entropy": 5.741696977615357,
      "epoch": 4.615838171262699,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030552125748651573,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21817321032285691,
      "num_tokens": 105636752.0,
      "step": 50885
    },
    {
      "entropy": 5.761196899414062,
      "epoch": 4.6162917271407835,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030548919982756914,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22304200530052185,
      "num_tokens": 105647476.0,
      "step": 50890
    },
    {
      "entropy": 5.687077474594116,
      "epoch": 4.616745283018868,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030545714153809805,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22941347807645798,
      "num_tokens": 105658135.0,
      "step": 50895
    },
    {
      "entropy": 5.6511882781982425,
      "epoch": 4.617198838896952,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003054250826187655,
      "loss": 4.79,
      "mean_token_accuracy": 0.23458317220211028,
      "num_tokens": 105667802.0,
      "step": 50900
    },
    {
      "entropy": 5.751392078399658,
      "epoch": 4.617652394775036,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003053930230702346,
      "loss": 4.9153,
      "mean_token_accuracy": 0.2316810443997383,
      "num_tokens": 105679161.0,
      "step": 50905
    },
    {
      "entropy": 5.7327817440032955,
      "epoch": 4.618105950653121,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003053609628931682,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2137141466140747,
      "num_tokens": 105688093.0,
      "step": 50910
    },
    {
      "entropy": 5.739538478851318,
      "epoch": 4.618559506531205,
      "grad_norm": 1.375,
      "learning_rate": 0.0003053289020882293,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2192021906375885,
      "num_tokens": 105697982.0,
      "step": 50915
    },
    {
      "entropy": 5.697785758972168,
      "epoch": 4.619013062409289,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030529684065608103,
      "loss": 4.7986,
      "mean_token_accuracy": 0.22995615750551224,
      "num_tokens": 105708729.0,
      "step": 50920
    },
    {
      "entropy": 5.632616758346558,
      "epoch": 4.619466618287373,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030526477859738635,
      "loss": 4.8011,
      "mean_token_accuracy": 0.22863494008779525,
      "num_tokens": 105718919.0,
      "step": 50925
    },
    {
      "entropy": 5.681674146652222,
      "epoch": 4.619920174165458,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003052327159128083,
      "loss": 4.7942,
      "mean_token_accuracy": 0.23521436899900436,
      "num_tokens": 105727787.0,
      "step": 50930
    },
    {
      "entropy": 5.630747842788696,
      "epoch": 4.620373730043541,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030520065260301003,
      "loss": 4.829,
      "mean_token_accuracy": 0.2197410359978676,
      "num_tokens": 105738282.0,
      "step": 50935
    },
    {
      "entropy": 5.697025489807129,
      "epoch": 4.620827285921625,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003051685886686546,
      "loss": 4.8426,
      "mean_token_accuracy": 0.22472249418497087,
      "num_tokens": 105748786.0,
      "step": 50940
    },
    {
      "entropy": 5.707102203369141,
      "epoch": 4.6212808417997095,
      "grad_norm": 1.28125,
      "learning_rate": 0.000305136524110405,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2235704243183136,
      "num_tokens": 105759156.0,
      "step": 50945
    },
    {
      "entropy": 5.684208345413208,
      "epoch": 4.621734397677794,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003051044589289245,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21820244193077087,
      "num_tokens": 105769434.0,
      "step": 50950
    },
    {
      "entropy": 5.727300977706909,
      "epoch": 4.622187953555878,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030507239312487595,
      "loss": 4.9835,
      "mean_token_accuracy": 0.2165827989578247,
      "num_tokens": 105779408.0,
      "step": 50955
    },
    {
      "entropy": 5.743468284606934,
      "epoch": 4.622641509433962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030504032669892276,
      "loss": 4.9309,
      "mean_token_accuracy": 0.22200678884983063,
      "num_tokens": 105789517.0,
      "step": 50960
    },
    {
      "entropy": 5.733674573898315,
      "epoch": 4.623095065312047,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030500825965172797,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22870776653289795,
      "num_tokens": 105799823.0,
      "step": 50965
    },
    {
      "entropy": 5.688521146774292,
      "epoch": 4.623548621190131,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030497619198395466,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22698041051626205,
      "num_tokens": 105810038.0,
      "step": 50970
    },
    {
      "entropy": 5.695227003097534,
      "epoch": 4.624002177068215,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030494412369626615,
      "loss": 4.9034,
      "mean_token_accuracy": 0.22147733420133592,
      "num_tokens": 105819940.0,
      "step": 50975
    },
    {
      "entropy": 5.679834508895874,
      "epoch": 4.6244557329462985,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003049120547893255,
      "loss": 4.8398,
      "mean_token_accuracy": 0.2273387387394905,
      "num_tokens": 105829829.0,
      "step": 50980
    },
    {
      "entropy": 5.676961040496826,
      "epoch": 4.624909288824383,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003048799852637958,
      "loss": 4.8266,
      "mean_token_accuracy": 0.23475193977355957,
      "num_tokens": 105839463.0,
      "step": 50985
    },
    {
      "entropy": 5.7728523254394535,
      "epoch": 4.625362844702467,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030484791512034063,
      "loss": 5.0453,
      "mean_token_accuracy": 0.21351809054613113,
      "num_tokens": 105849916.0,
      "step": 50990
    },
    {
      "entropy": 5.662139797210694,
      "epoch": 4.625816400580551,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003048158443596228,
      "loss": 4.8453,
      "mean_token_accuracy": 0.2260385200381279,
      "num_tokens": 105860699.0,
      "step": 50995
    },
    {
      "entropy": 5.719467067718506,
      "epoch": 4.626269956458636,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003047837729823058,
      "loss": 4.841,
      "mean_token_accuracy": 0.22146389186382293,
      "num_tokens": 105870332.0,
      "step": 51000
    },
    {
      "epoch": 4.626269956458636,
      "eval_entropy": 5.39784821687241,
      "eval_loss": 4.957599639892578,
      "eval_mean_token_accuracy": 0.22998080086047293,
      "eval_num_tokens": 105870332.0,
      "eval_runtime": 20.675,
      "eval_samples_per_second": 1710.276,
      "eval_steps_per_second": 213.785,
      "step": 51000
    },
    {
      "entropy": 5.669802665710449,
      "epoch": 4.62672351233672,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003047517009890527,
      "loss": 4.8512,
      "mean_token_accuracy": 0.23374016880989074,
      "num_tokens": 105879364.0,
      "step": 51005
    },
    {
      "entropy": 5.746862888336182,
      "epoch": 4.627177068214804,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030471962838052693,
      "loss": 4.9762,
      "mean_token_accuracy": 0.22039295732975006,
      "num_tokens": 105889554.0,
      "step": 51010
    },
    {
      "entropy": 5.706625986099243,
      "epoch": 4.627630624092888,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003046875551573916,
      "loss": 4.7939,
      "mean_token_accuracy": 0.24061723351478576,
      "num_tokens": 105900426.0,
      "step": 51015
    },
    {
      "entropy": 5.563585567474365,
      "epoch": 4.628084179970973,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003046554813203101,
      "loss": 4.6877,
      "mean_token_accuracy": 0.23553241193294525,
      "num_tokens": 105909541.0,
      "step": 51020
    },
    {
      "entropy": 5.66141300201416,
      "epoch": 4.628537735849057,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003046234068699457,
      "loss": 4.9102,
      "mean_token_accuracy": 0.22208481431007385,
      "num_tokens": 105919951.0,
      "step": 51025
    },
    {
      "entropy": 5.64502363204956,
      "epoch": 4.628991291727141,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003045913318069616,
      "loss": 4.7995,
      "mean_token_accuracy": 0.23412237763404847,
      "num_tokens": 105929048.0,
      "step": 51030
    },
    {
      "entropy": 5.709818744659424,
      "epoch": 4.629444847605225,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003045592561320213,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23004454523324966,
      "num_tokens": 105938105.0,
      "step": 51035
    },
    {
      "entropy": 5.714500951766968,
      "epoch": 4.629898403483309,
      "grad_norm": 1.421875,
      "learning_rate": 0.000304527179845788,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21869143098592758,
      "num_tokens": 105948167.0,
      "step": 51040
    },
    {
      "entropy": 5.723334503173828,
      "epoch": 4.630351959361393,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003044951029489252,
      "loss": 4.8992,
      "mean_token_accuracy": 0.22694191932678223,
      "num_tokens": 105959387.0,
      "step": 51045
    },
    {
      "entropy": 5.600962972640991,
      "epoch": 4.630805515239477,
      "grad_norm": 1.375,
      "learning_rate": 0.000304463025442096,
      "loss": 4.7708,
      "mean_token_accuracy": 0.2324988678097725,
      "num_tokens": 105969436.0,
      "step": 51050
    },
    {
      "entropy": 5.7237160205841064,
      "epoch": 4.631259071117562,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003044309473259641,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2306859776377678,
      "num_tokens": 105979102.0,
      "step": 51055
    },
    {
      "entropy": 5.64754581451416,
      "epoch": 4.631712626995646,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030439886860119265,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23506016433238983,
      "num_tokens": 105989796.0,
      "step": 51060
    },
    {
      "entropy": 5.6681952476501465,
      "epoch": 4.63216618287373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000304366789268445,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22401145547628404,
      "num_tokens": 106000194.0,
      "step": 51065
    },
    {
      "entropy": 5.686945295333862,
      "epoch": 4.632619738751814,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003043347093283848,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22847085297107697,
      "num_tokens": 106009108.0,
      "step": 51070
    },
    {
      "entropy": 5.646271085739135,
      "epoch": 4.633073294629899,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030430262878167533,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22923977673053741,
      "num_tokens": 106019342.0,
      "step": 51075
    },
    {
      "entropy": 5.752689933776855,
      "epoch": 4.633526850507983,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003042705476289799,
      "loss": 4.924,
      "mean_token_accuracy": 0.21598901301622392,
      "num_tokens": 106029079.0,
      "step": 51080
    },
    {
      "entropy": 5.66182656288147,
      "epoch": 4.633980406386067,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003042384658709622,
      "loss": 4.8278,
      "mean_token_accuracy": 0.2351740464568138,
      "num_tokens": 106039406.0,
      "step": 51085
    },
    {
      "entropy": 5.696001291275024,
      "epoch": 4.634433962264151,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003042063835082856,
      "loss": 4.8857,
      "mean_token_accuracy": 0.22296128422021866,
      "num_tokens": 106049603.0,
      "step": 51090
    },
    {
      "entropy": 5.696039390563965,
      "epoch": 4.634887518142235,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003041743005416135,
      "loss": 4.8398,
      "mean_token_accuracy": 0.2324887990951538,
      "num_tokens": 106058692.0,
      "step": 51095
    },
    {
      "entropy": 5.62259316444397,
      "epoch": 4.635341074020319,
      "grad_norm": 1.375,
      "learning_rate": 0.0003041422169716094,
      "loss": 4.8308,
      "mean_token_accuracy": 0.2338256850838661,
      "num_tokens": 106069307.0,
      "step": 51100
    },
    {
      "entropy": 5.746149063110352,
      "epoch": 4.635794629898403,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003041101327989368,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2310529336333275,
      "num_tokens": 106080831.0,
      "step": 51105
    },
    {
      "entropy": 5.712434244155884,
      "epoch": 4.636248185776488,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003040780480242593,
      "loss": 4.886,
      "mean_token_accuracy": 0.22331262975931168,
      "num_tokens": 106090007.0,
      "step": 51110
    },
    {
      "entropy": 5.7120445728302,
      "epoch": 4.636701741654572,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030404596264824033,
      "loss": 4.8814,
      "mean_token_accuracy": 0.23163101971149444,
      "num_tokens": 106099790.0,
      "step": 51115
    },
    {
      "entropy": 5.679066801071167,
      "epoch": 4.637155297532656,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030401387667154347,
      "loss": 4.8114,
      "mean_token_accuracy": 0.23463165760040283,
      "num_tokens": 106110364.0,
      "step": 51120
    },
    {
      "entropy": 5.585643863677978,
      "epoch": 4.6376088534107405,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003039817900948322,
      "loss": 4.7756,
      "mean_token_accuracy": 0.23108186572790146,
      "num_tokens": 106121353.0,
      "step": 51125
    },
    {
      "entropy": 5.68495364189148,
      "epoch": 4.638062409288825,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003039497029187702,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2239231824874878,
      "num_tokens": 106130750.0,
      "step": 51130
    },
    {
      "entropy": 5.694917106628418,
      "epoch": 4.638515965166908,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030391761514402096,
      "loss": 4.8508,
      "mean_token_accuracy": 0.23063298761844636,
      "num_tokens": 106142014.0,
      "step": 51135
    },
    {
      "entropy": 5.694868898391723,
      "epoch": 4.638969521044992,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030388552677124804,
      "loss": 4.8595,
      "mean_token_accuracy": 0.23063510954380034,
      "num_tokens": 106152419.0,
      "step": 51140
    },
    {
      "entropy": 5.66571569442749,
      "epoch": 4.639423076923077,
      "grad_norm": 1.375,
      "learning_rate": 0.000303853437801115,
      "loss": 4.7898,
      "mean_token_accuracy": 0.23081966489553452,
      "num_tokens": 106163365.0,
      "step": 51145
    },
    {
      "entropy": 5.658709573745727,
      "epoch": 4.639876632801161,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003038213482342856,
      "loss": 4.8036,
      "mean_token_accuracy": 0.23678296357393264,
      "num_tokens": 106173324.0,
      "step": 51150
    },
    {
      "entropy": 5.783092784881592,
      "epoch": 4.640330188679245,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030378925807142334,
      "loss": 4.9541,
      "mean_token_accuracy": 0.2197749584913254,
      "num_tokens": 106183427.0,
      "step": 51155
    },
    {
      "entropy": 5.740220355987549,
      "epoch": 4.6407837445573294,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030375716731319193,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2168423131108284,
      "num_tokens": 106192894.0,
      "step": 51160
    },
    {
      "entropy": 5.671259260177612,
      "epoch": 4.641237300435414,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030372507596025493,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2237466275691986,
      "num_tokens": 106203185.0,
      "step": 51165
    },
    {
      "entropy": 5.766554689407348,
      "epoch": 4.641690856313498,
      "grad_norm": 1.421875,
      "learning_rate": 0.000303692984013276,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2177593767642975,
      "num_tokens": 106213817.0,
      "step": 51170
    },
    {
      "entropy": 5.641507530212403,
      "epoch": 4.642144412191582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003036608914729189,
      "loss": 4.8309,
      "mean_token_accuracy": 0.22922297269105912,
      "num_tokens": 106224367.0,
      "step": 51175
    },
    {
      "entropy": 5.582180595397949,
      "epoch": 4.6425979680696665,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003036287983398473,
      "loss": 4.7669,
      "mean_token_accuracy": 0.2271395966410637,
      "num_tokens": 106234146.0,
      "step": 51180
    },
    {
      "entropy": 5.766873550415039,
      "epoch": 4.643051523947751,
      "grad_norm": 1.375,
      "learning_rate": 0.00030359670461472475,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21937830597162247,
      "num_tokens": 106244910.0,
      "step": 51185
    },
    {
      "entropy": 5.790411520004272,
      "epoch": 4.643505079825834,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030356461029821516,
      "loss": 4.9641,
      "mean_token_accuracy": 0.2106051355600357,
      "num_tokens": 106254267.0,
      "step": 51190
    },
    {
      "entropy": 5.719942998886109,
      "epoch": 4.643958635703918,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003035325153909821,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22716279178857804,
      "num_tokens": 106264755.0,
      "step": 51195
    },
    {
      "entropy": 5.658880138397217,
      "epoch": 4.644412191582003,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003035004198936894,
      "loss": 4.851,
      "mean_token_accuracy": 0.22603063732385636,
      "num_tokens": 106273964.0,
      "step": 51200
    },
    {
      "entropy": 5.724995851516724,
      "epoch": 4.644865747460087,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003034683238070008,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22642871141433715,
      "num_tokens": 106283933.0,
      "step": 51205
    },
    {
      "entropy": 5.758076953887939,
      "epoch": 4.645319303338171,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030343622713157985,
      "loss": 4.8443,
      "mean_token_accuracy": 0.23019849210977555,
      "num_tokens": 106294075.0,
      "step": 51210
    },
    {
      "entropy": 5.69271183013916,
      "epoch": 4.6457728592162555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003034041298680906,
      "loss": 4.8315,
      "mean_token_accuracy": 0.23113490492105485,
      "num_tokens": 106304970.0,
      "step": 51215
    },
    {
      "entropy": 5.669328212738037,
      "epoch": 4.64622641509434,
      "grad_norm": 1.25,
      "learning_rate": 0.0003033720320171966,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2229427605867386,
      "num_tokens": 106315610.0,
      "step": 51220
    },
    {
      "entropy": 5.694483661651612,
      "epoch": 4.646679970972424,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003033399335795619,
      "loss": 4.8915,
      "mean_token_accuracy": 0.22542244344949722,
      "num_tokens": 106326302.0,
      "step": 51225
    },
    {
      "entropy": 5.680100154876709,
      "epoch": 4.647133526850508,
      "grad_norm": 1.375,
      "learning_rate": 0.00030330783455585,
      "loss": 4.8401,
      "mean_token_accuracy": 0.22401652932167054,
      "num_tokens": 106336350.0,
      "step": 51230
    },
    {
      "entropy": 5.688644981384277,
      "epoch": 4.647587082728592,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003032757349467249,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21996210515499115,
      "num_tokens": 106348233.0,
      "step": 51235
    },
    {
      "entropy": 5.665142917633057,
      "epoch": 4.648040638606677,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030324363475285046,
      "loss": 4.7835,
      "mean_token_accuracy": 0.2293636053800583,
      "num_tokens": 106357992.0,
      "step": 51240
    },
    {
      "entropy": 5.680606174468994,
      "epoch": 4.64849419448476,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003032115339748904,
      "loss": 4.7722,
      "mean_token_accuracy": 0.23482137620449067,
      "num_tokens": 106367683.0,
      "step": 51245
    },
    {
      "entropy": 5.66241307258606,
      "epoch": 4.6489477503628445,
      "grad_norm": 1.375,
      "learning_rate": 0.0003031794326135086,
      "loss": 4.8404,
      "mean_token_accuracy": 0.23036832809448243,
      "num_tokens": 106378275.0,
      "step": 51250
    },
    {
      "entropy": 5.675297784805298,
      "epoch": 4.649401306240929,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030314733066936903,
      "loss": 4.8663,
      "mean_token_accuracy": 0.23063841313123704,
      "num_tokens": 106387811.0,
      "step": 51255
    },
    {
      "entropy": 5.6799170017242435,
      "epoch": 4.649854862119013,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003031152281431355,
      "loss": 4.8865,
      "mean_token_accuracy": 0.22276319861412047,
      "num_tokens": 106398327.0,
      "step": 51260
    },
    {
      "entropy": 5.645149040222168,
      "epoch": 4.650308417997097,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030308312503547183,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22584296464920045,
      "num_tokens": 106408602.0,
      "step": 51265
    },
    {
      "entropy": 5.6509603500366214,
      "epoch": 4.6507619738751815,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000303051021347042,
      "loss": 4.814,
      "mean_token_accuracy": 0.23115438222885132,
      "num_tokens": 106418717.0,
      "step": 51270
    },
    {
      "entropy": 5.628789615631104,
      "epoch": 4.651215529753266,
      "grad_norm": 1.515625,
      "learning_rate": 0.00030301891707850987,
      "loss": 4.752,
      "mean_token_accuracy": 0.23340966552495956,
      "num_tokens": 106428311.0,
      "step": 51275
    },
    {
      "entropy": 5.710794305801391,
      "epoch": 4.65166908563135,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003029868122305395,
      "loss": 4.8729,
      "mean_token_accuracy": 0.2186696410179138,
      "num_tokens": 106438772.0,
      "step": 51280
    },
    {
      "entropy": 5.717220735549927,
      "epoch": 4.652122641509434,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030295470680379456,
      "loss": 4.8409,
      "mean_token_accuracy": 0.23524584919214248,
      "num_tokens": 106448895.0,
      "step": 51285
    },
    {
      "entropy": 5.680604696273804,
      "epoch": 4.652576197387518,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003029226007989393,
      "loss": 4.9338,
      "mean_token_accuracy": 0.225094273686409,
      "num_tokens": 106459711.0,
      "step": 51290
    },
    {
      "entropy": 5.707801485061646,
      "epoch": 4.653029753265602,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030289049421663744,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22908796966075898,
      "num_tokens": 106469691.0,
      "step": 51295
    },
    {
      "entropy": 5.690437936782837,
      "epoch": 4.653483309143686,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003028583870575531,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21950548589229585,
      "num_tokens": 106479894.0,
      "step": 51300
    },
    {
      "entropy": 5.662571048736572,
      "epoch": 4.6539368650217705,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003028262793223502,
      "loss": 4.821,
      "mean_token_accuracy": 0.23174362182617186,
      "num_tokens": 106491444.0,
      "step": 51305
    },
    {
      "entropy": 5.684965324401856,
      "epoch": 4.654390420899855,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030279417101169276,
      "loss": 4.8608,
      "mean_token_accuracy": 0.23137447386980056,
      "num_tokens": 106501633.0,
      "step": 51310
    },
    {
      "entropy": 5.661297416687011,
      "epoch": 4.654843976777939,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003027620621262448,
      "loss": 4.8167,
      "mean_token_accuracy": 0.2378679856657982,
      "num_tokens": 106511272.0,
      "step": 51315
    },
    {
      "entropy": 5.702607297897339,
      "epoch": 4.655297532656023,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003027299526666703,
      "loss": 4.947,
      "mean_token_accuracy": 0.2241317242383957,
      "num_tokens": 106521977.0,
      "step": 51320
    },
    {
      "entropy": 5.70430178642273,
      "epoch": 4.655751088534108,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030269784263363327,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22597798854112625,
      "num_tokens": 106532802.0,
      "step": 51325
    },
    {
      "entropy": 5.69038724899292,
      "epoch": 4.656204644412192,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030266573202779784,
      "loss": 4.869,
      "mean_token_accuracy": 0.2239525780081749,
      "num_tokens": 106542249.0,
      "step": 51330
    },
    {
      "entropy": 5.705550193786621,
      "epoch": 4.656658200290276,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030263362084982805,
      "loss": 4.867,
      "mean_token_accuracy": 0.22634542882442474,
      "num_tokens": 106552501.0,
      "step": 51335
    },
    {
      "entropy": 5.652243566513062,
      "epoch": 4.65711175616836,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030260150910038793,
      "loss": 4.8079,
      "mean_token_accuracy": 0.23635268062353135,
      "num_tokens": 106561901.0,
      "step": 51340
    },
    {
      "entropy": 5.636633729934692,
      "epoch": 4.657565312046444,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030256939678014155,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22904085069894792,
      "num_tokens": 106572104.0,
      "step": 51345
    },
    {
      "entropy": 5.688069486618042,
      "epoch": 4.658018867924528,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003025372838897529,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22042349725961685,
      "num_tokens": 106583436.0,
      "step": 51350
    },
    {
      "entropy": 5.716781854629517,
      "epoch": 4.658472423802612,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003025051704298864,
      "loss": 4.9441,
      "mean_token_accuracy": 0.21844100058078766,
      "num_tokens": 106593879.0,
      "step": 51355
    },
    {
      "entropy": 5.712018537521362,
      "epoch": 4.6589259796806965,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030247305640120585,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21754402816295623,
      "num_tokens": 106604774.0,
      "step": 51360
    },
    {
      "entropy": 5.738723278045654,
      "epoch": 4.659379535558781,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003024409418043756,
      "loss": 4.9562,
      "mean_token_accuracy": 0.22012415528297424,
      "num_tokens": 106616125.0,
      "step": 51365
    },
    {
      "entropy": 5.743148374557495,
      "epoch": 4.659833091436865,
      "grad_norm": 1.375,
      "learning_rate": 0.0003024088266400595,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2235090911388397,
      "num_tokens": 106625171.0,
      "step": 51370
    },
    {
      "entropy": 5.655279493331909,
      "epoch": 4.660286647314949,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030237671090892204,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22352489829063416,
      "num_tokens": 106636455.0,
      "step": 51375
    },
    {
      "entropy": 5.696791934967041,
      "epoch": 4.660740203193034,
      "grad_norm": 1.375,
      "learning_rate": 0.0003023445946116272,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2236808031797409,
      "num_tokens": 106645956.0,
      "step": 51380
    },
    {
      "entropy": 5.720179271697998,
      "epoch": 4.661193759071118,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030231247774883913,
      "loss": 4.9289,
      "mean_token_accuracy": 0.22432997673749924,
      "num_tokens": 106655562.0,
      "step": 51385
    },
    {
      "entropy": 5.702135753631592,
      "epoch": 4.661647314949201,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003022803603212221,
      "loss": 4.8636,
      "mean_token_accuracy": 0.2199104681611061,
      "num_tokens": 106666620.0,
      "step": 51390
    },
    {
      "entropy": 5.6934741020202635,
      "epoch": 4.662100870827286,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003022482423294403,
      "loss": 4.8612,
      "mean_token_accuracy": 0.22753432393074036,
      "num_tokens": 106676881.0,
      "step": 51395
    },
    {
      "entropy": 5.70339674949646,
      "epoch": 4.66255442670537,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003022161237741578,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2306442752480507,
      "num_tokens": 106687765.0,
      "step": 51400
    },
    {
      "entropy": 5.6788078308105465,
      "epoch": 4.663007982583454,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030218400465603895,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22433519810438157,
      "num_tokens": 106698796.0,
      "step": 51405
    },
    {
      "entropy": 5.7112500190734865,
      "epoch": 4.663461538461538,
      "grad_norm": 1.46875,
      "learning_rate": 0.000302151884975748,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23343796879053116,
      "num_tokens": 106709439.0,
      "step": 51410
    },
    {
      "entropy": 5.733988428115845,
      "epoch": 4.663915094339623,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003021197647339492,
      "loss": 4.9607,
      "mean_token_accuracy": 0.22006160169839858,
      "num_tokens": 106720325.0,
      "step": 51415
    },
    {
      "entropy": 5.726701402664185,
      "epoch": 4.664368650217707,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003020876439313066,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2302178904414177,
      "num_tokens": 106730526.0,
      "step": 51420
    },
    {
      "entropy": 5.701349401473999,
      "epoch": 4.664822206095791,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030205552256848476,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22798357158899307,
      "num_tokens": 106741142.0,
      "step": 51425
    },
    {
      "entropy": 5.7881358623504635,
      "epoch": 4.665275761973875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030202340064614783,
      "loss": 4.9615,
      "mean_token_accuracy": 0.215155129134655,
      "num_tokens": 106751843.0,
      "step": 51430
    },
    {
      "entropy": 5.629353046417236,
      "epoch": 4.66572931785196,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003019912781649599,
      "loss": 4.7225,
      "mean_token_accuracy": 0.23527796417474747,
      "num_tokens": 106761726.0,
      "step": 51435
    },
    {
      "entropy": 5.666103553771973,
      "epoch": 4.666182873730044,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030195915512558564,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23048738539218902,
      "num_tokens": 106771829.0,
      "step": 51440
    },
    {
      "entropy": 5.64337191581726,
      "epoch": 4.666636429608127,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003019270315286892,
      "loss": 4.7723,
      "mean_token_accuracy": 0.2336168259382248,
      "num_tokens": 106783457.0,
      "step": 51445
    },
    {
      "entropy": 5.689475774765015,
      "epoch": 4.667089985486212,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030189490737493477,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2355221062898636,
      "num_tokens": 106793290.0,
      "step": 51450
    },
    {
      "entropy": 5.670255470275879,
      "epoch": 4.667543541364296,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030186278266498677,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2240886077284813,
      "num_tokens": 106804412.0,
      "step": 51455
    },
    {
      "entropy": 5.750600147247314,
      "epoch": 4.66799709724238,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003018306573995097,
      "loss": 4.9651,
      "mean_token_accuracy": 0.22432412654161454,
      "num_tokens": 106815353.0,
      "step": 51460
    },
    {
      "entropy": 5.718930673599243,
      "epoch": 4.668450653120464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030179853157916777,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22483896613121032,
      "num_tokens": 106826350.0,
      "step": 51465
    },
    {
      "entropy": 5.748502492904663,
      "epoch": 4.668904208998549,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003017664052046253,
      "loss": 4.8974,
      "mean_token_accuracy": 0.22533434480428696,
      "num_tokens": 106837434.0,
      "step": 51470
    },
    {
      "entropy": 5.667345952987671,
      "epoch": 4.669357764876633,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030173427827654683,
      "loss": 4.8949,
      "mean_token_accuracy": 0.2244699016213417,
      "num_tokens": 106847829.0,
      "step": 51475
    },
    {
      "entropy": 5.7512421131134035,
      "epoch": 4.669811320754717,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003017021507955967,
      "loss": 4.9871,
      "mean_token_accuracy": 0.21415289640426635,
      "num_tokens": 106858519.0,
      "step": 51480
    },
    {
      "entropy": 5.743497562408447,
      "epoch": 4.670264876632801,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030167002276243915,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23480771631002426,
      "num_tokens": 106868614.0,
      "step": 51485
    },
    {
      "entropy": 5.754582166671753,
      "epoch": 4.670718432510886,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030163789417773885,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21363981813192368,
      "num_tokens": 106879592.0,
      "step": 51490
    },
    {
      "entropy": 5.711266326904297,
      "epoch": 4.67117198838897,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003016057650421601,
      "loss": 4.8843,
      "mean_token_accuracy": 0.225536048412323,
      "num_tokens": 106890083.0,
      "step": 51495
    },
    {
      "entropy": 5.749359464645385,
      "epoch": 4.671625544267053,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030157363535636726,
      "loss": 4.9417,
      "mean_token_accuracy": 0.22540417760610582,
      "num_tokens": 106900672.0,
      "step": 51500
    },
    {
      "entropy": 5.705791759490967,
      "epoch": 4.672079100145138,
      "grad_norm": 1.421875,
      "learning_rate": 0.000301541505121025,
      "loss": 4.8255,
      "mean_token_accuracy": 0.22961851954460144,
      "num_tokens": 106910396.0,
      "step": 51505
    },
    {
      "entropy": 5.724182176589966,
      "epoch": 4.672532656023222,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003015093743367976,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22040982097387313,
      "num_tokens": 106919742.0,
      "step": 51510
    },
    {
      "entropy": 5.6201235294342045,
      "epoch": 4.672986211901306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003014772430043495,
      "loss": 4.7879,
      "mean_token_accuracy": 0.23367086797952652,
      "num_tokens": 106930837.0,
      "step": 51515
    },
    {
      "entropy": 5.672926712036133,
      "epoch": 4.67343976777939,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030144511112434536,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23324193805456161,
      "num_tokens": 106942023.0,
      "step": 51520
    },
    {
      "entropy": 5.7488760471344,
      "epoch": 4.673893323657475,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030141297869744956,
      "loss": 4.9191,
      "mean_token_accuracy": 0.22036049664020538,
      "num_tokens": 106951374.0,
      "step": 51525
    },
    {
      "entropy": 5.802189588546753,
      "epoch": 4.674346879535559,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030138084572432663,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21683802157640458,
      "num_tokens": 106961714.0,
      "step": 51530
    },
    {
      "entropy": 5.724110317230225,
      "epoch": 4.674800435413643,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003013487122056411,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22790799140930176,
      "num_tokens": 106971783.0,
      "step": 51535
    },
    {
      "entropy": 5.660068798065185,
      "epoch": 4.6752539912917275,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003013165781420575,
      "loss": 4.8178,
      "mean_token_accuracy": 0.23308442085981368,
      "num_tokens": 106982047.0,
      "step": 51540
    },
    {
      "entropy": 5.695528554916382,
      "epoch": 4.675707547169811,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030128444353424034,
      "loss": 4.8758,
      "mean_token_accuracy": 0.2268896996974945,
      "num_tokens": 106993063.0,
      "step": 51545
    },
    {
      "entropy": 5.739959001541138,
      "epoch": 4.676161103047895,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030125230838285415,
      "loss": 4.8756,
      "mean_token_accuracy": 0.2244877427816391,
      "num_tokens": 107002865.0,
      "step": 51550
    },
    {
      "entropy": 5.776112794876099,
      "epoch": 4.676614658925979,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030122017268856353,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23449053466320038,
      "num_tokens": 107013007.0,
      "step": 51555
    },
    {
      "entropy": 5.740651607513428,
      "epoch": 4.677068214804064,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030118803645203315,
      "loss": 4.9203,
      "mean_token_accuracy": 0.2208094120025635,
      "num_tokens": 107023109.0,
      "step": 51560
    },
    {
      "entropy": 5.655166816711426,
      "epoch": 4.677521770682148,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003011558996739274,
      "loss": 4.8554,
      "mean_token_accuracy": 0.22683603465557098,
      "num_tokens": 107033824.0,
      "step": 51565
    },
    {
      "entropy": 5.695811748504639,
      "epoch": 4.677975326560232,
      "grad_norm": 1.375,
      "learning_rate": 0.000301123762354911,
      "loss": 4.8871,
      "mean_token_accuracy": 0.22363294959068297,
      "num_tokens": 107043529.0,
      "step": 51570
    },
    {
      "entropy": 5.737265777587891,
      "epoch": 4.6784288824383164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003010916244956486,
      "loss": 4.9185,
      "mean_token_accuracy": 0.23124850392341614,
      "num_tokens": 107054378.0,
      "step": 51575
    },
    {
      "entropy": 5.711061716079712,
      "epoch": 4.678882438316401,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003010594860968048,
      "loss": 4.7798,
      "mean_token_accuracy": 0.2294224739074707,
      "num_tokens": 107064300.0,
      "step": 51580
    },
    {
      "entropy": 5.7719213485717775,
      "epoch": 4.679335994194485,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000301027347159044,
      "loss": 4.9432,
      "mean_token_accuracy": 0.22296036332845687,
      "num_tokens": 107074560.0,
      "step": 51585
    },
    {
      "entropy": 5.730255746841431,
      "epoch": 4.679789550072569,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003009952076830312,
      "loss": 4.8509,
      "mean_token_accuracy": 0.22005038112401962,
      "num_tokens": 107084574.0,
      "step": 51590
    },
    {
      "entropy": 5.687801837921143,
      "epoch": 4.6802431059506535,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003009630676694308,
      "loss": 4.856,
      "mean_token_accuracy": 0.22160498350858687,
      "num_tokens": 107095207.0,
      "step": 51595
    },
    {
      "entropy": 5.670245790481568,
      "epoch": 4.680696661828737,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030093092711890756,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22418876886367797,
      "num_tokens": 107106091.0,
      "step": 51600
    },
    {
      "entropy": 5.716653156280517,
      "epoch": 4.681150217706821,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003008987860321261,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22190040498971939,
      "num_tokens": 107117753.0,
      "step": 51605
    },
    {
      "entropy": 5.652648496627807,
      "epoch": 4.681603773584905,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003008666444097512,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2351503327488899,
      "num_tokens": 107128131.0,
      "step": 51610
    },
    {
      "entropy": 5.747717666625976,
      "epoch": 4.68205732946299,
      "grad_norm": 1.375,
      "learning_rate": 0.0003008345022524475,
      "loss": 5.0164,
      "mean_token_accuracy": 0.2131069391965866,
      "num_tokens": 107139447.0,
      "step": 51615
    },
    {
      "entropy": 5.729649257659912,
      "epoch": 4.682510885341074,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030080235956087963,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2259477660059929,
      "num_tokens": 107149721.0,
      "step": 51620
    },
    {
      "entropy": 5.704457235336304,
      "epoch": 4.682964441219158,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003007702163357125,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2241818428039551,
      "num_tokens": 107159814.0,
      "step": 51625
    },
    {
      "entropy": 5.679449081420898,
      "epoch": 4.6834179970972425,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030073807257761074,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2302435517311096,
      "num_tokens": 107169012.0,
      "step": 51630
    },
    {
      "entropy": 5.754698133468628,
      "epoch": 4.683871552975327,
      "grad_norm": 1.375,
      "learning_rate": 0.00030070592828723903,
      "loss": 4.8997,
      "mean_token_accuracy": 0.222773739695549,
      "num_tokens": 107179571.0,
      "step": 51635
    },
    {
      "entropy": 5.70891547203064,
      "epoch": 4.684325108853411,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003006737834652621,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22693364024162294,
      "num_tokens": 107190535.0,
      "step": 51640
    },
    {
      "entropy": 5.739522218704224,
      "epoch": 4.684778664731495,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003006416381123449,
      "loss": 4.8777,
      "mean_token_accuracy": 0.23051415830850602,
      "num_tokens": 107200361.0,
      "step": 51645
    },
    {
      "entropy": 5.6945044040679935,
      "epoch": 4.68523222060958,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030060949222915205,
      "loss": 4.8798,
      "mean_token_accuracy": 0.2261993870139122,
      "num_tokens": 107211080.0,
      "step": 51650
    },
    {
      "entropy": 5.710424184799194,
      "epoch": 4.685685776487663,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003005773458163484,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22517562806606292,
      "num_tokens": 107220556.0,
      "step": 51655
    },
    {
      "entropy": 5.710369873046875,
      "epoch": 4.686139332365747,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030054519887459864,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22381639629602432,
      "num_tokens": 107230693.0,
      "step": 51660
    },
    {
      "entropy": 5.759099054336548,
      "epoch": 4.6865928882438315,
      "grad_norm": 1.375,
      "learning_rate": 0.00030051305140456775,
      "loss": 4.9287,
      "mean_token_accuracy": 0.2214125096797943,
      "num_tokens": 107240926.0,
      "step": 51665
    },
    {
      "entropy": 5.708201599121094,
      "epoch": 4.687046444121916,
      "grad_norm": 1.25,
      "learning_rate": 0.00030048090340692034,
      "loss": 4.9111,
      "mean_token_accuracy": 0.22321535497903824,
      "num_tokens": 107251425.0,
      "step": 51670
    },
    {
      "entropy": 5.711727094650269,
      "epoch": 4.6875,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030044875488232145,
      "loss": 4.8696,
      "mean_token_accuracy": 0.23406895995140076,
      "num_tokens": 107262046.0,
      "step": 51675
    },
    {
      "entropy": 5.625852298736572,
      "epoch": 4.687953555878084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003004166058314358,
      "loss": 4.8269,
      "mean_token_accuracy": 0.23338100612163543,
      "num_tokens": 107273141.0,
      "step": 51680
    },
    {
      "entropy": 5.71053638458252,
      "epoch": 4.6884071117561685,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003003844562549282,
      "loss": 4.9049,
      "mean_token_accuracy": 0.22485740333795548,
      "num_tokens": 107282497.0,
      "step": 51685
    },
    {
      "entropy": 5.6593469142913815,
      "epoch": 4.688860667634253,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030035230615346373,
      "loss": 4.8198,
      "mean_token_accuracy": 0.23240915834903716,
      "num_tokens": 107292710.0,
      "step": 51690
    },
    {
      "entropy": 5.638056755065918,
      "epoch": 4.689314223512337,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030032015552770695,
      "loss": 4.741,
      "mean_token_accuracy": 0.23705403059720992,
      "num_tokens": 107303157.0,
      "step": 51695
    },
    {
      "entropy": 5.679092693328857,
      "epoch": 4.68976777939042,
      "grad_norm": 1.375,
      "learning_rate": 0.00030028800437832296,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22418105602264404,
      "num_tokens": 107313503.0,
      "step": 51700
    },
    {
      "entropy": 5.662060976028442,
      "epoch": 4.690221335268505,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003002558527059766,
      "loss": 4.8521,
      "mean_token_accuracy": 0.2306606724858284,
      "num_tokens": 107322276.0,
      "step": 51705
    },
    {
      "entropy": 5.695643091201783,
      "epoch": 4.690674891146589,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003002237005113328,
      "loss": 4.9196,
      "mean_token_accuracy": 0.219387686252594,
      "num_tokens": 107332876.0,
      "step": 51710
    },
    {
      "entropy": 5.692871522903443,
      "epoch": 4.691128447024673,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030019154779505644,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22704126685857773,
      "num_tokens": 107343262.0,
      "step": 51715
    },
    {
      "entropy": 5.620898246765137,
      "epoch": 4.6915820029027575,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003001593945578124,
      "loss": 4.8372,
      "mean_token_accuracy": 0.22025711685419083,
      "num_tokens": 107354281.0,
      "step": 51720
    },
    {
      "entropy": 5.618070983886719,
      "epoch": 4.692035558780842,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003001272408002657,
      "loss": 4.8246,
      "mean_token_accuracy": 0.23399486839771272,
      "num_tokens": 107364011.0,
      "step": 51725
    },
    {
      "entropy": 5.681484508514404,
      "epoch": 4.692489114658926,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030009508652308124,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22898947894573213,
      "num_tokens": 107372943.0,
      "step": 51730
    },
    {
      "entropy": 5.703304195404053,
      "epoch": 4.69294267053701,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030006293172692405,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22835515141487123,
      "num_tokens": 107382838.0,
      "step": 51735
    },
    {
      "entropy": 5.759661722183227,
      "epoch": 4.693396226415095,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030003077641245897,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22728771567344666,
      "num_tokens": 107393250.0,
      "step": 51740
    },
    {
      "entropy": 5.740446758270264,
      "epoch": 4.693849782293179,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029999862058035116,
      "loss": 4.7648,
      "mean_token_accuracy": 0.23679504543542862,
      "num_tokens": 107403153.0,
      "step": 51745
    },
    {
      "entropy": 5.728807544708252,
      "epoch": 4.694303338171263,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029996646423126544,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22529579252004622,
      "num_tokens": 107413092.0,
      "step": 51750
    },
    {
      "entropy": 5.721133089065551,
      "epoch": 4.6947568940493465,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002999343073658669,
      "loss": 4.8889,
      "mean_token_accuracy": 0.2239679679274559,
      "num_tokens": 107423561.0,
      "step": 51755
    },
    {
      "entropy": 5.744008779525757,
      "epoch": 4.695210449927431,
      "grad_norm": 1.375,
      "learning_rate": 0.00029990214998482046,
      "loss": 4.9063,
      "mean_token_accuracy": 0.22527667880058289,
      "num_tokens": 107433954.0,
      "step": 51760
    },
    {
      "entropy": 5.725955772399902,
      "epoch": 4.695664005805515,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002998699920887913,
      "loss": 4.8151,
      "mean_token_accuracy": 0.2316053569316864,
      "num_tokens": 107443150.0,
      "step": 51765
    },
    {
      "entropy": 5.654349756240845,
      "epoch": 4.696117561683599,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002998378336784443,
      "loss": 4.8126,
      "mean_token_accuracy": 0.2267543137073517,
      "num_tokens": 107454250.0,
      "step": 51770
    },
    {
      "entropy": 5.59835729598999,
      "epoch": 4.6965711175616836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002998056747544446,
      "loss": 4.7908,
      "mean_token_accuracy": 0.23501149117946624,
      "num_tokens": 107465058.0,
      "step": 51775
    },
    {
      "entropy": 5.618649101257324,
      "epoch": 4.697024673439768,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002997735153174572,
      "loss": 4.8117,
      "mean_token_accuracy": 0.23399557620286943,
      "num_tokens": 107475449.0,
      "step": 51780
    },
    {
      "entropy": 5.834081029891967,
      "epoch": 4.697478229317852,
      "grad_norm": 1.375,
      "learning_rate": 0.00029974135536814725,
      "loss": 5.0666,
      "mean_token_accuracy": 0.21283139884471894,
      "num_tokens": 107486645.0,
      "step": 51785
    },
    {
      "entropy": 5.704679918289185,
      "epoch": 4.697931785195936,
      "grad_norm": 1.375,
      "learning_rate": 0.0002997091949071797,
      "loss": 4.8719,
      "mean_token_accuracy": 0.22688257992267608,
      "num_tokens": 107497242.0,
      "step": 51790
    },
    {
      "entropy": 5.729033136367798,
      "epoch": 4.698385341074021,
      "grad_norm": 1.375,
      "learning_rate": 0.0002996770339352197,
      "loss": 4.8972,
      "mean_token_accuracy": 0.2236238017678261,
      "num_tokens": 107507596.0,
      "step": 51795
    },
    {
      "entropy": 5.740897989273071,
      "epoch": 4.698838896952105,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002996448724529324,
      "loss": 4.9547,
      "mean_token_accuracy": 0.2185393750667572,
      "num_tokens": 107517795.0,
      "step": 51800
    },
    {
      "entropy": 5.720307064056397,
      "epoch": 4.699292452830189,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029961271046098284,
      "loss": 4.8499,
      "mean_token_accuracy": 0.2285489097237587,
      "num_tokens": 107527805.0,
      "step": 51805
    },
    {
      "entropy": 5.70620059967041,
      "epoch": 4.6997460087082725,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029958054796003615,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2281572312116623,
      "num_tokens": 107537396.0,
      "step": 51810
    },
    {
      "entropy": 5.664787292480469,
      "epoch": 4.700199564586357,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002995483849507575,
      "loss": 4.873,
      "mean_token_accuracy": 0.22679208368062972,
      "num_tokens": 107546876.0,
      "step": 51815
    },
    {
      "entropy": 5.7024781703948975,
      "epoch": 4.700653120464441,
      "grad_norm": 1.515625,
      "learning_rate": 0.000299516221433812,
      "loss": 4.8401,
      "mean_token_accuracy": 0.2315427467226982,
      "num_tokens": 107556865.0,
      "step": 51820
    },
    {
      "entropy": 5.663261270523071,
      "epoch": 4.701106676342525,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029948405740986466,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22021955996751785,
      "num_tokens": 107566305.0,
      "step": 51825
    },
    {
      "entropy": 5.762229919433594,
      "epoch": 4.70156023222061,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002994518928795809,
      "loss": 4.9301,
      "mean_token_accuracy": 0.22024478167295455,
      "num_tokens": 107575640.0,
      "step": 51830
    },
    {
      "entropy": 5.7421904563903805,
      "epoch": 4.702013788098694,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029941972784362574,
      "loss": 4.9411,
      "mean_token_accuracy": 0.23112618178129196,
      "num_tokens": 107585955.0,
      "step": 51835
    },
    {
      "entropy": 5.754872989654541,
      "epoch": 4.702467343976778,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029938756230266436,
      "loss": 4.8952,
      "mean_token_accuracy": 0.22437931150197982,
      "num_tokens": 107596911.0,
      "step": 51840
    },
    {
      "entropy": 5.653985691070557,
      "epoch": 4.702920899854862,
      "grad_norm": 1.4375,
      "learning_rate": 0.000299355396257362,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22944838404655457,
      "num_tokens": 107606117.0,
      "step": 51845
    },
    {
      "entropy": 5.650228977203369,
      "epoch": 4.703374455732947,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002993232297083838,
      "loss": 4.7159,
      "mean_token_accuracy": 0.24623163789510727,
      "num_tokens": 107616363.0,
      "step": 51850
    },
    {
      "entropy": 5.666198825836181,
      "epoch": 4.70382801161103,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002992910626563951,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22974806129932404,
      "num_tokens": 107627402.0,
      "step": 51855
    },
    {
      "entropy": 5.685579681396485,
      "epoch": 4.704281567489114,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029925889510206094,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22676150500774384,
      "num_tokens": 107637446.0,
      "step": 51860
    },
    {
      "entropy": 5.641146659851074,
      "epoch": 4.704735123367199,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002992267270460467,
      "loss": 4.8278,
      "mean_token_accuracy": 0.23108793348073958,
      "num_tokens": 107647386.0,
      "step": 51865
    },
    {
      "entropy": 5.710530853271484,
      "epoch": 4.705188679245283,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002991945584890175,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22982815206050872,
      "num_tokens": 107657881.0,
      "step": 51870
    },
    {
      "entropy": 5.725927400588989,
      "epoch": 4.705642235123367,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029916238943163874,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2323443815112114,
      "num_tokens": 107668478.0,
      "step": 51875
    },
    {
      "entropy": 5.804004430770874,
      "epoch": 4.706095791001451,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002991302198745756,
      "loss": 4.9929,
      "mean_token_accuracy": 0.22051699161529542,
      "num_tokens": 107679149.0,
      "step": 51880
    },
    {
      "entropy": 5.723533678054809,
      "epoch": 4.706549346879536,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002990980498184934,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2196463406085968,
      "num_tokens": 107689471.0,
      "step": 51885
    },
    {
      "entropy": 5.700410270690918,
      "epoch": 4.70700290275762,
      "grad_norm": 1.375,
      "learning_rate": 0.00029906587926405734,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23685257285833358,
      "num_tokens": 107700604.0,
      "step": 51890
    },
    {
      "entropy": 5.596528148651123,
      "epoch": 4.707456458635704,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002990337082119327,
      "loss": 4.7993,
      "mean_token_accuracy": 0.233617202937603,
      "num_tokens": 107710268.0,
      "step": 51895
    },
    {
      "entropy": 5.725350046157837,
      "epoch": 4.707910014513788,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029900153666278503,
      "loss": 4.874,
      "mean_token_accuracy": 0.2210090398788452,
      "num_tokens": 107720328.0,
      "step": 51900
    },
    {
      "entropy": 5.708228588104248,
      "epoch": 4.708363570391873,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029896936461727924,
      "loss": 4.8436,
      "mean_token_accuracy": 0.2278873473405838,
      "num_tokens": 107730295.0,
      "step": 51905
    },
    {
      "entropy": 5.685615825653076,
      "epoch": 4.708817126269956,
      "grad_norm": 1.5,
      "learning_rate": 0.00029893719207608105,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22217898815870285,
      "num_tokens": 107741334.0,
      "step": 51910
    },
    {
      "entropy": 5.695149183273315,
      "epoch": 4.70927068214804,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029890501903985555,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2237737387418747,
      "num_tokens": 107752060.0,
      "step": 51915
    },
    {
      "entropy": 5.707188749313355,
      "epoch": 4.709724238026125,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029887284550926817,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2277616083621979,
      "num_tokens": 107762188.0,
      "step": 51920
    },
    {
      "entropy": 5.707287073135376,
      "epoch": 4.710177793904209,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002988406714849843,
      "loss": 4.889,
      "mean_token_accuracy": 0.21933306157588958,
      "num_tokens": 107771670.0,
      "step": 51925
    },
    {
      "entropy": 5.694632911682129,
      "epoch": 4.710631349782293,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002988084969676692,
      "loss": 4.8346,
      "mean_token_accuracy": 0.23011205047369004,
      "num_tokens": 107781734.0,
      "step": 51930
    },
    {
      "entropy": 5.759563159942627,
      "epoch": 4.711084905660377,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029877632195798843,
      "loss": 4.8839,
      "mean_token_accuracy": 0.2181293249130249,
      "num_tokens": 107792159.0,
      "step": 51935
    },
    {
      "entropy": 5.702487277984619,
      "epoch": 4.711538461538462,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029874414645660713,
      "loss": 4.8762,
      "mean_token_accuracy": 0.229290147125721,
      "num_tokens": 107803153.0,
      "step": 51940
    },
    {
      "entropy": 5.654756116867065,
      "epoch": 4.711992017416546,
      "grad_norm": 1.375,
      "learning_rate": 0.0002987119704641909,
      "loss": 4.8282,
      "mean_token_accuracy": 0.22932239174842833,
      "num_tokens": 107813214.0,
      "step": 51945
    },
    {
      "entropy": 5.676151847839355,
      "epoch": 4.71244557329463,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002986797939814051,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22841181606054306,
      "num_tokens": 107823976.0,
      "step": 51950
    },
    {
      "entropy": 5.715548658370972,
      "epoch": 4.712899129172714,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029864761700891507,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2285393714904785,
      "num_tokens": 107834127.0,
      "step": 51955
    },
    {
      "entropy": 5.766420841217041,
      "epoch": 4.713352685050799,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029861543954738635,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2132691413164139,
      "num_tokens": 107843753.0,
      "step": 51960
    },
    {
      "entropy": 5.753912258148193,
      "epoch": 4.713806240928882,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029858326159748434,
      "loss": 4.8768,
      "mean_token_accuracy": 0.23405207097530364,
      "num_tokens": 107854690.0,
      "step": 51965
    },
    {
      "entropy": 5.6941605567932125,
      "epoch": 4.714259796806966,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029855108315987437,
      "loss": 4.855,
      "mean_token_accuracy": 0.22638794928789138,
      "num_tokens": 107865235.0,
      "step": 51970
    },
    {
      "entropy": 5.685709142684937,
      "epoch": 4.714713352685051,
      "grad_norm": 1.1875,
      "learning_rate": 0.000298518904235222,
      "loss": 4.8519,
      "mean_token_accuracy": 0.226410673558712,
      "num_tokens": 107876052.0,
      "step": 51975
    },
    {
      "entropy": 5.652086162567139,
      "epoch": 4.715166908563135,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002984867248241928,
      "loss": 4.8667,
      "mean_token_accuracy": 0.2300359770655632,
      "num_tokens": 107886187.0,
      "step": 51980
    },
    {
      "entropy": 5.701600360870361,
      "epoch": 4.715620464441219,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002984545449274521,
      "loss": 4.894,
      "mean_token_accuracy": 0.22958866208791734,
      "num_tokens": 107896995.0,
      "step": 51985
    },
    {
      "entropy": 5.736516380310059,
      "epoch": 4.7160740203193035,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002984223645456655,
      "loss": 4.9122,
      "mean_token_accuracy": 0.2275230571627617,
      "num_tokens": 107907868.0,
      "step": 51990
    },
    {
      "entropy": 5.733256816864014,
      "epoch": 4.716527576197388,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029839018367949825,
      "loss": 4.8941,
      "mean_token_accuracy": 0.2257614925503731,
      "num_tokens": 107917564.0,
      "step": 51995
    },
    {
      "entropy": 5.720840358734131,
      "epoch": 4.716981132075472,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002983580023296162,
      "loss": 4.8285,
      "mean_token_accuracy": 0.232545106112957,
      "num_tokens": 107926853.0,
      "step": 52000
    },
    {
      "entropy": 5.657673072814942,
      "epoch": 4.717434687953556,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029832582049668465,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2222544804215431,
      "num_tokens": 107937440.0,
      "step": 52005
    },
    {
      "entropy": 5.69452657699585,
      "epoch": 4.71788824383164,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029829363818136917,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22474961578845978,
      "num_tokens": 107947750.0,
      "step": 52010
    },
    {
      "entropy": 5.74138765335083,
      "epoch": 4.718341799709724,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002982614553843354,
      "loss": 4.9006,
      "mean_token_accuracy": 0.22536349594593047,
      "num_tokens": 107959646.0,
      "step": 52015
    },
    {
      "entropy": 5.7532861709594725,
      "epoch": 4.718795355587808,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029822927210624873,
      "loss": 4.9117,
      "mean_token_accuracy": 0.22418395429849625,
      "num_tokens": 107969678.0,
      "step": 52020
    },
    {
      "entropy": 5.762997436523437,
      "epoch": 4.719248911465892,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029819708834777483,
      "loss": 4.9014,
      "mean_token_accuracy": 0.23029086887836456,
      "num_tokens": 107979589.0,
      "step": 52025
    },
    {
      "entropy": 5.631584787368775,
      "epoch": 4.719702467343977,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002981649041095792,
      "loss": 4.828,
      "mean_token_accuracy": 0.23045136630535126,
      "num_tokens": 107990682.0,
      "step": 52030
    },
    {
      "entropy": 5.693318796157837,
      "epoch": 4.720156023222061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002981327193923275,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2252933070063591,
      "num_tokens": 108000999.0,
      "step": 52035
    },
    {
      "entropy": 5.646811866760254,
      "epoch": 4.720609579100145,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002981005341966852,
      "loss": 4.7852,
      "mean_token_accuracy": 0.22625019997358323,
      "num_tokens": 108010709.0,
      "step": 52040
    },
    {
      "entropy": 5.640130805969238,
      "epoch": 4.7210631349782295,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000298068348523318,
      "loss": 4.8059,
      "mean_token_accuracy": 0.23199942260980605,
      "num_tokens": 108020522.0,
      "step": 52045
    },
    {
      "entropy": 5.639171314239502,
      "epoch": 4.721516690856314,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002980361623728915,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23442708998918532,
      "num_tokens": 108031387.0,
      "step": 52050
    },
    {
      "entropy": 5.693453931808472,
      "epoch": 4.721970246734398,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002980039757460713,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22383427768945693,
      "num_tokens": 108041653.0,
      "step": 52055
    },
    {
      "entropy": 5.783263492584228,
      "epoch": 4.722423802612482,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029797178864352306,
      "loss": 5.0024,
      "mean_token_accuracy": 0.21857763975858688,
      "num_tokens": 108051969.0,
      "step": 52060
    },
    {
      "entropy": 5.7032817840576175,
      "epoch": 4.722877358490566,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029793960106591226,
      "loss": 4.8671,
      "mean_token_accuracy": 0.23025840222835542,
      "num_tokens": 108061364.0,
      "step": 52065
    },
    {
      "entropy": 5.687718152999878,
      "epoch": 4.72333091436865,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002979074130139047,
      "loss": 4.9286,
      "mean_token_accuracy": 0.22460275441408156,
      "num_tokens": 108071140.0,
      "step": 52070
    },
    {
      "entropy": 5.689372396469116,
      "epoch": 4.723784470246734,
      "grad_norm": 1.25,
      "learning_rate": 0.00029787522448816603,
      "loss": 4.9491,
      "mean_token_accuracy": 0.22298672199249267,
      "num_tokens": 108082599.0,
      "step": 52075
    },
    {
      "entropy": 5.749588489532471,
      "epoch": 4.7242380261248185,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029784303548936183,
      "loss": 4.895,
      "mean_token_accuracy": 0.2268347382545471,
      "num_tokens": 108092017.0,
      "step": 52080
    },
    {
      "entropy": 5.77444281578064,
      "epoch": 4.724691582002903,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002978108460181579,
      "loss": 4.9533,
      "mean_token_accuracy": 0.2234375461935997,
      "num_tokens": 108103853.0,
      "step": 52085
    },
    {
      "entropy": 5.688105964660645,
      "epoch": 4.725145137880987,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002977786560752198,
      "loss": 4.8719,
      "mean_token_accuracy": 0.23306871950626373,
      "num_tokens": 108114727.0,
      "step": 52090
    },
    {
      "entropy": 5.672855806350708,
      "epoch": 4.725598693759071,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002977464656612133,
      "loss": 4.8398,
      "mean_token_accuracy": 0.23611743450164796,
      "num_tokens": 108125653.0,
      "step": 52095
    },
    {
      "entropy": 5.72394871711731,
      "epoch": 4.7260522496371555,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029771427477680415,
      "loss": 4.8684,
      "mean_token_accuracy": 0.23263166546821595,
      "num_tokens": 108135281.0,
      "step": 52100
    },
    {
      "entropy": 5.742738199234009,
      "epoch": 4.72650580551524,
      "grad_norm": 1.375,
      "learning_rate": 0.00029768208342265796,
      "loss": 4.879,
      "mean_token_accuracy": 0.22593287229537964,
      "num_tokens": 108145005.0,
      "step": 52105
    },
    {
      "entropy": 5.750942373275757,
      "epoch": 4.726959361393323,
      "grad_norm": 1.25,
      "learning_rate": 0.0002976498915994406,
      "loss": 4.9462,
      "mean_token_accuracy": 0.22204003781080245,
      "num_tokens": 108156477.0,
      "step": 52110
    },
    {
      "entropy": 5.7403322696685795,
      "epoch": 4.727412917271408,
      "grad_norm": 1.53125,
      "learning_rate": 0.00029761769930781757,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22446776777505875,
      "num_tokens": 108166947.0,
      "step": 52115
    },
    {
      "entropy": 5.66885895729065,
      "epoch": 4.727866473149492,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002975855065484548,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2429202154278755,
      "num_tokens": 108175889.0,
      "step": 52120
    },
    {
      "entropy": 5.794272232055664,
      "epoch": 4.728320029027576,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000297553313322018,
      "loss": 4.9543,
      "mean_token_accuracy": 0.22142204344272615,
      "num_tokens": 108186276.0,
      "step": 52125
    },
    {
      "entropy": 5.679504776000977,
      "epoch": 4.72877358490566,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029752111962917297,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2245083585381508,
      "num_tokens": 108196960.0,
      "step": 52130
    },
    {
      "entropy": 5.730292749404907,
      "epoch": 4.7292271407837445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002974889254705854,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22244356721639633,
      "num_tokens": 108207442.0,
      "step": 52135
    },
    {
      "entropy": 5.7272477626800535,
      "epoch": 4.729680696661829,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029745673084692116,
      "loss": 4.8368,
      "mean_token_accuracy": 0.23940708935260774,
      "num_tokens": 108217919.0,
      "step": 52140
    },
    {
      "entropy": 5.652689695358276,
      "epoch": 4.730134252539913,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000297424535758846,
      "loss": 4.814,
      "mean_token_accuracy": 0.23497195690870284,
      "num_tokens": 108228106.0,
      "step": 52145
    },
    {
      "entropy": 5.673973846435547,
      "epoch": 4.730587808417997,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029739234020702564,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2275964081287384,
      "num_tokens": 108238398.0,
      "step": 52150
    },
    {
      "entropy": 5.699950313568115,
      "epoch": 4.731041364296082,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029736014419212605,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22216071635484697,
      "num_tokens": 108248825.0,
      "step": 52155
    },
    {
      "entropy": 5.641301250457763,
      "epoch": 4.731494920174166,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029732794771481305,
      "loss": 4.8153,
      "mean_token_accuracy": 0.23114029914140702,
      "num_tokens": 108258989.0,
      "step": 52160
    },
    {
      "entropy": 5.638158130645752,
      "epoch": 4.731948476052249,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002972957507757523,
      "loss": 4.8173,
      "mean_token_accuracy": 0.2276649922132492,
      "num_tokens": 108268251.0,
      "step": 52165
    },
    {
      "entropy": 5.703626251220703,
      "epoch": 4.7324020319303335,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029726355337560985,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2223575934767723,
      "num_tokens": 108278290.0,
      "step": 52170
    },
    {
      "entropy": 5.7311327934265135,
      "epoch": 4.732855587808418,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029723135551505133,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22717294842004776,
      "num_tokens": 108288727.0,
      "step": 52175
    },
    {
      "entropy": 5.709416198730469,
      "epoch": 4.733309143686502,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029719915719474274,
      "loss": 4.8632,
      "mean_token_accuracy": 0.228701388835907,
      "num_tokens": 108299781.0,
      "step": 52180
    },
    {
      "entropy": 5.754968976974487,
      "epoch": 4.733762699564586,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002971669584153499,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22096543610095978,
      "num_tokens": 108310703.0,
      "step": 52185
    },
    {
      "entropy": 5.660548686981201,
      "epoch": 4.7342162554426706,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002971347591775388,
      "loss": 4.8384,
      "mean_token_accuracy": 0.22919217348098755,
      "num_tokens": 108321096.0,
      "step": 52190
    },
    {
      "entropy": 5.699263620376587,
      "epoch": 4.734669811320755,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002971025594819751,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22457738369703292,
      "num_tokens": 108331899.0,
      "step": 52195
    },
    {
      "entropy": 5.597525882720947,
      "epoch": 4.735123367198839,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029707035932932495,
      "loss": 4.7395,
      "mean_token_accuracy": 0.23539411276578903,
      "num_tokens": 108342248.0,
      "step": 52200
    },
    {
      "entropy": 5.694772863388062,
      "epoch": 4.735576923076923,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029703815872025413,
      "loss": 4.9296,
      "mean_token_accuracy": 0.22604646384716034,
      "num_tokens": 108355264.0,
      "step": 52205
    },
    {
      "entropy": 5.663049411773682,
      "epoch": 4.736030478955008,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002970059576554285,
      "loss": 4.8577,
      "mean_token_accuracy": 0.22374349534511567,
      "num_tokens": 108365929.0,
      "step": 52210
    },
    {
      "entropy": 5.719920825958252,
      "epoch": 4.736484034833092,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002969737561355142,
      "loss": 4.8496,
      "mean_token_accuracy": 0.2240307241678238,
      "num_tokens": 108376874.0,
      "step": 52215
    },
    {
      "entropy": 5.6895068168640135,
      "epoch": 4.736937590711175,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002969415541611769,
      "loss": 4.8266,
      "mean_token_accuracy": 0.22603156864643098,
      "num_tokens": 108387277.0,
      "step": 52220
    },
    {
      "entropy": 5.740444946289062,
      "epoch": 4.7373911465892595,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002969093517330827,
      "loss": 4.987,
      "mean_token_accuracy": 0.21798019707202912,
      "num_tokens": 108399088.0,
      "step": 52225
    },
    {
      "entropy": 5.759555435180664,
      "epoch": 4.737844702467344,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002968771488518975,
      "loss": 4.968,
      "mean_token_accuracy": 0.2231164425611496,
      "num_tokens": 108409875.0,
      "step": 52230
    },
    {
      "entropy": 5.739425754547119,
      "epoch": 4.738298258345428,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002968449455182874,
      "loss": 4.9067,
      "mean_token_accuracy": 0.225621035695076,
      "num_tokens": 108421133.0,
      "step": 52235
    },
    {
      "entropy": 5.850949096679687,
      "epoch": 4.738751814223512,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029681274173291817,
      "loss": 5.0754,
      "mean_token_accuracy": 0.21042024940252305,
      "num_tokens": 108433408.0,
      "step": 52240
    },
    {
      "entropy": 5.692927598953247,
      "epoch": 4.739205370101597,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029678053749645583,
      "loss": 4.7937,
      "mean_token_accuracy": 0.23546018004417418,
      "num_tokens": 108443783.0,
      "step": 52245
    },
    {
      "entropy": 5.579117965698242,
      "epoch": 4.739658925979681,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002967483328095665,
      "loss": 4.7534,
      "mean_token_accuracy": 0.23290360867977142,
      "num_tokens": 108454770.0,
      "step": 52250
    },
    {
      "entropy": 5.659149122238159,
      "epoch": 4.740112481857765,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002967161276729161,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23259420543909073,
      "num_tokens": 108464812.0,
      "step": 52255
    },
    {
      "entropy": 5.722370767593384,
      "epoch": 4.740566037735849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002966839220871707,
      "loss": 4.8978,
      "mean_token_accuracy": 0.21800289005041124,
      "num_tokens": 108474642.0,
      "step": 52260
    },
    {
      "entropy": 5.655215501785278,
      "epoch": 4.741019593613933,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002966517160529962,
      "loss": 4.8424,
      "mean_token_accuracy": 0.2305043339729309,
      "num_tokens": 108484143.0,
      "step": 52265
    },
    {
      "entropy": 5.740351724624634,
      "epoch": 4.741473149492017,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002966195095710587,
      "loss": 4.9979,
      "mean_token_accuracy": 0.21649323105812074,
      "num_tokens": 108494850.0,
      "step": 52270
    },
    {
      "entropy": 5.7345781326293945,
      "epoch": 4.741926705370101,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002965873026420242,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2188183456659317,
      "num_tokens": 108505441.0,
      "step": 52275
    },
    {
      "entropy": 5.758806943893433,
      "epoch": 4.742380261248186,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029655509526655885,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21559732854366304,
      "num_tokens": 108517350.0,
      "step": 52280
    },
    {
      "entropy": 5.663479423522949,
      "epoch": 4.74283381712627,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029652288744532856,
      "loss": 4.8476,
      "mean_token_accuracy": 0.22637329250574112,
      "num_tokens": 108528472.0,
      "step": 52285
    },
    {
      "entropy": 5.6780132293701175,
      "epoch": 4.743287373004354,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002964906791789995,
      "loss": 4.9509,
      "mean_token_accuracy": 0.22134366631507874,
      "num_tokens": 108539834.0,
      "step": 52290
    },
    {
      "entropy": 5.7488175392150875,
      "epoch": 4.743740928882438,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002964584704682377,
      "loss": 4.9718,
      "mean_token_accuracy": 0.22798196524381636,
      "num_tokens": 108551478.0,
      "step": 52295
    },
    {
      "entropy": 5.63799262046814,
      "epoch": 4.744194484760523,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029642626131370936,
      "loss": 4.7384,
      "mean_token_accuracy": 0.2367408514022827,
      "num_tokens": 108560957.0,
      "step": 52300
    },
    {
      "entropy": 5.725354623794556,
      "epoch": 4.744648040638607,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002963940517160804,
      "loss": 4.9012,
      "mean_token_accuracy": 0.22656871378421783,
      "num_tokens": 108571146.0,
      "step": 52305
    },
    {
      "entropy": 5.7113714694976805,
      "epoch": 4.745101596516691,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029636184167601695,
      "loss": 4.8273,
      "mean_token_accuracy": 0.229578597843647,
      "num_tokens": 108581085.0,
      "step": 52310
    },
    {
      "entropy": 5.6662393569946286,
      "epoch": 4.745555152394775,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00029632963119418526,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22988783270120622,
      "num_tokens": 108591103.0,
      "step": 52315
    },
    {
      "entropy": 5.710867929458618,
      "epoch": 4.746008708272859,
      "grad_norm": 1.375,
      "learning_rate": 0.00029629742027125134,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22651751041412355,
      "num_tokens": 108601433.0,
      "step": 52320
    },
    {
      "entropy": 5.6881287574768065,
      "epoch": 4.746462264150943,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029626520890788127,
      "loss": 4.881,
      "mean_token_accuracy": 0.2258609801530838,
      "num_tokens": 108611728.0,
      "step": 52325
    },
    {
      "entropy": 5.7142034530639645,
      "epoch": 4.746915820029027,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029623299710474134,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22255601733922958,
      "num_tokens": 108621521.0,
      "step": 52330
    },
    {
      "entropy": 5.664261245727539,
      "epoch": 4.747369375907112,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029620078486249763,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2229413330554962,
      "num_tokens": 108632921.0,
      "step": 52335
    },
    {
      "entropy": 5.692826414108277,
      "epoch": 4.747822931785196,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002961685721818163,
      "loss": 4.8823,
      "mean_token_accuracy": 0.23165314346551896,
      "num_tokens": 108643244.0,
      "step": 52340
    },
    {
      "entropy": 5.6616472721099855,
      "epoch": 4.74827648766328,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002961363590633634,
      "loss": 4.8625,
      "mean_token_accuracy": 0.23320806324481963,
      "num_tokens": 108653036.0,
      "step": 52345
    },
    {
      "entropy": 5.7584147453308105,
      "epoch": 4.748730043541364,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002961041455078053,
      "loss": 4.913,
      "mean_token_accuracy": 0.2292940065264702,
      "num_tokens": 108663317.0,
      "step": 52350
    },
    {
      "entropy": 5.681039953231812,
      "epoch": 4.749183599419449,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002960719315158081,
      "loss": 4.8099,
      "mean_token_accuracy": 0.22573661357164382,
      "num_tokens": 108673657.0,
      "step": 52355
    },
    {
      "entropy": 5.677040767669678,
      "epoch": 4.749637155297533,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002960397170880378,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22310486286878586,
      "num_tokens": 108683831.0,
      "step": 52360
    },
    {
      "entropy": 5.659259891510009,
      "epoch": 4.750090711175617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029600750222516097,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22849397212266923,
      "num_tokens": 108694062.0,
      "step": 52365
    },
    {
      "entropy": 5.6789881706237795,
      "epoch": 4.7505442670537015,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029597528692784365,
      "loss": 4.8532,
      "mean_token_accuracy": 0.23046325892210007,
      "num_tokens": 108704712.0,
      "step": 52370
    },
    {
      "entropy": 5.693850898742676,
      "epoch": 4.750997822931785,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029594307119675195,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22850783467292785,
      "num_tokens": 108714741.0,
      "step": 52375
    },
    {
      "entropy": 5.745883369445801,
      "epoch": 4.751451378809869,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002959108550325523,
      "loss": 4.9485,
      "mean_token_accuracy": 0.22994166612625122,
      "num_tokens": 108724860.0,
      "step": 52380
    },
    {
      "entropy": 5.683001852035522,
      "epoch": 4.751904934687953,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029587863843591076,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23930975347757338,
      "num_tokens": 108735527.0,
      "step": 52385
    },
    {
      "entropy": 5.711326026916504,
      "epoch": 4.752358490566038,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029584642140749375,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21327580362558365,
      "num_tokens": 108745748.0,
      "step": 52390
    },
    {
      "entropy": 5.723109340667724,
      "epoch": 4.752812046444122,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029581420394796733,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22603173404932023,
      "num_tokens": 108756635.0,
      "step": 52395
    },
    {
      "entropy": 5.674608612060547,
      "epoch": 4.753265602322206,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029578198605799796,
      "loss": 4.815,
      "mean_token_accuracy": 0.23320177495479583,
      "num_tokens": 108765912.0,
      "step": 52400
    },
    {
      "entropy": 5.713481664657593,
      "epoch": 4.7537191582002905,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029574976773825176,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22743331044912338,
      "num_tokens": 108775831.0,
      "step": 52405
    },
    {
      "entropy": 5.702878379821778,
      "epoch": 4.754172714078375,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000295717548989395,
      "loss": 4.8323,
      "mean_token_accuracy": 0.22578840404748918,
      "num_tokens": 108785937.0,
      "step": 52410
    },
    {
      "entropy": 5.676595687866211,
      "epoch": 4.754626269956459,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002956853298120942,
      "loss": 4.8463,
      "mean_token_accuracy": 0.22898517549037933,
      "num_tokens": 108796289.0,
      "step": 52415
    },
    {
      "entropy": 5.725243139266968,
      "epoch": 4.755079825834542,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029565311020701544,
      "loss": 4.9307,
      "mean_token_accuracy": 0.21802253723144532,
      "num_tokens": 108806040.0,
      "step": 52420
    },
    {
      "entropy": 5.727774238586425,
      "epoch": 4.755533381712627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029562089017482515,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22284581661224365,
      "num_tokens": 108815440.0,
      "step": 52425
    },
    {
      "entropy": 5.751468753814697,
      "epoch": 4.755986937590711,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002955886697161895,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21906500160694123,
      "num_tokens": 108826059.0,
      "step": 52430
    },
    {
      "entropy": 5.699904584884644,
      "epoch": 4.756440493468795,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029555644883177496,
      "loss": 4.8426,
      "mean_token_accuracy": 0.23133622705936432,
      "num_tokens": 108836373.0,
      "step": 52435
    },
    {
      "entropy": 5.62133674621582,
      "epoch": 4.756894049346879,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029552422752224787,
      "loss": 4.7941,
      "mean_token_accuracy": 0.23761999905109404,
      "num_tokens": 108847364.0,
      "step": 52440
    },
    {
      "entropy": 5.63627347946167,
      "epoch": 4.757347605224964,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029549200578827444,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22325572073459626,
      "num_tokens": 108856893.0,
      "step": 52445
    },
    {
      "entropy": 5.729235410690308,
      "epoch": 4.757801161103048,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002954597836305212,
      "loss": 4.8192,
      "mean_token_accuracy": 0.2344656839966774,
      "num_tokens": 108867218.0,
      "step": 52450
    },
    {
      "entropy": 5.694946336746216,
      "epoch": 4.758254716981132,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029542756104965434,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22605782002210617,
      "num_tokens": 108876938.0,
      "step": 52455
    },
    {
      "entropy": 5.722285175323487,
      "epoch": 4.7587082728592165,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029539533804634027,
      "loss": 4.8753,
      "mean_token_accuracy": 0.2258369356393814,
      "num_tokens": 108887831.0,
      "step": 52460
    },
    {
      "entropy": 5.692723751068115,
      "epoch": 4.759161828737301,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029536311462124553,
      "loss": 4.8311,
      "mean_token_accuracy": 0.23753968626260757,
      "num_tokens": 108897634.0,
      "step": 52465
    },
    {
      "entropy": 5.656403875350952,
      "epoch": 4.759615384615385,
      "grad_norm": 1.375,
      "learning_rate": 0.0002953308907750363,
      "loss": 4.8178,
      "mean_token_accuracy": 0.2306310921907425,
      "num_tokens": 108907976.0,
      "step": 52470
    },
    {
      "entropy": 5.675820207595825,
      "epoch": 4.760068940493468,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002952986665083792,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2252734437584877,
      "num_tokens": 108918755.0,
      "step": 52475
    },
    {
      "entropy": 5.694291830062866,
      "epoch": 4.760522496371553,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002952664418219404,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2342644527554512,
      "num_tokens": 108928349.0,
      "step": 52480
    },
    {
      "entropy": 5.69525179862976,
      "epoch": 4.760976052249637,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029523421671638644,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21468786150217056,
      "num_tokens": 108938429.0,
      "step": 52485
    },
    {
      "entropy": 5.706098031997681,
      "epoch": 4.761429608127721,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002952019911923837,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22395166158676147,
      "num_tokens": 108948378.0,
      "step": 52490
    },
    {
      "entropy": 5.638829565048217,
      "epoch": 4.7618831640058055,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029516976525059864,
      "loss": 4.8042,
      "mean_token_accuracy": 0.23193731009960175,
      "num_tokens": 108958667.0,
      "step": 52495
    },
    {
      "entropy": 5.7732401371002195,
      "epoch": 4.76233671988389,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002951375388916977,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21860934495925904,
      "num_tokens": 108968955.0,
      "step": 52500
    },
    {
      "entropy": 5.728277206420898,
      "epoch": 4.762790275761974,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002951053121163474,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2291712298989296,
      "num_tokens": 108979565.0,
      "step": 52505
    },
    {
      "entropy": 5.711278343200684,
      "epoch": 4.763243831640058,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000295073084925214,
      "loss": 4.8962,
      "mean_token_accuracy": 0.22517583519220352,
      "num_tokens": 108989547.0,
      "step": 52510
    },
    {
      "entropy": 5.671773862838745,
      "epoch": 4.7636973875181425,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029504085731896413,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2282327890396118,
      "num_tokens": 109000286.0,
      "step": 52515
    },
    {
      "entropy": 5.749588966369629,
      "epoch": 4.764150943396227,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002950086292982642,
      "loss": 4.821,
      "mean_token_accuracy": 0.2338106617331505,
      "num_tokens": 109010676.0,
      "step": 52520
    },
    {
      "entropy": 5.7736889839172365,
      "epoch": 4.764604499274311,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029497640086378077,
      "loss": 4.935,
      "mean_token_accuracy": 0.21653926074504853,
      "num_tokens": 109020800.0,
      "step": 52525
    },
    {
      "entropy": 5.716938829421997,
      "epoch": 4.7650580551523944,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029494417201618027,
      "loss": 4.976,
      "mean_token_accuracy": 0.2205507665872574,
      "num_tokens": 109031158.0,
      "step": 52530
    },
    {
      "entropy": 5.707615900039673,
      "epoch": 4.765511611030479,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002949119427561292,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22886426299810408,
      "num_tokens": 109041664.0,
      "step": 52535
    },
    {
      "entropy": 5.714785814285278,
      "epoch": 4.765965166908563,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029487971308429415,
      "loss": 4.8852,
      "mean_token_accuracy": 0.23214801251888276,
      "num_tokens": 109052633.0,
      "step": 52540
    },
    {
      "entropy": 5.709836101531982,
      "epoch": 4.766418722786647,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002948474830013414,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2332943558692932,
      "num_tokens": 109062687.0,
      "step": 52545
    },
    {
      "entropy": 5.759007930755615,
      "epoch": 4.7668722786647315,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029481525250793777,
      "loss": 4.911,
      "mean_token_accuracy": 0.22893718183040618,
      "num_tokens": 109073348.0,
      "step": 52550
    },
    {
      "entropy": 5.67171802520752,
      "epoch": 4.767325834542816,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002947830216047497,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22057821452617646,
      "num_tokens": 109083276.0,
      "step": 52555
    },
    {
      "entropy": 5.655615758895874,
      "epoch": 4.7677793904209,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029475079029244356,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2280133292078972,
      "num_tokens": 109094342.0,
      "step": 52560
    },
    {
      "entropy": 5.722983694076538,
      "epoch": 4.768232946298984,
      "grad_norm": 1.25,
      "learning_rate": 0.00029471855857168616,
      "loss": 4.8937,
      "mean_token_accuracy": 0.23006550818681717,
      "num_tokens": 109105178.0,
      "step": 52565
    },
    {
      "entropy": 5.770494890213013,
      "epoch": 4.768686502177069,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002946863264431439,
      "loss": 4.9757,
      "mean_token_accuracy": 0.21710769236087799,
      "num_tokens": 109116205.0,
      "step": 52570
    },
    {
      "entropy": 5.676611280441284,
      "epoch": 4.769140058055152,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002946540939074834,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22194610089063643,
      "num_tokens": 109126299.0,
      "step": 52575
    },
    {
      "entropy": 5.739670372009277,
      "epoch": 4.769593613933236,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029462186096537116,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2243604525923729,
      "num_tokens": 109137675.0,
      "step": 52580
    },
    {
      "entropy": 5.703912973403931,
      "epoch": 4.7700471698113205,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029458962761747394,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22587060183286667,
      "num_tokens": 109147716.0,
      "step": 52585
    },
    {
      "entropy": 5.706764030456543,
      "epoch": 4.770500725689405,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002945573938644582,
      "loss": 4.851,
      "mean_token_accuracy": 0.22461943477392196,
      "num_tokens": 109159211.0,
      "step": 52590
    },
    {
      "entropy": 5.692456293106079,
      "epoch": 4.770954281567489,
      "grad_norm": 1.25,
      "learning_rate": 0.00029452515970699054,
      "loss": 4.918,
      "mean_token_accuracy": 0.22108344584703446,
      "num_tokens": 109169490.0,
      "step": 52595
    },
    {
      "entropy": 5.758542680740357,
      "epoch": 4.771407837445573,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029449292514573767,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2179384022951126,
      "num_tokens": 109179687.0,
      "step": 52600
    },
    {
      "entropy": 5.722743225097656,
      "epoch": 4.7718613933236576,
      "grad_norm": 1.515625,
      "learning_rate": 0.00029446069018136613,
      "loss": 4.8658,
      "mean_token_accuracy": 0.22773549258708953,
      "num_tokens": 109189415.0,
      "step": 52605
    },
    {
      "entropy": 5.670944690704346,
      "epoch": 4.772314949201742,
      "grad_norm": 1.375,
      "learning_rate": 0.0002944284548145424,
      "loss": 4.7451,
      "mean_token_accuracy": 0.23052756637334823,
      "num_tokens": 109199281.0,
      "step": 52610
    },
    {
      "entropy": 5.684368658065796,
      "epoch": 4.772768505079826,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002943962190459335,
      "loss": 4.8233,
      "mean_token_accuracy": 0.22915125638246536,
      "num_tokens": 109208347.0,
      "step": 52615
    },
    {
      "entropy": 5.662523651123047,
      "epoch": 4.77322206095791,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029436398287620573,
      "loss": 4.8759,
      "mean_token_accuracy": 0.22414501011371613,
      "num_tokens": 109218026.0,
      "step": 52620
    },
    {
      "entropy": 5.651474094390869,
      "epoch": 4.773675616835995,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002943317463060259,
      "loss": 4.843,
      "mean_token_accuracy": 0.2254362016916275,
      "num_tokens": 109229457.0,
      "step": 52625
    },
    {
      "entropy": 5.6916430473327635,
      "epoch": 4.774129172714078,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002942995093360606,
      "loss": 4.833,
      "mean_token_accuracy": 0.22889277040958406,
      "num_tokens": 109239504.0,
      "step": 52630
    },
    {
      "entropy": 5.618703842163086,
      "epoch": 4.774582728592162,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002942672719669765,
      "loss": 4.7614,
      "mean_token_accuracy": 0.23176366686820984,
      "num_tokens": 109249745.0,
      "step": 52635
    },
    {
      "entropy": 5.6538464546203615,
      "epoch": 4.7750362844702465,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002942350341994404,
      "loss": 4.8279,
      "mean_token_accuracy": 0.22823248207569122,
      "num_tokens": 109259767.0,
      "step": 52640
    },
    {
      "entropy": 5.8229269027709964,
      "epoch": 4.775489840348331,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029420279603411875,
      "loss": 5.0426,
      "mean_token_accuracy": 0.21315561830997468,
      "num_tokens": 109269611.0,
      "step": 52645
    },
    {
      "entropy": 5.8259439945220945,
      "epoch": 4.775943396226415,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002941705574716785,
      "loss": 4.9738,
      "mean_token_accuracy": 0.21583335250616073,
      "num_tokens": 109281396.0,
      "step": 52650
    },
    {
      "entropy": 5.73730502128601,
      "epoch": 4.776396952104499,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002941383185127862,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22996749877929687,
      "num_tokens": 109292346.0,
      "step": 52655
    },
    {
      "entropy": 5.771154117584229,
      "epoch": 4.776850507982584,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029410607915810854,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21828571557998658,
      "num_tokens": 109302966.0,
      "step": 52660
    },
    {
      "entropy": 5.62310791015625,
      "epoch": 4.777304063860668,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002940738394083124,
      "loss": 4.8172,
      "mean_token_accuracy": 0.2289861962199211,
      "num_tokens": 109312558.0,
      "step": 52665
    },
    {
      "entropy": 5.717225074768066,
      "epoch": 4.777757619738752,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029404159926406436,
      "loss": 4.9318,
      "mean_token_accuracy": 0.22840534150600433,
      "num_tokens": 109322364.0,
      "step": 52670
    },
    {
      "entropy": 5.792484617233276,
      "epoch": 4.7782111756168355,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002940093587260312,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2220743864774704,
      "num_tokens": 109332449.0,
      "step": 52675
    },
    {
      "entropy": 5.756906127929687,
      "epoch": 4.778664731494921,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029397711779487954,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23047149777412415,
      "num_tokens": 109343262.0,
      "step": 52680
    },
    {
      "entropy": 5.672053861618042,
      "epoch": 4.779118287373004,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029394487647127645,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23126582503318788,
      "num_tokens": 109352951.0,
      "step": 52685
    },
    {
      "entropy": 5.644257688522339,
      "epoch": 4.779571843251088,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029391263475588844,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22318521440029143,
      "num_tokens": 109363913.0,
      "step": 52690
    },
    {
      "entropy": 5.658127069473267,
      "epoch": 4.780025399129173,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029388039264938224,
      "loss": 4.844,
      "mean_token_accuracy": 0.23034267574548722,
      "num_tokens": 109374405.0,
      "step": 52695
    },
    {
      "entropy": 5.756059360504151,
      "epoch": 4.780478955007257,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002938481501524247,
      "loss": 4.931,
      "mean_token_accuracy": 0.22311010658740998,
      "num_tokens": 109385696.0,
      "step": 52700
    },
    {
      "entropy": 5.707730531692505,
      "epoch": 4.780932510885341,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029381590726568267,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2254899710416794,
      "num_tokens": 109395061.0,
      "step": 52705
    },
    {
      "entropy": 5.710212850570679,
      "epoch": 4.781386066763425,
      "grad_norm": 1.25,
      "learning_rate": 0.00029378366398982286,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22357667982578278,
      "num_tokens": 109406152.0,
      "step": 52710
    },
    {
      "entropy": 5.782017660140991,
      "epoch": 4.78183962264151,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029375142032551216,
      "loss": 4.929,
      "mean_token_accuracy": 0.2212611109018326,
      "num_tokens": 109416596.0,
      "step": 52715
    },
    {
      "entropy": 5.837281131744385,
      "epoch": 4.782293178519594,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029371917627341726,
      "loss": 4.9462,
      "mean_token_accuracy": 0.2203601136803627,
      "num_tokens": 109426169.0,
      "step": 52720
    },
    {
      "entropy": 5.765199184417725,
      "epoch": 4.782746734397678,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000293686931834205,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22857267558574676,
      "num_tokens": 109436359.0,
      "step": 52725
    },
    {
      "entropy": 5.706815528869629,
      "epoch": 4.7832002902757615,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002936546870085422,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22328441441059113,
      "num_tokens": 109446050.0,
      "step": 52730
    },
    {
      "entropy": 5.701969909667969,
      "epoch": 4.783653846153846,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029362244179709576,
      "loss": 4.919,
      "mean_token_accuracy": 0.2262421190738678,
      "num_tokens": 109457124.0,
      "step": 52735
    },
    {
      "entropy": 5.649980974197388,
      "epoch": 4.78410740203193,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029359019620053253,
      "loss": 4.7902,
      "mean_token_accuracy": 0.22887579947710038,
      "num_tokens": 109467642.0,
      "step": 52740
    },
    {
      "entropy": 5.740623331069946,
      "epoch": 4.784560957910014,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002935579502195192,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2217440128326416,
      "num_tokens": 109477877.0,
      "step": 52745
    },
    {
      "entropy": 5.714398670196533,
      "epoch": 4.785014513788099,
      "grad_norm": 1.25,
      "learning_rate": 0.00029352570385472276,
      "loss": 4.8473,
      "mean_token_accuracy": 0.22672291547060014,
      "num_tokens": 109489302.0,
      "step": 52750
    },
    {
      "entropy": 5.700951623916626,
      "epoch": 4.785468069666183,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002934934571068101,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2265818178653717,
      "num_tokens": 109500023.0,
      "step": 52755
    },
    {
      "entropy": 5.67012882232666,
      "epoch": 4.785921625544267,
      "grad_norm": 1.328125,
      "learning_rate": 0.000293461209976448,
      "loss": 4.878,
      "mean_token_accuracy": 0.22981180250644684,
      "num_tokens": 109510588.0,
      "step": 52760
    },
    {
      "entropy": 5.741567134857178,
      "epoch": 4.786375181422351,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002934289624643034,
      "loss": 4.9124,
      "mean_token_accuracy": 0.224912229180336,
      "num_tokens": 109520162.0,
      "step": 52765
    },
    {
      "entropy": 5.725140810012817,
      "epoch": 4.786828737300436,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002933967145710431,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2199263632297516,
      "num_tokens": 109530155.0,
      "step": 52770
    },
    {
      "entropy": 5.713819456100464,
      "epoch": 4.78728229317852,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002933644662973341,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22447557449340821,
      "num_tokens": 109539897.0,
      "step": 52775
    },
    {
      "entropy": 5.669476270675659,
      "epoch": 4.787735849056604,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029333221764384326,
      "loss": 4.8629,
      "mean_token_accuracy": 0.2251901760697365,
      "num_tokens": 109549995.0,
      "step": 52780
    },
    {
      "entropy": 5.744611358642578,
      "epoch": 4.788189404934688,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029329996861123745,
      "loss": 4.9019,
      "mean_token_accuracy": 0.22449939846992492,
      "num_tokens": 109560446.0,
      "step": 52785
    },
    {
      "entropy": 5.658577251434326,
      "epoch": 4.788642960812772,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029326771920018367,
      "loss": 4.8481,
      "mean_token_accuracy": 0.23171466290950776,
      "num_tokens": 109570864.0,
      "step": 52790
    },
    {
      "entropy": 5.761103916168213,
      "epoch": 4.789096516690856,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029323546941134875,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2290053889155388,
      "num_tokens": 109581015.0,
      "step": 52795
    },
    {
      "entropy": 5.684720706939697,
      "epoch": 4.78955007256894,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029320321924539974,
      "loss": 4.8655,
      "mean_token_accuracy": 0.225576052069664,
      "num_tokens": 109591695.0,
      "step": 52800
    },
    {
      "entropy": 5.63878607749939,
      "epoch": 4.790003628447025,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002931709687030035,
      "loss": 4.8234,
      "mean_token_accuracy": 0.23024071007966995,
      "num_tokens": 109602274.0,
      "step": 52805
    },
    {
      "entropy": 5.731855773925782,
      "epoch": 4.790457184325109,
      "grad_norm": 1.484375,
      "learning_rate": 0.00029313871778482693,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2288328304886818,
      "num_tokens": 109612995.0,
      "step": 52810
    },
    {
      "entropy": 5.712751817703247,
      "epoch": 4.790910740203193,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002931064664915371,
      "loss": 4.8369,
      "mean_token_accuracy": 0.22629590928554535,
      "num_tokens": 109623133.0,
      "step": 52815
    },
    {
      "entropy": 5.6302742004394535,
      "epoch": 4.7913642960812775,
      "grad_norm": 1.234375,
      "learning_rate": 0.000293074214823801,
      "loss": 4.8307,
      "mean_token_accuracy": 0.2278583362698555,
      "num_tokens": 109634312.0,
      "step": 52820
    },
    {
      "entropy": 5.722249317169189,
      "epoch": 4.791817851959362,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029304196278228544,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2155337378382683,
      "num_tokens": 109644333.0,
      "step": 52825
    },
    {
      "entropy": 5.6774139404296875,
      "epoch": 4.792271407837445,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002930097103676575,
      "loss": 4.8039,
      "mean_token_accuracy": 0.2241603583097458,
      "num_tokens": 109654730.0,
      "step": 52830
    },
    {
      "entropy": 5.659193706512451,
      "epoch": 4.79272496371553,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029297745758058414,
      "loss": 4.8206,
      "mean_token_accuracy": 0.22981105148792266,
      "num_tokens": 109664867.0,
      "step": 52835
    },
    {
      "entropy": 5.726649951934815,
      "epoch": 4.793178519593614,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029294520442173244,
      "loss": 4.9162,
      "mean_token_accuracy": 0.2279348224401474,
      "num_tokens": 109675090.0,
      "step": 52840
    },
    {
      "entropy": 5.666755676269531,
      "epoch": 4.793632075471698,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002929129508917693,
      "loss": 4.8194,
      "mean_token_accuracy": 0.22896580547094345,
      "num_tokens": 109685046.0,
      "step": 52845
    },
    {
      "entropy": 5.798278331756592,
      "epoch": 4.794085631349782,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002928806969913618,
      "loss": 4.9342,
      "mean_token_accuracy": 0.22219759970903397,
      "num_tokens": 109696240.0,
      "step": 52850
    },
    {
      "entropy": 5.6901181697845455,
      "epoch": 4.794539187227866,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002928484427211769,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22160701751708983,
      "num_tokens": 109707676.0,
      "step": 52855
    },
    {
      "entropy": 5.768873500823974,
      "epoch": 4.794992743105951,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002928161880818817,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21958851367235183,
      "num_tokens": 109718168.0,
      "step": 52860
    },
    {
      "entropy": 5.6786865234375,
      "epoch": 4.795446298984035,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002927839330741431,
      "loss": 4.8303,
      "mean_token_accuracy": 0.2235743522644043,
      "num_tokens": 109728624.0,
      "step": 52865
    },
    {
      "entropy": 5.813884449005127,
      "epoch": 4.795899854862119,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029275167769862834,
      "loss": 4.9646,
      "mean_token_accuracy": 0.2128518670797348,
      "num_tokens": 109738301.0,
      "step": 52870
    },
    {
      "entropy": 5.743026494979858,
      "epoch": 4.7963534107402035,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002927194219560043,
      "loss": 4.9695,
      "mean_token_accuracy": 0.2203739508986473,
      "num_tokens": 109749269.0,
      "step": 52875
    },
    {
      "entropy": 5.690475749969482,
      "epoch": 4.796806966618288,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029268716584693804,
      "loss": 4.8348,
      "mean_token_accuracy": 0.22430347353219987,
      "num_tokens": 109758335.0,
      "step": 52880
    },
    {
      "entropy": 5.715195083618164,
      "epoch": 4.797260522496371,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002926549093720968,
      "loss": 4.8092,
      "mean_token_accuracy": 0.23409441262483596,
      "num_tokens": 109768848.0,
      "step": 52885
    },
    {
      "entropy": 5.669817733764648,
      "epoch": 4.797714078374455,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029262265253214754,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22385039776563645,
      "num_tokens": 109779331.0,
      "step": 52890
    },
    {
      "entropy": 5.70377516746521,
      "epoch": 4.79816763425254,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029259039532775725,
      "loss": 4.8499,
      "mean_token_accuracy": 0.22774562686681749,
      "num_tokens": 109789331.0,
      "step": 52895
    },
    {
      "entropy": 5.71776008605957,
      "epoch": 4.798621190130624,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002925581377595931,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23032056391239167,
      "num_tokens": 109798630.0,
      "step": 52900
    },
    {
      "entropy": 5.695995759963989,
      "epoch": 4.799074746008708,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029252587982832225,
      "loss": 4.8622,
      "mean_token_accuracy": 0.23020026832818985,
      "num_tokens": 109808501.0,
      "step": 52905
    },
    {
      "entropy": 5.708749914169312,
      "epoch": 4.7995283018867925,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002924936215346117,
      "loss": 4.8699,
      "mean_token_accuracy": 0.23062758296728134,
      "num_tokens": 109818847.0,
      "step": 52910
    },
    {
      "entropy": 5.724177694320678,
      "epoch": 4.799981857764877,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029246136287912865,
      "loss": 4.9249,
      "mean_token_accuracy": 0.22241527885198592,
      "num_tokens": 109830017.0,
      "step": 52915
    },
    {
      "entropy": 5.684803438186646,
      "epoch": 4.800435413642961,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002924291038625402,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2242955818772316,
      "num_tokens": 109839795.0,
      "step": 52920
    },
    {
      "entropy": 5.703842735290527,
      "epoch": 4.800888969521045,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029239684448551336,
      "loss": 4.8331,
      "mean_token_accuracy": 0.23117814511060714,
      "num_tokens": 109850202.0,
      "step": 52925
    },
    {
      "entropy": 5.666717052459717,
      "epoch": 4.8013425253991295,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002923645847487153,
      "loss": 4.893,
      "mean_token_accuracy": 0.2276853695511818,
      "num_tokens": 109860625.0,
      "step": 52930
    },
    {
      "entropy": 5.713709831237793,
      "epoch": 4.801796081277214,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002923323246528133,
      "loss": 4.8581,
      "mean_token_accuracy": 0.23624363243579866,
      "num_tokens": 109871241.0,
      "step": 52935
    },
    {
      "entropy": 5.70577220916748,
      "epoch": 4.802249637155297,
      "grad_norm": 1.375,
      "learning_rate": 0.00029230006419847434,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22462093085050583,
      "num_tokens": 109881627.0,
      "step": 52940
    },
    {
      "entropy": 5.746870136260986,
      "epoch": 4.8027031930333814,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002922678033863656,
      "loss": 4.9401,
      "mean_token_accuracy": 0.222772952914238,
      "num_tokens": 109893247.0,
      "step": 52945
    },
    {
      "entropy": 5.712557601928711,
      "epoch": 4.803156748911466,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002922355422171544,
      "loss": 4.899,
      "mean_token_accuracy": 0.22306635677814485,
      "num_tokens": 109904112.0,
      "step": 52950
    },
    {
      "entropy": 5.667606019973755,
      "epoch": 4.80361030478955,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029220328069150774,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2350955128669739,
      "num_tokens": 109913854.0,
      "step": 52955
    },
    {
      "entropy": 5.767124271392822,
      "epoch": 4.804063860667634,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029217101881009284,
      "loss": 4.9331,
      "mean_token_accuracy": 0.22184567898511887,
      "num_tokens": 109923947.0,
      "step": 52960
    },
    {
      "entropy": 5.744030237197876,
      "epoch": 4.8045174165457185,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002921387565735769,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22664909213781356,
      "num_tokens": 109934590.0,
      "step": 52965
    },
    {
      "entropy": 5.709019756317138,
      "epoch": 4.804970972423803,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002921064939826272,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22135301381349565,
      "num_tokens": 109944224.0,
      "step": 52970
    },
    {
      "entropy": 5.686956357955933,
      "epoch": 4.805424528301887,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002920742310379108,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22261488437652588,
      "num_tokens": 109954414.0,
      "step": 52975
    },
    {
      "entropy": 5.712692165374756,
      "epoch": 4.805878084179971,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029204196774009476,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22638538032770156,
      "num_tokens": 109964304.0,
      "step": 52980
    },
    {
      "entropy": 5.727761459350586,
      "epoch": 4.806331640058055,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029200970408984673,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21328738927841187,
      "num_tokens": 109975401.0,
      "step": 52985
    },
    {
      "entropy": 5.7732932567596436,
      "epoch": 4.806785195936139,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002919774400878336,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22519029676914215,
      "num_tokens": 109985964.0,
      "step": 52990
    },
    {
      "entropy": 5.764254474639893,
      "epoch": 4.807238751814223,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002919451757347226,
      "loss": 4.9209,
      "mean_token_accuracy": 0.22534904927015303,
      "num_tokens": 109995476.0,
      "step": 52995
    },
    {
      "entropy": 5.7477428913116455,
      "epoch": 4.8076923076923075,
      "grad_norm": 1.5,
      "learning_rate": 0.0002919129110311812,
      "loss": 4.8319,
      "mean_token_accuracy": 0.228172142803669,
      "num_tokens": 110005238.0,
      "step": 53000
    },
    {
      "entropy": 5.707856702804565,
      "epoch": 4.808145863570392,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002918806459778764,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2128169760107994,
      "num_tokens": 110016120.0,
      "step": 53005
    },
    {
      "entropy": 5.682864952087402,
      "epoch": 4.808599419448476,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002918483805754756,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22305786609649658,
      "num_tokens": 110026739.0,
      "step": 53010
    },
    {
      "entropy": 5.764877557754517,
      "epoch": 4.80905297532656,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002918161148246459,
      "loss": 4.9783,
      "mean_token_accuracy": 0.21849198937416076,
      "num_tokens": 110037491.0,
      "step": 53015
    },
    {
      "entropy": 5.725549030303955,
      "epoch": 4.8095065312046446,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029178384872605465,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22133791744709014,
      "num_tokens": 110048550.0,
      "step": 53020
    },
    {
      "entropy": 5.705082130432129,
      "epoch": 4.809960087082729,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029175158228036924,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2249337688088417,
      "num_tokens": 110058472.0,
      "step": 53025
    },
    {
      "entropy": 5.684668302536011,
      "epoch": 4.810413642960813,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002917193154882568,
      "loss": 4.8137,
      "mean_token_accuracy": 0.23527472615242004,
      "num_tokens": 110070766.0,
      "step": 53030
    },
    {
      "entropy": 5.736445951461792,
      "epoch": 4.810867198838897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002916870483503847,
      "loss": 4.8965,
      "mean_token_accuracy": 0.22753387689590454,
      "num_tokens": 110081739.0,
      "step": 53035
    },
    {
      "entropy": 5.715646982192993,
      "epoch": 4.811320754716981,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029165478086742006,
      "loss": 4.8757,
      "mean_token_accuracy": 0.22525137960910796,
      "num_tokens": 110091075.0,
      "step": 53040
    },
    {
      "entropy": 5.737630414962768,
      "epoch": 4.811774310595065,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029162251304003044,
      "loss": 4.8368,
      "mean_token_accuracy": 0.2309018149971962,
      "num_tokens": 110100530.0,
      "step": 53045
    },
    {
      "entropy": 5.6480035305023195,
      "epoch": 4.812227866473149,
      "grad_norm": 1.375,
      "learning_rate": 0.0002915902448688829,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2233763724565506,
      "num_tokens": 110110866.0,
      "step": 53050
    },
    {
      "entropy": 5.660047340393066,
      "epoch": 4.8126814223512335,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029155797635464493,
      "loss": 4.9127,
      "mean_token_accuracy": 0.23183491080999374,
      "num_tokens": 110121344.0,
      "step": 53055
    },
    {
      "entropy": 5.626062536239624,
      "epoch": 4.813134978229318,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002915257074979838,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2262982115149498,
      "num_tokens": 110130529.0,
      "step": 53060
    },
    {
      "entropy": 5.692236375808716,
      "epoch": 4.813588534107402,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002914934382995668,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22667412012815474,
      "num_tokens": 110141536.0,
      "step": 53065
    },
    {
      "entropy": 5.766056776046753,
      "epoch": 4.814042089985486,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002914611687600613,
      "loss": 4.9256,
      "mean_token_accuracy": 0.22511058300733566,
      "num_tokens": 110152364.0,
      "step": 53070
    },
    {
      "entropy": 5.707515716552734,
      "epoch": 4.814495645863571,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002914288988801347,
      "loss": 4.8976,
      "mean_token_accuracy": 0.22976793795824052,
      "num_tokens": 110163800.0,
      "step": 53075
    },
    {
      "entropy": 5.656997346878052,
      "epoch": 4.814949201741655,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002913966286604541,
      "loss": 4.8093,
      "mean_token_accuracy": 0.22386230230331422,
      "num_tokens": 110174590.0,
      "step": 53080
    },
    {
      "entropy": 5.757140350341797,
      "epoch": 4.815402757619739,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029136435810168726,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22157079577445984,
      "num_tokens": 110185142.0,
      "step": 53085
    },
    {
      "entropy": 5.639705991744995,
      "epoch": 4.815856313497823,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002913320872045012,
      "loss": 4.7575,
      "mean_token_accuracy": 0.24027238190174102,
      "num_tokens": 110194013.0,
      "step": 53090
    },
    {
      "entropy": 5.688896656036377,
      "epoch": 4.816309869375907,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029129981596956353,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23747388422489166,
      "num_tokens": 110204087.0,
      "step": 53095
    },
    {
      "entropy": 5.7392237186431885,
      "epoch": 4.816763425253991,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029126754439754137,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22796111404895783,
      "num_tokens": 110214066.0,
      "step": 53100
    },
    {
      "entropy": 5.718297100067138,
      "epoch": 4.817216981132075,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002912352724891023,
      "loss": 4.8605,
      "mean_token_accuracy": 0.23046908527612686,
      "num_tokens": 110224378.0,
      "step": 53105
    },
    {
      "entropy": 5.652319812774659,
      "epoch": 4.81767053701016,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002912030002449137,
      "loss": 4.7984,
      "mean_token_accuracy": 0.22663492560386658,
      "num_tokens": 110235493.0,
      "step": 53110
    },
    {
      "entropy": 5.663365411758423,
      "epoch": 4.818124092888244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029117072766564285,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21813922226428986,
      "num_tokens": 110246953.0,
      "step": 53115
    },
    {
      "entropy": 5.653994369506836,
      "epoch": 4.818577648766328,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002911384547519573,
      "loss": 4.7671,
      "mean_token_accuracy": 0.2376271143555641,
      "num_tokens": 110257387.0,
      "step": 53120
    },
    {
      "entropy": 5.707979202270508,
      "epoch": 4.819031204644412,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029110618150452437,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22715743482112885,
      "num_tokens": 110268152.0,
      "step": 53125
    },
    {
      "entropy": 5.747511768341065,
      "epoch": 4.819484760522497,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029107390792401145,
      "loss": 4.8745,
      "mean_token_accuracy": 0.225829815864563,
      "num_tokens": 110277933.0,
      "step": 53130
    },
    {
      "entropy": 5.666027498245239,
      "epoch": 4.819938316400581,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002910416340110861,
      "loss": 4.8005,
      "mean_token_accuracy": 0.23051194846630096,
      "num_tokens": 110288219.0,
      "step": 53135
    },
    {
      "entropy": 5.691755580902099,
      "epoch": 4.820391872278664,
      "grad_norm": 1.375,
      "learning_rate": 0.0002910093597664156,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22596426904201508,
      "num_tokens": 110299209.0,
      "step": 53140
    },
    {
      "entropy": 5.7016407489776615,
      "epoch": 4.8208454281567485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002909770851906675,
      "loss": 4.8619,
      "mean_token_accuracy": 0.22626949697732926,
      "num_tokens": 110309482.0,
      "step": 53145
    },
    {
      "entropy": 5.708537673950195,
      "epoch": 4.821298984034833,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029094481028450926,
      "loss": 4.8442,
      "mean_token_accuracy": 0.23105167597532272,
      "num_tokens": 110319499.0,
      "step": 53150
    },
    {
      "entropy": 5.70291428565979,
      "epoch": 4.821752539912917,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002909125350486082,
      "loss": 4.7924,
      "mean_token_accuracy": 0.2370670810341835,
      "num_tokens": 110329453.0,
      "step": 53155
    },
    {
      "entropy": 5.678890514373779,
      "epoch": 4.822206095791001,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002908802594836319,
      "loss": 4.9274,
      "mean_token_accuracy": 0.22061172276735305,
      "num_tokens": 110339343.0,
      "step": 53160
    },
    {
      "entropy": 5.72106351852417,
      "epoch": 4.822659651669086,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002908479835902477,
      "loss": 4.9629,
      "mean_token_accuracy": 0.22094863206148146,
      "num_tokens": 110349101.0,
      "step": 53165
    },
    {
      "entropy": 5.667288970947266,
      "epoch": 4.82311320754717,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002908157073691233,
      "loss": 4.8449,
      "mean_token_accuracy": 0.23003908097743989,
      "num_tokens": 110359377.0,
      "step": 53170
    },
    {
      "entropy": 5.706157875061035,
      "epoch": 4.823566763425254,
      "grad_norm": 1.375,
      "learning_rate": 0.00029078343082092594,
      "loss": 4.8415,
      "mean_token_accuracy": 0.23168621361255645,
      "num_tokens": 110370132.0,
      "step": 53175
    },
    {
      "entropy": 5.675251865386963,
      "epoch": 4.824020319303338,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029075115394632325,
      "loss": 4.8582,
      "mean_token_accuracy": 0.22084247767925264,
      "num_tokens": 110381046.0,
      "step": 53180
    },
    {
      "entropy": 5.749770259857177,
      "epoch": 4.824473875181423,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029071887674598265,
      "loss": 4.9345,
      "mean_token_accuracy": 0.219432233273983,
      "num_tokens": 110392309.0,
      "step": 53185
    },
    {
      "entropy": 5.780007934570312,
      "epoch": 4.824927431059507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002906865992205717,
      "loss": 4.9949,
      "mean_token_accuracy": 0.21967292875051497,
      "num_tokens": 110402464.0,
      "step": 53190
    },
    {
      "entropy": 5.6408935546875,
      "epoch": 4.82538098693759,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002906543213707579,
      "loss": 4.7664,
      "mean_token_accuracy": 0.23759431540966033,
      "num_tokens": 110412272.0,
      "step": 53195
    },
    {
      "entropy": 5.700955724716186,
      "epoch": 4.825834542815675,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029062204319720865,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22535383850336074,
      "num_tokens": 110423425.0,
      "step": 53200
    },
    {
      "entropy": 5.677987432479858,
      "epoch": 4.826288098693759,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002905897647005917,
      "loss": 4.8836,
      "mean_token_accuracy": 0.22114178091287612,
      "num_tokens": 110434469.0,
      "step": 53205
    },
    {
      "entropy": 5.703193473815918,
      "epoch": 4.826741654571843,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002905574858815744,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2307879939675331,
      "num_tokens": 110443777.0,
      "step": 53210
    },
    {
      "entropy": 5.685116195678711,
      "epoch": 4.827195210449927,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002905252067408242,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21065299212932587,
      "num_tokens": 110454481.0,
      "step": 53215
    },
    {
      "entropy": 5.733447122573852,
      "epoch": 4.827648766328012,
      "grad_norm": 1.375,
      "learning_rate": 0.0002904929272790089,
      "loss": 4.9611,
      "mean_token_accuracy": 0.2224806636571884,
      "num_tokens": 110464789.0,
      "step": 53220
    },
    {
      "entropy": 5.6415880680084225,
      "epoch": 4.828102322206096,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002904606474967959,
      "loss": 4.7548,
      "mean_token_accuracy": 0.23687005788087845,
      "num_tokens": 110474276.0,
      "step": 53225
    },
    {
      "entropy": 5.736977386474609,
      "epoch": 4.82855587808418,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029042836739485273,
      "loss": 4.9709,
      "mean_token_accuracy": 0.22200947403907775,
      "num_tokens": 110486193.0,
      "step": 53230
    },
    {
      "entropy": 5.745906639099121,
      "epoch": 4.8290094339622645,
      "grad_norm": 1.25,
      "learning_rate": 0.000290396086973847,
      "loss": 4.8916,
      "mean_token_accuracy": 0.22743541449308396,
      "num_tokens": 110496619.0,
      "step": 53235
    },
    {
      "entropy": 5.719669151306152,
      "epoch": 4.829462989840349,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029036380623444636,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2190220147371292,
      "num_tokens": 110507103.0,
      "step": 53240
    },
    {
      "entropy": 5.763865947723389,
      "epoch": 4.829916545718433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002903315251773182,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22651223540306092,
      "num_tokens": 110516855.0,
      "step": 53245
    },
    {
      "entropy": 5.709060573577881,
      "epoch": 4.830370101596516,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029029924380313024,
      "loss": 4.8614,
      "mean_token_accuracy": 0.231406070291996,
      "num_tokens": 110526220.0,
      "step": 53250
    },
    {
      "entropy": 5.7124556541442875,
      "epoch": 4.830823657474601,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029026696211255,
      "loss": 4.9432,
      "mean_token_accuracy": 0.22026798129081726,
      "num_tokens": 110537029.0,
      "step": 53255
    },
    {
      "entropy": 5.7129669189453125,
      "epoch": 4.831277213352685,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029023468010624503,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22641316652297974,
      "num_tokens": 110548334.0,
      "step": 53260
    },
    {
      "entropy": 5.72319712638855,
      "epoch": 4.831730769230769,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029020239778488303,
      "loss": 4.9131,
      "mean_token_accuracy": 0.22993522882461548,
      "num_tokens": 110557999.0,
      "step": 53265
    },
    {
      "entropy": 5.739657974243164,
      "epoch": 4.832184325108853,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002901701151491316,
      "loss": 4.9983,
      "mean_token_accuracy": 0.22574182450771332,
      "num_tokens": 110568743.0,
      "step": 53270
    },
    {
      "entropy": 5.75551471710205,
      "epoch": 4.832637880986938,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029013783219965835,
      "loss": 4.9022,
      "mean_token_accuracy": 0.23180636912584304,
      "num_tokens": 110578961.0,
      "step": 53275
    },
    {
      "entropy": 5.754303693771362,
      "epoch": 4.833091436865022,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029010554893713086,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2186148002743721,
      "num_tokens": 110590496.0,
      "step": 53280
    },
    {
      "entropy": 5.7678557395935055,
      "epoch": 4.833544992743106,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002900732653622167,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2317003220319748,
      "num_tokens": 110601018.0,
      "step": 53285
    },
    {
      "entropy": 5.6918281555175785,
      "epoch": 4.8339985486211905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002900409814755836,
      "loss": 4.867,
      "mean_token_accuracy": 0.21962923556566238,
      "num_tokens": 110611369.0,
      "step": 53290
    },
    {
      "entropy": 5.720493745803833,
      "epoch": 4.834452104499274,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029000869727789923,
      "loss": 4.9053,
      "mean_token_accuracy": 0.227529139816761,
      "num_tokens": 110621223.0,
      "step": 53295
    },
    {
      "entropy": 5.6726336002349855,
      "epoch": 4.834905660377358,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028997641276983107,
      "loss": 4.805,
      "mean_token_accuracy": 0.22905991971492767,
      "num_tokens": 110631033.0,
      "step": 53300
    },
    {
      "entropy": 5.67767186164856,
      "epoch": 4.835359216255442,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028994412795204695,
      "loss": 4.8513,
      "mean_token_accuracy": 0.22878595143556596,
      "num_tokens": 110642368.0,
      "step": 53305
    },
    {
      "entropy": 5.661333465576172,
      "epoch": 4.835812772133527,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002899118428252144,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22592115849256517,
      "num_tokens": 110653317.0,
      "step": 53310
    },
    {
      "entropy": 5.683364486694336,
      "epoch": 4.836266328011611,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028987955739000123,
      "loss": 4.8722,
      "mean_token_accuracy": 0.2261163741350174,
      "num_tokens": 110663143.0,
      "step": 53315
    },
    {
      "entropy": 5.717665147781372,
      "epoch": 4.836719883889695,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000289847271647075,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22522483617067338,
      "num_tokens": 110673383.0,
      "step": 53320
    },
    {
      "entropy": 5.706106328964234,
      "epoch": 4.8371734397677795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002898149855971034,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2243417978286743,
      "num_tokens": 110683576.0,
      "step": 53325
    },
    {
      "entropy": 5.643167352676391,
      "epoch": 4.837626995645864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028978269924075404,
      "loss": 4.8307,
      "mean_token_accuracy": 0.22620574682950972,
      "num_tokens": 110693986.0,
      "step": 53330
    },
    {
      "entropy": 5.658590984344483,
      "epoch": 4.838080551523948,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002897504125786948,
      "loss": 4.8307,
      "mean_token_accuracy": 0.23188216090202332,
      "num_tokens": 110703724.0,
      "step": 53335
    },
    {
      "entropy": 5.662581110000611,
      "epoch": 4.838534107402032,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028971812561159324,
      "loss": 4.8954,
      "mean_token_accuracy": 0.22620427906513213,
      "num_tokens": 110714251.0,
      "step": 53340
    },
    {
      "entropy": 5.713614749908447,
      "epoch": 4.8389876632801165,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028968583834011714,
      "loss": 4.9185,
      "mean_token_accuracy": 0.22135642617940904,
      "num_tokens": 110724011.0,
      "step": 53345
    },
    {
      "entropy": 5.765115690231323,
      "epoch": 4.8394412191582,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028965355076493405,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22582901865243912,
      "num_tokens": 110733627.0,
      "step": 53350
    },
    {
      "entropy": 5.7468451976776125,
      "epoch": 4.839894775036284,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002896212628867118,
      "loss": 4.8827,
      "mean_token_accuracy": 0.21982210278511047,
      "num_tokens": 110744514.0,
      "step": 53355
    },
    {
      "entropy": 5.763471555709839,
      "epoch": 4.8403483309143684,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028958897470611825,
      "loss": 4.9385,
      "mean_token_accuracy": 0.22468570321798326,
      "num_tokens": 110754257.0,
      "step": 53360
    },
    {
      "entropy": 5.669404792785644,
      "epoch": 4.840801886792453,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002895566862238208,
      "loss": 4.8551,
      "mean_token_accuracy": 0.22839585244655608,
      "num_tokens": 110764383.0,
      "step": 53365
    },
    {
      "entropy": 5.766780281066895,
      "epoch": 4.841255442670537,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002895243974404875,
      "loss": 4.9359,
      "mean_token_accuracy": 0.224902506172657,
      "num_tokens": 110775006.0,
      "step": 53370
    },
    {
      "entropy": 5.769671821594239,
      "epoch": 4.841708998548621,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002894921083567859,
      "loss": 4.8934,
      "mean_token_accuracy": 0.2240691974759102,
      "num_tokens": 110784764.0,
      "step": 53375
    },
    {
      "entropy": 5.732567024230957,
      "epoch": 4.8421625544267055,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002894598189733838,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2360507845878601,
      "num_tokens": 110794942.0,
      "step": 53380
    },
    {
      "entropy": 5.6754542827606205,
      "epoch": 4.84261611030479,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000289427529290949,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23812349289655685,
      "num_tokens": 110804696.0,
      "step": 53385
    },
    {
      "entropy": 5.629809713363647,
      "epoch": 4.843069666182874,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028939523931014924,
      "loss": 4.7734,
      "mean_token_accuracy": 0.24304719120264054,
      "num_tokens": 110814867.0,
      "step": 53390
    },
    {
      "entropy": 5.707390832901001,
      "epoch": 4.843523222060957,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028936294903165224,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22382638305425645,
      "num_tokens": 110824738.0,
      "step": 53395
    },
    {
      "entropy": 5.7175374031066895,
      "epoch": 4.843976777939043,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002893306584561257,
      "loss": 4.9363,
      "mean_token_accuracy": 0.2249444082379341,
      "num_tokens": 110834646.0,
      "step": 53400
    },
    {
      "entropy": 5.76894097328186,
      "epoch": 4.844430333817126,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028929836758423763,
      "loss": 4.9482,
      "mean_token_accuracy": 0.22279154062271117,
      "num_tokens": 110843907.0,
      "step": 53405
    },
    {
      "entropy": 5.717739534378052,
      "epoch": 4.84488388969521,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002892660764166556,
      "loss": 4.9002,
      "mean_token_accuracy": 0.22525620013475417,
      "num_tokens": 110854930.0,
      "step": 53410
    },
    {
      "entropy": 5.647493553161621,
      "epoch": 4.8453374455732945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028923378495404747,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22745131999254226,
      "num_tokens": 110865256.0,
      "step": 53415
    },
    {
      "entropy": 5.742908239364624,
      "epoch": 4.845791001451379,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028920149319708105,
      "loss": 4.9552,
      "mean_token_accuracy": 0.21954317688941954,
      "num_tokens": 110875600.0,
      "step": 53420
    },
    {
      "entropy": 5.801014089584351,
      "epoch": 4.846244557329463,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028916920114642415,
      "loss": 4.9966,
      "mean_token_accuracy": 0.22216343134641647,
      "num_tokens": 110886856.0,
      "step": 53425
    },
    {
      "entropy": 5.72771954536438,
      "epoch": 4.846698113207547,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028913690880274454,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22906962186098098,
      "num_tokens": 110896908.0,
      "step": 53430
    },
    {
      "entropy": 5.699478340148926,
      "epoch": 4.8471516690856316,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028910461616671004,
      "loss": 4.8531,
      "mean_token_accuracy": 0.23159646689891816,
      "num_tokens": 110908987.0,
      "step": 53435
    },
    {
      "entropy": 5.695361852645874,
      "epoch": 4.847605224963716,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028907232323898855,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23401235938072204,
      "num_tokens": 110920036.0,
      "step": 53440
    },
    {
      "entropy": 5.738633394241333,
      "epoch": 4.8480587808418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028904003002024775,
      "loss": 4.93,
      "mean_token_accuracy": 0.22125989347696304,
      "num_tokens": 110931170.0,
      "step": 53445
    },
    {
      "entropy": 5.761282348632813,
      "epoch": 4.8485123367198835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002890077365111556,
      "loss": 4.9139,
      "mean_token_accuracy": 0.22150086760520935,
      "num_tokens": 110942698.0,
      "step": 53450
    },
    {
      "entropy": 5.707619762420654,
      "epoch": 4.848965892597968,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028897544271237985,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22391979694366454,
      "num_tokens": 110953400.0,
      "step": 53455
    },
    {
      "entropy": 5.748131322860718,
      "epoch": 4.849419448476052,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028894314862458844,
      "loss": 4.8948,
      "mean_token_accuracy": 0.22750435918569564,
      "num_tokens": 110963792.0,
      "step": 53460
    },
    {
      "entropy": 5.767831611633301,
      "epoch": 4.849873004354136,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028891085424844913,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22269899100065232,
      "num_tokens": 110973636.0,
      "step": 53465
    },
    {
      "entropy": 5.708690404891968,
      "epoch": 4.8503265602322205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002888785595846298,
      "loss": 4.8862,
      "mean_token_accuracy": 0.23166489750146865,
      "num_tokens": 110983163.0,
      "step": 53470
    },
    {
      "entropy": 5.66907229423523,
      "epoch": 4.850780116110305,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002888462646337983,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22453228235244752,
      "num_tokens": 110993344.0,
      "step": 53475
    },
    {
      "entropy": 5.718780422210694,
      "epoch": 4.851233671988389,
      "grad_norm": 1.375,
      "learning_rate": 0.00028881396939662256,
      "loss": 4.906,
      "mean_token_accuracy": 0.22094124406576157,
      "num_tokens": 111004380.0,
      "step": 53480
    },
    {
      "entropy": 5.708864641189575,
      "epoch": 4.851687227866473,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002887816738737704,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23130547404289245,
      "num_tokens": 111015661.0,
      "step": 53485
    },
    {
      "entropy": 5.698126792907715,
      "epoch": 4.852140783744558,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002887493780659097,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22605706453323365,
      "num_tokens": 111026496.0,
      "step": 53490
    },
    {
      "entropy": 5.661305093765259,
      "epoch": 4.852594339622642,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028871708197370844,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23275681138038634,
      "num_tokens": 111036356.0,
      "step": 53495
    },
    {
      "entropy": 5.718493986129761,
      "epoch": 4.853047895500726,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028868478559783433,
      "loss": 4.9415,
      "mean_token_accuracy": 0.22666852921247482,
      "num_tokens": 111045967.0,
      "step": 53500
    },
    {
      "entropy": 5.718503618240357,
      "epoch": 4.8535014513788095,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028865248893895543,
      "loss": 4.8738,
      "mean_token_accuracy": 0.22280971109867095,
      "num_tokens": 111056459.0,
      "step": 53505
    },
    {
      "entropy": 5.693415212631225,
      "epoch": 4.853955007256894,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028862019199773955,
      "loss": 4.8193,
      "mean_token_accuracy": 0.22341233491897583,
      "num_tokens": 111065901.0,
      "step": 53510
    },
    {
      "entropy": 5.659722757339478,
      "epoch": 4.854408563134978,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002885878947748546,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2273734837770462,
      "num_tokens": 111075721.0,
      "step": 53515
    },
    {
      "entropy": 5.677297067642212,
      "epoch": 4.854862119013062,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028855559727096854,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2332039386034012,
      "num_tokens": 111086453.0,
      "step": 53520
    },
    {
      "entropy": 5.730919075012207,
      "epoch": 4.855315674891147,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028852329948674926,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21830133646726607,
      "num_tokens": 111099130.0,
      "step": 53525
    },
    {
      "entropy": 5.6874536037445065,
      "epoch": 4.855769230769231,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002884910014228648,
      "loss": 4.8514,
      "mean_token_accuracy": 0.22493163049221038,
      "num_tokens": 111108304.0,
      "step": 53530
    },
    {
      "entropy": 5.750700330734253,
      "epoch": 4.856222786647315,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002884587030799829,
      "loss": 4.8356,
      "mean_token_accuracy": 0.2276545375585556,
      "num_tokens": 111118369.0,
      "step": 53535
    },
    {
      "entropy": 5.6577966690063475,
      "epoch": 4.856676342525399,
      "grad_norm": 1.5,
      "learning_rate": 0.00028842640445877153,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22726404964923858,
      "num_tokens": 111128550.0,
      "step": 53540
    },
    {
      "entropy": 5.737801647186279,
      "epoch": 4.857129898403484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028839410555989885,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21648889482021333,
      "num_tokens": 111140824.0,
      "step": 53545
    },
    {
      "entropy": 5.716819524765015,
      "epoch": 4.857583454281567,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028836180638403257,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22313021421432494,
      "num_tokens": 111152241.0,
      "step": 53550
    },
    {
      "entropy": 5.748799991607666,
      "epoch": 4.858037010159651,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002883295069318407,
      "loss": 4.933,
      "mean_token_accuracy": 0.2197170674800873,
      "num_tokens": 111161897.0,
      "step": 53555
    },
    {
      "entropy": 5.747142696380616,
      "epoch": 4.8584905660377355,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002882972072039914,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22953276336193085,
      "num_tokens": 111172586.0,
      "step": 53560
    },
    {
      "entropy": 5.738012647628784,
      "epoch": 4.85894412191582,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028826490720115224,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21973629891872407,
      "num_tokens": 111182734.0,
      "step": 53565
    },
    {
      "entropy": 5.661239242553711,
      "epoch": 4.859397677793904,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028823260692399156,
      "loss": 4.7865,
      "mean_token_accuracy": 0.2367113471031189,
      "num_tokens": 111193551.0,
      "step": 53570
    },
    {
      "entropy": 5.6758044242858885,
      "epoch": 4.859851233671988,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002882003063731771,
      "loss": 4.8433,
      "mean_token_accuracy": 0.22650264650583268,
      "num_tokens": 111202604.0,
      "step": 53575
    },
    {
      "entropy": 5.648223876953125,
      "epoch": 4.860304789550073,
      "grad_norm": 1.25,
      "learning_rate": 0.00028816800554937703,
      "loss": 4.8465,
      "mean_token_accuracy": 0.2342176094651222,
      "num_tokens": 111213675.0,
      "step": 53580
    },
    {
      "entropy": 5.724376106262207,
      "epoch": 4.860758345428157,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028813570445325915,
      "loss": 4.833,
      "mean_token_accuracy": 0.233545158803463,
      "num_tokens": 111223354.0,
      "step": 53585
    },
    {
      "entropy": 5.717669582366943,
      "epoch": 4.861211901306241,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028810340308549164,
      "loss": 4.8039,
      "mean_token_accuracy": 0.23384129703044892,
      "num_tokens": 111232865.0,
      "step": 53590
    },
    {
      "entropy": 5.645920753479004,
      "epoch": 4.861665457184325,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028807110144674243,
      "loss": 4.8453,
      "mean_token_accuracy": 0.23224321454763414,
      "num_tokens": 111242924.0,
      "step": 53595
    },
    {
      "entropy": 5.697003889083862,
      "epoch": 4.86211901306241,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002880387995376794,
      "loss": 4.805,
      "mean_token_accuracy": 0.23652424067258834,
      "num_tokens": 111252921.0,
      "step": 53600
    },
    {
      "entropy": 5.71504430770874,
      "epoch": 4.862572568940493,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002880064973589708,
      "loss": 4.9186,
      "mean_token_accuracy": 0.22686822712421417,
      "num_tokens": 111263176.0,
      "step": 53605
    },
    {
      "entropy": 5.703498506546021,
      "epoch": 4.863026124818577,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028797419491128446,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22136844247579573,
      "num_tokens": 111273043.0,
      "step": 53610
    },
    {
      "entropy": 5.717957019805908,
      "epoch": 4.863479680696662,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028794189219528846,
      "loss": 4.8487,
      "mean_token_accuracy": 0.22567281723022461,
      "num_tokens": 111282720.0,
      "step": 53615
    },
    {
      "entropy": 5.604652404785156,
      "epoch": 4.863933236574746,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028790958921165086,
      "loss": 4.8728,
      "mean_token_accuracy": 0.21556784808635712,
      "num_tokens": 111293946.0,
      "step": 53620
    },
    {
      "entropy": 5.713226556777954,
      "epoch": 4.86438679245283,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002878772859610396,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21980482786893846,
      "num_tokens": 111305052.0,
      "step": 53625
    },
    {
      "entropy": 5.759779548645019,
      "epoch": 4.864840348330914,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028784498244412284,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2251374065876007,
      "num_tokens": 111314216.0,
      "step": 53630
    },
    {
      "entropy": 5.680251884460449,
      "epoch": 4.865293904208999,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028781267866156854,
      "loss": 4.828,
      "mean_token_accuracy": 0.22952256351709366,
      "num_tokens": 111324939.0,
      "step": 53635
    },
    {
      "entropy": 5.752964830398559,
      "epoch": 4.865747460087083,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028778037461404477,
      "loss": 4.8724,
      "mean_token_accuracy": 0.22276898026466369,
      "num_tokens": 111334908.0,
      "step": 53640
    },
    {
      "entropy": 5.660997533798218,
      "epoch": 4.866201015965167,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028774807030221966,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22553714662790297,
      "num_tokens": 111345560.0,
      "step": 53645
    },
    {
      "entropy": 5.6748041152954105,
      "epoch": 4.8666545718432515,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002877157657267611,
      "loss": 4.8079,
      "mean_token_accuracy": 0.2311113342642784,
      "num_tokens": 111355955.0,
      "step": 53650
    },
    {
      "entropy": 5.736548995971679,
      "epoch": 4.867108127721336,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028768346088833737,
      "loss": 4.9359,
      "mean_token_accuracy": 0.22651538997888565,
      "num_tokens": 111365567.0,
      "step": 53655
    },
    {
      "entropy": 5.648615884780884,
      "epoch": 4.867561683599419,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002876511557876164,
      "loss": 4.796,
      "mean_token_accuracy": 0.23362772911787033,
      "num_tokens": 111375740.0,
      "step": 53660
    },
    {
      "entropy": 5.71203351020813,
      "epoch": 4.868015239477503,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028761885042526635,
      "loss": 4.8663,
      "mean_token_accuracy": 0.22828697711229323,
      "num_tokens": 111385593.0,
      "step": 53665
    },
    {
      "entropy": 5.751468563079834,
      "epoch": 4.868468795355588,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028758654480195517,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21549118757247926,
      "num_tokens": 111395731.0,
      "step": 53670
    },
    {
      "entropy": 5.738149404525757,
      "epoch": 4.868922351233672,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002875542389183511,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23398682475090027,
      "num_tokens": 111406140.0,
      "step": 53675
    },
    {
      "entropy": 5.706876039505005,
      "epoch": 4.869375907111756,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028752193277512217,
      "loss": 4.8974,
      "mean_token_accuracy": 0.22574634701013566,
      "num_tokens": 111416853.0,
      "step": 53680
    },
    {
      "entropy": 5.693897104263305,
      "epoch": 4.86982946298984,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002874896263729364,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22748258709907532,
      "num_tokens": 111427043.0,
      "step": 53685
    },
    {
      "entropy": 5.804768323898315,
      "epoch": 4.870283018867925,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002874573197124621,
      "loss": 4.9373,
      "mean_token_accuracy": 0.22401622384786607,
      "num_tokens": 111436791.0,
      "step": 53690
    },
    {
      "entropy": 5.681873369216919,
      "epoch": 4.870736574746009,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002874250127943672,
      "loss": 4.8151,
      "mean_token_accuracy": 0.2400381252169609,
      "num_tokens": 111446603.0,
      "step": 53695
    },
    {
      "entropy": 5.662153863906861,
      "epoch": 4.871190130624093,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028739270561931984,
      "loss": 4.7866,
      "mean_token_accuracy": 0.23792233616113662,
      "num_tokens": 111456942.0,
      "step": 53700
    },
    {
      "entropy": 5.680538558959961,
      "epoch": 4.871643686502177,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028736039818798813,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22305906414985657,
      "num_tokens": 111467671.0,
      "step": 53705
    },
    {
      "entropy": 5.729675436019898,
      "epoch": 4.872097242380261,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002873280905010403,
      "loss": 4.9644,
      "mean_token_accuracy": 0.22017888575792313,
      "num_tokens": 111476495.0,
      "step": 53710
    },
    {
      "entropy": 5.678573513031006,
      "epoch": 4.872550798258345,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028729578255914446,
      "loss": 4.7437,
      "mean_token_accuracy": 0.23702103346586229,
      "num_tokens": 111487491.0,
      "step": 53715
    },
    {
      "entropy": 5.670679903030395,
      "epoch": 4.873004354136429,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002872634743629686,
      "loss": 4.7994,
      "mean_token_accuracy": 0.22724130600690842,
      "num_tokens": 111497970.0,
      "step": 53720
    },
    {
      "entropy": 5.696907615661621,
      "epoch": 4.873457910014514,
      "grad_norm": 1.3125,
      "learning_rate": 0.000287231165913181,
      "loss": 4.8964,
      "mean_token_accuracy": 0.224970343708992,
      "num_tokens": 111507635.0,
      "step": 53725
    },
    {
      "entropy": 5.674832963943482,
      "epoch": 4.873911465892598,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002871988572104498,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22727602273225783,
      "num_tokens": 111516751.0,
      "step": 53730
    },
    {
      "entropy": 5.685202169418335,
      "epoch": 4.874365021770682,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028716654825544316,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22562188655138016,
      "num_tokens": 111527748.0,
      "step": 53735
    },
    {
      "entropy": 5.734482765197754,
      "epoch": 4.8748185776487665,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002871342390488291,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22178497016429902,
      "num_tokens": 111537718.0,
      "step": 53740
    },
    {
      "entropy": 5.626764011383057,
      "epoch": 4.875272133526851,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028710192959127597,
      "loss": 4.7571,
      "mean_token_accuracy": 0.23437585979700087,
      "num_tokens": 111546631.0,
      "step": 53745
    },
    {
      "entropy": 5.692741107940674,
      "epoch": 4.875725689404935,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002870696198834518,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22681108117103577,
      "num_tokens": 111557276.0,
      "step": 53750
    },
    {
      "entropy": 5.6999351501464846,
      "epoch": 4.876179245283019,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028703730992602486,
      "loss": 4.8606,
      "mean_token_accuracy": 0.22272623777389527,
      "num_tokens": 111567146.0,
      "step": 53755
    },
    {
      "entropy": 5.713692045211792,
      "epoch": 4.876632801161103,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028700499971966327,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2216823846101761,
      "num_tokens": 111576887.0,
      "step": 53760
    },
    {
      "entropy": 5.746707820892334,
      "epoch": 4.877086357039187,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028697268926503527,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21884811222553252,
      "num_tokens": 111588212.0,
      "step": 53765
    },
    {
      "entropy": 5.703935718536377,
      "epoch": 4.877539912917271,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028694037856280894,
      "loss": 4.8712,
      "mean_token_accuracy": 0.2273002415895462,
      "num_tokens": 111598308.0,
      "step": 53770
    },
    {
      "entropy": 5.724487733840943,
      "epoch": 4.8779934687953554,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002869080676136525,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22822787016630172,
      "num_tokens": 111608252.0,
      "step": 53775
    },
    {
      "entropy": 5.696218729019165,
      "epoch": 4.87844702467344,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002868757564182343,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2224739521741867,
      "num_tokens": 111619555.0,
      "step": 53780
    },
    {
      "entropy": 5.649436950683594,
      "epoch": 4.878900580551524,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002868434449772223,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23114476948976517,
      "num_tokens": 111629578.0,
      "step": 53785
    },
    {
      "entropy": 5.713950157165527,
      "epoch": 4.879354136429608,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002868111332912849,
      "loss": 4.8805,
      "mean_token_accuracy": 0.22169079333543779,
      "num_tokens": 111639389.0,
      "step": 53790
    },
    {
      "entropy": 5.791858673095703,
      "epoch": 4.8798076923076925,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002867788213610903,
      "loss": 4.9439,
      "mean_token_accuracy": 0.22163280546665193,
      "num_tokens": 111649855.0,
      "step": 53795
    },
    {
      "entropy": 5.783744287490845,
      "epoch": 4.880261248185777,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002867465091873066,
      "loss": 4.9612,
      "mean_token_accuracy": 0.22596526890993118,
      "num_tokens": 111660315.0,
      "step": 53800
    },
    {
      "entropy": 5.713084936141968,
      "epoch": 4.880714804063861,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002867141967706021,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2163013696670532,
      "num_tokens": 111669883.0,
      "step": 53805
    },
    {
      "entropy": 5.664013862609863,
      "epoch": 4.881168359941945,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028668188411164506,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21635933667421342,
      "num_tokens": 111680598.0,
      "step": 53810
    },
    {
      "entropy": 5.736945772171021,
      "epoch": 4.881621915820029,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028664957121110374,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22265576124191283,
      "num_tokens": 111689805.0,
      "step": 53815
    },
    {
      "entropy": 5.749247312545776,
      "epoch": 4.882075471698113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028661725806964624,
      "loss": 4.8174,
      "mean_token_accuracy": 0.2343645066022873,
      "num_tokens": 111700313.0,
      "step": 53820
    },
    {
      "entropy": 5.74837474822998,
      "epoch": 4.882529027576197,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002865849446879409,
      "loss": 4.926,
      "mean_token_accuracy": 0.22320168763399123,
      "num_tokens": 111711301.0,
      "step": 53825
    },
    {
      "entropy": 5.684029054641724,
      "epoch": 4.8829825834542815,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002865526310666559,
      "loss": 4.8262,
      "mean_token_accuracy": 0.2272129699587822,
      "num_tokens": 111720800.0,
      "step": 53830
    },
    {
      "entropy": 5.691203975677491,
      "epoch": 4.883436139332366,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028652031720645963,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2302948847413063,
      "num_tokens": 111731665.0,
      "step": 53835
    },
    {
      "entropy": 5.698927545547486,
      "epoch": 4.88388969521045,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002864880031080202,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22241023927927017,
      "num_tokens": 111742149.0,
      "step": 53840
    },
    {
      "entropy": 5.708329582214356,
      "epoch": 4.884343251088534,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028645568877200603,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23040664345026016,
      "num_tokens": 111753089.0,
      "step": 53845
    },
    {
      "entropy": 5.655672168731689,
      "epoch": 4.8847968069666186,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002864233741990852,
      "loss": 4.833,
      "mean_token_accuracy": 0.23219386339187623,
      "num_tokens": 111762937.0,
      "step": 53850
    },
    {
      "entropy": 5.649271154403687,
      "epoch": 4.885250362844703,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002863910593899261,
      "loss": 4.8105,
      "mean_token_accuracy": 0.2320755124092102,
      "num_tokens": 111772920.0,
      "step": 53855
    },
    {
      "entropy": 5.6976982116699215,
      "epoch": 4.885703918722786,
      "grad_norm": 1.34375,
      "learning_rate": 0.000286358744345197,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22410765439271926,
      "num_tokens": 111783117.0,
      "step": 53860
    },
    {
      "entropy": 5.709683036804199,
      "epoch": 4.8861574746008705,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002863264290655662,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2189591720700264,
      "num_tokens": 111795033.0,
      "step": 53865
    },
    {
      "entropy": 5.682132911682129,
      "epoch": 4.886611030478955,
      "grad_norm": 1.25,
      "learning_rate": 0.00028629411355170193,
      "loss": 4.8498,
      "mean_token_accuracy": 0.22528148740530013,
      "num_tokens": 111805156.0,
      "step": 53870
    },
    {
      "entropy": 5.751446199417114,
      "epoch": 4.887064586357039,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028626179780427246,
      "loss": 4.9402,
      "mean_token_accuracy": 0.22120103538036345,
      "num_tokens": 111815363.0,
      "step": 53875
    },
    {
      "entropy": 5.739631223678589,
      "epoch": 4.887518142235123,
      "grad_norm": 1.640625,
      "learning_rate": 0.00028622948182394623,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2263425827026367,
      "num_tokens": 111824983.0,
      "step": 53880
    },
    {
      "entropy": 5.626009368896485,
      "epoch": 4.8879716981132075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028619716561139143,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2258175104856491,
      "num_tokens": 111836468.0,
      "step": 53885
    },
    {
      "entropy": 5.728803110122681,
      "epoch": 4.888425253991292,
      "grad_norm": 1.25,
      "learning_rate": 0.0002861648491672764,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22371973991394042,
      "num_tokens": 111847659.0,
      "step": 53890
    },
    {
      "entropy": 5.786420679092407,
      "epoch": 4.888878809869376,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028613253249226944,
      "loss": 4.9563,
      "mean_token_accuracy": 0.21959042698144912,
      "num_tokens": 111858742.0,
      "step": 53895
    },
    {
      "entropy": 5.733302354812622,
      "epoch": 4.88933236574746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028610021558703885,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2256762683391571,
      "num_tokens": 111869692.0,
      "step": 53900
    },
    {
      "entropy": 5.695660257339478,
      "epoch": 4.889785921625545,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028606789845225295,
      "loss": 4.8597,
      "mean_token_accuracy": 0.22561905086040496,
      "num_tokens": 111880532.0,
      "step": 53905
    },
    {
      "entropy": 5.763547945022583,
      "epoch": 4.890239477503629,
      "grad_norm": 1.375,
      "learning_rate": 0.00028603558108858016,
      "loss": 4.9603,
      "mean_token_accuracy": 0.21911047101020814,
      "num_tokens": 111891178.0,
      "step": 53910
    },
    {
      "entropy": 5.750147533416748,
      "epoch": 4.890693033381712,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002860032634966887,
      "loss": 4.931,
      "mean_token_accuracy": 0.22735134810209273,
      "num_tokens": 111902043.0,
      "step": 53915
    },
    {
      "entropy": 5.719081163406372,
      "epoch": 4.8911465892597965,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000285970945677247,
      "loss": 4.8372,
      "mean_token_accuracy": 0.2283968970179558,
      "num_tokens": 111912170.0,
      "step": 53920
    },
    {
      "entropy": 5.648064994812012,
      "epoch": 4.891600145137881,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028593862763092327,
      "loss": 4.8564,
      "mean_token_accuracy": 0.23223772943019866,
      "num_tokens": 111922415.0,
      "step": 53925
    },
    {
      "entropy": 5.751843023300171,
      "epoch": 4.892053701015965,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028590630935838596,
      "loss": 4.9395,
      "mean_token_accuracy": 0.22198187410831452,
      "num_tokens": 111933338.0,
      "step": 53930
    },
    {
      "entropy": 5.702044439315796,
      "epoch": 4.892507256894049,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002858739908603034,
      "loss": 4.8685,
      "mean_token_accuracy": 0.2306683748960495,
      "num_tokens": 111942225.0,
      "step": 53935
    },
    {
      "entropy": 5.709688711166382,
      "epoch": 4.892960812772134,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002858416721373439,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2347833201289177,
      "num_tokens": 111953356.0,
      "step": 53940
    },
    {
      "entropy": 5.692207145690918,
      "epoch": 4.893414368650218,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028580935319017593,
      "loss": 4.8022,
      "mean_token_accuracy": 0.22532621622085572,
      "num_tokens": 111964275.0,
      "step": 53945
    },
    {
      "entropy": 5.711901521682739,
      "epoch": 4.893867924528302,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028577703401946777,
      "loss": 4.855,
      "mean_token_accuracy": 0.23261095434427262,
      "num_tokens": 111974006.0,
      "step": 53950
    },
    {
      "entropy": 5.726830387115479,
      "epoch": 4.894321480406386,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028574471462588776,
      "loss": 4.9754,
      "mean_token_accuracy": 0.21775874495506287,
      "num_tokens": 111985085.0,
      "step": 53955
    },
    {
      "entropy": 5.743198013305664,
      "epoch": 4.89477503628447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002857123950101043,
      "loss": 4.9751,
      "mean_token_accuracy": 0.2131907433271408,
      "num_tokens": 111995162.0,
      "step": 53960
    },
    {
      "entropy": 5.76284704208374,
      "epoch": 4.895228592162555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002856800751727858,
      "loss": 4.903,
      "mean_token_accuracy": 0.2220376119017601,
      "num_tokens": 112005625.0,
      "step": 53965
    },
    {
      "entropy": 5.697068595886231,
      "epoch": 4.895682148040638,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028564775511460066,
      "loss": 4.851,
      "mean_token_accuracy": 0.23063214123249054,
      "num_tokens": 112015977.0,
      "step": 53970
    },
    {
      "entropy": 5.690370512008667,
      "epoch": 4.8961357039187225,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002856154348362173,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22830039411783218,
      "num_tokens": 112025655.0,
      "step": 53975
    },
    {
      "entropy": 5.629462623596192,
      "epoch": 4.896589259796807,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028558311433830386,
      "loss": 4.8329,
      "mean_token_accuracy": 0.23692049086093903,
      "num_tokens": 112036986.0,
      "step": 53980
    },
    {
      "entropy": 5.681453132629395,
      "epoch": 4.897042815674891,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000285550793621529,
      "loss": 4.9042,
      "mean_token_accuracy": 0.22871854454278945,
      "num_tokens": 112047541.0,
      "step": 53985
    },
    {
      "entropy": 5.712277746200561,
      "epoch": 4.897496371552975,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002855184726865611,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22325896471738815,
      "num_tokens": 112058125.0,
      "step": 53990
    },
    {
      "entropy": 5.794654321670532,
      "epoch": 4.89794992743106,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028548615153406844,
      "loss": 4.9237,
      "mean_token_accuracy": 0.22130759358406066,
      "num_tokens": 112068712.0,
      "step": 53995
    },
    {
      "entropy": 5.683501052856445,
      "epoch": 4.898403483309144,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028545383016471954,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22819461524486542,
      "num_tokens": 112078529.0,
      "step": 54000
    },
    {
      "epoch": 4.898403483309144,
      "eval_entropy": 5.439466955543104,
      "eval_loss": 4.949206352233887,
      "eval_mean_token_accuracy": 0.23169961707725514,
      "eval_num_tokens": 112078529.0,
      "eval_runtime": 20.6686,
      "eval_samples_per_second": 1710.811,
      "eval_steps_per_second": 213.851,
      "step": 54000
    },
    {
      "entropy": 5.786762332916259,
      "epoch": 4.898857039187228,
      "grad_norm": 1.375,
      "learning_rate": 0.0002854215085791826,
      "loss": 4.9472,
      "mean_token_accuracy": 0.21876634806394576,
      "num_tokens": 112088814.0,
      "step": 54005
    },
    {
      "entropy": 5.692005300521851,
      "epoch": 4.899310595065312,
      "grad_norm": 1.375,
      "learning_rate": 0.00028538918677812643,
      "loss": 4.8598,
      "mean_token_accuracy": 0.23113337755203248,
      "num_tokens": 112099406.0,
      "step": 54010
    },
    {
      "entropy": 5.680268716812134,
      "epoch": 4.899764150943396,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028535686476221915,
      "loss": 4.873,
      "mean_token_accuracy": 0.22273847013711928,
      "num_tokens": 112110762.0,
      "step": 54015
    },
    {
      "entropy": 5.792736530303955,
      "epoch": 4.90021770682148,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002853245425321292,
      "loss": 5.0497,
      "mean_token_accuracy": 0.21029496937990189,
      "num_tokens": 112120636.0,
      "step": 54020
    },
    {
      "entropy": 5.686125469207764,
      "epoch": 4.900671262699564,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028529222008852513,
      "loss": 4.8428,
      "mean_token_accuracy": 0.22900496572256088,
      "num_tokens": 112130756.0,
      "step": 54025
    },
    {
      "entropy": 5.751194953918457,
      "epoch": 4.901124818577649,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002852598974320753,
      "loss": 4.9145,
      "mean_token_accuracy": 0.23226490765810012,
      "num_tokens": 112140881.0,
      "step": 54030
    },
    {
      "entropy": 5.687699842453003,
      "epoch": 4.901578374455733,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028522757456344824,
      "loss": 4.8445,
      "mean_token_accuracy": 0.2290465459227562,
      "num_tokens": 112152337.0,
      "step": 54035
    },
    {
      "entropy": 5.719067096710205,
      "epoch": 4.902031930333817,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002851952514833123,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22088975161314012,
      "num_tokens": 112164199.0,
      "step": 54040
    },
    {
      "entropy": 5.714598512649536,
      "epoch": 4.902485486211901,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028516292819233597,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22964525669813157,
      "num_tokens": 112173594.0,
      "step": 54045
    },
    {
      "entropy": 5.707909870147705,
      "epoch": 4.902939042089986,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028513060469118767,
      "loss": 4.8689,
      "mean_token_accuracy": 0.23154743015766144,
      "num_tokens": 112183261.0,
      "step": 54050
    },
    {
      "entropy": 5.704583597183228,
      "epoch": 4.90339259796807,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028509828098053594,
      "loss": 4.8418,
      "mean_token_accuracy": 0.2308574825525284,
      "num_tokens": 112193506.0,
      "step": 54055
    },
    {
      "entropy": 5.68382716178894,
      "epoch": 4.903846153846154,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002850659570610491,
      "loss": 4.8391,
      "mean_token_accuracy": 0.23197389245033265,
      "num_tokens": 112204665.0,
      "step": 54060
    },
    {
      "entropy": 5.671403980255127,
      "epoch": 4.9042997097242385,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002850336329333958,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22861335575580596,
      "num_tokens": 112214933.0,
      "step": 54065
    },
    {
      "entropy": 5.683179521560669,
      "epoch": 4.904753265602322,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002850013085982443,
      "loss": 4.8404,
      "mean_token_accuracy": 0.23129152953624726,
      "num_tokens": 112224525.0,
      "step": 54070
    },
    {
      "entropy": 5.674804782867431,
      "epoch": 4.905206821480406,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028496898405626326,
      "loss": 4.8058,
      "mean_token_accuracy": 0.23222305327653886,
      "num_tokens": 112234363.0,
      "step": 54075
    },
    {
      "entropy": 5.690914726257324,
      "epoch": 4.90566037735849,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028493665930812113,
      "loss": 4.8172,
      "mean_token_accuracy": 0.22917517870664597,
      "num_tokens": 112245721.0,
      "step": 54080
    },
    {
      "entropy": 5.704526901245117,
      "epoch": 4.906113933236575,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002849043343544863,
      "loss": 4.8716,
      "mean_token_accuracy": 0.234432652592659,
      "num_tokens": 112255775.0,
      "step": 54085
    },
    {
      "entropy": 5.7118744373321535,
      "epoch": 4.906567489114659,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002848720091960273,
      "loss": 4.936,
      "mean_token_accuracy": 0.22226140946149825,
      "num_tokens": 112266049.0,
      "step": 54090
    },
    {
      "entropy": 5.69004487991333,
      "epoch": 4.907021044992743,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028483968383341264,
      "loss": 4.8537,
      "mean_token_accuracy": 0.2194112077355385,
      "num_tokens": 112276663.0,
      "step": 54095
    },
    {
      "entropy": 5.666248226165772,
      "epoch": 4.907474600870827,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002848073582673109,
      "loss": 4.8272,
      "mean_token_accuracy": 0.22625916004180907,
      "num_tokens": 112287065.0,
      "step": 54100
    },
    {
      "entropy": 5.704418420791626,
      "epoch": 4.907928156748912,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002847750324983903,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22276561856269836,
      "num_tokens": 112297053.0,
      "step": 54105
    },
    {
      "entropy": 5.6816990852355955,
      "epoch": 4.908381712626996,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028474270652731966,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22981341779232026,
      "num_tokens": 112306001.0,
      "step": 54110
    },
    {
      "entropy": 5.712575531005859,
      "epoch": 4.908835268505079,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028471038035476735,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2267809435725212,
      "num_tokens": 112315865.0,
      "step": 54115
    },
    {
      "entropy": 5.752542495727539,
      "epoch": 4.9092888243831645,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002846780539814019,
      "loss": 4.9771,
      "mean_token_accuracy": 0.2161294028162956,
      "num_tokens": 112326768.0,
      "step": 54120
    },
    {
      "entropy": 5.66358437538147,
      "epoch": 4.909742380261248,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028464572740789177,
      "loss": 4.7549,
      "mean_token_accuracy": 0.2291353702545166,
      "num_tokens": 112337272.0,
      "step": 54125
    },
    {
      "entropy": 5.765954494476318,
      "epoch": 4.910195936139332,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002846134006349057,
      "loss": 4.9064,
      "mean_token_accuracy": 0.22091783136129378,
      "num_tokens": 112346288.0,
      "step": 54130
    },
    {
      "entropy": 5.653571367263794,
      "epoch": 4.910649492017416,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002845810736631119,
      "loss": 4.7499,
      "mean_token_accuracy": 0.23299187570810317,
      "num_tokens": 112356829.0,
      "step": 54135
    },
    {
      "entropy": 5.628326845169068,
      "epoch": 4.911103047895501,
      "grad_norm": 1.375,
      "learning_rate": 0.00028454874649317915,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23223993927240372,
      "num_tokens": 112367301.0,
      "step": 54140
    },
    {
      "entropy": 5.7186401844024655,
      "epoch": 4.911556603773585,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028451641912577587,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2299736261367798,
      "num_tokens": 112377638.0,
      "step": 54145
    },
    {
      "entropy": 5.731436491012573,
      "epoch": 4.912010159651669,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028448409156157055,
      "loss": 4.8865,
      "mean_token_accuracy": 0.23088494688272476,
      "num_tokens": 112387629.0,
      "step": 54150
    },
    {
      "entropy": 5.742110109329223,
      "epoch": 4.9124637155297535,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002844517638012318,
      "loss": 4.915,
      "mean_token_accuracy": 0.22940946370363235,
      "num_tokens": 112397216.0,
      "step": 54155
    },
    {
      "entropy": 5.7544715881347654,
      "epoch": 4.912917271407838,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028441943584542827,
      "loss": 4.877,
      "mean_token_accuracy": 0.23322973698377608,
      "num_tokens": 112407438.0,
      "step": 54160
    },
    {
      "entropy": 5.694133567810058,
      "epoch": 4.913370827285922,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028438710769482836,
      "loss": 4.7868,
      "mean_token_accuracy": 0.24277843534946442,
      "num_tokens": 112417328.0,
      "step": 54165
    },
    {
      "entropy": 5.711171579360962,
      "epoch": 4.913824383164005,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002843547793501007,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21728071570396423,
      "num_tokens": 112427711.0,
      "step": 54170
    },
    {
      "entropy": 5.705705547332764,
      "epoch": 4.91427793904209,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002843224508119137,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22821189910173417,
      "num_tokens": 112438953.0,
      "step": 54175
    },
    {
      "entropy": 5.723622798919678,
      "epoch": 4.914731494920174,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002842901220809362,
      "loss": 4.9208,
      "mean_token_accuracy": 0.22657449543476105,
      "num_tokens": 112449029.0,
      "step": 54180
    },
    {
      "entropy": 5.653174638748169,
      "epoch": 4.915185050798258,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002842577931578365,
      "loss": 4.7838,
      "mean_token_accuracy": 0.23529146164655684,
      "num_tokens": 112458790.0,
      "step": 54185
    },
    {
      "entropy": 5.667581367492676,
      "epoch": 4.9156386066763424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002842254640432834,
      "loss": 4.869,
      "mean_token_accuracy": 0.22392934560775757,
      "num_tokens": 112469723.0,
      "step": 54190
    },
    {
      "entropy": 5.721121406555175,
      "epoch": 4.916092162554427,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028419313473794516,
      "loss": 4.8759,
      "mean_token_accuracy": 0.23030205816030502,
      "num_tokens": 112480448.0,
      "step": 54195
    },
    {
      "entropy": 5.682301425933838,
      "epoch": 4.916545718432511,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028416080524249073,
      "loss": 4.8135,
      "mean_token_accuracy": 0.22227472215890884,
      "num_tokens": 112490107.0,
      "step": 54200
    },
    {
      "entropy": 5.685569190979004,
      "epoch": 4.916999274310595,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028412847555758835,
      "loss": 4.8713,
      "mean_token_accuracy": 0.23042784929275512,
      "num_tokens": 112499670.0,
      "step": 54205
    },
    {
      "entropy": 5.6859087467193605,
      "epoch": 4.9174528301886795,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002840961456839069,
      "loss": 4.8032,
      "mean_token_accuracy": 0.23523691892623902,
      "num_tokens": 112509087.0,
      "step": 54210
    },
    {
      "entropy": 5.694583988189697,
      "epoch": 4.917906386066764,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002840638156221148,
      "loss": 4.8278,
      "mean_token_accuracy": 0.22525808662176133,
      "num_tokens": 112518506.0,
      "step": 54215
    },
    {
      "entropy": 5.7247706890106205,
      "epoch": 4.918359941944848,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028403148537288063,
      "loss": 4.8887,
      "mean_token_accuracy": 0.22729354947805405,
      "num_tokens": 112528474.0,
      "step": 54220
    },
    {
      "entropy": 5.702308893203735,
      "epoch": 4.918813497822931,
      "grad_norm": 1.5,
      "learning_rate": 0.0002839991549368731,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2236451029777527,
      "num_tokens": 112538513.0,
      "step": 54225
    },
    {
      "entropy": 5.771905088424683,
      "epoch": 4.919267053701016,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028396682431476074,
      "loss": 4.965,
      "mean_token_accuracy": 0.2248364195227623,
      "num_tokens": 112548228.0,
      "step": 54230
    },
    {
      "entropy": 5.7269540309906,
      "epoch": 4.9197206095791,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028393449350721216,
      "loss": 4.8502,
      "mean_token_accuracy": 0.23014936596155167,
      "num_tokens": 112558218.0,
      "step": 54235
    },
    {
      "entropy": 5.7233977794647215,
      "epoch": 4.920174165457184,
      "grad_norm": 1.3125,
      "learning_rate": 0.000283902162514896,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22143557369709016,
      "num_tokens": 112568103.0,
      "step": 54240
    },
    {
      "entropy": 5.6609611988067625,
      "epoch": 4.9206277213352685,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002838698313384809,
      "loss": 4.8081,
      "mean_token_accuracy": 0.23033802956342697,
      "num_tokens": 112577729.0,
      "step": 54245
    },
    {
      "entropy": 5.692463636398315,
      "epoch": 4.921081277213353,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002838374999786353,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2290833130478859,
      "num_tokens": 112589103.0,
      "step": 54250
    },
    {
      "entropy": 5.696480703353882,
      "epoch": 4.921534833091437,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028380516843602797,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23101602047681807,
      "num_tokens": 112598935.0,
      "step": 54255
    },
    {
      "entropy": 5.70030403137207,
      "epoch": 4.921988388969521,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028377283671132756,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2311462014913559,
      "num_tokens": 112609837.0,
      "step": 54260
    },
    {
      "entropy": 5.706631755828857,
      "epoch": 4.9224419448476056,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002837405048052027,
      "loss": 4.9052,
      "mean_token_accuracy": 0.22495495527982712,
      "num_tokens": 112620159.0,
      "step": 54265
    },
    {
      "entropy": 5.675590324401855,
      "epoch": 4.922895500725689,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028370817271832193,
      "loss": 4.8473,
      "mean_token_accuracy": 0.2314787358045578,
      "num_tokens": 112630779.0,
      "step": 54270
    },
    {
      "entropy": 5.637400579452515,
      "epoch": 4.923349056603773,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002836758404513539,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22709064334630966,
      "num_tokens": 112640702.0,
      "step": 54275
    },
    {
      "entropy": 5.732302331924439,
      "epoch": 4.9238026124818575,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002836435080049673,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22644711583852767,
      "num_tokens": 112651294.0,
      "step": 54280
    },
    {
      "entropy": 5.747742700576782,
      "epoch": 4.924256168359942,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002836111753798308,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21619609445333482,
      "num_tokens": 112661570.0,
      "step": 54285
    },
    {
      "entropy": 5.689810705184937,
      "epoch": 4.924709724238026,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028357884257661285,
      "loss": 4.9026,
      "mean_token_accuracy": 0.2255192518234253,
      "num_tokens": 112671996.0,
      "step": 54290
    },
    {
      "entropy": 5.674630260467529,
      "epoch": 4.92516328011611,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002835465095959824,
      "loss": 4.764,
      "mean_token_accuracy": 0.23342894464731218,
      "num_tokens": 112681661.0,
      "step": 54295
    },
    {
      "entropy": 5.740693664550781,
      "epoch": 4.9256168359941945,
      "grad_norm": 1.5,
      "learning_rate": 0.0002835141764386079,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2313784569501877,
      "num_tokens": 112692026.0,
      "step": 54300
    },
    {
      "entropy": 5.761195850372315,
      "epoch": 4.926070391872279,
      "grad_norm": 1.171875,
      "learning_rate": 0.000283481843105158,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21760229021310806,
      "num_tokens": 112703011.0,
      "step": 54305
    },
    {
      "entropy": 5.742293119430542,
      "epoch": 4.926523947750363,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002834495095963015,
      "loss": 4.9343,
      "mean_token_accuracy": 0.22416833490133287,
      "num_tokens": 112712687.0,
      "step": 54310
    },
    {
      "entropy": 5.777934265136719,
      "epoch": 4.926977503628447,
      "grad_norm": 1.359375,
      "learning_rate": 0.000283417175912707,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22527682036161423,
      "num_tokens": 112723422.0,
      "step": 54315
    },
    {
      "entropy": 5.698883962631226,
      "epoch": 4.927431059506532,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002833848420550432,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22811296582221985,
      "num_tokens": 112733793.0,
      "step": 54320
    },
    {
      "entropy": 5.6670241355896,
      "epoch": 4.927884615384615,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028335250802397863,
      "loss": 4.7856,
      "mean_token_accuracy": 0.237040738761425,
      "num_tokens": 112744175.0,
      "step": 54325
    },
    {
      "entropy": 5.737780284881592,
      "epoch": 4.928338171262699,
      "grad_norm": 1.375,
      "learning_rate": 0.00028332017382018203,
      "loss": 4.9631,
      "mean_token_accuracy": 0.22308721095323564,
      "num_tokens": 112754484.0,
      "step": 54330
    },
    {
      "entropy": 5.634085988998413,
      "epoch": 4.9287917271407835,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002832878394443222,
      "loss": 4.7297,
      "mean_token_accuracy": 0.2443850725889206,
      "num_tokens": 112764245.0,
      "step": 54335
    },
    {
      "entropy": 5.693640661239624,
      "epoch": 4.929245283018868,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028325550489706773,
      "loss": 4.8731,
      "mean_token_accuracy": 0.229867322742939,
      "num_tokens": 112773629.0,
      "step": 54340
    },
    {
      "entropy": 5.690195846557617,
      "epoch": 4.929698838896952,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002832231701790873,
      "loss": 4.8387,
      "mean_token_accuracy": 0.22893824130296708,
      "num_tokens": 112784091.0,
      "step": 54345
    },
    {
      "entropy": 5.6803254127502445,
      "epoch": 4.930152394775036,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028319083529104966,
      "loss": 4.8145,
      "mean_token_accuracy": 0.22334740459918975,
      "num_tokens": 112794116.0,
      "step": 54350
    },
    {
      "entropy": 5.699283266067505,
      "epoch": 4.930605950653121,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028315850023362345,
      "loss": 4.8652,
      "mean_token_accuracy": 0.2248288318514824,
      "num_tokens": 112804717.0,
      "step": 54355
    },
    {
      "entropy": 5.670443391799926,
      "epoch": 4.931059506531205,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028312616500747727,
      "loss": 4.8217,
      "mean_token_accuracy": 0.22858214527368545,
      "num_tokens": 112815384.0,
      "step": 54360
    },
    {
      "entropy": 5.709795761108398,
      "epoch": 4.931513062409289,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028309382961328003,
      "loss": 4.9206,
      "mean_token_accuracy": 0.2262149080634117,
      "num_tokens": 112825813.0,
      "step": 54365
    },
    {
      "entropy": 5.681831026077271,
      "epoch": 4.931966618287373,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028306149405170035,
      "loss": 4.8127,
      "mean_token_accuracy": 0.2383423015475273,
      "num_tokens": 112836032.0,
      "step": 54370
    },
    {
      "entropy": 5.669189977645874,
      "epoch": 4.932420174165458,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002830291583234068,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22222245931625367,
      "num_tokens": 112846042.0,
      "step": 54375
    },
    {
      "entropy": 5.693417263031006,
      "epoch": 4.932873730043541,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028299682242906825,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22258464694023133,
      "num_tokens": 112856409.0,
      "step": 54380
    },
    {
      "entropy": 5.728625345230102,
      "epoch": 4.933327285921625,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028296448636935345,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21991202384233474,
      "num_tokens": 112866788.0,
      "step": 54385
    },
    {
      "entropy": 5.6594569206237795,
      "epoch": 4.9337808417997095,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028293215014493095,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23661718070507048,
      "num_tokens": 112876670.0,
      "step": 54390
    },
    {
      "entropy": 5.697341108322144,
      "epoch": 4.934234397677794,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002828998137564697,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22278357148170472,
      "num_tokens": 112887337.0,
      "step": 54395
    },
    {
      "entropy": 5.631769847869873,
      "epoch": 4.934687953555878,
      "grad_norm": 1.375,
      "learning_rate": 0.0002828674772046382,
      "loss": 4.8945,
      "mean_token_accuracy": 0.21878768503665924,
      "num_tokens": 112897693.0,
      "step": 54400
    },
    {
      "entropy": 5.700870227813721,
      "epoch": 4.935141509433962,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002828351404901052,
      "loss": 4.8648,
      "mean_token_accuracy": 0.21976656466722488,
      "num_tokens": 112907848.0,
      "step": 54405
    },
    {
      "entropy": 5.724103879928589,
      "epoch": 4.935595065312047,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002828028036135396,
      "loss": 4.8859,
      "mean_token_accuracy": 0.23390974700450898,
      "num_tokens": 112917344.0,
      "step": 54410
    },
    {
      "entropy": 5.702811765670776,
      "epoch": 4.936048621190131,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002827704665756101,
      "loss": 4.8139,
      "mean_token_accuracy": 0.23413980156183242,
      "num_tokens": 112928004.0,
      "step": 54415
    },
    {
      "entropy": 5.703136873245239,
      "epoch": 4.936502177068215,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028273812937698524,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23214667290449142,
      "num_tokens": 112938999.0,
      "step": 54420
    },
    {
      "entropy": 5.699092102050781,
      "epoch": 4.9369557329462985,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028270579201833385,
      "loss": 4.846,
      "mean_token_accuracy": 0.22940158247947692,
      "num_tokens": 112948246.0,
      "step": 54425
    },
    {
      "entropy": 5.729854965209961,
      "epoch": 4.937409288824383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002826734545003248,
      "loss": 4.8958,
      "mean_token_accuracy": 0.22712025791406631,
      "num_tokens": 112958430.0,
      "step": 54430
    },
    {
      "entropy": 5.6736925601959225,
      "epoch": 4.937862844702467,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002826411168236268,
      "loss": 4.8477,
      "mean_token_accuracy": 0.2314803794026375,
      "num_tokens": 112969327.0,
      "step": 54435
    },
    {
      "entropy": 5.723709964752198,
      "epoch": 4.938316400580551,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028260877898890843,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22638148218393325,
      "num_tokens": 112980413.0,
      "step": 54440
    },
    {
      "entropy": 5.663726472854615,
      "epoch": 4.938769956458636,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028257644099683864,
      "loss": 4.8281,
      "mean_token_accuracy": 0.23045842200517655,
      "num_tokens": 112990924.0,
      "step": 54445
    },
    {
      "entropy": 5.708080863952636,
      "epoch": 4.93922351233672,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002825441028480862,
      "loss": 4.9118,
      "mean_token_accuracy": 0.21522739231586457,
      "num_tokens": 113001949.0,
      "step": 54450
    },
    {
      "entropy": 5.774792814254761,
      "epoch": 4.939677068214804,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002825117645433196,
      "loss": 4.9804,
      "mean_token_accuracy": 0.22294891327619554,
      "num_tokens": 113012149.0,
      "step": 54455
    },
    {
      "entropy": 5.70626482963562,
      "epoch": 4.940130624092888,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028247942608320797,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22979073226451874,
      "num_tokens": 113022783.0,
      "step": 54460
    },
    {
      "entropy": 5.677342653274536,
      "epoch": 4.940584179970973,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028244708746841984,
      "loss": 4.8841,
      "mean_token_accuracy": 0.23036383986473083,
      "num_tokens": 113032486.0,
      "step": 54465
    },
    {
      "entropy": 5.760522413253784,
      "epoch": 4.941037735849057,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002824147486996241,
      "loss": 4.9391,
      "mean_token_accuracy": 0.23161107152700425,
      "num_tokens": 113042430.0,
      "step": 54470
    },
    {
      "entropy": 5.708906793594361,
      "epoch": 4.941491291727141,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028238240977748936,
      "loss": 4.864,
      "mean_token_accuracy": 0.21951953023672105,
      "num_tokens": 113052917.0,
      "step": 54475
    },
    {
      "entropy": 5.702584218978882,
      "epoch": 4.941944847605225,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028235007070268454,
      "loss": 4.9079,
      "mean_token_accuracy": 0.2248315006494522,
      "num_tokens": 113063786.0,
      "step": 54480
    },
    {
      "entropy": 5.727828407287598,
      "epoch": 4.942398403483309,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028231773147587843,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2265966549515724,
      "num_tokens": 113073979.0,
      "step": 54485
    },
    {
      "entropy": 5.6551214218139645,
      "epoch": 4.942851959361393,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002822853920977397,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23026318699121476,
      "num_tokens": 113085439.0,
      "step": 54490
    },
    {
      "entropy": 5.687733125686646,
      "epoch": 4.943305515239477,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002822530525689373,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2236199527978897,
      "num_tokens": 113096825.0,
      "step": 54495
    },
    {
      "entropy": 5.623260068893432,
      "epoch": 4.943759071117562,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028222071289013984,
      "loss": 4.7848,
      "mean_token_accuracy": 0.23784786760807036,
      "num_tokens": 113107267.0,
      "step": 54500
    },
    {
      "entropy": 5.658058452606201,
      "epoch": 4.944212626995646,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028218837306201623,
      "loss": 4.8043,
      "mean_token_accuracy": 0.22736668288707734,
      "num_tokens": 113118438.0,
      "step": 54505
    },
    {
      "entropy": 5.720398330688477,
      "epoch": 4.94466618287373,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028215603308523517,
      "loss": 4.8444,
      "mean_token_accuracy": 0.23482178151607513,
      "num_tokens": 113128361.0,
      "step": 54510
    },
    {
      "entropy": 5.728179311752319,
      "epoch": 4.945119738751814,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028212369296046566,
      "loss": 4.9169,
      "mean_token_accuracy": 0.22803172022104262,
      "num_tokens": 113138907.0,
      "step": 54515
    },
    {
      "entropy": 5.624498271942139,
      "epoch": 4.945573294629899,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028209135268837627,
      "loss": 4.9046,
      "mean_token_accuracy": 0.22451532930135726,
      "num_tokens": 113149107.0,
      "step": 54520
    },
    {
      "entropy": 5.698498106002807,
      "epoch": 4.946026850507983,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028205901226963586,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22430721670389175,
      "num_tokens": 113159300.0,
      "step": 54525
    },
    {
      "entropy": 5.653660917282105,
      "epoch": 4.946480406386067,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028202667170491337,
      "loss": 4.7977,
      "mean_token_accuracy": 0.23024680614471435,
      "num_tokens": 113168911.0,
      "step": 54530
    },
    {
      "entropy": 5.683440017700195,
      "epoch": 4.946933962264151,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002819943309948775,
      "loss": 4.84,
      "mean_token_accuracy": 0.22778553515672684,
      "num_tokens": 113179514.0,
      "step": 54535
    },
    {
      "entropy": 5.683540296554566,
      "epoch": 4.947387518142235,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000281961990140197,
      "loss": 4.8443,
      "mean_token_accuracy": 0.2294665053486824,
      "num_tokens": 113189732.0,
      "step": 54540
    },
    {
      "entropy": 5.723318147659302,
      "epoch": 4.947841074020319,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028192964914154086,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22808950245380402,
      "num_tokens": 113200012.0,
      "step": 54545
    },
    {
      "entropy": 5.714090299606323,
      "epoch": 4.948294629898403,
      "grad_norm": 1.375,
      "learning_rate": 0.0002818973079995778,
      "loss": 4.9139,
      "mean_token_accuracy": 0.23177824020385743,
      "num_tokens": 113210504.0,
      "step": 54550
    },
    {
      "entropy": 5.689174747467041,
      "epoch": 4.948748185776488,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002818649667149766,
      "loss": 4.8474,
      "mean_token_accuracy": 0.22797897458076477,
      "num_tokens": 113220520.0,
      "step": 54555
    },
    {
      "entropy": 5.645603275299072,
      "epoch": 4.949201741654572,
      "grad_norm": 1.375,
      "learning_rate": 0.00028183262528840616,
      "loss": 4.7644,
      "mean_token_accuracy": 0.23811232000589372,
      "num_tokens": 113230355.0,
      "step": 54560
    },
    {
      "entropy": 5.7030433177947994,
      "epoch": 4.949655297532656,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028180028372053537,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21602719277143478,
      "num_tokens": 113240184.0,
      "step": 54565
    },
    {
      "entropy": 5.755028533935547,
      "epoch": 4.9501088534107405,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028176794201203293,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22618512958288192,
      "num_tokens": 113249748.0,
      "step": 54570
    },
    {
      "entropy": 5.75802845954895,
      "epoch": 4.950562409288825,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028173560016356763,
      "loss": 4.856,
      "mean_token_accuracy": 0.22724371999502183,
      "num_tokens": 113260018.0,
      "step": 54575
    },
    {
      "entropy": 5.678590536117554,
      "epoch": 4.951015965166908,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002817032581758085,
      "loss": 4.7956,
      "mean_token_accuracy": 0.23465741872787477,
      "num_tokens": 113269682.0,
      "step": 54580
    },
    {
      "entropy": 5.700878095626831,
      "epoch": 4.951469521044992,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028167091604942423,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2220049500465393,
      "num_tokens": 113279485.0,
      "step": 54585
    },
    {
      "entropy": 5.770608615875244,
      "epoch": 4.951923076923077,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028163857378508386,
      "loss": 4.9387,
      "mean_token_accuracy": 0.22448277175426484,
      "num_tokens": 113290559.0,
      "step": 54590
    },
    {
      "entropy": 5.753852510452271,
      "epoch": 4.952376632801161,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028160623138345585,
      "loss": 4.9535,
      "mean_token_accuracy": 0.22358934879302977,
      "num_tokens": 113300581.0,
      "step": 54595
    },
    {
      "entropy": 5.713739204406738,
      "epoch": 4.952830188679245,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028157388884520953,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2239859491586685,
      "num_tokens": 113311261.0,
      "step": 54600
    },
    {
      "entropy": 5.686870813369751,
      "epoch": 4.9532837445573294,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028154154617101337,
      "loss": 4.8037,
      "mean_token_accuracy": 0.231002976000309,
      "num_tokens": 113321701.0,
      "step": 54605
    },
    {
      "entropy": 5.680002117156983,
      "epoch": 4.953737300435414,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028150920336153637,
      "loss": 4.8594,
      "mean_token_accuracy": 0.2291154980659485,
      "num_tokens": 113332364.0,
      "step": 54610
    },
    {
      "entropy": 5.640918922424317,
      "epoch": 4.954190856313498,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002814768604174475,
      "loss": 4.8172,
      "mean_token_accuracy": 0.22883844524621963,
      "num_tokens": 113342920.0,
      "step": 54615
    },
    {
      "entropy": 5.740123605728149,
      "epoch": 4.954644412191582,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002814445173394154,
      "loss": 4.8749,
      "mean_token_accuracy": 0.2275213047862053,
      "num_tokens": 113353602.0,
      "step": 54620
    },
    {
      "entropy": 5.686970043182373,
      "epoch": 4.9550979680696665,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028141217412810905,
      "loss": 4.7863,
      "mean_token_accuracy": 0.23204198032617568,
      "num_tokens": 113362413.0,
      "step": 54625
    },
    {
      "entropy": 5.703030633926391,
      "epoch": 4.955551523947751,
      "grad_norm": 1.5,
      "learning_rate": 0.0002813798307841973,
      "loss": 4.8521,
      "mean_token_accuracy": 0.22893642336130143,
      "num_tokens": 113373016.0,
      "step": 54630
    },
    {
      "entropy": 5.661010026931763,
      "epoch": 4.956005079825834,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002813474873083491,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22425884902477264,
      "num_tokens": 113384798.0,
      "step": 54635
    },
    {
      "entropy": 5.744677257537842,
      "epoch": 4.956458635703918,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002813151437012331,
      "loss": 4.905,
      "mean_token_accuracy": 0.22854529023170472,
      "num_tokens": 113396266.0,
      "step": 54640
    },
    {
      "entropy": 5.721941900253296,
      "epoch": 4.956912191582003,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002812827999635185,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22364733815193177,
      "num_tokens": 113406475.0,
      "step": 54645
    },
    {
      "entropy": 5.691723728179932,
      "epoch": 4.957365747460087,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028125045609587383,
      "loss": 4.8303,
      "mean_token_accuracy": 0.23171280324459076,
      "num_tokens": 113415835.0,
      "step": 54650
    },
    {
      "entropy": 5.7426961898803714,
      "epoch": 4.957819303338171,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002812181120989683,
      "loss": 4.8316,
      "mean_token_accuracy": 0.2333318278193474,
      "num_tokens": 113426359.0,
      "step": 54655
    },
    {
      "entropy": 5.74349570274353,
      "epoch": 4.9582728592162555,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002811857679734705,
      "loss": 4.9722,
      "mean_token_accuracy": 0.2200889214873314,
      "num_tokens": 113437467.0,
      "step": 54660
    },
    {
      "entropy": 5.717677307128906,
      "epoch": 4.95872641509434,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002811534237200494,
      "loss": 4.948,
      "mean_token_accuracy": 0.2235303446650505,
      "num_tokens": 113447510.0,
      "step": 54665
    },
    {
      "entropy": 5.721703243255615,
      "epoch": 4.959179970972424,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002811210793393741,
      "loss": 4.8572,
      "mean_token_accuracy": 0.2208423137664795,
      "num_tokens": 113457559.0,
      "step": 54670
    },
    {
      "entropy": 5.758690118789673,
      "epoch": 4.959633526850508,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028108873483211316,
      "loss": 4.9628,
      "mean_token_accuracy": 0.22915026694536209,
      "num_tokens": 113468589.0,
      "step": 54675
    },
    {
      "entropy": 5.6979289054870605,
      "epoch": 4.960087082728592,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028105639019893573,
      "loss": 4.8458,
      "mean_token_accuracy": 0.229690982401371,
      "num_tokens": 113478585.0,
      "step": 54680
    },
    {
      "entropy": 5.724532175064087,
      "epoch": 4.960540638606677,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028102404544051056,
      "loss": 4.8528,
      "mean_token_accuracy": 0.2330579489469528,
      "num_tokens": 113487980.0,
      "step": 54685
    },
    {
      "entropy": 5.723384094238281,
      "epoch": 4.96099419448476,
      "grad_norm": 1.375,
      "learning_rate": 0.00028099170055750667,
      "loss": 4.8697,
      "mean_token_accuracy": 0.23216424584388734,
      "num_tokens": 113498123.0,
      "step": 54690
    },
    {
      "entropy": 5.69852876663208,
      "epoch": 4.9614477503628445,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028095935555059277,
      "loss": 4.8305,
      "mean_token_accuracy": 0.22908719182014464,
      "num_tokens": 113508685.0,
      "step": 54695
    },
    {
      "entropy": 5.677050256729126,
      "epoch": 4.961901306240929,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028092701042043794,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22664470821619034,
      "num_tokens": 113519704.0,
      "step": 54700
    },
    {
      "entropy": 5.7334198474884035,
      "epoch": 4.962354862119013,
      "grad_norm": 1.3125,
      "learning_rate": 0.000280894665167711,
      "loss": 4.9767,
      "mean_token_accuracy": 0.2181650161743164,
      "num_tokens": 113529770.0,
      "step": 54705
    },
    {
      "entropy": 5.652707290649414,
      "epoch": 4.962808417997097,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002808623197930809,
      "loss": 4.7405,
      "mean_token_accuracy": 0.2406319111585617,
      "num_tokens": 113538667.0,
      "step": 54710
    },
    {
      "entropy": 5.745785140991211,
      "epoch": 4.9632619738751815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002808299742972165,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22426744401454926,
      "num_tokens": 113549420.0,
      "step": 54715
    },
    {
      "entropy": 5.721398115158081,
      "epoch": 4.963715529753266,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028079762868078685,
      "loss": 4.877,
      "mean_token_accuracy": 0.22337627410888672,
      "num_tokens": 113559325.0,
      "step": 54720
    },
    {
      "entropy": 5.706426191329956,
      "epoch": 4.96416908563135,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002807652829444607,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2314179480075836,
      "num_tokens": 113570152.0,
      "step": 54725
    },
    {
      "entropy": 5.648687505722046,
      "epoch": 4.964622641509434,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028073293708890697,
      "loss": 4.7974,
      "mean_token_accuracy": 0.234486323595047,
      "num_tokens": 113580302.0,
      "step": 54730
    },
    {
      "entropy": 5.742233324050903,
      "epoch": 4.965076197387518,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028070059111479475,
      "loss": 4.9181,
      "mean_token_accuracy": 0.2213611587882042,
      "num_tokens": 113590731.0,
      "step": 54735
    },
    {
      "entropy": 5.735830068588257,
      "epoch": 4.965529753265602,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002806682450227928,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22391819655895234,
      "num_tokens": 113600850.0,
      "step": 54740
    },
    {
      "entropy": 5.713379907608032,
      "epoch": 4.965983309143686,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028063589881357007,
      "loss": 4.891,
      "mean_token_accuracy": 0.22778348475694657,
      "num_tokens": 113611559.0,
      "step": 54745
    },
    {
      "entropy": 5.647102785110474,
      "epoch": 4.9664368650217705,
      "grad_norm": 1.375,
      "learning_rate": 0.0002806035524877955,
      "loss": 4.7823,
      "mean_token_accuracy": 0.2340170532464981,
      "num_tokens": 113621172.0,
      "step": 54750
    },
    {
      "entropy": 5.7102171897888185,
      "epoch": 4.966890420899855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002805712060461381,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22603833824396133,
      "num_tokens": 113631573.0,
      "step": 54755
    },
    {
      "entropy": 5.800774002075196,
      "epoch": 4.967343976777939,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028053885948926664,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21874116063117982,
      "num_tokens": 113641990.0,
      "step": 54760
    },
    {
      "entropy": 5.7508808135986325,
      "epoch": 4.967797532656023,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002805065128178503,
      "loss": 4.8931,
      "mean_token_accuracy": 0.22997388541698455,
      "num_tokens": 113651937.0,
      "step": 54765
    },
    {
      "entropy": 5.758263969421387,
      "epoch": 4.968251088534108,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028047416603255783,
      "loss": 4.9413,
      "mean_token_accuracy": 0.2262187123298645,
      "num_tokens": 113661150.0,
      "step": 54770
    },
    {
      "entropy": 5.701646184921264,
      "epoch": 4.968704644412192,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002804418191340582,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22493981719017028,
      "num_tokens": 113671636.0,
      "step": 54775
    },
    {
      "entropy": 5.689225482940674,
      "epoch": 4.969158200290276,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028040947212302033,
      "loss": 4.9324,
      "mean_token_accuracy": 0.2221432149410248,
      "num_tokens": 113682051.0,
      "step": 54780
    },
    {
      "entropy": 5.693001985549927,
      "epoch": 4.96961175616836,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002803771250001132,
      "loss": 4.7874,
      "mean_token_accuracy": 0.23342819809913634,
      "num_tokens": 113691682.0,
      "step": 54785
    },
    {
      "entropy": 5.803101110458374,
      "epoch": 4.970065312046444,
      "grad_norm": 1.375,
      "learning_rate": 0.00028034477776600576,
      "loss": 4.9769,
      "mean_token_accuracy": 0.22133006155490875,
      "num_tokens": 113702976.0,
      "step": 54790
    },
    {
      "entropy": 5.652866983413697,
      "epoch": 4.970518867924528,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028031243042136696,
      "loss": 4.7967,
      "mean_token_accuracy": 0.23591292649507523,
      "num_tokens": 113712871.0,
      "step": 54795
    },
    {
      "entropy": 5.703150749206543,
      "epoch": 4.970972423802612,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002802800829668657,
      "loss": 4.8593,
      "mean_token_accuracy": 0.22659465074539184,
      "num_tokens": 113722425.0,
      "step": 54800
    },
    {
      "entropy": 5.727939558029175,
      "epoch": 4.9714259796806965,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028024773540317105,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22905452698469161,
      "num_tokens": 113731851.0,
      "step": 54805
    },
    {
      "entropy": 5.684409093856812,
      "epoch": 4.971879535558781,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028021538773095184,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22176220268011093,
      "num_tokens": 113742649.0,
      "step": 54810
    },
    {
      "entropy": 5.697668790817261,
      "epoch": 4.972333091436865,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002801830399508771,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2269762024283409,
      "num_tokens": 113752620.0,
      "step": 54815
    },
    {
      "entropy": 5.6240521430969235,
      "epoch": 4.972786647314949,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028015069206361584,
      "loss": 4.8332,
      "mean_token_accuracy": 0.23651812821626664,
      "num_tokens": 113762165.0,
      "step": 54820
    },
    {
      "entropy": 5.739464521408081,
      "epoch": 4.973240203193034,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028011834406983696,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2276887983083725,
      "num_tokens": 113773342.0,
      "step": 54825
    },
    {
      "entropy": 5.702677774429321,
      "epoch": 4.973693759071118,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002800859959702093,
      "loss": 4.8307,
      "mean_token_accuracy": 0.22720603942871093,
      "num_tokens": 113783159.0,
      "step": 54830
    },
    {
      "entropy": 5.678329086303711,
      "epoch": 4.974147314949201,
      "grad_norm": 1.375,
      "learning_rate": 0.000280053647765402,
      "loss": 4.8729,
      "mean_token_accuracy": 0.2275085523724556,
      "num_tokens": 113794089.0,
      "step": 54835
    },
    {
      "entropy": 5.7038603782653805,
      "epoch": 4.974600870827286,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028002129945608404,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22189982682466508,
      "num_tokens": 113804459.0,
      "step": 54840
    },
    {
      "entropy": 5.646104431152343,
      "epoch": 4.97505442670537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002799889510429242,
      "loss": 4.8129,
      "mean_token_accuracy": 0.24125673919916152,
      "num_tokens": 113815032.0,
      "step": 54845
    },
    {
      "entropy": 5.669192123413086,
      "epoch": 4.975507982583454,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002799566025265917,
      "loss": 4.855,
      "mean_token_accuracy": 0.23042384684085845,
      "num_tokens": 113825731.0,
      "step": 54850
    },
    {
      "entropy": 5.741806602478027,
      "epoch": 4.975961538461538,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002799242539077554,
      "loss": 4.8355,
      "mean_token_accuracy": 0.22400842010974883,
      "num_tokens": 113835752.0,
      "step": 54855
    },
    {
      "entropy": 5.738182020187378,
      "epoch": 4.976415094339623,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002798919051870842,
      "loss": 4.8671,
      "mean_token_accuracy": 0.23055519312620162,
      "num_tokens": 113845241.0,
      "step": 54860
    },
    {
      "entropy": 5.6997700214385985,
      "epoch": 4.976868650217707,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002798595563652472,
      "loss": 4.9157,
      "mean_token_accuracy": 0.22412057518959044,
      "num_tokens": 113856066.0,
      "step": 54865
    },
    {
      "entropy": 5.66191725730896,
      "epoch": 4.977322206095791,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002798272074429133,
      "loss": 4.8072,
      "mean_token_accuracy": 0.22775765806436538,
      "num_tokens": 113866532.0,
      "step": 54870
    },
    {
      "entropy": 5.645719909667969,
      "epoch": 4.977775761973875,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002797948584207515,
      "loss": 4.8346,
      "mean_token_accuracy": 0.22618357986211776,
      "num_tokens": 113876197.0,
      "step": 54875
    },
    {
      "entropy": 5.707626628875732,
      "epoch": 4.97822931785196,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002797625092994309,
      "loss": 4.8638,
      "mean_token_accuracy": 0.23036608397960662,
      "num_tokens": 113886606.0,
      "step": 54880
    },
    {
      "entropy": 5.723944044113159,
      "epoch": 4.978682873730044,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027973016007962025,
      "loss": 4.8553,
      "mean_token_accuracy": 0.23215206265449523,
      "num_tokens": 113897010.0,
      "step": 54885
    },
    {
      "entropy": 5.746036195755005,
      "epoch": 4.979136429608127,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002796978107619889,
      "loss": 4.8786,
      "mean_token_accuracy": 0.23613009303808213,
      "num_tokens": 113908762.0,
      "step": 54890
    },
    {
      "entropy": 5.72992696762085,
      "epoch": 4.979589985486212,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002796654613472054,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2215655967593193,
      "num_tokens": 113919164.0,
      "step": 54895
    },
    {
      "entropy": 5.753871202468872,
      "epoch": 4.980043541364296,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002796331118359391,
      "loss": 4.9572,
      "mean_token_accuracy": 0.22109295427799225,
      "num_tokens": 113928975.0,
      "step": 54900
    },
    {
      "entropy": 5.71602611541748,
      "epoch": 4.98049709724238,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002796007622288588,
      "loss": 4.8199,
      "mean_token_accuracy": 0.2368995025753975,
      "num_tokens": 113937884.0,
      "step": 54905
    },
    {
      "entropy": 5.753722381591797,
      "epoch": 4.980950653120464,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002795684125266336,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22873757779598236,
      "num_tokens": 113948750.0,
      "step": 54910
    },
    {
      "entropy": 5.7624541282653805,
      "epoch": 4.981404208998549,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002795360627299324,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2273629441857338,
      "num_tokens": 113960166.0,
      "step": 54915
    },
    {
      "entropy": 5.751711511611939,
      "epoch": 4.981857764876633,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027950371283942443,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2226872131228447,
      "num_tokens": 113971279.0,
      "step": 54920
    },
    {
      "entropy": 5.73383092880249,
      "epoch": 4.982311320754717,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027947136285577846,
      "loss": 4.9253,
      "mean_token_accuracy": 0.2313312515616417,
      "num_tokens": 113980810.0,
      "step": 54925
    },
    {
      "entropy": 5.680556964874268,
      "epoch": 4.982764876632801,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027943901277966343,
      "loss": 4.8086,
      "mean_token_accuracy": 0.23163387775421143,
      "num_tokens": 113990773.0,
      "step": 54930
    },
    {
      "entropy": 5.695731735229492,
      "epoch": 4.983218432510886,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002794066626117486,
      "loss": 4.8357,
      "mean_token_accuracy": 0.2355637788772583,
      "num_tokens": 114001401.0,
      "step": 54935
    },
    {
      "entropy": 5.7366289615631105,
      "epoch": 4.98367198838897,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002793743123527029,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22409427911043167,
      "num_tokens": 114012382.0,
      "step": 54940
    },
    {
      "entropy": 5.725373363494873,
      "epoch": 4.984125544267053,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002793419620031952,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22574714869260787,
      "num_tokens": 114022395.0,
      "step": 54945
    },
    {
      "entropy": 5.709998369216919,
      "epoch": 4.984579100145138,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027930961156389473,
      "loss": 4.8509,
      "mean_token_accuracy": 0.21616513282060623,
      "num_tokens": 114032629.0,
      "step": 54950
    },
    {
      "entropy": 5.721265983581543,
      "epoch": 4.985032656023222,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027927726103547034,
      "loss": 4.864,
      "mean_token_accuracy": 0.22772633731365205,
      "num_tokens": 114042642.0,
      "step": 54955
    },
    {
      "entropy": 5.712122917175293,
      "epoch": 4.985486211901306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002792449104185911,
      "loss": 4.909,
      "mean_token_accuracy": 0.21963929235935212,
      "num_tokens": 114052972.0,
      "step": 54960
    },
    {
      "entropy": 5.717755079269409,
      "epoch": 4.98593976777939,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027921255971392607,
      "loss": 4.9574,
      "mean_token_accuracy": 0.22279605865478516,
      "num_tokens": 114064422.0,
      "step": 54965
    },
    {
      "entropy": 5.763202953338623,
      "epoch": 4.986393323657475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002791802089221442,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2231297805905342,
      "num_tokens": 114075611.0,
      "step": 54970
    },
    {
      "entropy": 5.764481019973755,
      "epoch": 4.986846879535559,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027914785804391455,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2289043754339218,
      "num_tokens": 114086686.0,
      "step": 54975
    },
    {
      "entropy": 5.722176551818848,
      "epoch": 4.987300435413643,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027911550707990614,
      "loss": 4.8218,
      "mean_token_accuracy": 0.22980393171310426,
      "num_tokens": 114097968.0,
      "step": 54980
    },
    {
      "entropy": 5.648860597610474,
      "epoch": 4.9877539912917275,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027908315603078796,
      "loss": 4.8511,
      "mean_token_accuracy": 0.2354019731283188,
      "num_tokens": 114108323.0,
      "step": 54985
    },
    {
      "entropy": 5.601229095458985,
      "epoch": 4.988207547169811,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027905080489722914,
      "loss": 4.8318,
      "mean_token_accuracy": 0.22978938668966292,
      "num_tokens": 114118129.0,
      "step": 54990
    },
    {
      "entropy": 5.69426646232605,
      "epoch": 4.988661103047895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002790184536798985,
      "loss": 4.9561,
      "mean_token_accuracy": 0.22143328487873076,
      "num_tokens": 114127426.0,
      "step": 54995
    },
    {
      "entropy": 5.682356452941894,
      "epoch": 4.989114658925979,
      "grad_norm": 1.375,
      "learning_rate": 0.00027898610237946537,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23246012032032012,
      "num_tokens": 114137216.0,
      "step": 55000
    },
    {
      "entropy": 5.698159885406494,
      "epoch": 4.989568214804064,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002789537509965986,
      "loss": 4.9069,
      "mean_token_accuracy": 0.224359692633152,
      "num_tokens": 114148617.0,
      "step": 55005
    },
    {
      "entropy": 5.761729383468628,
      "epoch": 4.990021770682148,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002789213995319672,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22824790179729462,
      "num_tokens": 114159512.0,
      "step": 55010
    },
    {
      "entropy": 5.690060043334961,
      "epoch": 4.990475326560232,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002788890479862403,
      "loss": 4.8153,
      "mean_token_accuracy": 0.22937567234039308,
      "num_tokens": 114169249.0,
      "step": 55015
    },
    {
      "entropy": 5.716419410705567,
      "epoch": 4.9909288824383164,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027885669636008685,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21868611574172975,
      "num_tokens": 114179622.0,
      "step": 55020
    },
    {
      "entropy": 5.675246143341065,
      "epoch": 4.991382438316401,
      "grad_norm": 1.546875,
      "learning_rate": 0.000278824344654176,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23450643718242645,
      "num_tokens": 114189712.0,
      "step": 55025
    },
    {
      "entropy": 5.739132881164551,
      "epoch": 4.991835994194485,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027879199286917665,
      "loss": 4.9749,
      "mean_token_accuracy": 0.21294585466384888,
      "num_tokens": 114200074.0,
      "step": 55030
    },
    {
      "entropy": 5.715860986709595,
      "epoch": 4.992289550072569,
      "grad_norm": 1.34375,
      "learning_rate": 0.000278759641005758,
      "loss": 4.9098,
      "mean_token_accuracy": 0.2274160534143448,
      "num_tokens": 114209659.0,
      "step": 55035
    },
    {
      "entropy": 5.703530931472779,
      "epoch": 4.9927431059506535,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000278727289064589,
      "loss": 4.8165,
      "mean_token_accuracy": 0.2316439062356949,
      "num_tokens": 114219765.0,
      "step": 55040
    },
    {
      "entropy": 5.716529464721679,
      "epoch": 4.993196661828737,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002786949370463387,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22556371986865997,
      "num_tokens": 114229654.0,
      "step": 55045
    },
    {
      "entropy": 5.70538821220398,
      "epoch": 4.993650217706821,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002786625849516762,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22296198606491088,
      "num_tokens": 114240504.0,
      "step": 55050
    },
    {
      "entropy": 5.721671867370605,
      "epoch": 4.994103773584905,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002786302327812704,
      "loss": 4.9337,
      "mean_token_accuracy": 0.21554198265075683,
      "num_tokens": 114250748.0,
      "step": 55055
    },
    {
      "entropy": 5.736050939559936,
      "epoch": 4.99455732946299,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027859788053579055,
      "loss": 4.9268,
      "mean_token_accuracy": 0.21946865916252137,
      "num_tokens": 114261655.0,
      "step": 55060
    },
    {
      "entropy": 5.682221460342407,
      "epoch": 4.995010885341074,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027856552821590557,
      "loss": 4.8906,
      "mean_token_accuracy": 0.22851268351078033,
      "num_tokens": 114271193.0,
      "step": 55065
    },
    {
      "entropy": 5.684272909164429,
      "epoch": 4.995464441219158,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002785331758222846,
      "loss": 4.8558,
      "mean_token_accuracy": 0.2274901643395424,
      "num_tokens": 114281461.0,
      "step": 55070
    },
    {
      "entropy": 5.725293779373169,
      "epoch": 4.9959179970972425,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002785008233555967,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2194614052772522,
      "num_tokens": 114290865.0,
      "step": 55075
    },
    {
      "entropy": 5.758294105529785,
      "epoch": 4.996371552975327,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002784684708165107,
      "loss": 4.9166,
      "mean_token_accuracy": 0.224168661236763,
      "num_tokens": 114301811.0,
      "step": 55080
    },
    {
      "entropy": 5.7851954936981205,
      "epoch": 4.996825108853411,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027843611820569595,
      "loss": 4.8958,
      "mean_token_accuracy": 0.2225022092461586,
      "num_tokens": 114313070.0,
      "step": 55085
    },
    {
      "entropy": 5.74291877746582,
      "epoch": 4.997278664731495,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00027840376552382145,
      "loss": 4.9408,
      "mean_token_accuracy": 0.22102911472320558,
      "num_tokens": 114323670.0,
      "step": 55090
    },
    {
      "entropy": 5.617269277572632,
      "epoch": 4.99773222060958,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027837141277155616,
      "loss": 4.8079,
      "mean_token_accuracy": 0.22541468888521193,
      "num_tokens": 114334367.0,
      "step": 55095
    },
    {
      "entropy": 5.650255107879639,
      "epoch": 4.998185776487663,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027833905994956913,
      "loss": 4.7912,
      "mean_token_accuracy": 0.23505220264196397,
      "num_tokens": 114344759.0,
      "step": 55100
    },
    {
      "entropy": 5.700997829437256,
      "epoch": 4.998639332365747,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002783067070585296,
      "loss": 4.848,
      "mean_token_accuracy": 0.2229651167988777,
      "num_tokens": 114355200.0,
      "step": 55105
    },
    {
      "entropy": 5.699042940139771,
      "epoch": 4.9990928882438315,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027827435409910644,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2372337371110916,
      "num_tokens": 114365333.0,
      "step": 55110
    },
    {
      "entropy": 5.697328329086304,
      "epoch": 4.999546444121916,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002782420010719689,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2282085672020912,
      "num_tokens": 114375965.0,
      "step": 55115
    },
    {
      "entropy": 5.741938209533691,
      "epoch": 5.0,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027820964797778587,
      "loss": 4.869,
      "mean_token_accuracy": 0.22616508603096008,
      "num_tokens": 114386175.0,
      "step": 55120
    },
    {
      "entropy": 5.718538379669189,
      "epoch": 5.000453555878084,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027817729481722657,
      "loss": 4.7873,
      "mean_token_accuracy": 0.237392595410347,
      "num_tokens": 114395139.0,
      "step": 55125
    },
    {
      "entropy": 5.687240695953369,
      "epoch": 5.0009071117561685,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027814494159095996,
      "loss": 4.7862,
      "mean_token_accuracy": 0.23585511595010758,
      "num_tokens": 114405378.0,
      "step": 55130
    },
    {
      "entropy": 5.709188747406006,
      "epoch": 5.001360667634253,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027811258829965515,
      "loss": 4.8715,
      "mean_token_accuracy": 0.23233377188444138,
      "num_tokens": 114415920.0,
      "step": 55135
    },
    {
      "entropy": 5.73164005279541,
      "epoch": 5.001814223512337,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002780802349439812,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22385970503091812,
      "num_tokens": 114426177.0,
      "step": 55140
    },
    {
      "entropy": 5.729484796524048,
      "epoch": 5.002267779390421,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027804788152460723,
      "loss": 4.8404,
      "mean_token_accuracy": 0.23033751100301741,
      "num_tokens": 114436221.0,
      "step": 55145
    },
    {
      "entropy": 5.581432914733886,
      "epoch": 5.002721335268505,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002780155280422023,
      "loss": 4.7191,
      "mean_token_accuracy": 0.2428567811846733,
      "num_tokens": 114445828.0,
      "step": 55150
    },
    {
      "entropy": 5.707047748565674,
      "epoch": 5.003174891146589,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027798317449743553,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22437370121479033,
      "num_tokens": 114454792.0,
      "step": 55155
    },
    {
      "entropy": 5.695884561538696,
      "epoch": 5.003628447024673,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002779508208909759,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23184185922145845,
      "num_tokens": 114465357.0,
      "step": 55160
    },
    {
      "entropy": 5.721657752990723,
      "epoch": 5.0040820029027575,
      "grad_norm": 1.25,
      "learning_rate": 0.0002779184672234925,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2352267637848854,
      "num_tokens": 114475947.0,
      "step": 55165
    },
    {
      "entropy": 5.728443384170532,
      "epoch": 5.004535558780842,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002778861134956545,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2308051347732544,
      "num_tokens": 114486335.0,
      "step": 55170
    },
    {
      "entropy": 5.779418420791626,
      "epoch": 5.004989114658926,
      "grad_norm": 1.4375,
      "learning_rate": 0.000277853759708131,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2279623806476593,
      "num_tokens": 114495959.0,
      "step": 55175
    },
    {
      "entropy": 5.761304807662964,
      "epoch": 5.00544267053701,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002778214058615909,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2293373167514801,
      "num_tokens": 114506328.0,
      "step": 55180
    },
    {
      "entropy": 5.68112268447876,
      "epoch": 5.005896226415095,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002777890519567035,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23931930661201478,
      "num_tokens": 114515368.0,
      "step": 55185
    },
    {
      "entropy": 5.657574319839478,
      "epoch": 5.006349782293179,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027775669799413774,
      "loss": 4.8666,
      "mean_token_accuracy": 0.22688290029764174,
      "num_tokens": 114525894.0,
      "step": 55190
    },
    {
      "entropy": 5.698281288146973,
      "epoch": 5.006803338171263,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027772434397456284,
      "loss": 4.9371,
      "mean_token_accuracy": 0.22825067937374116,
      "num_tokens": 114535739.0,
      "step": 55195
    },
    {
      "entropy": 5.652894020080566,
      "epoch": 5.0072568940493465,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027769198989864764,
      "loss": 4.7703,
      "mean_token_accuracy": 0.2353666380047798,
      "num_tokens": 114546519.0,
      "step": 55200
    },
    {
      "entropy": 5.713939476013183,
      "epoch": 5.007710449927431,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002776596357670616,
      "loss": 4.8011,
      "mean_token_accuracy": 0.2356634885072708,
      "num_tokens": 114556523.0,
      "step": 55205
    },
    {
      "entropy": 5.777362489700318,
      "epoch": 5.008164005805515,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027762728158047353,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22805403620004655,
      "num_tokens": 114566255.0,
      "step": 55210
    },
    {
      "entropy": 5.708388805389404,
      "epoch": 5.008617561683599,
      "grad_norm": 1.25,
      "learning_rate": 0.00027759492733955257,
      "loss": 4.8278,
      "mean_token_accuracy": 0.23463455885648726,
      "num_tokens": 114576686.0,
      "step": 55215
    },
    {
      "entropy": 5.676296329498291,
      "epoch": 5.0090711175616836,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002775625730449679,
      "loss": 4.8766,
      "mean_token_accuracy": 0.22399553507566453,
      "num_tokens": 114586606.0,
      "step": 55220
    },
    {
      "entropy": 5.673249053955078,
      "epoch": 5.009524673439768,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027753021869738855,
      "loss": 4.8045,
      "mean_token_accuracy": 0.22540598213672638,
      "num_tokens": 114598450.0,
      "step": 55225
    },
    {
      "entropy": 5.770160102844239,
      "epoch": 5.009978229317852,
      "grad_norm": 1.375,
      "learning_rate": 0.00027749786429748366,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22198348939418794,
      "num_tokens": 114608761.0,
      "step": 55230
    },
    {
      "entropy": 5.745623397827148,
      "epoch": 5.010431785195936,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002774655098459223,
      "loss": 4.8449,
      "mean_token_accuracy": 0.23027018904685975,
      "num_tokens": 114619100.0,
      "step": 55235
    },
    {
      "entropy": 5.779316091537476,
      "epoch": 5.010885341074021,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002774331553433735,
      "loss": 4.9342,
      "mean_token_accuracy": 0.22044748216867446,
      "num_tokens": 114629594.0,
      "step": 55240
    },
    {
      "entropy": 5.662737798690796,
      "epoch": 5.011338896952105,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002774008007905065,
      "loss": 4.7944,
      "mean_token_accuracy": 0.22424897998571397,
      "num_tokens": 114639090.0,
      "step": 55245
    },
    {
      "entropy": 5.6730114936828615,
      "epoch": 5.011792452830188,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027736844618799026,
      "loss": 4.8087,
      "mean_token_accuracy": 0.23465257585048677,
      "num_tokens": 114648707.0,
      "step": 55250
    },
    {
      "entropy": 5.678273868560791,
      "epoch": 5.0122460087082725,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002773360915364941,
      "loss": 4.8694,
      "mean_token_accuracy": 0.2231345698237419,
      "num_tokens": 114659534.0,
      "step": 55255
    },
    {
      "entropy": 5.697882556915284,
      "epoch": 5.012699564586357,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002773037368366868,
      "loss": 4.8175,
      "mean_token_accuracy": 0.22925879508256913,
      "num_tokens": 114670094.0,
      "step": 55260
    },
    {
      "entropy": 5.699722003936768,
      "epoch": 5.013153120464441,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002772713820892377,
      "loss": 4.8771,
      "mean_token_accuracy": 0.23352526426315307,
      "num_tokens": 114680807.0,
      "step": 55265
    },
    {
      "entropy": 5.720627307891846,
      "epoch": 5.013606676342525,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002772390272948159,
      "loss": 4.8398,
      "mean_token_accuracy": 0.23341410458087922,
      "num_tokens": 114691334.0,
      "step": 55270
    },
    {
      "entropy": 5.682211017608642,
      "epoch": 5.01406023222061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027720667245409035,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22281283885240555,
      "num_tokens": 114702519.0,
      "step": 55275
    },
    {
      "entropy": 5.698017549514771,
      "epoch": 5.014513788098694,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027717431756773024,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23064726293087007,
      "num_tokens": 114712893.0,
      "step": 55280
    },
    {
      "entropy": 5.738060331344604,
      "epoch": 5.014967343976778,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002771419626364048,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2161671906709671,
      "num_tokens": 114722101.0,
      "step": 55285
    },
    {
      "entropy": 5.773082637786866,
      "epoch": 5.015420899854862,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027710960766078296,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22851771712303162,
      "num_tokens": 114733184.0,
      "step": 55290
    },
    {
      "entropy": 5.714853143692016,
      "epoch": 5.015874455732947,
      "grad_norm": 1.25,
      "learning_rate": 0.00027707725264153386,
      "loss": 4.7446,
      "mean_token_accuracy": 0.23414382338523865,
      "num_tokens": 114743553.0,
      "step": 55295
    },
    {
      "entropy": 5.727458953857422,
      "epoch": 5.01632801161103,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027704489757932667,
      "loss": 4.8506,
      "mean_token_accuracy": 0.233202762901783,
      "num_tokens": 114754655.0,
      "step": 55300
    },
    {
      "entropy": 5.728694772720337,
      "epoch": 5.016781567489114,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002770125424748305,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21823073029518128,
      "num_tokens": 114764918.0,
      "step": 55305
    },
    {
      "entropy": 5.725438833236694,
      "epoch": 5.017235123367199,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002769801873287145,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22964864373207092,
      "num_tokens": 114775968.0,
      "step": 55310
    },
    {
      "entropy": 5.727709102630615,
      "epoch": 5.017688679245283,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027694783214164757,
      "loss": 4.897,
      "mean_token_accuracy": 0.2264543056488037,
      "num_tokens": 114786363.0,
      "step": 55315
    },
    {
      "entropy": 5.728764581680298,
      "epoch": 5.018142235123367,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002769154769142991,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22628896832466125,
      "num_tokens": 114797034.0,
      "step": 55320
    },
    {
      "entropy": 5.712536478042603,
      "epoch": 5.018595791001451,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002768831216473381,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23059484660625457,
      "num_tokens": 114806493.0,
      "step": 55325
    },
    {
      "entropy": 5.706564712524414,
      "epoch": 5.019049346879536,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027685076634143354,
      "loss": 4.93,
      "mean_token_accuracy": 0.219291290640831,
      "num_tokens": 114816986.0,
      "step": 55330
    },
    {
      "entropy": 5.686685419082641,
      "epoch": 5.01950290275762,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002768184109972547,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2229609966278076,
      "num_tokens": 114827592.0,
      "step": 55335
    },
    {
      "entropy": 5.716079902648926,
      "epoch": 5.019956458635704,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027678605561547073,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22621632367372513,
      "num_tokens": 114839080.0,
      "step": 55340
    },
    {
      "entropy": 5.746042251586914,
      "epoch": 5.020410014513788,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002767537001967506,
      "loss": 4.8411,
      "mean_token_accuracy": 0.23090214729309083,
      "num_tokens": 114848816.0,
      "step": 55345
    },
    {
      "entropy": 5.718955945968628,
      "epoch": 5.020863570391873,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027672134474176347,
      "loss": 4.8639,
      "mean_token_accuracy": 0.23284552693367006,
      "num_tokens": 114858781.0,
      "step": 55350
    },
    {
      "entropy": 5.671054315567017,
      "epoch": 5.021317126269956,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027668898925117856,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2295094147324562,
      "num_tokens": 114869257.0,
      "step": 55355
    },
    {
      "entropy": 5.657876873016358,
      "epoch": 5.02177068214804,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002766566337256649,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2399832308292389,
      "num_tokens": 114879878.0,
      "step": 55360
    },
    {
      "entropy": 5.726356744766235,
      "epoch": 5.022224238026125,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00027662427816589163,
      "loss": 4.9325,
      "mean_token_accuracy": 0.22478135228157042,
      "num_tokens": 114890253.0,
      "step": 55365
    },
    {
      "entropy": 5.746557807922363,
      "epoch": 5.022677793904209,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002765919225725279,
      "loss": 4.8651,
      "mean_token_accuracy": 0.2253119319677353,
      "num_tokens": 114900715.0,
      "step": 55370
    },
    {
      "entropy": 5.709456110000611,
      "epoch": 5.023131349782293,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002765595669462428,
      "loss": 4.8184,
      "mean_token_accuracy": 0.22723291516304017,
      "num_tokens": 114910663.0,
      "step": 55375
    },
    {
      "entropy": 5.739911222457886,
      "epoch": 5.023584905660377,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002765272112877054,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2186341717839241,
      "num_tokens": 114920844.0,
      "step": 55380
    },
    {
      "entropy": 5.7050024509429935,
      "epoch": 5.024038461538462,
      "grad_norm": 1.3125,
      "learning_rate": 0.000276494855597585,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23099371641874314,
      "num_tokens": 114931701.0,
      "step": 55385
    },
    {
      "entropy": 5.719283246994019,
      "epoch": 5.024492017416546,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027646249987655047,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2320967584848404,
      "num_tokens": 114941764.0,
      "step": 55390
    },
    {
      "entropy": 5.706129646301269,
      "epoch": 5.02494557329463,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002764301441252712,
      "loss": 4.8079,
      "mean_token_accuracy": 0.2319590300321579,
      "num_tokens": 114952126.0,
      "step": 55395
    },
    {
      "entropy": 5.647757911682129,
      "epoch": 5.0253991291727145,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027639778834441606,
      "loss": 4.8198,
      "mean_token_accuracy": 0.23675525486469268,
      "num_tokens": 114961952.0,
      "step": 55400
    },
    {
      "entropy": 5.741489505767822,
      "epoch": 5.025852685050798,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002763654325346544,
      "loss": 4.9062,
      "mean_token_accuracy": 0.22678641378879547,
      "num_tokens": 114971338.0,
      "step": 55405
    },
    {
      "entropy": 5.719494390487671,
      "epoch": 5.026306240928882,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027633307669665524,
      "loss": 4.908,
      "mean_token_accuracy": 0.2285715952515602,
      "num_tokens": 114981792.0,
      "step": 55410
    },
    {
      "entropy": 5.674702167510986,
      "epoch": 5.026759796806966,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002763007208310876,
      "loss": 4.7628,
      "mean_token_accuracy": 0.23709815591573716,
      "num_tokens": 114992619.0,
      "step": 55415
    },
    {
      "entropy": 5.669103813171387,
      "epoch": 5.027213352685051,
      "grad_norm": 1.5,
      "learning_rate": 0.00027626836493862086,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23323239535093307,
      "num_tokens": 115003551.0,
      "step": 55420
    },
    {
      "entropy": 5.802165460586548,
      "epoch": 5.027666908563135,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027623600901992393,
      "loss": 4.911,
      "mean_token_accuracy": 0.22009456902742386,
      "num_tokens": 115013155.0,
      "step": 55425
    },
    {
      "entropy": 5.745601654052734,
      "epoch": 5.028120464441219,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027620365307566607,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22033227682113649,
      "num_tokens": 115023673.0,
      "step": 55430
    },
    {
      "entropy": 5.726300001144409,
      "epoch": 5.0285740203193035,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002761712971065163,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22854770123958587,
      "num_tokens": 115033597.0,
      "step": 55435
    },
    {
      "entropy": 5.666406869888306,
      "epoch": 5.029027576197388,
      "grad_norm": 1.5,
      "learning_rate": 0.00027613894111314384,
      "loss": 4.7933,
      "mean_token_accuracy": 0.23325251042842865,
      "num_tokens": 115043732.0,
      "step": 55440
    },
    {
      "entropy": 5.705982351303101,
      "epoch": 5.029481132075472,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027610658509621776,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23002560883760453,
      "num_tokens": 115054124.0,
      "step": 55445
    },
    {
      "entropy": 5.672350263595581,
      "epoch": 5.029934687953556,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002760742290564072,
      "loss": 4.7384,
      "mean_token_accuracy": 0.2473590388894081,
      "num_tokens": 115064513.0,
      "step": 55450
    },
    {
      "entropy": 5.721662902832032,
      "epoch": 5.03038824383164,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027604187299438146,
      "loss": 4.9323,
      "mean_token_accuracy": 0.23112096935510634,
      "num_tokens": 115074060.0,
      "step": 55455
    },
    {
      "entropy": 5.7489893436431885,
      "epoch": 5.030841799709724,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027600951691080937,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2176757723093033,
      "num_tokens": 115084615.0,
      "step": 55460
    },
    {
      "entropy": 5.703184795379639,
      "epoch": 5.031295355587808,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027597716080636026,
      "loss": 4.7616,
      "mean_token_accuracy": 0.2416100397706032,
      "num_tokens": 115095899.0,
      "step": 55465
    },
    {
      "entropy": 5.630411100387573,
      "epoch": 5.031748911465892,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002759448046817033,
      "loss": 4.7974,
      "mean_token_accuracy": 0.22903054654598237,
      "num_tokens": 115107121.0,
      "step": 55470
    },
    {
      "entropy": 5.7176109790802006,
      "epoch": 5.032202467343977,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002759124485375074,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22800477743148803,
      "num_tokens": 115117842.0,
      "step": 55475
    },
    {
      "entropy": 5.677669048309326,
      "epoch": 5.032656023222061,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002758800923744419,
      "loss": 4.7741,
      "mean_token_accuracy": 0.24051392525434495,
      "num_tokens": 115127268.0,
      "step": 55480
    },
    {
      "entropy": 5.69459981918335,
      "epoch": 5.033109579100145,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027584773619317583,
      "loss": 4.7837,
      "mean_token_accuracy": 0.23284126967191696,
      "num_tokens": 115137102.0,
      "step": 55485
    },
    {
      "entropy": 5.7319965839385985,
      "epoch": 5.0335631349782295,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002758153799943784,
      "loss": 4.9,
      "mean_token_accuracy": 0.2288999155163765,
      "num_tokens": 115147822.0,
      "step": 55490
    },
    {
      "entropy": 5.862299871444702,
      "epoch": 5.034016690856314,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027578302377871867,
      "loss": 5.058,
      "mean_token_accuracy": 0.21265411972999573,
      "num_tokens": 115158163.0,
      "step": 55495
    },
    {
      "entropy": 5.654470205307007,
      "epoch": 5.034470246734398,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027575066754686583,
      "loss": 4.846,
      "mean_token_accuracy": 0.22060960978269578,
      "num_tokens": 115169224.0,
      "step": 55500
    },
    {
      "entropy": 5.655987358093261,
      "epoch": 5.034923802612482,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027571831129948895,
      "loss": 4.7684,
      "mean_token_accuracy": 0.23679495900869368,
      "num_tokens": 115179357.0,
      "step": 55505
    },
    {
      "entropy": 5.733451986312867,
      "epoch": 5.035377358490566,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002756859550372573,
      "loss": 4.8771,
      "mean_token_accuracy": 0.2269522562623024,
      "num_tokens": 115190245.0,
      "step": 55510
    },
    {
      "entropy": 5.7673040390014645,
      "epoch": 5.03583091436865,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002756535987608398,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22739382237195968,
      "num_tokens": 115201106.0,
      "step": 55515
    },
    {
      "entropy": 5.739930009841919,
      "epoch": 5.036284470246734,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002756212424709058,
      "loss": 4.8511,
      "mean_token_accuracy": 0.2325811952352524,
      "num_tokens": 115211822.0,
      "step": 55520
    },
    {
      "entropy": 5.723788690567017,
      "epoch": 5.0367380261248185,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027558888616812435,
      "loss": 4.827,
      "mean_token_accuracy": 0.23230288326740264,
      "num_tokens": 115222662.0,
      "step": 55525
    },
    {
      "entropy": 5.734750843048095,
      "epoch": 5.037191582002903,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002755565298531645,
      "loss": 4.872,
      "mean_token_accuracy": 0.22850003093481064,
      "num_tokens": 115232534.0,
      "step": 55530
    },
    {
      "entropy": 5.748213195800782,
      "epoch": 5.037645137880987,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002755241735266956,
      "loss": 4.9216,
      "mean_token_accuracy": 0.2213672786951065,
      "num_tokens": 115243669.0,
      "step": 55535
    },
    {
      "entropy": 5.718574905395508,
      "epoch": 5.038098693759071,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002754918171893865,
      "loss": 4.808,
      "mean_token_accuracy": 0.24193204045295716,
      "num_tokens": 115253482.0,
      "step": 55540
    },
    {
      "entropy": 5.702405881881714,
      "epoch": 5.0385522496371555,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00027545946084190666,
      "loss": 4.9041,
      "mean_token_accuracy": 0.22435929477214814,
      "num_tokens": 115263037.0,
      "step": 55545
    },
    {
      "entropy": 5.760932445526123,
      "epoch": 5.03900580551524,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002754271044849249,
      "loss": 4.9147,
      "mean_token_accuracy": 0.2248095914721489,
      "num_tokens": 115273135.0,
      "step": 55550
    },
    {
      "entropy": 5.766715669631958,
      "epoch": 5.039459361393324,
      "grad_norm": 1.375,
      "learning_rate": 0.00027539474811911065,
      "loss": 4.88,
      "mean_token_accuracy": 0.22797966748476028,
      "num_tokens": 115283482.0,
      "step": 55555
    },
    {
      "entropy": 5.6656410694122314,
      "epoch": 5.0399129172714074,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002753623917451329,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23395129293203354,
      "num_tokens": 115293855.0,
      "step": 55560
    },
    {
      "entropy": 5.705261278152466,
      "epoch": 5.040366473149492,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002753300353636607,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22932448089122773,
      "num_tokens": 115304162.0,
      "step": 55565
    },
    {
      "entropy": 5.758790349960327,
      "epoch": 5.040820029027576,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027529767897536327,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2250407576560974,
      "num_tokens": 115315108.0,
      "step": 55570
    },
    {
      "entropy": 5.685592746734619,
      "epoch": 5.04127358490566,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027526532258090986,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2210848107933998,
      "num_tokens": 115325434.0,
      "step": 55575
    },
    {
      "entropy": 5.7098089218139645,
      "epoch": 5.0417271407837445,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027523296618096947,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22102674394845961,
      "num_tokens": 115335410.0,
      "step": 55580
    },
    {
      "entropy": 5.7147520065307615,
      "epoch": 5.042180696661829,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027520060977621126,
      "loss": 4.7866,
      "mean_token_accuracy": 0.23771567791700363,
      "num_tokens": 115345838.0,
      "step": 55585
    },
    {
      "entropy": 5.665270090103149,
      "epoch": 5.042634252539913,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002751682533673044,
      "loss": 4.7976,
      "mean_token_accuracy": 0.2445225387811661,
      "num_tokens": 115355988.0,
      "step": 55590
    },
    {
      "entropy": 5.6287659168243405,
      "epoch": 5.043087808417997,
      "grad_norm": 1.375,
      "learning_rate": 0.0002751358969549181,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22578785866498946,
      "num_tokens": 115366942.0,
      "step": 55595
    },
    {
      "entropy": 5.696264696121216,
      "epoch": 5.043541364296082,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027510354053972123,
      "loss": 4.7752,
      "mean_token_accuracy": 0.2469717502593994,
      "num_tokens": 115376948.0,
      "step": 55600
    },
    {
      "entropy": 5.728164529800415,
      "epoch": 5.043994920174166,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027507118412238324,
      "loss": 4.8971,
      "mean_token_accuracy": 0.23612573742866516,
      "num_tokens": 115387831.0,
      "step": 55605
    },
    {
      "entropy": 5.760399007797242,
      "epoch": 5.044448476052249,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002750388277035732,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23387691676616668,
      "num_tokens": 115397511.0,
      "step": 55610
    },
    {
      "entropy": 5.662650632858276,
      "epoch": 5.0449020319303335,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002750064712839601,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23215365111827851,
      "num_tokens": 115408049.0,
      "step": 55615
    },
    {
      "entropy": 5.68165693283081,
      "epoch": 5.045355587808418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027497411486421323,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22965006083250045,
      "num_tokens": 115418811.0,
      "step": 55620
    },
    {
      "entropy": 5.711132335662842,
      "epoch": 5.045809143686502,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027494175844500164,
      "loss": 4.7942,
      "mean_token_accuracy": 0.23035336434841155,
      "num_tokens": 115428324.0,
      "step": 55625
    },
    {
      "entropy": 5.683653116226196,
      "epoch": 5.046262699564586,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002749094020269945,
      "loss": 4.8147,
      "mean_token_accuracy": 0.22962941080331803,
      "num_tokens": 115437891.0,
      "step": 55630
    },
    {
      "entropy": 5.6653577327728275,
      "epoch": 5.0467162554426706,
      "grad_norm": 1.28125,
      "learning_rate": 0.000274877045610861,
      "loss": 4.8565,
      "mean_token_accuracy": 0.226553376019001,
      "num_tokens": 115448100.0,
      "step": 55635
    },
    {
      "entropy": 5.691584873199463,
      "epoch": 5.047169811320755,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002748446891972701,
      "loss": 4.829,
      "mean_token_accuracy": 0.2338498502969742,
      "num_tokens": 115458530.0,
      "step": 55640
    },
    {
      "entropy": 5.693638563156128,
      "epoch": 5.047623367198839,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027481233278689124,
      "loss": 4.8374,
      "mean_token_accuracy": 0.22933578491210938,
      "num_tokens": 115469222.0,
      "step": 55645
    },
    {
      "entropy": 5.752430772781372,
      "epoch": 5.048076923076923,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002747799763803932,
      "loss": 4.9083,
      "mean_token_accuracy": 0.22414713650941848,
      "num_tokens": 115478704.0,
      "step": 55650
    },
    {
      "entropy": 5.714435863494873,
      "epoch": 5.048530478955008,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027474761997844536,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22848111093044282,
      "num_tokens": 115488430.0,
      "step": 55655
    },
    {
      "entropy": 5.723462009429932,
      "epoch": 5.048984034833092,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002747152635817169,
      "loss": 4.8832,
      "mean_token_accuracy": 0.22308359295129776,
      "num_tokens": 115499010.0,
      "step": 55660
    },
    {
      "entropy": 5.69669394493103,
      "epoch": 5.049437590711175,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002746829071908768,
      "loss": 4.8502,
      "mean_token_accuracy": 0.2341110050678253,
      "num_tokens": 115508278.0,
      "step": 55665
    },
    {
      "entropy": 5.6871480464935305,
      "epoch": 5.0498911465892595,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002746505508065942,
      "loss": 4.7853,
      "mean_token_accuracy": 0.23497839719057084,
      "num_tokens": 115517977.0,
      "step": 55670
    },
    {
      "entropy": 5.6927917957305905,
      "epoch": 5.050344702467344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027461819442953846,
      "loss": 4.8081,
      "mean_token_accuracy": 0.229037906229496,
      "num_tokens": 115528533.0,
      "step": 55675
    },
    {
      "entropy": 5.705641651153565,
      "epoch": 5.050798258345428,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002745858380603784,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2268344983458519,
      "num_tokens": 115537930.0,
      "step": 55680
    },
    {
      "entropy": 5.757800340652466,
      "epoch": 5.051251814223512,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002745534816997834,
      "loss": 4.9148,
      "mean_token_accuracy": 0.216607204079628,
      "num_tokens": 115547436.0,
      "step": 55685
    },
    {
      "entropy": 5.681735181808472,
      "epoch": 5.051705370101597,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027452112534842246,
      "loss": 4.8106,
      "mean_token_accuracy": 0.2324461042881012,
      "num_tokens": 115559313.0,
      "step": 55690
    },
    {
      "entropy": 5.7597580909729,
      "epoch": 5.052158925979681,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002744887690069648,
      "loss": 4.8684,
      "mean_token_accuracy": 0.2309875100851059,
      "num_tokens": 115568773.0,
      "step": 55695
    },
    {
      "entropy": 5.713888788223267,
      "epoch": 5.052612481857765,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027445641267607947,
      "loss": 4.8868,
      "mean_token_accuracy": 0.22657170295715331,
      "num_tokens": 115579402.0,
      "step": 55700
    },
    {
      "entropy": 5.6802431583404545,
      "epoch": 5.053066037735849,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027442405635643574,
      "loss": 4.8841,
      "mean_token_accuracy": 0.23333456367254257,
      "num_tokens": 115588747.0,
      "step": 55705
    },
    {
      "entropy": 5.677075481414795,
      "epoch": 5.053519593613934,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002743917000487027,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22594254463911057,
      "num_tokens": 115599927.0,
      "step": 55710
    },
    {
      "entropy": 5.759659671783448,
      "epoch": 5.053973149492017,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027435934375354934,
      "loss": 4.9465,
      "mean_token_accuracy": 0.22579672783613206,
      "num_tokens": 115610120.0,
      "step": 55715
    },
    {
      "entropy": 5.717999696731567,
      "epoch": 5.054426705370101,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027432698747164496,
      "loss": 4.874,
      "mean_token_accuracy": 0.22308541238307952,
      "num_tokens": 115622039.0,
      "step": 55720
    },
    {
      "entropy": 5.677969884872437,
      "epoch": 5.054880261248186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027429463120365874,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2262391820549965,
      "num_tokens": 115633527.0,
      "step": 55725
    },
    {
      "entropy": 5.67539348602295,
      "epoch": 5.05533381712627,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002742622749502596,
      "loss": 4.852,
      "mean_token_accuracy": 0.2277738019824028,
      "num_tokens": 115644363.0,
      "step": 55730
    },
    {
      "entropy": 5.784001111984253,
      "epoch": 5.055787373004354,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002742299187121168,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21369970440864564,
      "num_tokens": 115654313.0,
      "step": 55735
    },
    {
      "entropy": 5.737823104858398,
      "epoch": 5.056240928882438,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027419756248989955,
      "loss": 4.7837,
      "mean_token_accuracy": 0.23601865023374557,
      "num_tokens": 115664724.0,
      "step": 55740
    },
    {
      "entropy": 5.697031116485595,
      "epoch": 5.056694484760523,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002741652062842769,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2286548525094986,
      "num_tokens": 115674037.0,
      "step": 55745
    },
    {
      "entropy": 5.713282346725464,
      "epoch": 5.057148040638607,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000274132850095918,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23129377365112305,
      "num_tokens": 115684431.0,
      "step": 55750
    },
    {
      "entropy": 5.685303974151611,
      "epoch": 5.057601596516691,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027410049392549197,
      "loss": 4.8404,
      "mean_token_accuracy": 0.22676413059234618,
      "num_tokens": 115695293.0,
      "step": 55755
    },
    {
      "entropy": 5.684050416946411,
      "epoch": 5.058055152394775,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027406813777366794,
      "loss": 4.827,
      "mean_token_accuracy": 0.23389339745044707,
      "num_tokens": 115705906.0,
      "step": 55760
    },
    {
      "entropy": 5.679602909088135,
      "epoch": 5.058508708272859,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027403578164111517,
      "loss": 4.7895,
      "mean_token_accuracy": 0.24183788299560546,
      "num_tokens": 115716157.0,
      "step": 55765
    },
    {
      "entropy": 5.785975408554077,
      "epoch": 5.058962264150943,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002740034255285025,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22712966203689575,
      "num_tokens": 115727691.0,
      "step": 55770
    },
    {
      "entropy": 5.698011684417724,
      "epoch": 5.059415820029027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002739710694364994,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22534691095352172,
      "num_tokens": 115738984.0,
      "step": 55775
    },
    {
      "entropy": 5.680123615264892,
      "epoch": 5.059869375907112,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027393871336577476,
      "loss": 4.8158,
      "mean_token_accuracy": 0.2332799553871155,
      "num_tokens": 115749040.0,
      "step": 55780
    },
    {
      "entropy": 5.720378732681274,
      "epoch": 5.060322931785196,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027390635731699774,
      "loss": 4.946,
      "mean_token_accuracy": 0.21684779673814775,
      "num_tokens": 115760125.0,
      "step": 55785
    },
    {
      "entropy": 5.714204502105713,
      "epoch": 5.06077648766328,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002738740012908376,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2262340307235718,
      "num_tokens": 115770780.0,
      "step": 55790
    },
    {
      "entropy": 5.692262601852417,
      "epoch": 5.061230043541364,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027384164528796345,
      "loss": 4.8253,
      "mean_token_accuracy": 0.23305843472480775,
      "num_tokens": 115782051.0,
      "step": 55795
    },
    {
      "entropy": 5.613987398147583,
      "epoch": 5.061683599419449,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002738092893090443,
      "loss": 4.7799,
      "mean_token_accuracy": 0.23495042771101,
      "num_tokens": 115791692.0,
      "step": 55800
    },
    {
      "entropy": 5.687026739120483,
      "epoch": 5.062137155297533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002737769333547494,
      "loss": 4.8103,
      "mean_token_accuracy": 0.24219475090503692,
      "num_tokens": 115802738.0,
      "step": 55805
    },
    {
      "entropy": 5.725786447525024,
      "epoch": 5.062590711175617,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002737445774257478,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22314288169145585,
      "num_tokens": 115813073.0,
      "step": 55810
    },
    {
      "entropy": 5.760724973678589,
      "epoch": 5.063044267053701,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002737122215227087,
      "loss": 4.9012,
      "mean_token_accuracy": 0.22280044704675675,
      "num_tokens": 115823976.0,
      "step": 55815
    },
    {
      "entropy": 5.690623807907104,
      "epoch": 5.063497822931785,
      "grad_norm": 1.375,
      "learning_rate": 0.0002736798656463011,
      "loss": 4.822,
      "mean_token_accuracy": 0.2285018026828766,
      "num_tokens": 115833332.0,
      "step": 55820
    },
    {
      "entropy": 5.670272302627564,
      "epoch": 5.063951378809869,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027364750979719433,
      "loss": 4.8239,
      "mean_token_accuracy": 0.2293476238846779,
      "num_tokens": 115844053.0,
      "step": 55825
    },
    {
      "entropy": 5.682009840011597,
      "epoch": 5.064404934687953,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002736151539760573,
      "loss": 4.8059,
      "mean_token_accuracy": 0.24087414741516114,
      "num_tokens": 115854860.0,
      "step": 55830
    },
    {
      "entropy": 5.71255669593811,
      "epoch": 5.064858490566038,
      "grad_norm": 1.578125,
      "learning_rate": 0.00027358279818355926,
      "loss": 4.7461,
      "mean_token_accuracy": 0.23685603737831115,
      "num_tokens": 115865523.0,
      "step": 55835
    },
    {
      "entropy": 5.6893268585205075,
      "epoch": 5.065312046444122,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027355044242036934,
      "loss": 4.8594,
      "mean_token_accuracy": 0.23167635947465898,
      "num_tokens": 115876033.0,
      "step": 55840
    },
    {
      "entropy": 5.710326766967773,
      "epoch": 5.065765602322206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002735180866871567,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2341604694724083,
      "num_tokens": 115886575.0,
      "step": 55845
    },
    {
      "entropy": 5.699028062820434,
      "epoch": 5.0662191582002905,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002734857309845903,
      "loss": 4.8825,
      "mean_token_accuracy": 0.22394295781850815,
      "num_tokens": 115897628.0,
      "step": 55850
    },
    {
      "entropy": 5.678902053833008,
      "epoch": 5.066672714078375,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002734533753133394,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23187558352947235,
      "num_tokens": 115908472.0,
      "step": 55855
    },
    {
      "entropy": 5.688889503479004,
      "epoch": 5.067126269956459,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027342101967407306,
      "loss": 4.7671,
      "mean_token_accuracy": 0.23627837002277374,
      "num_tokens": 115918798.0,
      "step": 55860
    },
    {
      "entropy": 5.666551494598389,
      "epoch": 5.067579825834543,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002733886640674605,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23636932224035262,
      "num_tokens": 115928825.0,
      "step": 55865
    },
    {
      "entropy": 5.66332745552063,
      "epoch": 5.068033381712627,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002733563084941707,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23408484756946563,
      "num_tokens": 115939315.0,
      "step": 55870
    },
    {
      "entropy": 5.652817630767823,
      "epoch": 5.068486937590711,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027332395295487296,
      "loss": 4.8651,
      "mean_token_accuracy": 0.2320214331150055,
      "num_tokens": 115949870.0,
      "step": 55875
    },
    {
      "entropy": 5.781927013397217,
      "epoch": 5.068940493468795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027329159745023625,
      "loss": 5.0003,
      "mean_token_accuracy": 0.21632659882307054,
      "num_tokens": 115961108.0,
      "step": 55880
    },
    {
      "entropy": 5.691059255599976,
      "epoch": 5.069394049346879,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027325924198092974,
      "loss": 4.8341,
      "mean_token_accuracy": 0.22736490070819854,
      "num_tokens": 115972032.0,
      "step": 55885
    },
    {
      "entropy": 5.691657400131225,
      "epoch": 5.069847605224964,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002732268865476226,
      "loss": 4.8392,
      "mean_token_accuracy": 0.23208522498607637,
      "num_tokens": 115982819.0,
      "step": 55890
    },
    {
      "entropy": 5.692436170578003,
      "epoch": 5.070301161103048,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002731945311509838,
      "loss": 4.873,
      "mean_token_accuracy": 0.22811790108680724,
      "num_tokens": 115992856.0,
      "step": 55895
    },
    {
      "entropy": 5.728944778442383,
      "epoch": 5.070754716981132,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002731621757916826,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22873034030199052,
      "num_tokens": 116004122.0,
      "step": 55900
    },
    {
      "entropy": 5.691617584228515,
      "epoch": 5.0712082728592165,
      "grad_norm": 1.453125,
      "learning_rate": 0.000273129820470388,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2370896354317665,
      "num_tokens": 116014238.0,
      "step": 55905
    },
    {
      "entropy": 5.641591119766235,
      "epoch": 5.071661828737301,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027309746518776936,
      "loss": 4.703,
      "mean_token_accuracy": 0.2380621314048767,
      "num_tokens": 116024310.0,
      "step": 55910
    },
    {
      "entropy": 5.718996524810791,
      "epoch": 5.072115384615385,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002730651099444955,
      "loss": 4.8427,
      "mean_token_accuracy": 0.23280981183052063,
      "num_tokens": 116034648.0,
      "step": 55915
    },
    {
      "entropy": 5.694766855239868,
      "epoch": 5.072568940493468,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002730327547412357,
      "loss": 4.8559,
      "mean_token_accuracy": 0.23043699860572814,
      "num_tokens": 116045178.0,
      "step": 55920
    },
    {
      "entropy": 5.691390323638916,
      "epoch": 5.073022496371553,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027300039957865903,
      "loss": 4.8814,
      "mean_token_accuracy": 0.2313838541507721,
      "num_tokens": 116056429.0,
      "step": 55925
    },
    {
      "entropy": 5.649953079223633,
      "epoch": 5.073476052249637,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002729680444574346,
      "loss": 4.8233,
      "mean_token_accuracy": 0.23370919227600098,
      "num_tokens": 116066151.0,
      "step": 55930
    },
    {
      "entropy": 5.72604193687439,
      "epoch": 5.073929608127721,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027293568937823156,
      "loss": 4.846,
      "mean_token_accuracy": 0.22674172818660737,
      "num_tokens": 116075330.0,
      "step": 55935
    },
    {
      "entropy": 5.719258451461792,
      "epoch": 5.0743831640058055,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002729033343417189,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22484124451875687,
      "num_tokens": 116086108.0,
      "step": 55940
    },
    {
      "entropy": 5.701149320602417,
      "epoch": 5.07483671988389,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027287097934856593,
      "loss": 4.8559,
      "mean_token_accuracy": 0.21981702893972396,
      "num_tokens": 116097046.0,
      "step": 55945
    },
    {
      "entropy": 5.676800918579102,
      "epoch": 5.075290275761974,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027283862439944157,
      "loss": 4.82,
      "mean_token_accuracy": 0.23172362595796586,
      "num_tokens": 116108048.0,
      "step": 55950
    },
    {
      "entropy": 5.727983999252319,
      "epoch": 5.075743831640058,
      "grad_norm": 1.515625,
      "learning_rate": 0.000272806269495015,
      "loss": 4.8892,
      "mean_token_accuracy": 0.23617778420448304,
      "num_tokens": 116117417.0,
      "step": 55955
    },
    {
      "entropy": 5.814570569992066,
      "epoch": 5.0761973875181425,
      "grad_norm": 1.515625,
      "learning_rate": 0.00027277391463595543,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2117897778749466,
      "num_tokens": 116128462.0,
      "step": 55960
    },
    {
      "entropy": 5.855157709121704,
      "epoch": 5.076650943396227,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002727415598229318,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21706285029649736,
      "num_tokens": 116139908.0,
      "step": 55965
    },
    {
      "entropy": 5.7705741882324215,
      "epoch": 5.07710449927431,
      "grad_norm": 1.5,
      "learning_rate": 0.0002727092050566133,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22928963005542755,
      "num_tokens": 116149310.0,
      "step": 55970
    },
    {
      "entropy": 5.718259477615357,
      "epoch": 5.0775580551523944,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000272676850337669,
      "loss": 4.8346,
      "mean_token_accuracy": 0.2349010542035103,
      "num_tokens": 116160010.0,
      "step": 55975
    },
    {
      "entropy": 5.697924947738647,
      "epoch": 5.078011611030479,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002726444956667681,
      "loss": 4.8805,
      "mean_token_accuracy": 0.2286814421415329,
      "num_tokens": 116170042.0,
      "step": 55980
    },
    {
      "entropy": 5.692703294754028,
      "epoch": 5.078465166908563,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002726121410445796,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22586879581212999,
      "num_tokens": 116180842.0,
      "step": 55985
    },
    {
      "entropy": 5.75346188545227,
      "epoch": 5.078918722786647,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002725797864717726,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22639603465795516,
      "num_tokens": 116191496.0,
      "step": 55990
    },
    {
      "entropy": 5.688834381103516,
      "epoch": 5.0793722786647315,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002725474319490162,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23388054817914963,
      "num_tokens": 116201451.0,
      "step": 55995
    },
    {
      "entropy": 5.69865779876709,
      "epoch": 5.079825834542816,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002725150774769796,
      "loss": 4.7823,
      "mean_token_accuracy": 0.22746256589889527,
      "num_tokens": 116211327.0,
      "step": 56000
    },
    {
      "entropy": 5.750326156616211,
      "epoch": 5.0802793904209,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002724827230563319,
      "loss": 4.8859,
      "mean_token_accuracy": 0.23544124364852906,
      "num_tokens": 116221676.0,
      "step": 56005
    },
    {
      "entropy": 5.715517854690551,
      "epoch": 5.080732946298984,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027245036868774205,
      "loss": 4.8526,
      "mean_token_accuracy": 0.22663965672254563,
      "num_tokens": 116232778.0,
      "step": 56010
    },
    {
      "entropy": 5.68361644744873,
      "epoch": 5.081186502177069,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002724180143718792,
      "loss": 4.7609,
      "mean_token_accuracy": 0.23288994878530503,
      "num_tokens": 116242198.0,
      "step": 56015
    },
    {
      "entropy": 5.640819931030274,
      "epoch": 5.081640058055152,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027238566010941243,
      "loss": 4.8065,
      "mean_token_accuracy": 0.2345404803752899,
      "num_tokens": 116252171.0,
      "step": 56020
    },
    {
      "entropy": 5.631471729278564,
      "epoch": 5.082093613933236,
      "grad_norm": 1.34375,
      "learning_rate": 0.000272353305901011,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2385095849633217,
      "num_tokens": 116263014.0,
      "step": 56025
    },
    {
      "entropy": 5.6867390155792235,
      "epoch": 5.0825471698113205,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002723209517473438,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2245911628007889,
      "num_tokens": 116273126.0,
      "step": 56030
    },
    {
      "entropy": 5.72823657989502,
      "epoch": 5.083000725689405,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027228859764908,
      "loss": 4.8468,
      "mean_token_accuracy": 0.23192966282367705,
      "num_tokens": 116283174.0,
      "step": 56035
    },
    {
      "entropy": 5.697193717956543,
      "epoch": 5.083454281567489,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00027225624360688864,
      "loss": 4.8828,
      "mean_token_accuracy": 0.23186166882514953,
      "num_tokens": 116293015.0,
      "step": 56040
    },
    {
      "entropy": 5.707467937469483,
      "epoch": 5.083907837445573,
      "grad_norm": 1.296875,
      "learning_rate": 0.000272223889621439,
      "loss": 4.917,
      "mean_token_accuracy": 0.22470275163650513,
      "num_tokens": 116304578.0,
      "step": 56045
    },
    {
      "entropy": 5.723817968368531,
      "epoch": 5.0843613933236576,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002721915356933999,
      "loss": 4.861,
      "mean_token_accuracy": 0.2288716271519661,
      "num_tokens": 116314666.0,
      "step": 56050
    },
    {
      "entropy": 5.729741334915161,
      "epoch": 5.084814949201742,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027215918182344064,
      "loss": 4.8198,
      "mean_token_accuracy": 0.23083318620920182,
      "num_tokens": 116324972.0,
      "step": 56055
    },
    {
      "entropy": 5.702785015106201,
      "epoch": 5.085268505079826,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002721268280122302,
      "loss": 4.8318,
      "mean_token_accuracy": 0.22968674451112747,
      "num_tokens": 116335653.0,
      "step": 56060
    },
    {
      "entropy": 5.786572790145874,
      "epoch": 5.08572206095791,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002720944742604377,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22505755871534347,
      "num_tokens": 116345522.0,
      "step": 56065
    },
    {
      "entropy": 5.665349960327148,
      "epoch": 5.086175616835995,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027206212056873213,
      "loss": 4.8433,
      "mean_token_accuracy": 0.23174792975187303,
      "num_tokens": 116354889.0,
      "step": 56070
    },
    {
      "entropy": 5.627313327789307,
      "epoch": 5.086629172714078,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002720297669377828,
      "loss": 4.7754,
      "mean_token_accuracy": 0.24208427518606185,
      "num_tokens": 116364619.0,
      "step": 56075
    },
    {
      "entropy": 5.631125974655151,
      "epoch": 5.087082728592162,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027199741336825857,
      "loss": 4.7864,
      "mean_token_accuracy": 0.2362619996070862,
      "num_tokens": 116374346.0,
      "step": 56080
    },
    {
      "entropy": 5.734328079223633,
      "epoch": 5.0875362844702465,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002719650598608285,
      "loss": 4.8871,
      "mean_token_accuracy": 0.23654995262622833,
      "num_tokens": 116385780.0,
      "step": 56085
    },
    {
      "entropy": 5.784533882141114,
      "epoch": 5.087989840348331,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002719327064161619,
      "loss": 4.8391,
      "mean_token_accuracy": 0.2279491662979126,
      "num_tokens": 116395232.0,
      "step": 56090
    },
    {
      "entropy": 5.692733907699585,
      "epoch": 5.088443396226415,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002719003530349277,
      "loss": 4.8202,
      "mean_token_accuracy": 0.234047532081604,
      "num_tokens": 116404478.0,
      "step": 56095
    },
    {
      "entropy": 5.6740124225616455,
      "epoch": 5.088896952104499,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002718679997177949,
      "loss": 4.7766,
      "mean_token_accuracy": 0.24028661847114563,
      "num_tokens": 116414493.0,
      "step": 56100
    },
    {
      "entropy": 5.748485326766968,
      "epoch": 5.089350507982584,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027183564646543273,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2211135908961296,
      "num_tokens": 116424722.0,
      "step": 56105
    },
    {
      "entropy": 5.624870347976684,
      "epoch": 5.089804063860668,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027180329327851024,
      "loss": 4.7661,
      "mean_token_accuracy": 0.23358000814914703,
      "num_tokens": 116435639.0,
      "step": 56110
    },
    {
      "entropy": 5.613701295852661,
      "epoch": 5.090257619738752,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027177094015769644,
      "loss": 4.7631,
      "mean_token_accuracy": 0.23153149336576462,
      "num_tokens": 116444908.0,
      "step": 56115
    },
    {
      "entropy": 5.727848768234253,
      "epoch": 5.090711175616836,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027173858710366033,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23921380043029786,
      "num_tokens": 116454965.0,
      "step": 56120
    },
    {
      "entropy": 5.678770065307617,
      "epoch": 5.09116473149492,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027170623411707114,
      "loss": 4.7981,
      "mean_token_accuracy": 0.22587355077266694,
      "num_tokens": 116465861.0,
      "step": 56125
    },
    {
      "entropy": 5.696830320358276,
      "epoch": 5.091618287373004,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027167388119859787,
      "loss": 4.9398,
      "mean_token_accuracy": 0.22230857908725737,
      "num_tokens": 116476367.0,
      "step": 56130
    },
    {
      "entropy": 5.686172914505005,
      "epoch": 5.092071843251088,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027164152834890953,
      "loss": 4.8502,
      "mean_token_accuracy": 0.2322307586669922,
      "num_tokens": 116486492.0,
      "step": 56135
    },
    {
      "entropy": 5.695814895629883,
      "epoch": 5.092525399129173,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027160917556867534,
      "loss": 4.7961,
      "mean_token_accuracy": 0.23033812493085862,
      "num_tokens": 116496280.0,
      "step": 56140
    },
    {
      "entropy": 5.795706129074096,
      "epoch": 5.092978955007257,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002715768228585642,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21780968606472015,
      "num_tokens": 116505942.0,
      "step": 56145
    },
    {
      "entropy": 5.7448444843292235,
      "epoch": 5.093432510885341,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002715444702192453,
      "loss": 4.823,
      "mean_token_accuracy": 0.22737322896718978,
      "num_tokens": 116515749.0,
      "step": 56150
    },
    {
      "entropy": 5.679000997543335,
      "epoch": 5.093886066763425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002715121176513875,
      "loss": 4.8242,
      "mean_token_accuracy": 0.23253775686025618,
      "num_tokens": 116525569.0,
      "step": 56155
    },
    {
      "entropy": 5.765509414672851,
      "epoch": 5.09433962264151,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002714797651556601,
      "loss": 4.9301,
      "mean_token_accuracy": 0.230079185962677,
      "num_tokens": 116535837.0,
      "step": 56160
    },
    {
      "entropy": 5.667415714263916,
      "epoch": 5.094793178519594,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027144741273273207,
      "loss": 4.8362,
      "mean_token_accuracy": 0.2259022206068039,
      "num_tokens": 116546299.0,
      "step": 56165
    },
    {
      "entropy": 5.733115100860596,
      "epoch": 5.095246734397678,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027141506038327233,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2331210970878601,
      "num_tokens": 116556725.0,
      "step": 56170
    },
    {
      "entropy": 5.711272525787353,
      "epoch": 5.0957002902757615,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002713827081079502,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2276226282119751,
      "num_tokens": 116566853.0,
      "step": 56175
    },
    {
      "entropy": 5.6981127738952635,
      "epoch": 5.096153846153846,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002713503559074345,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22828152328729628,
      "num_tokens": 116578347.0,
      "step": 56180
    },
    {
      "entropy": 5.723492813110352,
      "epoch": 5.09660740203193,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027131800378239437,
      "loss": 4.9125,
      "mean_token_accuracy": 0.22235750257968903,
      "num_tokens": 116588836.0,
      "step": 56185
    },
    {
      "entropy": 5.675451946258545,
      "epoch": 5.097060957910014,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002712856517334989,
      "loss": 4.762,
      "mean_token_accuracy": 0.23205420821905137,
      "num_tokens": 116599463.0,
      "step": 56190
    },
    {
      "entropy": 5.6995988368988035,
      "epoch": 5.097514513788099,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027125329976141713,
      "loss": 4.8712,
      "mean_token_accuracy": 0.22195915132761002,
      "num_tokens": 116610329.0,
      "step": 56195
    },
    {
      "entropy": 5.730882024765014,
      "epoch": 5.097968069666183,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000271220947866818,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23379059731960297,
      "num_tokens": 116619954.0,
      "step": 56200
    },
    {
      "entropy": 5.729697132110596,
      "epoch": 5.098421625544267,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002711885960503706,
      "loss": 4.8823,
      "mean_token_accuracy": 0.2308156043291092,
      "num_tokens": 116629347.0,
      "step": 56205
    },
    {
      "entropy": 5.724252462387085,
      "epoch": 5.098875181422351,
      "grad_norm": 1.375,
      "learning_rate": 0.0002711562443127441,
      "loss": 4.8049,
      "mean_token_accuracy": 0.23415262252092361,
      "num_tokens": 116639674.0,
      "step": 56210
    },
    {
      "entropy": 5.674929666519165,
      "epoch": 5.099328737300436,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002711238926546074,
      "loss": 4.8131,
      "mean_token_accuracy": 0.22462995648384093,
      "num_tokens": 116650889.0,
      "step": 56215
    },
    {
      "entropy": 5.626111268997192,
      "epoch": 5.09978229317852,
      "grad_norm": 1.375,
      "learning_rate": 0.00027109154107662957,
      "loss": 4.8157,
      "mean_token_accuracy": 0.22627450823783873,
      "num_tokens": 116661213.0,
      "step": 56220
    },
    {
      "entropy": 5.650832843780518,
      "epoch": 5.100235849056604,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002710591895794797,
      "loss": 4.7821,
      "mean_token_accuracy": 0.23107851445674896,
      "num_tokens": 116672064.0,
      "step": 56225
    },
    {
      "entropy": 5.749910020828247,
      "epoch": 5.100689404934688,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002710268381638268,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2181069314479828,
      "num_tokens": 116681616.0,
      "step": 56230
    },
    {
      "entropy": 5.689093971252442,
      "epoch": 5.101142960812772,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002709944868303399,
      "loss": 4.7872,
      "mean_token_accuracy": 0.2307741343975067,
      "num_tokens": 116691427.0,
      "step": 56235
    },
    {
      "entropy": 5.688955593109131,
      "epoch": 5.101596516690856,
      "grad_norm": 1.359375,
      "learning_rate": 0.000270962135579688,
      "loss": 4.8287,
      "mean_token_accuracy": 0.2313108667731285,
      "num_tokens": 116701326.0,
      "step": 56240
    },
    {
      "entropy": 5.701787614822388,
      "epoch": 5.10205007256894,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002709297844125402,
      "loss": 4.8674,
      "mean_token_accuracy": 0.23137038499116896,
      "num_tokens": 116712353.0,
      "step": 56245
    },
    {
      "entropy": 5.709525203704834,
      "epoch": 5.102503628447025,
      "grad_norm": 1.5,
      "learning_rate": 0.0002708974333295655,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2318824827671051,
      "num_tokens": 116721906.0,
      "step": 56250
    },
    {
      "entropy": 5.706891870498657,
      "epoch": 5.102957184325109,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027086508233143287,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22848699986934662,
      "num_tokens": 116732684.0,
      "step": 56255
    },
    {
      "entropy": 5.662374687194824,
      "epoch": 5.103410740203193,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002708327314188114,
      "loss": 4.845,
      "mean_token_accuracy": 0.23235982954502105,
      "num_tokens": 116743071.0,
      "step": 56260
    },
    {
      "entropy": 5.7236809730529785,
      "epoch": 5.1038642960812775,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002708003805923702,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22792622298002244,
      "num_tokens": 116752756.0,
      "step": 56265
    },
    {
      "entropy": 5.698482894897461,
      "epoch": 5.104317851959362,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027076802985277813,
      "loss": 4.8279,
      "mean_token_accuracy": 0.22845792025327682,
      "num_tokens": 116762878.0,
      "step": 56270
    },
    {
      "entropy": 5.73944935798645,
      "epoch": 5.104771407837446,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002707356792007042,
      "loss": 4.8663,
      "mean_token_accuracy": 0.23018293231725692,
      "num_tokens": 116772953.0,
      "step": 56275
    },
    {
      "entropy": 5.795046710968018,
      "epoch": 5.105224963715529,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027070332863681764,
      "loss": 4.8803,
      "mean_token_accuracy": 0.22600976675748824,
      "num_tokens": 116783530.0,
      "step": 56280
    },
    {
      "entropy": 5.710184097290039,
      "epoch": 5.105678519593614,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00027067097816178734,
      "loss": 4.8341,
      "mean_token_accuracy": 0.22741353809833526,
      "num_tokens": 116792740.0,
      "step": 56285
    },
    {
      "entropy": 5.673601675033569,
      "epoch": 5.106132075471698,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002706386277762822,
      "loss": 4.8072,
      "mean_token_accuracy": 0.23143598437309265,
      "num_tokens": 116802846.0,
      "step": 56290
    },
    {
      "entropy": 5.738059568405151,
      "epoch": 5.106585631349782,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002706062774809714,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22531277686357498,
      "num_tokens": 116812787.0,
      "step": 56295
    },
    {
      "entropy": 5.741108655929565,
      "epoch": 5.107039187227866,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000270573927276524,
      "loss": 4.906,
      "mean_token_accuracy": 0.22059136927127837,
      "num_tokens": 116823840.0,
      "step": 56300
    },
    {
      "entropy": 5.688503980636597,
      "epoch": 5.107492743105951,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002705415771636088,
      "loss": 4.7979,
      "mean_token_accuracy": 0.23244229108095169,
      "num_tokens": 116833971.0,
      "step": 56305
    },
    {
      "entropy": 5.65808629989624,
      "epoch": 5.107946298984035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027050922714289493,
      "loss": 4.7697,
      "mean_token_accuracy": 0.2310914844274521,
      "num_tokens": 116844292.0,
      "step": 56310
    },
    {
      "entropy": 5.678363037109375,
      "epoch": 5.108399854862119,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027047687721505143,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22711152136325835,
      "num_tokens": 116854399.0,
      "step": 56315
    },
    {
      "entropy": 5.693557405471802,
      "epoch": 5.1088534107402035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027044452738074726,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23008902370929718,
      "num_tokens": 116865418.0,
      "step": 56320
    },
    {
      "entropy": 5.63359956741333,
      "epoch": 5.109306966618288,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027041217764065136,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23620689809322357,
      "num_tokens": 116875225.0,
      "step": 56325
    },
    {
      "entropy": 5.748524141311646,
      "epoch": 5.109760522496371,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027037982799543286,
      "loss": 4.8811,
      "mean_token_accuracy": 0.225726018846035,
      "num_tokens": 116885062.0,
      "step": 56330
    },
    {
      "entropy": 5.733633232116699,
      "epoch": 5.110214078374455,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002703474784457607,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22905084490776062,
      "num_tokens": 116894705.0,
      "step": 56335
    },
    {
      "entropy": 5.702228355407715,
      "epoch": 5.11066763425254,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027031512899230374,
      "loss": 4.8597,
      "mean_token_accuracy": 0.22119402140378952,
      "num_tokens": 116904441.0,
      "step": 56340
    },
    {
      "entropy": 5.676147079467773,
      "epoch": 5.111121190130624,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002702827796357313,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23149157464504241,
      "num_tokens": 116915008.0,
      "step": 56345
    },
    {
      "entropy": 5.715675258636475,
      "epoch": 5.111574746008708,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002702504303767121,
      "loss": 4.9194,
      "mean_token_accuracy": 0.22455388754606248,
      "num_tokens": 116926035.0,
      "step": 56350
    },
    {
      "entropy": 5.7627115726470945,
      "epoch": 5.1120283018867925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002702180812159152,
      "loss": 4.8931,
      "mean_token_accuracy": 0.23121010661125183,
      "num_tokens": 116936263.0,
      "step": 56355
    },
    {
      "entropy": 5.678738927841186,
      "epoch": 5.112481857764877,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027018573215400955,
      "loss": 4.7118,
      "mean_token_accuracy": 0.23757259845733641,
      "num_tokens": 116947080.0,
      "step": 56360
    },
    {
      "entropy": 5.723044300079346,
      "epoch": 5.112935413642961,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027015338319166424,
      "loss": 4.7828,
      "mean_token_accuracy": 0.2306107237935066,
      "num_tokens": 116958224.0,
      "step": 56365
    },
    {
      "entropy": 5.673953056335449,
      "epoch": 5.113388969521045,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002701210343295482,
      "loss": 4.887,
      "mean_token_accuracy": 0.23122036457061768,
      "num_tokens": 116968385.0,
      "step": 56370
    },
    {
      "entropy": 5.677678823471069,
      "epoch": 5.1138425253991295,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027008868556833044,
      "loss": 4.8704,
      "mean_token_accuracy": 0.22587008625268937,
      "num_tokens": 116979309.0,
      "step": 56375
    },
    {
      "entropy": 5.745979738235474,
      "epoch": 5.114296081277214,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027005633690867985,
      "loss": 4.89,
      "mean_token_accuracy": 0.2306362584233284,
      "num_tokens": 116990739.0,
      "step": 56380
    },
    {
      "entropy": 5.736185598373413,
      "epoch": 5.114749637155297,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002700239883512655,
      "loss": 4.8651,
      "mean_token_accuracy": 0.2287585273385048,
      "num_tokens": 117000820.0,
      "step": 56385
    },
    {
      "entropy": 5.665032625198364,
      "epoch": 5.1152031930333814,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002699916398967563,
      "loss": 4.8,
      "mean_token_accuracy": 0.23648434281349182,
      "num_tokens": 117011319.0,
      "step": 56390
    },
    {
      "entropy": 5.707212162017822,
      "epoch": 5.115656748911466,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002699592915458214,
      "loss": 4.8272,
      "mean_token_accuracy": 0.23061859160661696,
      "num_tokens": 117021480.0,
      "step": 56395
    },
    {
      "entropy": 5.753375720977783,
      "epoch": 5.11611030478955,
      "grad_norm": 1.125,
      "learning_rate": 0.00026992694329912954,
      "loss": 4.8965,
      "mean_token_accuracy": 0.22275896817445756,
      "num_tokens": 117033237.0,
      "step": 56400
    },
    {
      "entropy": 5.7101329326629635,
      "epoch": 5.116563860667634,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026989459515734974,
      "loss": 4.8189,
      "mean_token_accuracy": 0.2324957862496376,
      "num_tokens": 117044237.0,
      "step": 56405
    },
    {
      "entropy": 5.741513061523437,
      "epoch": 5.1170174165457185,
      "grad_norm": 1.375,
      "learning_rate": 0.00026986224712115107,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22783341854810715,
      "num_tokens": 117054783.0,
      "step": 56410
    },
    {
      "entropy": 5.706694650650024,
      "epoch": 5.117470972423803,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002698298991912024,
      "loss": 4.87,
      "mean_token_accuracy": 0.2330836534500122,
      "num_tokens": 117064865.0,
      "step": 56415
    },
    {
      "entropy": 5.715668201446533,
      "epoch": 5.117924528301887,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002697975513681728,
      "loss": 4.8526,
      "mean_token_accuracy": 0.22513079941272734,
      "num_tokens": 117074968.0,
      "step": 56420
    },
    {
      "entropy": 5.680552959442139,
      "epoch": 5.118378084179971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002697652036527311,
      "loss": 4.8337,
      "mean_token_accuracy": 0.233208529651165,
      "num_tokens": 117086574.0,
      "step": 56425
    },
    {
      "entropy": 5.633585500717163,
      "epoch": 5.118831640058056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026973285604554647,
      "loss": 4.6879,
      "mean_token_accuracy": 0.24286247044801712,
      "num_tokens": 117097212.0,
      "step": 56430
    },
    {
      "entropy": 5.6884448528289795,
      "epoch": 5.119285195936139,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026970050854728756,
      "loss": 4.7691,
      "mean_token_accuracy": 0.2376386970281601,
      "num_tokens": 117107544.0,
      "step": 56435
    },
    {
      "entropy": 5.725967168807983,
      "epoch": 5.119738751814223,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026966816115862354,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21855146586894988,
      "num_tokens": 117118300.0,
      "step": 56440
    },
    {
      "entropy": 5.735312795639038,
      "epoch": 5.1201923076923075,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002696358138802233,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2401987999677658,
      "num_tokens": 117128702.0,
      "step": 56445
    },
    {
      "entropy": 5.626700687408447,
      "epoch": 5.120645863570392,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026960346671275583,
      "loss": 4.748,
      "mean_token_accuracy": 0.2407726138830185,
      "num_tokens": 117139485.0,
      "step": 56450
    },
    {
      "entropy": 5.778329801559448,
      "epoch": 5.121099419448476,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026957111965688994,
      "loss": 4.9,
      "mean_token_accuracy": 0.22534102946519852,
      "num_tokens": 117149895.0,
      "step": 56455
    },
    {
      "entropy": 5.727129936218262,
      "epoch": 5.12155297532656,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026953877271329476,
      "loss": 4.928,
      "mean_token_accuracy": 0.2206023469567299,
      "num_tokens": 117160901.0,
      "step": 56460
    },
    {
      "entropy": 5.635831737518311,
      "epoch": 5.1220065312046446,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026950642588263914,
      "loss": 4.7333,
      "mean_token_accuracy": 0.24159741550683975,
      "num_tokens": 117170820.0,
      "step": 56465
    },
    {
      "entropy": 5.677332353591919,
      "epoch": 5.122460087082729,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026947407916559196,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23181381225585937,
      "num_tokens": 117181581.0,
      "step": 56470
    },
    {
      "entropy": 5.672154569625855,
      "epoch": 5.122913642960813,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026944173256282234,
      "loss": 4.7941,
      "mean_token_accuracy": 0.23202822208404542,
      "num_tokens": 117190817.0,
      "step": 56475
    },
    {
      "entropy": 5.724192142486572,
      "epoch": 5.123367198838897,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026940938607499903,
      "loss": 4.8786,
      "mean_token_accuracy": 0.23483936935663224,
      "num_tokens": 117201459.0,
      "step": 56480
    },
    {
      "entropy": 5.644607210159302,
      "epoch": 5.123820754716981,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002693770397027911,
      "loss": 4.7698,
      "mean_token_accuracy": 0.22521475702524185,
      "num_tokens": 117211941.0,
      "step": 56485
    },
    {
      "entropy": 5.7010821342468265,
      "epoch": 5.124274310595065,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026934469344686734,
      "loss": 4.873,
      "mean_token_accuracy": 0.22005630135536194,
      "num_tokens": 117222010.0,
      "step": 56490
    },
    {
      "entropy": 5.696380758285523,
      "epoch": 5.124727866473149,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002693123473078968,
      "loss": 4.8622,
      "mean_token_accuracy": 0.2202014833688736,
      "num_tokens": 117231453.0,
      "step": 56495
    },
    {
      "entropy": 5.724776268005371,
      "epoch": 5.1251814223512335,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026928000128654845,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22877732962369918,
      "num_tokens": 117241348.0,
      "step": 56500
    },
    {
      "entropy": 5.6771176815032955,
      "epoch": 5.125634978229318,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000269247655383491,
      "loss": 4.7874,
      "mean_token_accuracy": 0.2321414604783058,
      "num_tokens": 117251472.0,
      "step": 56505
    },
    {
      "entropy": 5.685848236083984,
      "epoch": 5.126088534107402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026921530959939354,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23266687095165253,
      "num_tokens": 117262007.0,
      "step": 56510
    },
    {
      "entropy": 5.644178438186645,
      "epoch": 5.126542089985486,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000269182963934925,
      "loss": 4.8172,
      "mean_token_accuracy": 0.22645535171031952,
      "num_tokens": 117272573.0,
      "step": 56515
    },
    {
      "entropy": 5.720179176330566,
      "epoch": 5.126995645863571,
      "grad_norm": 1.375,
      "learning_rate": 0.0002691506183907542,
      "loss": 4.9114,
      "mean_token_accuracy": 0.22720131874084473,
      "num_tokens": 117283832.0,
      "step": 56520
    },
    {
      "entropy": 5.738627243041992,
      "epoch": 5.127449201741655,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026911827296755014,
      "loss": 4.899,
      "mean_token_accuracy": 0.23013959228992462,
      "num_tokens": 117295287.0,
      "step": 56525
    },
    {
      "entropy": 5.715682792663574,
      "epoch": 5.127902757619739,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002690859276659817,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23542941659688948,
      "num_tokens": 117305585.0,
      "step": 56530
    },
    {
      "entropy": 5.726154136657715,
      "epoch": 5.1283563134978225,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002690535824867178,
      "loss": 4.8847,
      "mean_token_accuracy": 0.23002250790596007,
      "num_tokens": 117316149.0,
      "step": 56535
    },
    {
      "entropy": 5.705884838104248,
      "epoch": 5.128809869375907,
      "grad_norm": 1.375,
      "learning_rate": 0.0002690212374304273,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23441451340913771,
      "num_tokens": 117325777.0,
      "step": 56540
    },
    {
      "entropy": 5.756582260131836,
      "epoch": 5.129263425253991,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002689888924977791,
      "loss": 4.9385,
      "mean_token_accuracy": 0.22556937634944915,
      "num_tokens": 117336303.0,
      "step": 56545
    },
    {
      "entropy": 5.684505605697632,
      "epoch": 5.129716981132075,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002689565476894422,
      "loss": 4.7982,
      "mean_token_accuracy": 0.23973448872566222,
      "num_tokens": 117347169.0,
      "step": 56550
    },
    {
      "entropy": 5.699363374710083,
      "epoch": 5.13017053701016,
      "grad_norm": 1.5,
      "learning_rate": 0.00026892420300608533,
      "loss": 4.8037,
      "mean_token_accuracy": 0.23977982252836227,
      "num_tokens": 117357461.0,
      "step": 56555
    },
    {
      "entropy": 5.647726774215698,
      "epoch": 5.130624092888244,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026889185844837754,
      "loss": 4.8713,
      "mean_token_accuracy": 0.2276977241039276,
      "num_tokens": 117367915.0,
      "step": 56560
    },
    {
      "entropy": 5.6865458488464355,
      "epoch": 5.131077648766328,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002688595140169877,
      "loss": 4.907,
      "mean_token_accuracy": 0.2246981143951416,
      "num_tokens": 117378352.0,
      "step": 56565
    },
    {
      "entropy": 5.718582534790039,
      "epoch": 5.131531204644412,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002688271697125847,
      "loss": 4.8615,
      "mean_token_accuracy": 0.2224150627851486,
      "num_tokens": 117389074.0,
      "step": 56570
    },
    {
      "entropy": 5.791985702514649,
      "epoch": 5.131984760522497,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002687948255358373,
      "loss": 4.8499,
      "mean_token_accuracy": 0.2307104930281639,
      "num_tokens": 117399643.0,
      "step": 56575
    },
    {
      "entropy": 5.713716506958008,
      "epoch": 5.132438316400581,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026876248148741464,
      "loss": 4.7886,
      "mean_token_accuracy": 0.2316012680530548,
      "num_tokens": 117410015.0,
      "step": 56580
    },
    {
      "entropy": 5.663673448562622,
      "epoch": 5.132891872278664,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026873013756798533,
      "loss": 4.804,
      "mean_token_accuracy": 0.23234913349151612,
      "num_tokens": 117419517.0,
      "step": 56585
    },
    {
      "entropy": 5.63922905921936,
      "epoch": 5.1333454281567485,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026869779377821837,
      "loss": 4.7987,
      "mean_token_accuracy": 0.22960458993911742,
      "num_tokens": 117430450.0,
      "step": 56590
    },
    {
      "entropy": 5.680904340744019,
      "epoch": 5.133798984034833,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002686654501187827,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22652235478162766,
      "num_tokens": 117440403.0,
      "step": 56595
    },
    {
      "entropy": 5.673957777023316,
      "epoch": 5.134252539912917,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002686331065903471,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2202273800969124,
      "num_tokens": 117451014.0,
      "step": 56600
    },
    {
      "entropy": 5.717944097518921,
      "epoch": 5.134706095791001,
      "grad_norm": 1.375,
      "learning_rate": 0.0002686007631935805,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22518347203731537,
      "num_tokens": 117461238.0,
      "step": 56605
    },
    {
      "entropy": 5.742697525024414,
      "epoch": 5.135159651669086,
      "grad_norm": 1.375,
      "learning_rate": 0.0002685684199291518,
      "loss": 4.851,
      "mean_token_accuracy": 0.22594875246286392,
      "num_tokens": 117472298.0,
      "step": 56610
    },
    {
      "entropy": 5.776615571975708,
      "epoch": 5.13561320754717,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026853607679772973,
      "loss": 4.8975,
      "mean_token_accuracy": 0.23054667860269545,
      "num_tokens": 117483010.0,
      "step": 56615
    },
    {
      "entropy": 5.6410377502441404,
      "epoch": 5.136066763425254,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026850373379998325,
      "loss": 4.7554,
      "mean_token_accuracy": 0.23934103101491927,
      "num_tokens": 117494325.0,
      "step": 56620
    },
    {
      "entropy": 5.713347005844116,
      "epoch": 5.136520319303338,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026847139093658123,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2250442996621132,
      "num_tokens": 117505426.0,
      "step": 56625
    },
    {
      "entropy": 5.729852867126465,
      "epoch": 5.136973875181423,
      "grad_norm": 1.5,
      "learning_rate": 0.00026843904820819247,
      "loss": 4.8376,
      "mean_token_accuracy": 0.2256636068224907,
      "num_tokens": 117515920.0,
      "step": 56630
    },
    {
      "entropy": 5.68405704498291,
      "epoch": 5.137427431059507,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026840670561548597,
      "loss": 4.813,
      "mean_token_accuracy": 0.2261823058128357,
      "num_tokens": 117525454.0,
      "step": 56635
    },
    {
      "entropy": 5.67120394706726,
      "epoch": 5.13788098693759,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002683743631591303,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23871116936206818,
      "num_tokens": 117536234.0,
      "step": 56640
    },
    {
      "entropy": 5.704612588882446,
      "epoch": 5.138334542815675,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026834202083979465,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22866308689117432,
      "num_tokens": 117546639.0,
      "step": 56645
    },
    {
      "entropy": 5.690085458755493,
      "epoch": 5.138788098693759,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026830967865814763,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22593240290880204,
      "num_tokens": 117557104.0,
      "step": 56650
    },
    {
      "entropy": 5.783428525924682,
      "epoch": 5.139241654571843,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026827733661485825,
      "loss": 4.9098,
      "mean_token_accuracy": 0.23345725536346434,
      "num_tokens": 117568909.0,
      "step": 56655
    },
    {
      "entropy": 5.700820636749268,
      "epoch": 5.139695210449927,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026824499471059514,
      "loss": 4.8393,
      "mean_token_accuracy": 0.22866545915603637,
      "num_tokens": 117579987.0,
      "step": 56660
    },
    {
      "entropy": 5.623366975784302,
      "epoch": 5.140148766328012,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026821265294602736,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2268975093960762,
      "num_tokens": 117591152.0,
      "step": 56665
    },
    {
      "entropy": 5.6758771419525145,
      "epoch": 5.140602322206096,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002681803113218236,
      "loss": 4.7826,
      "mean_token_accuracy": 0.2389973908662796,
      "num_tokens": 117601219.0,
      "step": 56670
    },
    {
      "entropy": 5.767596197128296,
      "epoch": 5.14105587808418,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002681479698386527,
      "loss": 4.8867,
      "mean_token_accuracy": 0.23281267881393433,
      "num_tokens": 117611670.0,
      "step": 56675
    },
    {
      "entropy": 5.729133749008179,
      "epoch": 5.1415094339622645,
      "grad_norm": 1.5,
      "learning_rate": 0.0002681156284971836,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22892612218856812,
      "num_tokens": 117622016.0,
      "step": 56680
    },
    {
      "entropy": 5.694206380844117,
      "epoch": 5.141962989840349,
      "grad_norm": 1.421875,
      "learning_rate": 0.000268083287298085,
      "loss": 4.8287,
      "mean_token_accuracy": 0.22915857434272766,
      "num_tokens": 117631891.0,
      "step": 56685
    },
    {
      "entropy": 5.625613737106323,
      "epoch": 5.142416545718432,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002680509462420258,
      "loss": 4.7605,
      "mean_token_accuracy": 0.23909955024719237,
      "num_tokens": 117642333.0,
      "step": 56690
    },
    {
      "entropy": 5.6427100658416744,
      "epoch": 5.142870101596516,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026801860532967486,
      "loss": 4.8144,
      "mean_token_accuracy": 0.23759718388319015,
      "num_tokens": 117652761.0,
      "step": 56695
    },
    {
      "entropy": 5.699489974975586,
      "epoch": 5.143323657474601,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002679862645617009,
      "loss": 4.8099,
      "mean_token_accuracy": 0.2402305155992508,
      "num_tokens": 117661660.0,
      "step": 56700
    },
    {
      "entropy": 5.709484338760376,
      "epoch": 5.143777213352685,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002679539239387728,
      "loss": 4.8567,
      "mean_token_accuracy": 0.22012127786874772,
      "num_tokens": 117672083.0,
      "step": 56705
    },
    {
      "entropy": 5.6730035781860355,
      "epoch": 5.144230769230769,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026792158346155924,
      "loss": 4.8057,
      "mean_token_accuracy": 0.23787874728441238,
      "num_tokens": 117682923.0,
      "step": 56710
    },
    {
      "entropy": 5.61964054107666,
      "epoch": 5.144684325108853,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026788924313072925,
      "loss": 4.7864,
      "mean_token_accuracy": 0.2322809413075447,
      "num_tokens": 117694409.0,
      "step": 56715
    },
    {
      "entropy": 5.694318723678589,
      "epoch": 5.145137880986938,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026785690294695155,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23078823685646058,
      "num_tokens": 117704185.0,
      "step": 56720
    },
    {
      "entropy": 5.729047012329102,
      "epoch": 5.145591436865022,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026782456291089477,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22190772593021393,
      "num_tokens": 117715067.0,
      "step": 56725
    },
    {
      "entropy": 5.645669603347779,
      "epoch": 5.146044992743106,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026779222302322796,
      "loss": 4.8129,
      "mean_token_accuracy": 0.2355114459991455,
      "num_tokens": 117724258.0,
      "step": 56730
    },
    {
      "entropy": 5.743412256240845,
      "epoch": 5.1464985486211905,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002677598832846198,
      "loss": 4.9171,
      "mean_token_accuracy": 0.2268160253763199,
      "num_tokens": 117733334.0,
      "step": 56735
    },
    {
      "entropy": 5.70098876953125,
      "epoch": 5.146952104499274,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026772754369573914,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2384485587477684,
      "num_tokens": 117744482.0,
      "step": 56740
    },
    {
      "entropy": 5.743811225891113,
      "epoch": 5.147405660377358,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026769520425725464,
      "loss": 4.9221,
      "mean_token_accuracy": 0.22720155864953995,
      "num_tokens": 117755313.0,
      "step": 56745
    },
    {
      "entropy": 5.661306667327881,
      "epoch": 5.147859216255442,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026766286496983523,
      "loss": 4.8492,
      "mean_token_accuracy": 0.23370615392923355,
      "num_tokens": 117765892.0,
      "step": 56750
    },
    {
      "entropy": 5.6500006198883055,
      "epoch": 5.148312772133527,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026763052583414964,
      "loss": 4.8026,
      "mean_token_accuracy": 0.2296631708741188,
      "num_tokens": 117776132.0,
      "step": 56755
    },
    {
      "entropy": 5.716532754898071,
      "epoch": 5.148766328011611,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002675981868508666,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22846895158290864,
      "num_tokens": 117787194.0,
      "step": 56760
    },
    {
      "entropy": 5.685942554473877,
      "epoch": 5.149219883889695,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026756584802065504,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23133328557014465,
      "num_tokens": 117797991.0,
      "step": 56765
    },
    {
      "entropy": 5.767144632339478,
      "epoch": 5.1496734397677795,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002675335093441836,
      "loss": 4.9522,
      "mean_token_accuracy": 0.22231999337673186,
      "num_tokens": 117808114.0,
      "step": 56770
    },
    {
      "entropy": 5.708118009567261,
      "epoch": 5.150126995645864,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002675011708221211,
      "loss": 4.8638,
      "mean_token_accuracy": 0.23232703953981398,
      "num_tokens": 117818413.0,
      "step": 56775
    },
    {
      "entropy": 5.660328960418701,
      "epoch": 5.150580551523948,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026746883245513627,
      "loss": 4.7799,
      "mean_token_accuracy": 0.22923831939697265,
      "num_tokens": 117828926.0,
      "step": 56780
    },
    {
      "entropy": 5.709826135635376,
      "epoch": 5.151034107402032,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002674364942438979,
      "loss": 4.8196,
      "mean_token_accuracy": 0.22869331985712052,
      "num_tokens": 117841051.0,
      "step": 56785
    },
    {
      "entropy": 5.653927516937256,
      "epoch": 5.1514876632801165,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026740415618907475,
      "loss": 4.8118,
      "mean_token_accuracy": 0.22923027127981185,
      "num_tokens": 117852010.0,
      "step": 56790
    },
    {
      "entropy": 5.736068153381348,
      "epoch": 5.1519412191582,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002673718182913356,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22681669443845748,
      "num_tokens": 117860873.0,
      "step": 56795
    },
    {
      "entropy": 5.693616104125977,
      "epoch": 5.152394775036284,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002673394805513492,
      "loss": 4.881,
      "mean_token_accuracy": 0.22894809246063233,
      "num_tokens": 117871417.0,
      "step": 56800
    },
    {
      "entropy": 5.710697841644287,
      "epoch": 5.1528483309143684,
      "grad_norm": 1.25,
      "learning_rate": 0.00026730714296978433,
      "loss": 4.8159,
      "mean_token_accuracy": 0.22861335277557374,
      "num_tokens": 117881115.0,
      "step": 56805
    },
    {
      "entropy": 5.738108444213867,
      "epoch": 5.153301886792453,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002672748055473096,
      "loss": 4.8258,
      "mean_token_accuracy": 0.229445880651474,
      "num_tokens": 117891341.0,
      "step": 56810
    },
    {
      "entropy": 5.669263648986816,
      "epoch": 5.153755442670537,
      "grad_norm": 1.453125,
      "learning_rate": 0.000267242468284594,
      "loss": 4.8168,
      "mean_token_accuracy": 0.22961983680725098,
      "num_tokens": 117902304.0,
      "step": 56815
    },
    {
      "entropy": 5.627924156188965,
      "epoch": 5.154208998548621,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026721013118230605,
      "loss": 4.8071,
      "mean_token_accuracy": 0.23406459391117096,
      "num_tokens": 117913367.0,
      "step": 56820
    },
    {
      "entropy": 5.666586542129517,
      "epoch": 5.1546625544267055,
      "grad_norm": 1.375,
      "learning_rate": 0.00026717779424111456,
      "loss": 4.7827,
      "mean_token_accuracy": 0.22826813608407975,
      "num_tokens": 117922750.0,
      "step": 56825
    },
    {
      "entropy": 5.680392503738403,
      "epoch": 5.15511611030479,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026714545746168826,
      "loss": 4.8263,
      "mean_token_accuracy": 0.22714938521385192,
      "num_tokens": 117933950.0,
      "step": 56830
    },
    {
      "entropy": 5.675910949707031,
      "epoch": 5.155569666182874,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000267113120844696,
      "loss": 4.8671,
      "mean_token_accuracy": 0.2258892461657524,
      "num_tokens": 117944985.0,
      "step": 56835
    },
    {
      "entropy": 5.618929719924926,
      "epoch": 5.156023222060958,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002670807843908064,
      "loss": 4.7276,
      "mean_token_accuracy": 0.23721825778484346,
      "num_tokens": 117954560.0,
      "step": 56840
    },
    {
      "entropy": 5.66965765953064,
      "epoch": 5.156476777939042,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026704844810068805,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23372475504875184,
      "num_tokens": 117965690.0,
      "step": 56845
    },
    {
      "entropy": 5.720085334777832,
      "epoch": 5.156930333817126,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002670161119750099,
      "loss": 4.8344,
      "mean_token_accuracy": 0.2268162712454796,
      "num_tokens": 117975856.0,
      "step": 56850
    },
    {
      "entropy": 5.698156261444092,
      "epoch": 5.15738388969521,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002669837760144407,
      "loss": 4.8087,
      "mean_token_accuracy": 0.2320525199174881,
      "num_tokens": 117985673.0,
      "step": 56855
    },
    {
      "entropy": 5.67565221786499,
      "epoch": 5.1578374455732945,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026695144021964895,
      "loss": 4.7921,
      "mean_token_accuracy": 0.23237836509943008,
      "num_tokens": 117995866.0,
      "step": 56860
    },
    {
      "entropy": 5.656952285766602,
      "epoch": 5.158291001451379,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026691910459130346,
      "loss": 4.8469,
      "mean_token_accuracy": 0.2281578838825226,
      "num_tokens": 118006756.0,
      "step": 56865
    },
    {
      "entropy": 5.704979372024536,
      "epoch": 5.158744557329463,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026688676913007303,
      "loss": 4.9141,
      "mean_token_accuracy": 0.23199657201766968,
      "num_tokens": 118016506.0,
      "step": 56870
    },
    {
      "entropy": 5.700165843963623,
      "epoch": 5.159198113207547,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026685443383662615,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23043746054172515,
      "num_tokens": 118026681.0,
      "step": 56875
    },
    {
      "entropy": 5.675029277801514,
      "epoch": 5.1596516690856316,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026682209871163175,
      "loss": 4.8294,
      "mean_token_accuracy": 0.2316351816058159,
      "num_tokens": 118036376.0,
      "step": 56880
    },
    {
      "entropy": 5.7270715713500975,
      "epoch": 5.160105224963716,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002667897637557584,
      "loss": 4.8796,
      "mean_token_accuracy": 0.23103978335857392,
      "num_tokens": 118047029.0,
      "step": 56885
    },
    {
      "entropy": 5.7259503364562985,
      "epoch": 5.1605587808418,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002667574289696748,
      "loss": 4.8202,
      "mean_token_accuracy": 0.22742191553115845,
      "num_tokens": 118056887.0,
      "step": 56890
    },
    {
      "entropy": 5.733806896209717,
      "epoch": 5.1610123367198835,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026672509435404974,
      "loss": 4.808,
      "mean_token_accuracy": 0.22473210096359253,
      "num_tokens": 118067036.0,
      "step": 56895
    },
    {
      "entropy": 5.687627172470092,
      "epoch": 5.161465892597968,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026669275990955175,
      "loss": 4.8612,
      "mean_token_accuracy": 0.23310279101133347,
      "num_tokens": 118077439.0,
      "step": 56900
    },
    {
      "entropy": 5.691139888763428,
      "epoch": 5.161919448476052,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026666042563684966,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2213638484477997,
      "num_tokens": 118089009.0,
      "step": 56905
    },
    {
      "entropy": 5.675304841995239,
      "epoch": 5.162373004354136,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002666280915366121,
      "loss": 4.8738,
      "mean_token_accuracy": 0.22702086418867112,
      "num_tokens": 118098820.0,
      "step": 56910
    },
    {
      "entropy": 5.70185112953186,
      "epoch": 5.1628265602322205,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026659575760950774,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23308684080839157,
      "num_tokens": 118108896.0,
      "step": 56915
    },
    {
      "entropy": 5.781920576095581,
      "epoch": 5.163280116110305,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002665634238562052,
      "loss": 4.9451,
      "mean_token_accuracy": 0.2269973188638687,
      "num_tokens": 118118481.0,
      "step": 56920
    },
    {
      "entropy": 5.779246616363525,
      "epoch": 5.163733671988389,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002665310902773733,
      "loss": 4.9098,
      "mean_token_accuracy": 0.23016159236431122,
      "num_tokens": 118128236.0,
      "step": 56925
    },
    {
      "entropy": 5.707146453857422,
      "epoch": 5.164187227866473,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026649875687368053,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23316035121679307,
      "num_tokens": 118138967.0,
      "step": 56930
    },
    {
      "entropy": 5.738507795333862,
      "epoch": 5.164640783744558,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026646642364579567,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2205555334687233,
      "num_tokens": 118148920.0,
      "step": 56935
    },
    {
      "entropy": 5.6975922107696535,
      "epoch": 5.165094339622642,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026643409059438726,
      "loss": 4.8089,
      "mean_token_accuracy": 0.23055629730224608,
      "num_tokens": 118158184.0,
      "step": 56940
    },
    {
      "entropy": 5.7067248821258545,
      "epoch": 5.165547895500726,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026640175772012415,
      "loss": 4.8251,
      "mean_token_accuracy": 0.22593393325805664,
      "num_tokens": 118168785.0,
      "step": 56945
    },
    {
      "entropy": 5.705708980560303,
      "epoch": 5.1660014513788095,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026636942502367486,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2210922732949257,
      "num_tokens": 118179713.0,
      "step": 56950
    },
    {
      "entropy": 5.712290573120117,
      "epoch": 5.166455007256894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000266337092505708,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22563852965831757,
      "num_tokens": 118190226.0,
      "step": 56955
    },
    {
      "entropy": 5.709011697769165,
      "epoch": 5.166908563134978,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002663047601668923,
      "loss": 4.8099,
      "mean_token_accuracy": 0.22988296896219254,
      "num_tokens": 118200649.0,
      "step": 56960
    },
    {
      "entropy": 5.720214796066284,
      "epoch": 5.167362119013062,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002662724280078964,
      "loss": 4.9303,
      "mean_token_accuracy": 0.22308055311441422,
      "num_tokens": 118211249.0,
      "step": 56965
    },
    {
      "entropy": 5.778474617004394,
      "epoch": 5.167815674891147,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002662400960293889,
      "loss": 4.9488,
      "mean_token_accuracy": 0.22481218576431275,
      "num_tokens": 118221734.0,
      "step": 56970
    },
    {
      "entropy": 5.740991640090942,
      "epoch": 5.168269230769231,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026620776423203846,
      "loss": 4.8668,
      "mean_token_accuracy": 0.23388909548521042,
      "num_tokens": 118232371.0,
      "step": 56975
    },
    {
      "entropy": 5.764886665344238,
      "epoch": 5.168722786647315,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026617543261651374,
      "loss": 4.9335,
      "mean_token_accuracy": 0.22627049088478088,
      "num_tokens": 118242830.0,
      "step": 56980
    },
    {
      "entropy": 5.672119760513306,
      "epoch": 5.169176342525399,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002661431011834833,
      "loss": 4.7764,
      "mean_token_accuracy": 0.23186991661787032,
      "num_tokens": 118253415.0,
      "step": 56985
    },
    {
      "entropy": 5.746823978424072,
      "epoch": 5.169629898403484,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026611076993361574,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22523826360702515,
      "num_tokens": 118263744.0,
      "step": 56990
    },
    {
      "entropy": 5.688614416122436,
      "epoch": 5.170083454281568,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026607843886757964,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22544456869363785,
      "num_tokens": 118274563.0,
      "step": 56995
    },
    {
      "entropy": 5.702584266662598,
      "epoch": 5.170537010159651,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026604610798604386,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2243061766028404,
      "num_tokens": 118285152.0,
      "step": 57000
    },
    {
      "epoch": 5.170537010159651,
      "eval_entropy": 5.403809557241552,
      "eval_loss": 4.951836585998535,
      "eval_mean_token_accuracy": 0.2322682266911635,
      "eval_num_tokens": 118285152.0,
      "eval_runtime": 20.4634,
      "eval_samples_per_second": 1727.96,
      "eval_steps_per_second": 215.995,
      "step": 57000
    },
    {
      "entropy": 5.666970109939575,
      "epoch": 5.1709905660377355,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002660137772896768,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2308335691690445,
      "num_tokens": 118295783.0,
      "step": 57005
    },
    {
      "entropy": 5.747026062011718,
      "epoch": 5.17144412191582,
      "grad_norm": 1.25,
      "learning_rate": 0.00026598144677914703,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23196332305669784,
      "num_tokens": 118305432.0,
      "step": 57010
    },
    {
      "entropy": 5.7993861675262455,
      "epoch": 5.171897677793904,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002659491164551233,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2208962246775627,
      "num_tokens": 118316943.0,
      "step": 57015
    },
    {
      "entropy": 5.674861478805542,
      "epoch": 5.172351233671988,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002659167863182742,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22740925550460817,
      "num_tokens": 118326915.0,
      "step": 57020
    },
    {
      "entropy": 5.6732847690582275,
      "epoch": 5.172804789550073,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026588445636926815,
      "loss": 4.8806,
      "mean_token_accuracy": 0.227208511531353,
      "num_tokens": 118336925.0,
      "step": 57025
    },
    {
      "entropy": 5.71574034690857,
      "epoch": 5.173258345428157,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000265852126608774,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22519103437662125,
      "num_tokens": 118347681.0,
      "step": 57030
    },
    {
      "entropy": 5.731279945373535,
      "epoch": 5.173711901306241,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026581979703746016,
      "loss": 4.8605,
      "mean_token_accuracy": 0.2274943545460701,
      "num_tokens": 118359429.0,
      "step": 57035
    },
    {
      "entropy": 5.747083950042724,
      "epoch": 5.174165457184325,
      "grad_norm": 1.375,
      "learning_rate": 0.0002657874676559953,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2249818414449692,
      "num_tokens": 118369800.0,
      "step": 57040
    },
    {
      "entropy": 5.666087532043457,
      "epoch": 5.17461901306241,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026575513846504786,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2276519924402237,
      "num_tokens": 118381822.0,
      "step": 57045
    },
    {
      "entropy": 5.7898101806640625,
      "epoch": 5.175072568940493,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026572280946528655,
      "loss": 4.9699,
      "mean_token_accuracy": 0.22559009343385697,
      "num_tokens": 118393157.0,
      "step": 57050
    },
    {
      "entropy": 5.717793226242065,
      "epoch": 5.175526124818577,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002656904806573799,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2288466289639473,
      "num_tokens": 118403538.0,
      "step": 57055
    },
    {
      "entropy": 5.719566488265992,
      "epoch": 5.175979680696662,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026565815204199645,
      "loss": 4.7815,
      "mean_token_accuracy": 0.23608755320310593,
      "num_tokens": 118413756.0,
      "step": 57060
    },
    {
      "entropy": 5.718531560897827,
      "epoch": 5.176433236574746,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026562582361980486,
      "loss": 4.8482,
      "mean_token_accuracy": 0.23260662704706192,
      "num_tokens": 118423967.0,
      "step": 57065
    },
    {
      "entropy": 5.66935133934021,
      "epoch": 5.17688679245283,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002655934953914736,
      "loss": 4.8052,
      "mean_token_accuracy": 0.22826571613550187,
      "num_tokens": 118434004.0,
      "step": 57070
    },
    {
      "entropy": 5.7884856224060055,
      "epoch": 5.177340348330914,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002655611673576713,
      "loss": 4.975,
      "mean_token_accuracy": 0.22374993711709976,
      "num_tokens": 118445083.0,
      "step": 57075
    },
    {
      "entropy": 5.689521646499633,
      "epoch": 5.177793904208999,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026552883951906636,
      "loss": 4.7846,
      "mean_token_accuracy": 0.238901749253273,
      "num_tokens": 118454925.0,
      "step": 57080
    },
    {
      "entropy": 5.629383134841919,
      "epoch": 5.178247460087083,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002654965118763275,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2310573488473892,
      "num_tokens": 118465562.0,
      "step": 57085
    },
    {
      "entropy": 5.687942123413086,
      "epoch": 5.178701015965167,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002654641844301232,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22811280339956283,
      "num_tokens": 118476884.0,
      "step": 57090
    },
    {
      "entropy": 5.709527921676636,
      "epoch": 5.1791545718432515,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026543185718112193,
      "loss": 4.882,
      "mean_token_accuracy": 0.22673517167568208,
      "num_tokens": 118486874.0,
      "step": 57095
    },
    {
      "entropy": 5.770507955551148,
      "epoch": 5.179608127721336,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002653995301299924,
      "loss": 4.8962,
      "mean_token_accuracy": 0.23091472834348678,
      "num_tokens": 118496427.0,
      "step": 57100
    },
    {
      "entropy": 5.727494478225708,
      "epoch": 5.180061683599419,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000265367203277403,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23206431716680526,
      "num_tokens": 118506809.0,
      "step": 57105
    },
    {
      "entropy": 5.804040575027466,
      "epoch": 5.180515239477503,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002653348766240222,
      "loss": 4.9319,
      "mean_token_accuracy": 0.22154633998870848,
      "num_tokens": 118516525.0,
      "step": 57110
    },
    {
      "entropy": 5.596335172653198,
      "epoch": 5.180968795355588,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026530255017051876,
      "loss": 4.7418,
      "mean_token_accuracy": 0.23823448717594148,
      "num_tokens": 118527508.0,
      "step": 57115
    },
    {
      "entropy": 5.721756362915039,
      "epoch": 5.181422351233672,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026527022391756094,
      "loss": 4.89,
      "mean_token_accuracy": 0.22747301012277604,
      "num_tokens": 118538038.0,
      "step": 57120
    },
    {
      "entropy": 5.685556411743164,
      "epoch": 5.181875907111756,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026523789786581743,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2356083035469055,
      "num_tokens": 118548521.0,
      "step": 57125
    },
    {
      "entropy": 5.676804542541504,
      "epoch": 5.18232946298984,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026520557201595665,
      "loss": 4.7904,
      "mean_token_accuracy": 0.23269310146570205,
      "num_tokens": 118558703.0,
      "step": 57130
    },
    {
      "entropy": 5.784606313705444,
      "epoch": 5.182783018867925,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002651732463686471,
      "loss": 4.9329,
      "mean_token_accuracy": 0.22286900281906127,
      "num_tokens": 118568746.0,
      "step": 57135
    },
    {
      "entropy": 5.741765069961548,
      "epoch": 5.183236574746009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002651409209245574,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23169925659894944,
      "num_tokens": 118578822.0,
      "step": 57140
    },
    {
      "entropy": 5.730036067962646,
      "epoch": 5.183690130624093,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002651085956843559,
      "loss": 4.8846,
      "mean_token_accuracy": 0.23047399073839187,
      "num_tokens": 118588363.0,
      "step": 57145
    },
    {
      "entropy": 5.810389137268066,
      "epoch": 5.1841436865021775,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026507627064871123,
      "loss": 4.8736,
      "mean_token_accuracy": 0.22972043752670288,
      "num_tokens": 118597869.0,
      "step": 57150
    },
    {
      "entropy": 5.702063751220703,
      "epoch": 5.184597242380261,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002650439458182917,
      "loss": 4.7905,
      "mean_token_accuracy": 0.2351416453719139,
      "num_tokens": 118608196.0,
      "step": 57155
    },
    {
      "entropy": 5.672505617141724,
      "epoch": 5.185050798258345,
      "grad_norm": 1.453125,
      "learning_rate": 0.000265011621193766,
      "loss": 4.8035,
      "mean_token_accuracy": 0.2278994470834732,
      "num_tokens": 118617987.0,
      "step": 57160
    },
    {
      "entropy": 5.6560811519622805,
      "epoch": 5.185504354136429,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026497929677580254,
      "loss": 4.9178,
      "mean_token_accuracy": 0.21676027327775954,
      "num_tokens": 118629740.0,
      "step": 57165
    },
    {
      "entropy": 5.758957958221435,
      "epoch": 5.185957910014514,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002649469725650697,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22612751722335817,
      "num_tokens": 118640752.0,
      "step": 57170
    },
    {
      "entropy": 5.759427690505982,
      "epoch": 5.186411465892598,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026491464856223613,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22059738487005234,
      "num_tokens": 118650064.0,
      "step": 57175
    },
    {
      "entropy": 5.66896390914917,
      "epoch": 5.186865021770682,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002648823247679701,
      "loss": 4.8089,
      "mean_token_accuracy": 0.23043260276317595,
      "num_tokens": 118660332.0,
      "step": 57180
    },
    {
      "entropy": 5.658292627334594,
      "epoch": 5.1873185776487665,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026485000118294015,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23023627549409867,
      "num_tokens": 118671694.0,
      "step": 57185
    },
    {
      "entropy": 5.75336389541626,
      "epoch": 5.187772133526851,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026481767780781485,
      "loss": 4.9055,
      "mean_token_accuracy": 0.22390621453523635,
      "num_tokens": 118682103.0,
      "step": 57190
    },
    {
      "entropy": 5.717915344238281,
      "epoch": 5.188225689404935,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002647853546432626,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23198784589767457,
      "num_tokens": 118691995.0,
      "step": 57195
    },
    {
      "entropy": 5.705847835540771,
      "epoch": 5.188679245283019,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026475303168995167,
      "loss": 4.8919,
      "mean_token_accuracy": 0.22833572179079056,
      "num_tokens": 118702587.0,
      "step": 57200
    },
    {
      "entropy": 5.712657690048218,
      "epoch": 5.189132801161103,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026472070894855077,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22212835848331453,
      "num_tokens": 118712032.0,
      "step": 57205
    },
    {
      "entropy": 5.6584086418151855,
      "epoch": 5.189586357039187,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002646883864197282,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23706710785627366,
      "num_tokens": 118723266.0,
      "step": 57210
    },
    {
      "entropy": 5.700101613998413,
      "epoch": 5.190039912917271,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002646560641041524,
      "loss": 4.8387,
      "mean_token_accuracy": 0.23166913986206056,
      "num_tokens": 118733025.0,
      "step": 57215
    },
    {
      "entropy": 5.69539966583252,
      "epoch": 5.1904934687953554,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026462374200249186,
      "loss": 4.918,
      "mean_token_accuracy": 0.22974975556135177,
      "num_tokens": 118742690.0,
      "step": 57220
    },
    {
      "entropy": 5.719268035888672,
      "epoch": 5.19094702467344,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000264591420115415,
      "loss": 4.8535,
      "mean_token_accuracy": 0.23043246567249298,
      "num_tokens": 118753032.0,
      "step": 57225
    },
    {
      "entropy": 5.713577365875244,
      "epoch": 5.191400580551524,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002645590984435902,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23665496259927749,
      "num_tokens": 118763764.0,
      "step": 57230
    },
    {
      "entropy": 5.7820065975189205,
      "epoch": 5.191854136429608,
      "grad_norm": 1.375,
      "learning_rate": 0.0002645267769876859,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22527266442775726,
      "num_tokens": 118773236.0,
      "step": 57235
    },
    {
      "entropy": 5.672686672210693,
      "epoch": 5.1923076923076925,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002644944557483705,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22993765473365785,
      "num_tokens": 118783239.0,
      "step": 57240
    },
    {
      "entropy": 5.703679132461548,
      "epoch": 5.192761248185777,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002644621347263125,
      "loss": 4.7753,
      "mean_token_accuracy": 0.23824572414159775,
      "num_tokens": 118792514.0,
      "step": 57245
    },
    {
      "entropy": 5.648192262649536,
      "epoch": 5.193214804063861,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026442981392218014,
      "loss": 4.7375,
      "mean_token_accuracy": 0.2343438521027565,
      "num_tokens": 118802673.0,
      "step": 57250
    },
    {
      "entropy": 5.653634691238404,
      "epoch": 5.193668359941944,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000264397493336642,
      "loss": 4.8461,
      "mean_token_accuracy": 0.22533278912305832,
      "num_tokens": 118812917.0,
      "step": 57255
    },
    {
      "entropy": 5.614755344390869,
      "epoch": 5.194121915820029,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002643651729703664,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23312774002552034,
      "num_tokens": 118822907.0,
      "step": 57260
    },
    {
      "entropy": 5.7346995830535885,
      "epoch": 5.194575471698113,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002643328528240216,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22883837670087814,
      "num_tokens": 118832072.0,
      "step": 57265
    },
    {
      "entropy": 5.6736985206604,
      "epoch": 5.195029027576197,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002643005328982763,
      "loss": 4.8183,
      "mean_token_accuracy": 0.23603693395853043,
      "num_tokens": 118842499.0,
      "step": 57270
    },
    {
      "entropy": 5.735856342315674,
      "epoch": 5.1954825834542815,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026426821319379867,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2230727791786194,
      "num_tokens": 118853266.0,
      "step": 57275
    },
    {
      "entropy": 5.715337657928467,
      "epoch": 5.195936139332366,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026423589371125705,
      "loss": 4.8958,
      "mean_token_accuracy": 0.22506764084100722,
      "num_tokens": 118864052.0,
      "step": 57280
    },
    {
      "entropy": 5.713037300109863,
      "epoch": 5.19638969521045,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026420357445132,
      "loss": 4.8619,
      "mean_token_accuracy": 0.23014753609895705,
      "num_tokens": 118874301.0,
      "step": 57285
    },
    {
      "entropy": 5.725647735595703,
      "epoch": 5.196843251088534,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026417125541465574,
      "loss": 4.7921,
      "mean_token_accuracy": 0.22650639861822128,
      "num_tokens": 118884387.0,
      "step": 57290
    },
    {
      "entropy": 5.70687165260315,
      "epoch": 5.1972968069666186,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026413893660193265,
      "loss": 4.819,
      "mean_token_accuracy": 0.23510608375072478,
      "num_tokens": 118894612.0,
      "step": 57295
    },
    {
      "entropy": 5.741703796386719,
      "epoch": 5.197750362844703,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002641066180138191,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22246829122304917,
      "num_tokens": 118904185.0,
      "step": 57300
    },
    {
      "entropy": 5.8127734661102295,
      "epoch": 5.198203918722786,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026407429965098354,
      "loss": 4.9901,
      "mean_token_accuracy": 0.22216654419898987,
      "num_tokens": 118915013.0,
      "step": 57305
    },
    {
      "entropy": 5.759879064559937,
      "epoch": 5.1986574746008705,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026404198151409414,
      "loss": 4.8713,
      "mean_token_accuracy": 0.2410392567515373,
      "num_tokens": 118926520.0,
      "step": 57310
    },
    {
      "entropy": 5.600305128097534,
      "epoch": 5.199111030478955,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026400966360381946,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23337650746107103,
      "num_tokens": 118937601.0,
      "step": 57315
    },
    {
      "entropy": 5.681092596054077,
      "epoch": 5.199564586357039,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026397734592082766,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22794943153858185,
      "num_tokens": 118948030.0,
      "step": 57320
    },
    {
      "entropy": 5.697225141525268,
      "epoch": 5.200018142235123,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026394502846578715,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2327963501214981,
      "num_tokens": 118958670.0,
      "step": 57325
    },
    {
      "entropy": 5.803017282485962,
      "epoch": 5.2004716981132075,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026391271123936633,
      "loss": 4.9616,
      "mean_token_accuracy": 0.2303765371441841,
      "num_tokens": 118969211.0,
      "step": 57330
    },
    {
      "entropy": 5.765566444396972,
      "epoch": 5.200925253991292,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002638803942422335,
      "loss": 4.9253,
      "mean_token_accuracy": 0.22255240976810456,
      "num_tokens": 118980837.0,
      "step": 57335
    },
    {
      "entropy": 5.698023653030395,
      "epoch": 5.201378809869376,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026384807747505687,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2294198676943779,
      "num_tokens": 118991266.0,
      "step": 57340
    },
    {
      "entropy": 5.786822175979614,
      "epoch": 5.20183236574746,
      "grad_norm": 1.5,
      "learning_rate": 0.00026381576093850484,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2247697800397873,
      "num_tokens": 119001459.0,
      "step": 57345
    },
    {
      "entropy": 5.751399230957031,
      "epoch": 5.202285921625545,
      "grad_norm": 1.5,
      "learning_rate": 0.0002637834446332457,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2303353652358055,
      "num_tokens": 119012010.0,
      "step": 57350
    },
    {
      "entropy": 5.698359203338623,
      "epoch": 5.202739477503629,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026375112855994785,
      "loss": 4.8049,
      "mean_token_accuracy": 0.23279273509979248,
      "num_tokens": 119022371.0,
      "step": 57355
    },
    {
      "entropy": 5.799624013900757,
      "epoch": 5.203193033381712,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002637188127192795,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2252802088856697,
      "num_tokens": 119032457.0,
      "step": 57360
    },
    {
      "entropy": 5.645282983779907,
      "epoch": 5.2036465892597965,
      "grad_norm": 1.25,
      "learning_rate": 0.000263686497111909,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22395193725824356,
      "num_tokens": 119044133.0,
      "step": 57365
    },
    {
      "entropy": 5.659222936630249,
      "epoch": 5.204100145137881,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026365418173850466,
      "loss": 4.7341,
      "mean_token_accuracy": 0.23971920013427733,
      "num_tokens": 119054412.0,
      "step": 57370
    },
    {
      "entropy": 5.660951614379883,
      "epoch": 5.204553701015965,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026362186659973464,
      "loss": 4.7645,
      "mean_token_accuracy": 0.2362660586833954,
      "num_tokens": 119064950.0,
      "step": 57375
    },
    {
      "entropy": 5.724212551116944,
      "epoch": 5.205007256894049,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002635895516962673,
      "loss": 4.9429,
      "mean_token_accuracy": 0.22067004144191743,
      "num_tokens": 119074945.0,
      "step": 57380
    },
    {
      "entropy": 5.689139747619629,
      "epoch": 5.205460812772134,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026355723702877105,
      "loss": 4.8618,
      "mean_token_accuracy": 0.23066096752882004,
      "num_tokens": 119086193.0,
      "step": 57385
    },
    {
      "entropy": 5.712560033798217,
      "epoch": 5.205914368650218,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000263524922597914,
      "loss": 4.8388,
      "mean_token_accuracy": 0.23961153775453567,
      "num_tokens": 119096697.0,
      "step": 57390
    },
    {
      "entropy": 5.722845411300659,
      "epoch": 5.206367924528302,
      "grad_norm": 1.25,
      "learning_rate": 0.0002634926084043645,
      "loss": 4.8297,
      "mean_token_accuracy": 0.22532631754875182,
      "num_tokens": 119107833.0,
      "step": 57395
    },
    {
      "entropy": 5.74068660736084,
      "epoch": 5.206821480406386,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026346029444879074,
      "loss": 4.8417,
      "mean_token_accuracy": 0.229462893307209,
      "num_tokens": 119117602.0,
      "step": 57400
    },
    {
      "entropy": 5.733783817291259,
      "epoch": 5.207275036284471,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026342798073186106,
      "loss": 4.8713,
      "mean_token_accuracy": 0.23202442228794098,
      "num_tokens": 119128546.0,
      "step": 57405
    },
    {
      "entropy": 5.701044130325317,
      "epoch": 5.207728592162554,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026339566725424365,
      "loss": 4.848,
      "mean_token_accuracy": 0.22857709974050522,
      "num_tokens": 119138079.0,
      "step": 57410
    },
    {
      "entropy": 5.690371084213257,
      "epoch": 5.208182148040638,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026336335401660685,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23063485622406005,
      "num_tokens": 119148239.0,
      "step": 57415
    },
    {
      "entropy": 5.730324935913086,
      "epoch": 5.2086357039187225,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026333104101961886,
      "loss": 4.8303,
      "mean_token_accuracy": 0.23266777843236924,
      "num_tokens": 119159266.0,
      "step": 57420
    },
    {
      "entropy": 5.726632022857666,
      "epoch": 5.209089259796807,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002632987282639479,
      "loss": 4.8915,
      "mean_token_accuracy": 0.22761156558990478,
      "num_tokens": 119168766.0,
      "step": 57425
    },
    {
      "entropy": 5.658399343490601,
      "epoch": 5.209542815674891,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002632664157502622,
      "loss": 4.7581,
      "mean_token_accuracy": 0.2396991655230522,
      "num_tokens": 119179529.0,
      "step": 57430
    },
    {
      "entropy": 5.6761188983917235,
      "epoch": 5.209996371552975,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026323410347923,
      "loss": 4.8156,
      "mean_token_accuracy": 0.23034156262874603,
      "num_tokens": 119189796.0,
      "step": 57435
    },
    {
      "entropy": 5.733530187606812,
      "epoch": 5.21044992743106,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026320179145151957,
      "loss": 4.8928,
      "mean_token_accuracy": 0.2350971058011055,
      "num_tokens": 119199968.0,
      "step": 57440
    },
    {
      "entropy": 5.728397274017334,
      "epoch": 5.210903483309144,
      "grad_norm": 1.453125,
      "learning_rate": 0.000263169479667799,
      "loss": 4.9341,
      "mean_token_accuracy": 0.23067790567874907,
      "num_tokens": 119210179.0,
      "step": 57445
    },
    {
      "entropy": 5.728341960906983,
      "epoch": 5.211357039187228,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002631371681287368,
      "loss": 4.8996,
      "mean_token_accuracy": 0.23104715198278428,
      "num_tokens": 119222011.0,
      "step": 57450
    },
    {
      "entropy": 5.725264215469361,
      "epoch": 5.211810595065312,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00026310485683500085,
      "loss": 4.8475,
      "mean_token_accuracy": 0.23581180423498155,
      "num_tokens": 119232166.0,
      "step": 57455
    },
    {
      "entropy": 5.812628173828125,
      "epoch": 5.212264150943396,
      "grad_norm": 1.5,
      "learning_rate": 0.0002630725457872595,
      "loss": 4.9695,
      "mean_token_accuracy": 0.22253214120864867,
      "num_tokens": 119241473.0,
      "step": 57460
    },
    {
      "entropy": 5.764245510101318,
      "epoch": 5.21271770682148,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026304023498618093,
      "loss": 4.9127,
      "mean_token_accuracy": 0.23050932288169862,
      "num_tokens": 119252092.0,
      "step": 57465
    },
    {
      "entropy": 5.64822359085083,
      "epoch": 5.213171262699564,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026300792443243343,
      "loss": 4.7823,
      "mean_token_accuracy": 0.23318624198436738,
      "num_tokens": 119262473.0,
      "step": 57470
    },
    {
      "entropy": 5.719549131393433,
      "epoch": 5.213624818577649,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002629756141266851,
      "loss": 4.845,
      "mean_token_accuracy": 0.23259880095720292,
      "num_tokens": 119271639.0,
      "step": 57475
    },
    {
      "entropy": 5.680423784255981,
      "epoch": 5.214078374455733,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002629433040696041,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22717735171318054,
      "num_tokens": 119282463.0,
      "step": 57480
    },
    {
      "entropy": 5.74751992225647,
      "epoch": 5.214531930333817,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026291099426185865,
      "loss": 4.8731,
      "mean_token_accuracy": 0.23166370987892151,
      "num_tokens": 119292472.0,
      "step": 57485
    },
    {
      "entropy": 5.690866279602051,
      "epoch": 5.214985486211901,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002628786847041169,
      "loss": 4.8101,
      "mean_token_accuracy": 0.228443942964077,
      "num_tokens": 119302447.0,
      "step": 57490
    },
    {
      "entropy": 5.662809085845947,
      "epoch": 5.215439042089986,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026284637539704706,
      "loss": 4.7741,
      "mean_token_accuracy": 0.23160124719142913,
      "num_tokens": 119312864.0,
      "step": 57495
    },
    {
      "entropy": 5.745269727706909,
      "epoch": 5.21589259796807,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002628140663413172,
      "loss": 4.9198,
      "mean_token_accuracy": 0.22616743445396423,
      "num_tokens": 119322866.0,
      "step": 57500
    },
    {
      "entropy": 5.699090194702149,
      "epoch": 5.216346153846154,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002627817575375957,
      "loss": 4.7892,
      "mean_token_accuracy": 0.23682791143655776,
      "num_tokens": 119333185.0,
      "step": 57505
    },
    {
      "entropy": 5.643411159515381,
      "epoch": 5.2167997097242385,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026274944898655047,
      "loss": 4.7826,
      "mean_token_accuracy": 0.24360227584838867,
      "num_tokens": 119342955.0,
      "step": 57510
    },
    {
      "entropy": 5.683068513870239,
      "epoch": 5.217253265602322,
      "grad_norm": 1.375,
      "learning_rate": 0.00026271714068884977,
      "loss": 4.9138,
      "mean_token_accuracy": 0.22683630138635635,
      "num_tokens": 119353346.0,
      "step": 57515
    },
    {
      "entropy": 5.733803033828735,
      "epoch": 5.217706821480406,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026268483264516176,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22853785902261733,
      "num_tokens": 119364300.0,
      "step": 57520
    },
    {
      "entropy": 5.755802726745605,
      "epoch": 5.21816037735849,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026265252485615446,
      "loss": 4.8562,
      "mean_token_accuracy": 0.23656970560550689,
      "num_tokens": 119374742.0,
      "step": 57525
    },
    {
      "entropy": 5.711874914169312,
      "epoch": 5.218613933236575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002626202173224962,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21998124569654465,
      "num_tokens": 119384948.0,
      "step": 57530
    },
    {
      "entropy": 5.712575483322143,
      "epoch": 5.219067489114659,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026258791004485496,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22658462226390838,
      "num_tokens": 119395883.0,
      "step": 57535
    },
    {
      "entropy": 5.739665889739991,
      "epoch": 5.219521044992743,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026255560302389886,
      "loss": 4.8313,
      "mean_token_accuracy": 0.2295038029551506,
      "num_tokens": 119406285.0,
      "step": 57540
    },
    {
      "entropy": 5.65881667137146,
      "epoch": 5.219974600870827,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026252329626029616,
      "loss": 4.8293,
      "mean_token_accuracy": 0.23607510328292847,
      "num_tokens": 119417205.0,
      "step": 57545
    },
    {
      "entropy": 5.703415489196777,
      "epoch": 5.220428156748912,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026249098975471475,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22939011752605437,
      "num_tokens": 119427619.0,
      "step": 57550
    },
    {
      "entropy": 5.691246366500854,
      "epoch": 5.220881712626996,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002624586835078229,
      "loss": 4.8667,
      "mean_token_accuracy": 0.2319940522313118,
      "num_tokens": 119437020.0,
      "step": 57555
    },
    {
      "entropy": 5.736818456649781,
      "epoch": 5.22133526850508,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026242637752028875,
      "loss": 4.8993,
      "mean_token_accuracy": 0.22547044157981871,
      "num_tokens": 119447111.0,
      "step": 57560
    },
    {
      "entropy": 5.719141674041748,
      "epoch": 5.221788824383164,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026239407179278027,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23595301061868668,
      "num_tokens": 119458634.0,
      "step": 57565
    },
    {
      "entropy": 5.754668664932251,
      "epoch": 5.222242380261248,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026236176632596554,
      "loss": 4.8442,
      "mean_token_accuracy": 0.23053391128778458,
      "num_tokens": 119468193.0,
      "step": 57570
    },
    {
      "entropy": 5.679657793045044,
      "epoch": 5.222695936139332,
      "grad_norm": 1.375,
      "learning_rate": 0.00026232946112051276,
      "loss": 4.8408,
      "mean_token_accuracy": 0.2318178117275238,
      "num_tokens": 119477755.0,
      "step": 57575
    },
    {
      "entropy": 5.64786057472229,
      "epoch": 5.223149492017416,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026229715617708996,
      "loss": 4.7925,
      "mean_token_accuracy": 0.2341538891196251,
      "num_tokens": 119487943.0,
      "step": 57580
    },
    {
      "entropy": 5.729452133178711,
      "epoch": 5.223603047895501,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002622648514963652,
      "loss": 4.8574,
      "mean_token_accuracy": 0.21953269839286804,
      "num_tokens": 119498731.0,
      "step": 57585
    },
    {
      "entropy": 5.730173349380493,
      "epoch": 5.224056603773585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026223254707900655,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2394932985305786,
      "num_tokens": 119509991.0,
      "step": 57590
    },
    {
      "entropy": 5.737314176559448,
      "epoch": 5.224510159651669,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026220024292568204,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2277407631278038,
      "num_tokens": 119520208.0,
      "step": 57595
    },
    {
      "entropy": 5.763263320922851,
      "epoch": 5.2249637155297535,
      "grad_norm": 1.5,
      "learning_rate": 0.0002621679390370598,
      "loss": 4.9319,
      "mean_token_accuracy": 0.2230716571211815,
      "num_tokens": 119530061.0,
      "step": 57600
    },
    {
      "entropy": 5.828800678253174,
      "epoch": 5.225417271407838,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026213563541380787,
      "loss": 4.9015,
      "mean_token_accuracy": 0.22951372265815734,
      "num_tokens": 119540050.0,
      "step": 57605
    },
    {
      "entropy": 5.815857601165772,
      "epoch": 5.225870827285922,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002621033320565943,
      "loss": 4.9789,
      "mean_token_accuracy": 0.21991754919290543,
      "num_tokens": 119550526.0,
      "step": 57610
    },
    {
      "entropy": 5.7192024230957035,
      "epoch": 5.226324383164005,
      "grad_norm": 1.453125,
      "learning_rate": 0.000262071028966087,
      "loss": 4.9033,
      "mean_token_accuracy": 0.23080746978521346,
      "num_tokens": 119560625.0,
      "step": 57615
    },
    {
      "entropy": 5.666923236846924,
      "epoch": 5.22677793904209,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002620387261429543,
      "loss": 4.7926,
      "mean_token_accuracy": 0.2360827460885048,
      "num_tokens": 119571125.0,
      "step": 57620
    },
    {
      "entropy": 5.692579984664917,
      "epoch": 5.227231494920174,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026200642358786385,
      "loss": 4.8479,
      "mean_token_accuracy": 0.2261452630162239,
      "num_tokens": 119581629.0,
      "step": 57625
    },
    {
      "entropy": 5.695913076400757,
      "epoch": 5.227685050798258,
      "grad_norm": 1.234375,
      "learning_rate": 0.000261974121301484,
      "loss": 4.865,
      "mean_token_accuracy": 0.23024393022060394,
      "num_tokens": 119593421.0,
      "step": 57630
    },
    {
      "entropy": 5.6767181873321535,
      "epoch": 5.2281386066763424,
      "grad_norm": 1.375,
      "learning_rate": 0.00026194181928448256,
      "loss": 4.7835,
      "mean_token_accuracy": 0.2403876379132271,
      "num_tokens": 119602446.0,
      "step": 57635
    },
    {
      "entropy": 5.642608785629273,
      "epoch": 5.228592162554427,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002619095175375277,
      "loss": 4.7515,
      "mean_token_accuracy": 0.23983191698789597,
      "num_tokens": 119613071.0,
      "step": 57640
    },
    {
      "entropy": 5.690643739700318,
      "epoch": 5.229045718432511,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002618772160612873,
      "loss": 4.8557,
      "mean_token_accuracy": 0.2298905923962593,
      "num_tokens": 119623945.0,
      "step": 57645
    },
    {
      "entropy": 5.684172630310059,
      "epoch": 5.229499274310595,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002618449148564294,
      "loss": 4.8045,
      "mean_token_accuracy": 0.23713227361440659,
      "num_tokens": 119634491.0,
      "step": 57650
    },
    {
      "entropy": 5.808961200714111,
      "epoch": 5.2299528301886795,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002618126139236221,
      "loss": 5.0169,
      "mean_token_accuracy": 0.2199288785457611,
      "num_tokens": 119645076.0,
      "step": 57655
    },
    {
      "entropy": 5.68375334739685,
      "epoch": 5.230406386066764,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026178031326353325,
      "loss": 4.8223,
      "mean_token_accuracy": 0.23218177855014802,
      "num_tokens": 119655030.0,
      "step": 57660
    },
    {
      "entropy": 5.674128437042237,
      "epoch": 5.230859941944848,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002617480128768309,
      "loss": 4.8188,
      "mean_token_accuracy": 0.2302698612213135,
      "num_tokens": 119664914.0,
      "step": 57665
    },
    {
      "entropy": 5.631623888015747,
      "epoch": 5.231313497822931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026171571276418295,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2319004699587822,
      "num_tokens": 119676281.0,
      "step": 57670
    },
    {
      "entropy": 5.680263614654541,
      "epoch": 5.231767053701016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002616834129262575,
      "loss": 4.8177,
      "mean_token_accuracy": 0.2295519545674324,
      "num_tokens": 119686988.0,
      "step": 57675
    },
    {
      "entropy": 5.626495456695556,
      "epoch": 5.2322206095791,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026165111336372255,
      "loss": 4.7488,
      "mean_token_accuracy": 0.23669030517339706,
      "num_tokens": 119699327.0,
      "step": 57680
    },
    {
      "entropy": 5.6725311279296875,
      "epoch": 5.232674165457184,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026161881407724586,
      "loss": 4.8726,
      "mean_token_accuracy": 0.23071555197238922,
      "num_tokens": 119709527.0,
      "step": 57685
    },
    {
      "entropy": 5.6585142612457275,
      "epoch": 5.2331277213352685,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026158651506749556,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23109739124774933,
      "num_tokens": 119720237.0,
      "step": 57690
    },
    {
      "entropy": 5.708810806274414,
      "epoch": 5.233581277213353,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026155421633513947,
      "loss": 4.8708,
      "mean_token_accuracy": 0.22221488356590272,
      "num_tokens": 119731091.0,
      "step": 57695
    },
    {
      "entropy": 5.6741945266723635,
      "epoch": 5.234034833091437,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026152191788084573,
      "loss": 4.8364,
      "mean_token_accuracy": 0.23627713769674302,
      "num_tokens": 119740986.0,
      "step": 57700
    },
    {
      "entropy": 5.663184356689453,
      "epoch": 5.234488388969521,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026148961970528205,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23335488587617875,
      "num_tokens": 119751045.0,
      "step": 57705
    },
    {
      "entropy": 5.677868938446045,
      "epoch": 5.2349419448476056,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026145732180911667,
      "loss": 4.7974,
      "mean_token_accuracy": 0.22768507450819014,
      "num_tokens": 119761404.0,
      "step": 57710
    },
    {
      "entropy": 5.701984214782715,
      "epoch": 5.23539550072569,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002614250241930172,
      "loss": 4.8375,
      "mean_token_accuracy": 0.2291803315281868,
      "num_tokens": 119771827.0,
      "step": 57715
    },
    {
      "entropy": 5.708445072174072,
      "epoch": 5.235849056603773,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026139272685765164,
      "loss": 4.9005,
      "mean_token_accuracy": 0.226339390873909,
      "num_tokens": 119782688.0,
      "step": 57720
    },
    {
      "entropy": 5.740522003173828,
      "epoch": 5.2363026124818575,
      "grad_norm": 1.296875,
      "learning_rate": 0.000261360429803688,
      "loss": 4.8652,
      "mean_token_accuracy": 0.23290177434682846,
      "num_tokens": 119792498.0,
      "step": 57725
    },
    {
      "entropy": 5.642804908752441,
      "epoch": 5.236756168359942,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002613281330317941,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23163042664527894,
      "num_tokens": 119803855.0,
      "step": 57730
    },
    {
      "entropy": 5.647898197174072,
      "epoch": 5.237209724238026,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026129583654263793,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2356172978878021,
      "num_tokens": 119813280.0,
      "step": 57735
    },
    {
      "entropy": 5.708696174621582,
      "epoch": 5.23766328011611,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026126354033688745,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22803463339805602,
      "num_tokens": 119823966.0,
      "step": 57740
    },
    {
      "entropy": 5.807487154006958,
      "epoch": 5.2381168359941945,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002612312444152103,
      "loss": 4.9762,
      "mean_token_accuracy": 0.2253126546740532,
      "num_tokens": 119834134.0,
      "step": 57745
    },
    {
      "entropy": 5.678874588012695,
      "epoch": 5.238570391872279,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002611989487782746,
      "loss": 4.7783,
      "mean_token_accuracy": 0.234651517868042,
      "num_tokens": 119844134.0,
      "step": 57750
    },
    {
      "entropy": 5.717565631866455,
      "epoch": 5.239023947750363,
      "grad_norm": 1.375,
      "learning_rate": 0.00026116665342674816,
      "loss": 4.9308,
      "mean_token_accuracy": 0.2267521068453789,
      "num_tokens": 119855029.0,
      "step": 57755
    },
    {
      "entropy": 5.67842288017273,
      "epoch": 5.239477503628447,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026113435836129886,
      "loss": 4.807,
      "mean_token_accuracy": 0.22822314351797104,
      "num_tokens": 119865453.0,
      "step": 57760
    },
    {
      "entropy": 5.701947498321533,
      "epoch": 5.239931059506532,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026110206358259447,
      "loss": 4.8585,
      "mean_token_accuracy": 0.23687806129455566,
      "num_tokens": 119876866.0,
      "step": 57765
    },
    {
      "entropy": 5.701238489151001,
      "epoch": 5.240384615384615,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000261069769091303,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23509870767593383,
      "num_tokens": 119886944.0,
      "step": 57770
    },
    {
      "entropy": 5.7829407215118405,
      "epoch": 5.240838171262699,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002610374748880923,
      "loss": 4.9243,
      "mean_token_accuracy": 0.22465475499629975,
      "num_tokens": 119896599.0,
      "step": 57775
    },
    {
      "entropy": 5.738887929916382,
      "epoch": 5.2412917271407835,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026100518097363015,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2272268384695053,
      "num_tokens": 119907360.0,
      "step": 57780
    },
    {
      "entropy": 5.645131206512451,
      "epoch": 5.241745283018868,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026097288734858436,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2374293550848961,
      "num_tokens": 119917269.0,
      "step": 57785
    },
    {
      "entropy": 5.63355450630188,
      "epoch": 5.242198838896952,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000260940594013623,
      "loss": 4.7765,
      "mean_token_accuracy": 0.2316467970609665,
      "num_tokens": 119927234.0,
      "step": 57790
    },
    {
      "entropy": 5.676565408706665,
      "epoch": 5.242652394775036,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002609083009694136,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23482497036457062,
      "num_tokens": 119937208.0,
      "step": 57795
    },
    {
      "entropy": 5.710492515563965,
      "epoch": 5.243105950653121,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026087600821662416,
      "loss": 4.8727,
      "mean_token_accuracy": 0.23112597316503525,
      "num_tokens": 119948385.0,
      "step": 57800
    },
    {
      "entropy": 5.681622123718261,
      "epoch": 5.243559506531205,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002608437157559225,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23472214639186859,
      "num_tokens": 119958155.0,
      "step": 57805
    },
    {
      "entropy": 5.768026494979859,
      "epoch": 5.244013062409289,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026081142358797633,
      "loss": 4.8714,
      "mean_token_accuracy": 0.22825271338224412,
      "num_tokens": 119968822.0,
      "step": 57810
    },
    {
      "entropy": 5.642958879470825,
      "epoch": 5.244466618287373,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026077913171345356,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23452601581811905,
      "num_tokens": 119979363.0,
      "step": 57815
    },
    {
      "entropy": 5.678957080841064,
      "epoch": 5.244920174165458,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026074684013302203,
      "loss": 4.8556,
      "mean_token_accuracy": 0.2342766597867012,
      "num_tokens": 119989766.0,
      "step": 57820
    },
    {
      "entropy": 5.701813268661499,
      "epoch": 5.245373730043541,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002607145488473494,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2223648265004158,
      "num_tokens": 120000836.0,
      "step": 57825
    },
    {
      "entropy": 5.721605396270752,
      "epoch": 5.245827285921625,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026068225785710354,
      "loss": 4.8642,
      "mean_token_accuracy": 0.22931578308343886,
      "num_tokens": 120010747.0,
      "step": 57830
    },
    {
      "entropy": 5.714471817016602,
      "epoch": 5.2462808417997095,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002606499671629523,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2329900771379471,
      "num_tokens": 120020931.0,
      "step": 57835
    },
    {
      "entropy": 5.672157335281372,
      "epoch": 5.246734397677794,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002606176767655634,
      "loss": 4.8778,
      "mean_token_accuracy": 0.2234736531972885,
      "num_tokens": 120031631.0,
      "step": 57840
    },
    {
      "entropy": 5.764474487304687,
      "epoch": 5.247187953555878,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002605853866656046,
      "loss": 4.8899,
      "mean_token_accuracy": 0.2293737053871155,
      "num_tokens": 120041861.0,
      "step": 57845
    },
    {
      "entropy": 5.669917392730713,
      "epoch": 5.247641509433962,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002605530968637436,
      "loss": 4.7963,
      "mean_token_accuracy": 0.22929820716381072,
      "num_tokens": 120052007.0,
      "step": 57850
    },
    {
      "entropy": 5.726616048812867,
      "epoch": 5.248095065312047,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002605208073606483,
      "loss": 4.8794,
      "mean_token_accuracy": 0.2289353847503662,
      "num_tokens": 120062001.0,
      "step": 57855
    },
    {
      "entropy": 5.651793384552002,
      "epoch": 5.248548621190131,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002604885181569864,
      "loss": 4.8208,
      "mean_token_accuracy": 0.22828615307807923,
      "num_tokens": 120072709.0,
      "step": 57860
    },
    {
      "entropy": 5.692503929138184,
      "epoch": 5.249002177068215,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002604562292534256,
      "loss": 4.8679,
      "mean_token_accuracy": 0.2290812075138092,
      "num_tokens": 120083659.0,
      "step": 57865
    },
    {
      "entropy": 5.72118911743164,
      "epoch": 5.249455732946299,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002604239406506338,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22968175262212753,
      "num_tokens": 120093872.0,
      "step": 57870
    },
    {
      "entropy": 5.658806657791137,
      "epoch": 5.249909288824383,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026039165234927855,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23143462240695953,
      "num_tokens": 120103533.0,
      "step": 57875
    },
    {
      "entropy": 5.726127815246582,
      "epoch": 5.250362844702467,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002603593643500276,
      "loss": 4.8584,
      "mean_token_accuracy": 0.23279028683900832,
      "num_tokens": 120113006.0,
      "step": 57880
    },
    {
      "entropy": 5.71428599357605,
      "epoch": 5.250816400580551,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002603270766535488,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2278389200568199,
      "num_tokens": 120124449.0,
      "step": 57885
    },
    {
      "entropy": 5.715381669998169,
      "epoch": 5.251269956458636,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026029478926050976,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23291876614093782,
      "num_tokens": 120135100.0,
      "step": 57890
    },
    {
      "entropy": 5.7716001033782955,
      "epoch": 5.25172351233672,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026026250217157823,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22084630280733109,
      "num_tokens": 120146276.0,
      "step": 57895
    },
    {
      "entropy": 5.718128490447998,
      "epoch": 5.252177068214804,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00026023021538742184,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2341815188527107,
      "num_tokens": 120157028.0,
      "step": 57900
    },
    {
      "entropy": 5.701374053955078,
      "epoch": 5.252630624092888,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026019792890870846,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2438812658190727,
      "num_tokens": 120167027.0,
      "step": 57905
    },
    {
      "entropy": 5.794543600082397,
      "epoch": 5.253084179970973,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026016564273610563,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21652671992778777,
      "num_tokens": 120178060.0,
      "step": 57910
    },
    {
      "entropy": 5.67389669418335,
      "epoch": 5.253537735849057,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026013335687028117,
      "loss": 4.8616,
      "mean_token_accuracy": 0.23087759017944337,
      "num_tokens": 120187869.0,
      "step": 57915
    },
    {
      "entropy": 5.726092529296875,
      "epoch": 5.253991291727141,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026010107131190256,
      "loss": 4.8723,
      "mean_token_accuracy": 0.23231000304222107,
      "num_tokens": 120197701.0,
      "step": 57920
    },
    {
      "entropy": 5.626591587066651,
      "epoch": 5.254444847605225,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002600687860616376,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2401018798351288,
      "num_tokens": 120208978.0,
      "step": 57925
    },
    {
      "entropy": 5.656494426727295,
      "epoch": 5.254898403483309,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002600365011201541,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2390727087855339,
      "num_tokens": 120219021.0,
      "step": 57930
    },
    {
      "entropy": 5.677732992172241,
      "epoch": 5.255351959361393,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002600042164881194,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2131475493311882,
      "num_tokens": 120229357.0,
      "step": 57935
    },
    {
      "entropy": 5.625663995742798,
      "epoch": 5.255805515239477,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002599719321662014,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23607161790132522,
      "num_tokens": 120239975.0,
      "step": 57940
    },
    {
      "entropy": 5.6886616230010985,
      "epoch": 5.256259071117562,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002599396481550677,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23333188146352768,
      "num_tokens": 120250764.0,
      "step": 57945
    },
    {
      "entropy": 5.7071644306182865,
      "epoch": 5.256712626995646,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002599073644553858,
      "loss": 4.8279,
      "mean_token_accuracy": 0.22893238812685013,
      "num_tokens": 120260503.0,
      "step": 57950
    },
    {
      "entropy": 5.706077527999878,
      "epoch": 5.25716618287373,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002598750810678236,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23065200746059417,
      "num_tokens": 120270604.0,
      "step": 57955
    },
    {
      "entropy": 5.77266492843628,
      "epoch": 5.257619738751814,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002598427979930485,
      "loss": 4.8886,
      "mean_token_accuracy": 0.22615841329097747,
      "num_tokens": 120281601.0,
      "step": 57960
    },
    {
      "entropy": 5.728224420547486,
      "epoch": 5.258073294629899,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025981051523172823,
      "loss": 4.9188,
      "mean_token_accuracy": 0.22643936425447464,
      "num_tokens": 120292431.0,
      "step": 57965
    },
    {
      "entropy": 5.771369075775146,
      "epoch": 5.258526850507983,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002597782327845303,
      "loss": 4.916,
      "mean_token_accuracy": 0.23222908675670623,
      "num_tokens": 120303063.0,
      "step": 57970
    },
    {
      "entropy": 5.708464860916138,
      "epoch": 5.258980406386067,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002597459506521225,
      "loss": 4.892,
      "mean_token_accuracy": 0.21922605633735656,
      "num_tokens": 120313543.0,
      "step": 57975
    },
    {
      "entropy": 5.70140905380249,
      "epoch": 5.259433962264151,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025971366883517226,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2239258959889412,
      "num_tokens": 120322409.0,
      "step": 57980
    },
    {
      "entropy": 5.67016453742981,
      "epoch": 5.259887518142235,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002596813873343472,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23924376368522643,
      "num_tokens": 120332247.0,
      "step": 57985
    },
    {
      "entropy": 5.638071203231812,
      "epoch": 5.260341074020319,
      "grad_norm": 1.40625,
      "learning_rate": 0.000259649106150315,
      "loss": 4.8098,
      "mean_token_accuracy": 0.22817756682634355,
      "num_tokens": 120342224.0,
      "step": 57990
    },
    {
      "entropy": 5.7861591339111325,
      "epoch": 5.260794629898403,
      "grad_norm": 1.375,
      "learning_rate": 0.00025961682528374326,
      "loss": 4.9209,
      "mean_token_accuracy": 0.22359812259674072,
      "num_tokens": 120352093.0,
      "step": 57995
    },
    {
      "entropy": 5.779382514953613,
      "epoch": 5.261248185776488,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025958454473529944,
      "loss": 4.9669,
      "mean_token_accuracy": 0.21743486374616622,
      "num_tokens": 120363561.0,
      "step": 58000
    },
    {
      "entropy": 5.698659896850586,
      "epoch": 5.261701741654572,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025955226450565113,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22856882363557815,
      "num_tokens": 120374119.0,
      "step": 58005
    },
    {
      "entropy": 5.7975586414337155,
      "epoch": 5.262155297532656,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025951998459546597,
      "loss": 4.9888,
      "mean_token_accuracy": 0.21083358228206633,
      "num_tokens": 120384365.0,
      "step": 58010
    },
    {
      "entropy": 5.692000865936279,
      "epoch": 5.2626088534107405,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002594877050054115,
      "loss": 4.8289,
      "mean_token_accuracy": 0.2362939402461052,
      "num_tokens": 120394341.0,
      "step": 58015
    },
    {
      "entropy": 5.64760046005249,
      "epoch": 5.263062409288825,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025945542573615517,
      "loss": 4.7608,
      "mean_token_accuracy": 0.23408832401037216,
      "num_tokens": 120404995.0,
      "step": 58020
    },
    {
      "entropy": 5.696365880966186,
      "epoch": 5.263515965166908,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025942314678836467,
      "loss": 4.9113,
      "mean_token_accuracy": 0.22384830862283706,
      "num_tokens": 120416443.0,
      "step": 58025
    },
    {
      "entropy": 5.681481647491455,
      "epoch": 5.263969521044992,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002593908681627074,
      "loss": 4.8209,
      "mean_token_accuracy": 0.2345194935798645,
      "num_tokens": 120425915.0,
      "step": 58030
    },
    {
      "entropy": 5.782537603378296,
      "epoch": 5.264423076923077,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025935858985985097,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21704009771347046,
      "num_tokens": 120436885.0,
      "step": 58035
    },
    {
      "entropy": 5.664117908477783,
      "epoch": 5.264876632801161,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002593263118804629,
      "loss": 4.8063,
      "mean_token_accuracy": 0.23415485322475432,
      "num_tokens": 120447234.0,
      "step": 58040
    },
    {
      "entropy": 5.77064733505249,
      "epoch": 5.265330188679245,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002592940342252107,
      "loss": 4.8849,
      "mean_token_accuracy": 0.23179262429475783,
      "num_tokens": 120457385.0,
      "step": 58045
    },
    {
      "entropy": 5.703531217575073,
      "epoch": 5.2657837445573294,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025926175689476187,
      "loss": 4.7745,
      "mean_token_accuracy": 0.23364087641239167,
      "num_tokens": 120468495.0,
      "step": 58050
    },
    {
      "entropy": 5.700663661956787,
      "epoch": 5.266237300435414,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002592294798897839,
      "loss": 4.8089,
      "mean_token_accuracy": 0.23961519002914428,
      "num_tokens": 120478536.0,
      "step": 58055
    },
    {
      "entropy": 5.733733224868774,
      "epoch": 5.266690856313498,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025919720321094435,
      "loss": 4.9097,
      "mean_token_accuracy": 0.23143259286880494,
      "num_tokens": 120489872.0,
      "step": 58060
    },
    {
      "entropy": 5.680000495910645,
      "epoch": 5.267144412191582,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002591649268589106,
      "loss": 4.8034,
      "mean_token_accuracy": 0.2294098973274231,
      "num_tokens": 120499454.0,
      "step": 58065
    },
    {
      "entropy": 5.699082946777343,
      "epoch": 5.2675979680696665,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025913265083435025,
      "loss": 4.8622,
      "mean_token_accuracy": 0.2297005832195282,
      "num_tokens": 120509974.0,
      "step": 58070
    },
    {
      "entropy": 5.6640777587890625,
      "epoch": 5.268051523947751,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002591003751379307,
      "loss": 4.8207,
      "mean_token_accuracy": 0.2291363686323166,
      "num_tokens": 120520310.0,
      "step": 58075
    },
    {
      "entropy": 5.723127555847168,
      "epoch": 5.268505079825834,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002590680997703195,
      "loss": 4.8247,
      "mean_token_accuracy": 0.22809132784605027,
      "num_tokens": 120530662.0,
      "step": 58080
    },
    {
      "entropy": 5.748260164260865,
      "epoch": 5.268958635703918,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025903582473218404,
      "loss": 4.9149,
      "mean_token_accuracy": 0.22420405745506286,
      "num_tokens": 120541138.0,
      "step": 58085
    },
    {
      "entropy": 5.641852331161499,
      "epoch": 5.269412191582003,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025900355002419167,
      "loss": 4.7596,
      "mean_token_accuracy": 0.23728813827037812,
      "num_tokens": 120551758.0,
      "step": 58090
    },
    {
      "entropy": 5.682773160934448,
      "epoch": 5.269865747460087,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002589712756470101,
      "loss": 4.8288,
      "mean_token_accuracy": 0.22943220883607865,
      "num_tokens": 120562438.0,
      "step": 58095
    },
    {
      "entropy": 5.692621278762817,
      "epoch": 5.270319303338171,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002589390016013066,
      "loss": 4.8484,
      "mean_token_accuracy": 0.23640407174825667,
      "num_tokens": 120572032.0,
      "step": 58100
    },
    {
      "entropy": 5.751881837844849,
      "epoch": 5.2707728592162555,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002589067278877486,
      "loss": 4.8815,
      "mean_token_accuracy": 0.22771409898996353,
      "num_tokens": 120582415.0,
      "step": 58105
    },
    {
      "entropy": 5.733468151092529,
      "epoch": 5.27122641509434,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025887445450700357,
      "loss": 4.8595,
      "mean_token_accuracy": 0.23554737269878387,
      "num_tokens": 120593169.0,
      "step": 58110
    },
    {
      "entropy": 5.746434688568115,
      "epoch": 5.271679970972424,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000258842181459739,
      "loss": 4.7982,
      "mean_token_accuracy": 0.23303859829902648,
      "num_tokens": 120604472.0,
      "step": 58115
    },
    {
      "entropy": 5.716034841537476,
      "epoch": 5.272133526850508,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002588099087466222,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22715264111757277,
      "num_tokens": 120614286.0,
      "step": 58120
    },
    {
      "entropy": 5.688490676879883,
      "epoch": 5.2725870827285926,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002587776363683206,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23323942422866822,
      "num_tokens": 120624130.0,
      "step": 58125
    },
    {
      "entropy": 5.6925159931182865,
      "epoch": 5.273040638606676,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025874536432550163,
      "loss": 4.8944,
      "mean_token_accuracy": 0.22965086102485657,
      "num_tokens": 120634014.0,
      "step": 58130
    },
    {
      "entropy": 5.713029527664185,
      "epoch": 5.27349419448476,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002587130926188327,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2324983671307564,
      "num_tokens": 120645370.0,
      "step": 58135
    },
    {
      "entropy": 5.726689338684082,
      "epoch": 5.2739477503628445,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002586808212489811,
      "loss": 4.916,
      "mean_token_accuracy": 0.21882657408714296,
      "num_tokens": 120655967.0,
      "step": 58140
    },
    {
      "entropy": 5.706140041351318,
      "epoch": 5.274401306240929,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025864855021661434,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22847249507904052,
      "num_tokens": 120666223.0,
      "step": 58145
    },
    {
      "entropy": 5.646533679962158,
      "epoch": 5.274854862119013,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002586162795223996,
      "loss": 4.7639,
      "mean_token_accuracy": 0.2343052551150322,
      "num_tokens": 120676859.0,
      "step": 58150
    },
    {
      "entropy": 5.776551342010498,
      "epoch": 5.275308417997097,
      "grad_norm": 1.375,
      "learning_rate": 0.0002585840091670045,
      "loss": 4.9414,
      "mean_token_accuracy": 0.22696186751127242,
      "num_tokens": 120687930.0,
      "step": 58155
    },
    {
      "entropy": 5.684225130081177,
      "epoch": 5.2757619738751815,
      "grad_norm": 1.375,
      "learning_rate": 0.00025855173915109627,
      "loss": 4.814,
      "mean_token_accuracy": 0.22890155017375946,
      "num_tokens": 120698121.0,
      "step": 58160
    },
    {
      "entropy": 5.666386842727661,
      "epoch": 5.276215529753266,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025851946947534214,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22782896757125853,
      "num_tokens": 120708428.0,
      "step": 58165
    },
    {
      "entropy": 5.744906854629517,
      "epoch": 5.27666908563135,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025848720014040973,
      "loss": 4.9374,
      "mean_token_accuracy": 0.22855545729398727,
      "num_tokens": 120719726.0,
      "step": 58170
    },
    {
      "entropy": 5.707107210159302,
      "epoch": 5.277122641509434,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025845493114696615,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22698979526758195,
      "num_tokens": 120730698.0,
      "step": 58175
    },
    {
      "entropy": 5.750349235534668,
      "epoch": 5.277576197387518,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002584226624956789,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22382999807596207,
      "num_tokens": 120741122.0,
      "step": 58180
    },
    {
      "entropy": 5.733747243881226,
      "epoch": 5.278029753265602,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002583903941872151,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22846073508262635,
      "num_tokens": 120751623.0,
      "step": 58185
    },
    {
      "entropy": 5.699610853195191,
      "epoch": 5.278483309143686,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002583581262222422,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23093574941158296,
      "num_tokens": 120762203.0,
      "step": 58190
    },
    {
      "entropy": 5.663803052902222,
      "epoch": 5.2789368650217705,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025832585860142754,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22896911799907685,
      "num_tokens": 120772732.0,
      "step": 58195
    },
    {
      "entropy": 5.72712550163269,
      "epoch": 5.279390420899855,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002582935913254383,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23252967596054078,
      "num_tokens": 120782289.0,
      "step": 58200
    },
    {
      "entropy": 5.719165563583374,
      "epoch": 5.279843976777939,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002582613243949419,
      "loss": 4.8931,
      "mean_token_accuracy": 0.22640366107225418,
      "num_tokens": 120793730.0,
      "step": 58205
    },
    {
      "entropy": 5.725672054290771,
      "epoch": 5.280297532656023,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025822905781060565,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22655629962682725,
      "num_tokens": 120804322.0,
      "step": 58210
    },
    {
      "entropy": 5.65642409324646,
      "epoch": 5.280751088534108,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025819679157309665,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22809166163206102,
      "num_tokens": 120815095.0,
      "step": 58215
    },
    {
      "entropy": 5.705009174346924,
      "epoch": 5.281204644412192,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002581645256830823,
      "loss": 4.822,
      "mean_token_accuracy": 0.23072720021009446,
      "num_tokens": 120825941.0,
      "step": 58220
    },
    {
      "entropy": 5.682704830169678,
      "epoch": 5.281658200290276,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025813226014122985,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2291450545191765,
      "num_tokens": 120836512.0,
      "step": 58225
    },
    {
      "entropy": 5.779718637466431,
      "epoch": 5.28211175616836,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002580999949482066,
      "loss": 4.9047,
      "mean_token_accuracy": 0.23327054977416992,
      "num_tokens": 120846917.0,
      "step": 58230
    },
    {
      "entropy": 5.678488492965698,
      "epoch": 5.282565312046444,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002580677301046797,
      "loss": 4.8092,
      "mean_token_accuracy": 0.23986825346946716,
      "num_tokens": 120857084.0,
      "step": 58235
    },
    {
      "entropy": 5.688479900360107,
      "epoch": 5.283018867924528,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002580354656113165,
      "loss": 4.8303,
      "mean_token_accuracy": 0.23013900667428971,
      "num_tokens": 120867496.0,
      "step": 58240
    },
    {
      "entropy": 5.670460844039917,
      "epoch": 5.283472423802612,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002580032014687842,
      "loss": 4.822,
      "mean_token_accuracy": 0.23160935193300247,
      "num_tokens": 120877722.0,
      "step": 58245
    },
    {
      "entropy": 5.71226921081543,
      "epoch": 5.2839259796806965,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025797093767774995,
      "loss": 4.8894,
      "mean_token_accuracy": 0.230303817987442,
      "num_tokens": 120888225.0,
      "step": 58250
    },
    {
      "entropy": 5.770388555526734,
      "epoch": 5.284379535558781,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002579386742388811,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2243671089410782,
      "num_tokens": 120898664.0,
      "step": 58255
    },
    {
      "entropy": 5.683478546142578,
      "epoch": 5.284833091436865,
      "grad_norm": 1.375,
      "learning_rate": 0.00025790641115284486,
      "loss": 4.7729,
      "mean_token_accuracy": 0.2348209425806999,
      "num_tokens": 120909084.0,
      "step": 58260
    },
    {
      "entropy": 5.675636720657349,
      "epoch": 5.285286647314949,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002578741484203084,
      "loss": 4.805,
      "mean_token_accuracy": 0.2306385174393654,
      "num_tokens": 120919552.0,
      "step": 58265
    },
    {
      "entropy": 5.665475130081177,
      "epoch": 5.285740203193034,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002578418860419388,
      "loss": 4.8347,
      "mean_token_accuracy": 0.22804423570632934,
      "num_tokens": 120929918.0,
      "step": 58270
    },
    {
      "entropy": 5.634483385086059,
      "epoch": 5.286193759071118,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002578096240184035,
      "loss": 4.7846,
      "mean_token_accuracy": 0.2369912087917328,
      "num_tokens": 120940146.0,
      "step": 58275
    },
    {
      "entropy": 5.7366370677948,
      "epoch": 5.286647314949202,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002577773623503695,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2295157179236412,
      "num_tokens": 120950946.0,
      "step": 58280
    },
    {
      "entropy": 5.717037868499756,
      "epoch": 5.2871008708272855,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000257745101038504,
      "loss": 4.88,
      "mean_token_accuracy": 0.231857006251812,
      "num_tokens": 120961623.0,
      "step": 58285
    },
    {
      "entropy": 5.679444694519043,
      "epoch": 5.28755442670537,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002577128400834742,
      "loss": 4.7869,
      "mean_token_accuracy": 0.23429704010486602,
      "num_tokens": 120971798.0,
      "step": 58290
    },
    {
      "entropy": 5.725931167602539,
      "epoch": 5.288007982583454,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025768057948594733,
      "loss": 4.9271,
      "mean_token_accuracy": 0.22479304820299148,
      "num_tokens": 120982042.0,
      "step": 58295
    },
    {
      "entropy": 5.675510120391846,
      "epoch": 5.288461538461538,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002576483192465905,
      "loss": 4.8466,
      "mean_token_accuracy": 0.22793004512786866,
      "num_tokens": 120991650.0,
      "step": 58300
    },
    {
      "entropy": 5.752511405944825,
      "epoch": 5.288915094339623,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025761605936607077,
      "loss": 4.8542,
      "mean_token_accuracy": 0.23338786512613297,
      "num_tokens": 121001660.0,
      "step": 58305
    },
    {
      "entropy": 5.709050846099854,
      "epoch": 5.289368650217707,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002575837998450554,
      "loss": 4.8398,
      "mean_token_accuracy": 0.22893956750631334,
      "num_tokens": 121011065.0,
      "step": 58310
    },
    {
      "entropy": 5.590397882461548,
      "epoch": 5.289822206095791,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025755154068421156,
      "loss": 4.728,
      "mean_token_accuracy": 0.23665134906768798,
      "num_tokens": 121020520.0,
      "step": 58315
    },
    {
      "entropy": 5.653658771514893,
      "epoch": 5.290275761973875,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002575192818842062,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22057317793369294,
      "num_tokens": 121030525.0,
      "step": 58320
    },
    {
      "entropy": 5.655670309066773,
      "epoch": 5.29072931785196,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025748702344570657,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2331253930926323,
      "num_tokens": 121041285.0,
      "step": 58325
    },
    {
      "entropy": 5.723713827133179,
      "epoch": 5.291182873730044,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002574547653693798,
      "loss": 4.8649,
      "mean_token_accuracy": 0.22301072627305984,
      "num_tokens": 121050787.0,
      "step": 58330
    },
    {
      "entropy": 5.76524977684021,
      "epoch": 5.291636429608127,
      "grad_norm": 1.375,
      "learning_rate": 0.0002574225076558929,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21905867010354996,
      "num_tokens": 121061088.0,
      "step": 58335
    },
    {
      "entropy": 5.7390539169311525,
      "epoch": 5.292089985486212,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025739025030591307,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2165284350514412,
      "num_tokens": 121071111.0,
      "step": 58340
    },
    {
      "entropy": 5.810155057907105,
      "epoch": 5.292543541364296,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002573579933201073,
      "loss": 4.908,
      "mean_token_accuracy": 0.2227243348956108,
      "num_tokens": 121081339.0,
      "step": 58345
    },
    {
      "entropy": 5.634974193572998,
      "epoch": 5.29299709724238,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025732573669914286,
      "loss": 4.828,
      "mean_token_accuracy": 0.23103911578655242,
      "num_tokens": 121091830.0,
      "step": 58350
    },
    {
      "entropy": 5.568218803405761,
      "epoch": 5.293450653120464,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002572934804436865,
      "loss": 4.7586,
      "mean_token_accuracy": 0.23559923768043517,
      "num_tokens": 121102727.0,
      "step": 58355
    },
    {
      "entropy": 5.712048530578613,
      "epoch": 5.293904208998549,
      "grad_norm": 1.375,
      "learning_rate": 0.0002572612245544056,
      "loss": 4.9177,
      "mean_token_accuracy": 0.2209714263677597,
      "num_tokens": 121112845.0,
      "step": 58360
    },
    {
      "entropy": 5.663669157028198,
      "epoch": 5.294357764876633,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025722896903196705,
      "loss": 4.7393,
      "mean_token_accuracy": 0.23374312669038771,
      "num_tokens": 121123456.0,
      "step": 58365
    },
    {
      "entropy": 5.661744546890259,
      "epoch": 5.294811320754717,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002571967138770379,
      "loss": 4.7318,
      "mean_token_accuracy": 0.23469474017620087,
      "num_tokens": 121133185.0,
      "step": 58370
    },
    {
      "entropy": 5.639215135574341,
      "epoch": 5.295264876632801,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025716445909028537,
      "loss": 4.8186,
      "mean_token_accuracy": 0.23011112958192825,
      "num_tokens": 121143962.0,
      "step": 58375
    },
    {
      "entropy": 5.681608438491821,
      "epoch": 5.295718432510886,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002571322046723763,
      "loss": 4.8047,
      "mean_token_accuracy": 0.23593159765005112,
      "num_tokens": 121153184.0,
      "step": 58380
    },
    {
      "entropy": 5.667664813995361,
      "epoch": 5.29617198838897,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002570999506239778,
      "loss": 4.8617,
      "mean_token_accuracy": 0.22361389845609664,
      "num_tokens": 121165189.0,
      "step": 58385
    },
    {
      "entropy": 5.733178901672363,
      "epoch": 5.296625544267053,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025706769694575684,
      "loss": 4.8604,
      "mean_token_accuracy": 0.23134478479623793,
      "num_tokens": 121174787.0,
      "step": 58390
    },
    {
      "entropy": 5.699770402908325,
      "epoch": 5.297079100145138,
      "grad_norm": 1.375,
      "learning_rate": 0.0002570354436383805,
      "loss": 4.9491,
      "mean_token_accuracy": 0.2221537411212921,
      "num_tokens": 121185334.0,
      "step": 58395
    },
    {
      "entropy": 5.842860746383667,
      "epoch": 5.297532656023222,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002570031907025158,
      "loss": 4.9744,
      "mean_token_accuracy": 0.217634217441082,
      "num_tokens": 121195928.0,
      "step": 58400
    },
    {
      "entropy": 5.786369132995605,
      "epoch": 5.297986211901306,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025697093813882967,
      "loss": 4.9868,
      "mean_token_accuracy": 0.21811927407979964,
      "num_tokens": 121206369.0,
      "step": 58405
    },
    {
      "entropy": 5.739391136169433,
      "epoch": 5.29843976777939,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002569386859479891,
      "loss": 4.819,
      "mean_token_accuracy": 0.23459534645080565,
      "num_tokens": 121216573.0,
      "step": 58410
    },
    {
      "entropy": 5.654455518722534,
      "epoch": 5.298893323657475,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002569064341306612,
      "loss": 4.8085,
      "mean_token_accuracy": 0.2326110616326332,
      "num_tokens": 121226330.0,
      "step": 58415
    },
    {
      "entropy": 5.668958473205566,
      "epoch": 5.299346879535559,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025687418268751283,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23033961653709412,
      "num_tokens": 121236171.0,
      "step": 58420
    },
    {
      "entropy": 5.669463443756103,
      "epoch": 5.299800435413643,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025684193161921087,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22771447896957397,
      "num_tokens": 121246506.0,
      "step": 58425
    },
    {
      "entropy": 5.7308556079864506,
      "epoch": 5.3002539912917275,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025680968092642255,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22107288092374802,
      "num_tokens": 121257080.0,
      "step": 58430
    },
    {
      "entropy": 5.744137287139893,
      "epoch": 5.300707547169811,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025677743060981454,
      "loss": 4.9019,
      "mean_token_accuracy": 0.23493988960981368,
      "num_tokens": 121266349.0,
      "step": 58435
    },
    {
      "entropy": 5.659279680252075,
      "epoch": 5.301161103047895,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002567451806700539,
      "loss": 4.8344,
      "mean_token_accuracy": 0.2288001522421837,
      "num_tokens": 121278410.0,
      "step": 58440
    },
    {
      "entropy": 5.70210371017456,
      "epoch": 5.301614658925979,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002567129311078077,
      "loss": 4.7972,
      "mean_token_accuracy": 0.22994851768016816,
      "num_tokens": 121289334.0,
      "step": 58445
    },
    {
      "entropy": 5.738106060028076,
      "epoch": 5.302068214804064,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025668068192374264,
      "loss": 4.9102,
      "mean_token_accuracy": 0.22240374833345414,
      "num_tokens": 121299132.0,
      "step": 58450
    },
    {
      "entropy": 5.621947288513184,
      "epoch": 5.302521770682148,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025664843311852574,
      "loss": 4.7723,
      "mean_token_accuracy": 0.23223734200000762,
      "num_tokens": 121308905.0,
      "step": 58455
    },
    {
      "entropy": 5.686212348937988,
      "epoch": 5.302975326560232,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000256616184692824,
      "loss": 4.8443,
      "mean_token_accuracy": 0.2330305188894272,
      "num_tokens": 121318719.0,
      "step": 58460
    },
    {
      "entropy": 5.756199502944947,
      "epoch": 5.3034288824383164,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002565839366473042,
      "loss": 4.9435,
      "mean_token_accuracy": 0.2236429527401924,
      "num_tokens": 121328501.0,
      "step": 58465
    },
    {
      "entropy": 5.749545001983643,
      "epoch": 5.303882438316401,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025655168898263336,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2218639925122261,
      "num_tokens": 121339050.0,
      "step": 58470
    },
    {
      "entropy": 5.725380373001099,
      "epoch": 5.304335994194485,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002565194416994782,
      "loss": 4.8,
      "mean_token_accuracy": 0.23713091015815735,
      "num_tokens": 121347699.0,
      "step": 58475
    },
    {
      "entropy": 5.688079929351806,
      "epoch": 5.304789550072569,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002564871947985057,
      "loss": 4.821,
      "mean_token_accuracy": 0.2366761788725853,
      "num_tokens": 121357623.0,
      "step": 58480
    },
    {
      "entropy": 5.618593502044678,
      "epoch": 5.3052431059506535,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002564549482803828,
      "loss": 4.8098,
      "mean_token_accuracy": 0.23175611346960068,
      "num_tokens": 121367933.0,
      "step": 58485
    },
    {
      "entropy": 5.688368797302246,
      "epoch": 5.305696661828737,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025642270214577617,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22364383488893508,
      "num_tokens": 121377977.0,
      "step": 58490
    },
    {
      "entropy": 5.790189743041992,
      "epoch": 5.306150217706821,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025639045639535283,
      "loss": 4.9998,
      "mean_token_accuracy": 0.22102806270122527,
      "num_tokens": 121387866.0,
      "step": 58495
    },
    {
      "entropy": 5.701975917816162,
      "epoch": 5.306603773584905,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002563582110297797,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22630925476551056,
      "num_tokens": 121397659.0,
      "step": 58500
    },
    {
      "entropy": 5.707031106948852,
      "epoch": 5.30705732946299,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025632596604972346,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2371336415410042,
      "num_tokens": 121407907.0,
      "step": 58505
    },
    {
      "entropy": 5.632646703720093,
      "epoch": 5.307510885341074,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025629372145585095,
      "loss": 4.7363,
      "mean_token_accuracy": 0.24751432836055756,
      "num_tokens": 121417882.0,
      "step": 58510
    },
    {
      "entropy": 5.67032380104065,
      "epoch": 5.307964441219158,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002562614772488291,
      "loss": 4.8515,
      "mean_token_accuracy": 0.22700649350881577,
      "num_tokens": 121427522.0,
      "step": 58515
    },
    {
      "entropy": 5.703869771957398,
      "epoch": 5.3084179970972425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002562292334293247,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2335926979780197,
      "num_tokens": 121437796.0,
      "step": 58520
    },
    {
      "entropy": 5.692039108276367,
      "epoch": 5.308871552975327,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025619698999800446,
      "loss": 4.8319,
      "mean_token_accuracy": 0.23395728468894958,
      "num_tokens": 121448096.0,
      "step": 58525
    },
    {
      "entropy": 5.683729362487793,
      "epoch": 5.309325108853411,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025616474695553526,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23212100565433502,
      "num_tokens": 121458017.0,
      "step": 58530
    },
    {
      "entropy": 5.735240268707275,
      "epoch": 5.309778664731495,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025613250430258397,
      "loss": 4.9345,
      "mean_token_accuracy": 0.22094997465610505,
      "num_tokens": 121468662.0,
      "step": 58535
    },
    {
      "entropy": 5.721352434158325,
      "epoch": 5.31023222060958,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025610026203981724,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2288174107670784,
      "num_tokens": 121478988.0,
      "step": 58540
    },
    {
      "entropy": 5.681505966186523,
      "epoch": 5.310685776487663,
      "grad_norm": 1.5625,
      "learning_rate": 0.00025606802016790194,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23049531280994415,
      "num_tokens": 121487737.0,
      "step": 58545
    },
    {
      "entropy": 5.723097467422486,
      "epoch": 5.311139332365747,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002560357786875048,
      "loss": 4.9033,
      "mean_token_accuracy": 0.2316579431295395,
      "num_tokens": 121498670.0,
      "step": 58550
    },
    {
      "entropy": 5.761539077758789,
      "epoch": 5.3115928882438315,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002560035375992926,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22955431491136552,
      "num_tokens": 121508294.0,
      "step": 58555
    },
    {
      "entropy": 5.728103733062744,
      "epoch": 5.312046444121916,
      "grad_norm": 1.5,
      "learning_rate": 0.000255971296903932,
      "loss": 4.8811,
      "mean_token_accuracy": 0.22998746186494828,
      "num_tokens": 121518046.0,
      "step": 58560
    },
    {
      "entropy": 5.666566181182861,
      "epoch": 5.3125,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025593905660209004,
      "loss": 4.8103,
      "mean_token_accuracy": 0.23272323459386826,
      "num_tokens": 121527779.0,
      "step": 58565
    },
    {
      "entropy": 5.69438796043396,
      "epoch": 5.312953555878084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025590681669443303,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22071243077516556,
      "num_tokens": 121538365.0,
      "step": 58570
    },
    {
      "entropy": 5.73459267616272,
      "epoch": 5.3134071117561685,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000255874577181628,
      "loss": 4.8579,
      "mean_token_accuracy": 0.2245251476764679,
      "num_tokens": 121548668.0,
      "step": 58575
    },
    {
      "entropy": 5.724018812179565,
      "epoch": 5.313860667634253,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025584233806434165,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22095527946949006,
      "num_tokens": 121559276.0,
      "step": 58580
    },
    {
      "entropy": 5.672467422485352,
      "epoch": 5.314314223512337,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002558100993432405,
      "loss": 4.8298,
      "mean_token_accuracy": 0.2394853338599205,
      "num_tokens": 121569686.0,
      "step": 58585
    },
    {
      "entropy": 5.72370662689209,
      "epoch": 5.31476777939042,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002557778610189914,
      "loss": 4.8552,
      "mean_token_accuracy": 0.23641971349716187,
      "num_tokens": 121580443.0,
      "step": 58590
    },
    {
      "entropy": 5.714636278152466,
      "epoch": 5.315221335268505,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025574562309226115,
      "loss": 4.8766,
      "mean_token_accuracy": 0.22632632553577423,
      "num_tokens": 121592393.0,
      "step": 58595
    },
    {
      "entropy": 5.720839643478394,
      "epoch": 5.315674891146589,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002557133855637163,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22757550179958344,
      "num_tokens": 121603394.0,
      "step": 58600
    },
    {
      "entropy": 5.693944501876831,
      "epoch": 5.316128447024673,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002556811484340234,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2241050899028778,
      "num_tokens": 121613719.0,
      "step": 58605
    },
    {
      "entropy": 5.696455001831055,
      "epoch": 5.3165820029027575,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002556489117038493,
      "loss": 4.8277,
      "mean_token_accuracy": 0.2334457203745842,
      "num_tokens": 121623897.0,
      "step": 58610
    },
    {
      "entropy": 5.708748865127563,
      "epoch": 5.317035558780842,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002556166753738608,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2386556550860405,
      "num_tokens": 121633916.0,
      "step": 58615
    },
    {
      "entropy": 5.769975090026856,
      "epoch": 5.317489114658926,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025558443944472426,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22773462384939194,
      "num_tokens": 121643828.0,
      "step": 58620
    },
    {
      "entropy": 5.644172334671021,
      "epoch": 5.31794267053701,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002555522039171065,
      "loss": 4.789,
      "mean_token_accuracy": 0.2326523870229721,
      "num_tokens": 121652956.0,
      "step": 58625
    },
    {
      "entropy": 5.659831523895264,
      "epoch": 5.318396226415095,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025551996879167417,
      "loss": 4.8138,
      "mean_token_accuracy": 0.2303492546081543,
      "num_tokens": 121663199.0,
      "step": 58630
    },
    {
      "entropy": 5.644365930557251,
      "epoch": 5.318849782293179,
      "grad_norm": 1.6875,
      "learning_rate": 0.00025548773406909375,
      "loss": 4.7855,
      "mean_token_accuracy": 0.24135144352912902,
      "num_tokens": 121672731.0,
      "step": 58635
    },
    {
      "entropy": 5.752416753768921,
      "epoch": 5.319303338171263,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000255455499750032,
      "loss": 4.962,
      "mean_token_accuracy": 0.22224832326173782,
      "num_tokens": 121683307.0,
      "step": 58640
    },
    {
      "entropy": 5.735068273544312,
      "epoch": 5.3197568940493465,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025542326583515555,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22567862570285796,
      "num_tokens": 121693944.0,
      "step": 58645
    },
    {
      "entropy": 5.753876829147339,
      "epoch": 5.320210449927431,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002553910323251309,
      "loss": 4.8802,
      "mean_token_accuracy": 0.23487583100795745,
      "num_tokens": 121704930.0,
      "step": 58650
    },
    {
      "entropy": 5.744183921813965,
      "epoch": 5.320664005805515,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025535879922062465,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22093722820281983,
      "num_tokens": 121715924.0,
      "step": 58655
    },
    {
      "entropy": 5.6983246326446535,
      "epoch": 5.321117561683599,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002553265665223035,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2326543539762497,
      "num_tokens": 121726658.0,
      "step": 58660
    },
    {
      "entropy": 5.778060054779052,
      "epoch": 5.3215711175616836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025529433423083397,
      "loss": 4.9677,
      "mean_token_accuracy": 0.2212790697813034,
      "num_tokens": 121737019.0,
      "step": 58665
    },
    {
      "entropy": 5.7823652744293215,
      "epoch": 5.322024673439768,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002552621023468826,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22882489264011383,
      "num_tokens": 121748013.0,
      "step": 58670
    },
    {
      "entropy": 5.682180261611938,
      "epoch": 5.322478229317852,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025522987087111604,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23120439201593398,
      "num_tokens": 121758873.0,
      "step": 58675
    },
    {
      "entropy": 5.661458587646484,
      "epoch": 5.322931785195936,
      "grad_norm": 1.375,
      "learning_rate": 0.00025519763980420075,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2406239926815033,
      "num_tokens": 121769565.0,
      "step": 58680
    },
    {
      "entropy": 5.673756265640259,
      "epoch": 5.323385341074021,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025516540914680337,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22625633478164672,
      "num_tokens": 121780259.0,
      "step": 58685
    },
    {
      "entropy": 5.621939420700073,
      "epoch": 5.323838896952105,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025513317889959024,
      "loss": 4.8205,
      "mean_token_accuracy": 0.2360360398888588,
      "num_tokens": 121791608.0,
      "step": 58690
    },
    {
      "entropy": 5.637438583374023,
      "epoch": 5.324292452830189,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025510094906322815,
      "loss": 4.7973,
      "mean_token_accuracy": 0.2294296905398369,
      "num_tokens": 121800954.0,
      "step": 58695
    },
    {
      "entropy": 5.653806209564209,
      "epoch": 5.3247460087082725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002550687196383835,
      "loss": 4.7703,
      "mean_token_accuracy": 0.2314005747437477,
      "num_tokens": 121809780.0,
      "step": 58700
    },
    {
      "entropy": 5.6775308609008786,
      "epoch": 5.325199564586357,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025503649062572276,
      "loss": 4.8466,
      "mean_token_accuracy": 0.22455041706562043,
      "num_tokens": 121819941.0,
      "step": 58705
    },
    {
      "entropy": 5.703165721893311,
      "epoch": 5.325653120464441,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002550042620259125,
      "loss": 4.9491,
      "mean_token_accuracy": 0.21488214284181595,
      "num_tokens": 121831176.0,
      "step": 58710
    },
    {
      "entropy": 5.726931047439575,
      "epoch": 5.326106676342525,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025497203383961927,
      "loss": 4.8499,
      "mean_token_accuracy": 0.2284851536154747,
      "num_tokens": 121840832.0,
      "step": 58715
    },
    {
      "entropy": 5.7004083633422855,
      "epoch": 5.32656023222061,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002549398060675094,
      "loss": 4.9145,
      "mean_token_accuracy": 0.21840122044086457,
      "num_tokens": 121851976.0,
      "step": 58720
    },
    {
      "entropy": 5.642629623413086,
      "epoch": 5.327013788098694,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025490757871024944,
      "loss": 4.7781,
      "mean_token_accuracy": 0.22988748997449876,
      "num_tokens": 121862359.0,
      "step": 58725
    },
    {
      "entropy": 5.68431248664856,
      "epoch": 5.327467343976778,
      "grad_norm": 1.375,
      "learning_rate": 0.00025487535176850593,
      "loss": 4.7464,
      "mean_token_accuracy": 0.23772624880075455,
      "num_tokens": 121872306.0,
      "step": 58730
    },
    {
      "entropy": 5.726309394836425,
      "epoch": 5.327920899854862,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025484312524294523,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2257469415664673,
      "num_tokens": 121881711.0,
      "step": 58735
    },
    {
      "entropy": 5.652321815490723,
      "epoch": 5.328374455732947,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002548108991342338,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2295801192522049,
      "num_tokens": 121892781.0,
      "step": 58740
    },
    {
      "entropy": 5.696190547943115,
      "epoch": 5.32882801161103,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002547786734430382,
      "loss": 4.8751,
      "mean_token_accuracy": 0.23247999399900438,
      "num_tokens": 121902329.0,
      "step": 58745
    },
    {
      "entropy": 5.628053617477417,
      "epoch": 5.329281567489114,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002547464481700247,
      "loss": 4.7581,
      "mean_token_accuracy": 0.22987321466207505,
      "num_tokens": 121913410.0,
      "step": 58750
    },
    {
      "entropy": 5.6179266452789305,
      "epoch": 5.329735123367199,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002547142233158599,
      "loss": 4.7713,
      "mean_token_accuracy": 0.23512645661830903,
      "num_tokens": 121923959.0,
      "step": 58755
    },
    {
      "entropy": 5.698053359985352,
      "epoch": 5.330188679245283,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025468199888121015,
      "loss": 4.8015,
      "mean_token_accuracy": 0.234381066262722,
      "num_tokens": 121934988.0,
      "step": 58760
    },
    {
      "entropy": 5.774980640411377,
      "epoch": 5.330642235123367,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002546497748667418,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2246932253241539,
      "num_tokens": 121944848.0,
      "step": 58765
    },
    {
      "entropy": 5.772602844238281,
      "epoch": 5.331095791001451,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025461755127312123,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22814727276563646,
      "num_tokens": 121956085.0,
      "step": 58770
    },
    {
      "entropy": 5.768517065048218,
      "epoch": 5.331549346879536,
      "grad_norm": 1.421875,
      "learning_rate": 0.000254585328101015,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21710187792778016,
      "num_tokens": 121967165.0,
      "step": 58775
    },
    {
      "entropy": 5.737988090515136,
      "epoch": 5.33200290275762,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002545531053510893,
      "loss": 4.9014,
      "mean_token_accuracy": 0.22862596809864044,
      "num_tokens": 121978951.0,
      "step": 58780
    },
    {
      "entropy": 5.717339944839478,
      "epoch": 5.332456458635704,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002545208830240107,
      "loss": 4.8561,
      "mean_token_accuracy": 0.22495779991149903,
      "num_tokens": 121989130.0,
      "step": 58785
    },
    {
      "entropy": 5.761972332000733,
      "epoch": 5.332910014513788,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025448866112044526,
      "loss": 4.8605,
      "mean_token_accuracy": 0.23115548640489578,
      "num_tokens": 121999097.0,
      "step": 58790
    },
    {
      "entropy": 5.795695638656616,
      "epoch": 5.333363570391873,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025445643964105974,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21960892230272294,
      "num_tokens": 122009385.0,
      "step": 58795
    },
    {
      "entropy": 5.685347986221314,
      "epoch": 5.333817126269956,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002544242185865202,
      "loss": 4.8753,
      "mean_token_accuracy": 0.23309698551893235,
      "num_tokens": 122020331.0,
      "step": 58800
    },
    {
      "entropy": 5.693503046035767,
      "epoch": 5.33427068214804,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002543919979574931,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23010261952877045,
      "num_tokens": 122030968.0,
      "step": 58805
    },
    {
      "entropy": 5.720926713943482,
      "epoch": 5.334724238026125,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025435977775464466,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23165854662656785,
      "num_tokens": 122040759.0,
      "step": 58810
    },
    {
      "entropy": 5.746543979644775,
      "epoch": 5.335177793904209,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002543275579786413,
      "loss": 4.92,
      "mean_token_accuracy": 0.22917474657297135,
      "num_tokens": 122050986.0,
      "step": 58815
    },
    {
      "entropy": 5.700494003295899,
      "epoch": 5.335631349782293,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025429533863014934,
      "loss": 4.8631,
      "mean_token_accuracy": 0.226198011636734,
      "num_tokens": 122060602.0,
      "step": 58820
    },
    {
      "entropy": 5.730286884307861,
      "epoch": 5.336084905660377,
      "grad_norm": 1.53125,
      "learning_rate": 0.000254263119709835,
      "loss": 4.9239,
      "mean_token_accuracy": 0.22663835883140565,
      "num_tokens": 122070863.0,
      "step": 58825
    },
    {
      "entropy": 5.691616106033325,
      "epoch": 5.336538461538462,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002542309012183647,
      "loss": 4.918,
      "mean_token_accuracy": 0.2269996777176857,
      "num_tokens": 122080756.0,
      "step": 58830
    },
    {
      "entropy": 5.71538987159729,
      "epoch": 5.336992017416546,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002541986831564046,
      "loss": 4.8805,
      "mean_token_accuracy": 0.22546358555555343,
      "num_tokens": 122090695.0,
      "step": 58835
    },
    {
      "entropy": 5.726605796813965,
      "epoch": 5.33744557329463,
      "grad_norm": 1.375,
      "learning_rate": 0.00025416646552462103,
      "loss": 4.7939,
      "mean_token_accuracy": 0.2381649136543274,
      "num_tokens": 122101269.0,
      "step": 58840
    },
    {
      "entropy": 5.703877115249634,
      "epoch": 5.3378991291727145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002541342483236803,
      "loss": 4.8331,
      "mean_token_accuracy": 0.233405701816082,
      "num_tokens": 122111966.0,
      "step": 58845
    },
    {
      "entropy": 5.775815153121949,
      "epoch": 5.338352685050798,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025410203155424866,
      "loss": 4.8417,
      "mean_token_accuracy": 0.2373982086777687,
      "num_tokens": 122121936.0,
      "step": 58850
    },
    {
      "entropy": 5.7234282970428465,
      "epoch": 5.338806240928882,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025406981521699227,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22537937611341477,
      "num_tokens": 122131943.0,
      "step": 58855
    },
    {
      "entropy": 5.735448503494263,
      "epoch": 5.339259796806966,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025403759931257746,
      "loss": 4.8621,
      "mean_token_accuracy": 0.22661909610033035,
      "num_tokens": 122141613.0,
      "step": 58860
    },
    {
      "entropy": 5.729058265686035,
      "epoch": 5.339713352685051,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025400538384167044,
      "loss": 4.884,
      "mean_token_accuracy": 0.22789353132247925,
      "num_tokens": 122153141.0,
      "step": 58865
    },
    {
      "entropy": 5.682979726791382,
      "epoch": 5.340166908563135,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002539731688049374,
      "loss": 4.8,
      "mean_token_accuracy": 0.23194798231124877,
      "num_tokens": 122164556.0,
      "step": 58870
    },
    {
      "entropy": 5.793861293792725,
      "epoch": 5.340620464441219,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002539409542030446,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22470626682043077,
      "num_tokens": 122175504.0,
      "step": 58875
    },
    {
      "entropy": 5.685858201980591,
      "epoch": 5.3410740203193035,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025390874003665826,
      "loss": 4.8616,
      "mean_token_accuracy": 0.23276922553777696,
      "num_tokens": 122186346.0,
      "step": 58880
    },
    {
      "entropy": 5.713300180435181,
      "epoch": 5.341527576197388,
      "grad_norm": 1.375,
      "learning_rate": 0.0002538765263064446,
      "loss": 4.928,
      "mean_token_accuracy": 0.2223044976592064,
      "num_tokens": 122196647.0,
      "step": 58885
    },
    {
      "entropy": 5.705307149887085,
      "epoch": 5.341981132075472,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002538443130130697,
      "loss": 4.8666,
      "mean_token_accuracy": 0.22469670176506043,
      "num_tokens": 122209041.0,
      "step": 58890
    },
    {
      "entropy": 5.76189341545105,
      "epoch": 5.342434687953556,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00025381210015719977,
      "loss": 4.8756,
      "mean_token_accuracy": 0.23247870653867722,
      "num_tokens": 122218330.0,
      "step": 58895
    },
    {
      "entropy": 5.703264236450195,
      "epoch": 5.34288824383164,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025377988773950103,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22642079293727874,
      "num_tokens": 122227930.0,
      "step": 58900
    },
    {
      "entropy": 5.730687141418457,
      "epoch": 5.343341799709724,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025374767576063963,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22907356768846512,
      "num_tokens": 122238028.0,
      "step": 58905
    },
    {
      "entropy": 5.703137445449829,
      "epoch": 5.343795355587808,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025371546422128165,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23522734344005586,
      "num_tokens": 122249018.0,
      "step": 58910
    },
    {
      "entropy": 5.715253591537476,
      "epoch": 5.344248911465892,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025368325312209337,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22470647990703582,
      "num_tokens": 122258635.0,
      "step": 58915
    },
    {
      "entropy": 5.617026710510254,
      "epoch": 5.344702467343977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002536510424637409,
      "loss": 4.762,
      "mean_token_accuracy": 0.23799994289875032,
      "num_tokens": 122268370.0,
      "step": 58920
    },
    {
      "entropy": 5.637312364578247,
      "epoch": 5.345156023222061,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025361883224689027,
      "loss": 4.7909,
      "mean_token_accuracy": 0.23216337114572524,
      "num_tokens": 122279044.0,
      "step": 58925
    },
    {
      "entropy": 5.688122892379761,
      "epoch": 5.345609579100145,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025358662247220755,
      "loss": 4.8662,
      "mean_token_accuracy": 0.22479244321584702,
      "num_tokens": 122289539.0,
      "step": 58930
    },
    {
      "entropy": 5.680511569976806,
      "epoch": 5.3460631349782295,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025355441314035897,
      "loss": 4.7578,
      "mean_token_accuracy": 0.23657282143831254,
      "num_tokens": 122300263.0,
      "step": 58935
    },
    {
      "entropy": 5.7473657608032225,
      "epoch": 5.346516690856314,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002535222042520106,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2306331291794777,
      "num_tokens": 122310813.0,
      "step": 58940
    },
    {
      "entropy": 5.7003556251525875,
      "epoch": 5.346970246734398,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002534899958078285,
      "loss": 4.9063,
      "mean_token_accuracy": 0.22639621645212174,
      "num_tokens": 122321306.0,
      "step": 58945
    },
    {
      "entropy": 5.7414144515991214,
      "epoch": 5.347423802612482,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025345778780847884,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23092532455921172,
      "num_tokens": 122331355.0,
      "step": 58950
    },
    {
      "entropy": 5.713359355926514,
      "epoch": 5.347877358490566,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002534255802546276,
      "loss": 4.8384,
      "mean_token_accuracy": 0.23285816311836244,
      "num_tokens": 122341707.0,
      "step": 58955
    },
    {
      "entropy": 5.757581329345703,
      "epoch": 5.34833091436865,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002533933731469408,
      "loss": 4.9183,
      "mean_token_accuracy": 0.22660596966743468,
      "num_tokens": 122351680.0,
      "step": 58960
    },
    {
      "entropy": 5.730152797698975,
      "epoch": 5.348784470246734,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002533611664860846,
      "loss": 4.8458,
      "mean_token_accuracy": 0.22937985509634018,
      "num_tokens": 122361895.0,
      "step": 58965
    },
    {
      "entropy": 5.690565538406372,
      "epoch": 5.3492380261248185,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025332896027272493,
      "loss": 4.8377,
      "mean_token_accuracy": 0.2263054445385933,
      "num_tokens": 122372652.0,
      "step": 58970
    },
    {
      "entropy": 5.7116392135620115,
      "epoch": 5.349691582002903,
      "grad_norm": 1.5625,
      "learning_rate": 0.000253296754507528,
      "loss": 4.9032,
      "mean_token_accuracy": 0.22014779448509217,
      "num_tokens": 122383372.0,
      "step": 58975
    },
    {
      "entropy": 5.7215094566345215,
      "epoch": 5.350145137880987,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002532645491911596,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2312270075082779,
      "num_tokens": 122392544.0,
      "step": 58980
    },
    {
      "entropy": 5.742029094696045,
      "epoch": 5.350598693759071,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025323234432428587,
      "loss": 4.8481,
      "mean_token_accuracy": 0.2304393470287323,
      "num_tokens": 122402869.0,
      "step": 58985
    },
    {
      "entropy": 5.813335180282593,
      "epoch": 5.3510522496371555,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025320013990757297,
      "loss": 4.977,
      "mean_token_accuracy": 0.22416260838508606,
      "num_tokens": 122413910.0,
      "step": 58990
    },
    {
      "entropy": 5.750451231002808,
      "epoch": 5.35150580551524,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025316793594168656,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22275776714086531,
      "num_tokens": 122424322.0,
      "step": 58995
    },
    {
      "entropy": 5.752235555648804,
      "epoch": 5.351959361393324,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002531357324272929,
      "loss": 4.8573,
      "mean_token_accuracy": 0.2248907819390297,
      "num_tokens": 122434559.0,
      "step": 59000
    },
    {
      "entropy": 5.8155200481414795,
      "epoch": 5.3524129172714074,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002531035293650578,
      "loss": 4.9945,
      "mean_token_accuracy": 0.2146651864051819,
      "num_tokens": 122445641.0,
      "step": 59005
    },
    {
      "entropy": 5.806866931915283,
      "epoch": 5.352866473149492,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002530713267556473,
      "loss": 4.9232,
      "mean_token_accuracy": 0.22322617769241332,
      "num_tokens": 122457524.0,
      "step": 59010
    },
    {
      "entropy": 5.745232105255127,
      "epoch": 5.353320029027576,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002530391245997274,
      "loss": 4.863,
      "mean_token_accuracy": 0.22806279808282853,
      "num_tokens": 122468703.0,
      "step": 59015
    },
    {
      "entropy": 5.697064352035523,
      "epoch": 5.35377358490566,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002530069228979639,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22912262380123138,
      "num_tokens": 122479187.0,
      "step": 59020
    },
    {
      "entropy": 5.728501319885254,
      "epoch": 5.3542271407837445,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025297472165102284,
      "loss": 4.897,
      "mean_token_accuracy": 0.21767673641443253,
      "num_tokens": 122488893.0,
      "step": 59025
    },
    {
      "entropy": 5.631254100799561,
      "epoch": 5.354680696661829,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025294252085957024,
      "loss": 4.774,
      "mean_token_accuracy": 0.2269633322954178,
      "num_tokens": 122499596.0,
      "step": 59030
    },
    {
      "entropy": 5.705673789978027,
      "epoch": 5.355134252539913,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002529103205242719,
      "loss": 4.8306,
      "mean_token_accuracy": 0.23060255348682404,
      "num_tokens": 122509219.0,
      "step": 59035
    },
    {
      "entropy": 5.753299760818481,
      "epoch": 5.355587808417997,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025287812064579376,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21828963458538056,
      "num_tokens": 122519118.0,
      "step": 59040
    },
    {
      "entropy": 5.730877351760864,
      "epoch": 5.356041364296082,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002528459212248017,
      "loss": 4.88,
      "mean_token_accuracy": 0.23540834486484527,
      "num_tokens": 122528641.0,
      "step": 59045
    },
    {
      "entropy": 5.66904125213623,
      "epoch": 5.356494920174166,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002528137222619616,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23116123378276826,
      "num_tokens": 122539357.0,
      "step": 59050
    },
    {
      "entropy": 5.837612533569336,
      "epoch": 5.356948476052249,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025278152375793946,
      "loss": 4.9373,
      "mean_token_accuracy": 0.22821166664361953,
      "num_tokens": 122550218.0,
      "step": 59055
    },
    {
      "entropy": 5.785934734344482,
      "epoch": 5.3574020319303335,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025274932571340105,
      "loss": 4.9627,
      "mean_token_accuracy": 0.22070123255252838,
      "num_tokens": 122561196.0,
      "step": 59060
    },
    {
      "entropy": 5.7705528259277346,
      "epoch": 5.357855587808418,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025271712812901224,
      "loss": 4.916,
      "mean_token_accuracy": 0.22164162695407869,
      "num_tokens": 122572009.0,
      "step": 59065
    },
    {
      "entropy": 5.673266267776489,
      "epoch": 5.358309143686502,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025268493100543896,
      "loss": 4.7831,
      "mean_token_accuracy": 0.23193919956684111,
      "num_tokens": 122581677.0,
      "step": 59070
    },
    {
      "entropy": 5.647936630249023,
      "epoch": 5.358762699564586,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025265273434334684,
      "loss": 4.817,
      "mean_token_accuracy": 0.23277141600847245,
      "num_tokens": 122591503.0,
      "step": 59075
    },
    {
      "entropy": 5.659673070907592,
      "epoch": 5.3592162554426706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025262053814340197,
      "loss": 4.8134,
      "mean_token_accuracy": 0.22837260067462922,
      "num_tokens": 122602675.0,
      "step": 59080
    },
    {
      "entropy": 5.709323453903198,
      "epoch": 5.359669811320755,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002525883424062701,
      "loss": 4.8266,
      "mean_token_accuracy": 0.22359709143638612,
      "num_tokens": 122613832.0,
      "step": 59085
    },
    {
      "entropy": 5.734793758392334,
      "epoch": 5.360123367198839,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025255614713261693,
      "loss": 4.9142,
      "mean_token_accuracy": 0.2309286281466484,
      "num_tokens": 122623910.0,
      "step": 59090
    },
    {
      "entropy": 5.775254344940185,
      "epoch": 5.360576923076923,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025252395232310844,
      "loss": 4.8841,
      "mean_token_accuracy": 0.22800438702106476,
      "num_tokens": 122634634.0,
      "step": 59095
    },
    {
      "entropy": 5.7122992992401125,
      "epoch": 5.361030478955008,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002524917579784103,
      "loss": 4.8302,
      "mean_token_accuracy": 0.23010650873184205,
      "num_tokens": 122644555.0,
      "step": 59100
    },
    {
      "entropy": 5.676687431335449,
      "epoch": 5.361484034833092,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002524595640991884,
      "loss": 4.7848,
      "mean_token_accuracy": 0.23051672130823136,
      "num_tokens": 122654498.0,
      "step": 59105
    },
    {
      "entropy": 5.630251502990722,
      "epoch": 5.361937590711175,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002524273706861084,
      "loss": 4.7534,
      "mean_token_accuracy": 0.23220983296632766,
      "num_tokens": 122664540.0,
      "step": 59110
    },
    {
      "entropy": 5.700547313690185,
      "epoch": 5.3623911465892595,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002523951777398361,
      "loss": 4.8713,
      "mean_token_accuracy": 0.23249824494123458,
      "num_tokens": 122674372.0,
      "step": 59115
    },
    {
      "entropy": 5.792694139480591,
      "epoch": 5.362844702467344,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025236298526103737,
      "loss": 4.9534,
      "mean_token_accuracy": 0.22018318325281144,
      "num_tokens": 122686187.0,
      "step": 59120
    },
    {
      "entropy": 5.65504298210144,
      "epoch": 5.363298258345428,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002523307932503778,
      "loss": 4.788,
      "mean_token_accuracy": 0.2332022160291672,
      "num_tokens": 122697347.0,
      "step": 59125
    },
    {
      "entropy": 5.786406326293945,
      "epoch": 5.363751814223512,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025229860170852326,
      "loss": 4.9221,
      "mean_token_accuracy": 0.2200715497136116,
      "num_tokens": 122707164.0,
      "step": 59130
    },
    {
      "entropy": 5.706855249404907,
      "epoch": 5.364205370101597,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002522664106361394,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22454507499933243,
      "num_tokens": 122716275.0,
      "step": 59135
    },
    {
      "entropy": 5.65755877494812,
      "epoch": 5.364658925979681,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000252234220033892,
      "loss": 4.7912,
      "mean_token_accuracy": 0.2345229595899582,
      "num_tokens": 122726682.0,
      "step": 59140
    },
    {
      "entropy": 5.753113842010498,
      "epoch": 5.365112481857765,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002522020299024466,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21953305304050447,
      "num_tokens": 122738037.0,
      "step": 59145
    },
    {
      "entropy": 5.625066423416138,
      "epoch": 5.365566037735849,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025216984024246914,
      "loss": 4.7471,
      "mean_token_accuracy": 0.23695186227560044,
      "num_tokens": 122748342.0,
      "step": 59150
    },
    {
      "entropy": 5.628825521469116,
      "epoch": 5.366019593613933,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025213765105462524,
      "loss": 4.7145,
      "mean_token_accuracy": 0.24101561754941941,
      "num_tokens": 122759246.0,
      "step": 59155
    },
    {
      "entropy": 5.666534328460694,
      "epoch": 5.366473149492017,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002521054623395804,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2385580986738205,
      "num_tokens": 122769145.0,
      "step": 59160
    },
    {
      "entropy": 5.774560260772705,
      "epoch": 5.366926705370101,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002520732740980005,
      "loss": 4.9998,
      "mean_token_accuracy": 0.21818569600582122,
      "num_tokens": 122779765.0,
      "step": 59165
    },
    {
      "entropy": 5.731737232208252,
      "epoch": 5.367380261248186,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025204108633055116,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22870540022850036,
      "num_tokens": 122790708.0,
      "step": 59170
    },
    {
      "entropy": 5.714873456954956,
      "epoch": 5.36783381712627,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025200889903789796,
      "loss": 4.8239,
      "mean_token_accuracy": 0.23150569200515747,
      "num_tokens": 122800616.0,
      "step": 59175
    },
    {
      "entropy": 5.678650617599487,
      "epoch": 5.368287373004354,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002519767122207067,
      "loss": 4.8079,
      "mean_token_accuracy": 0.23374817222356797,
      "num_tokens": 122810385.0,
      "step": 59180
    },
    {
      "entropy": 5.6749499320983885,
      "epoch": 5.368740928882438,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025194452587964277,
      "loss": 4.8027,
      "mean_token_accuracy": 0.22948130071163178,
      "num_tokens": 122820375.0,
      "step": 59185
    },
    {
      "entropy": 5.762340974807739,
      "epoch": 5.369194484760523,
      "grad_norm": 1.28125,
      "learning_rate": 0.000251912340015372,
      "loss": 4.8569,
      "mean_token_accuracy": 0.22881208658218383,
      "num_tokens": 122831273.0,
      "step": 59190
    },
    {
      "entropy": 5.673686170578003,
      "epoch": 5.369648040638607,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025188015462855986,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23382017463445665,
      "num_tokens": 122842049.0,
      "step": 59195
    },
    {
      "entropy": 5.7355814456939695,
      "epoch": 5.370101596516691,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002518479697198721,
      "loss": 4.9407,
      "mean_token_accuracy": 0.22234368920326233,
      "num_tokens": 122854049.0,
      "step": 59200
    },
    {
      "entropy": 5.777974367141724,
      "epoch": 5.370555152394775,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002518157852899742,
      "loss": 4.9046,
      "mean_token_accuracy": 0.2313472345471382,
      "num_tokens": 122864528.0,
      "step": 59205
    },
    {
      "entropy": 5.73221640586853,
      "epoch": 5.371008708272859,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025178360133953166,
      "loss": 4.846,
      "mean_token_accuracy": 0.23153116703033447,
      "num_tokens": 122873975.0,
      "step": 59210
    },
    {
      "entropy": 5.790131092071533,
      "epoch": 5.371462264150943,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025175141786921024,
      "loss": 4.9606,
      "mean_token_accuracy": 0.22003549635410308,
      "num_tokens": 122884742.0,
      "step": 59215
    },
    {
      "entropy": 5.697367811203003,
      "epoch": 5.371915820029027,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002517192348796754,
      "loss": 4.8004,
      "mean_token_accuracy": 0.24087177813053132,
      "num_tokens": 122895300.0,
      "step": 59220
    },
    {
      "entropy": 5.6658769130706785,
      "epoch": 5.372369375907112,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002516870523715928,
      "loss": 4.8024,
      "mean_token_accuracy": 0.2374484583735466,
      "num_tokens": 122905805.0,
      "step": 59225
    },
    {
      "entropy": 5.642878484725952,
      "epoch": 5.372822931785196,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002516548703456278,
      "loss": 4.7853,
      "mean_token_accuracy": 0.2313608318567276,
      "num_tokens": 122916106.0,
      "step": 59230
    },
    {
      "entropy": 5.672796726226807,
      "epoch": 5.37327648766328,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000251622688802446,
      "loss": 4.832,
      "mean_token_accuracy": 0.22960762679576874,
      "num_tokens": 122926074.0,
      "step": 59235
    },
    {
      "entropy": 5.710990715026855,
      "epoch": 5.373730043541364,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002515905077427131,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22658197432756425,
      "num_tokens": 122936368.0,
      "step": 59240
    },
    {
      "entropy": 5.70075831413269,
      "epoch": 5.374183599419449,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025155832716709434,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2347483292222023,
      "num_tokens": 122946849.0,
      "step": 59245
    },
    {
      "entropy": 5.736695957183838,
      "epoch": 5.374637155297533,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025152614707625543,
      "loss": 4.8785,
      "mean_token_accuracy": 0.23170990645885467,
      "num_tokens": 122957093.0,
      "step": 59250
    },
    {
      "entropy": 5.6496748447418215,
      "epoch": 5.375090711175617,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002514939674708617,
      "loss": 4.7702,
      "mean_token_accuracy": 0.24135801643133165,
      "num_tokens": 122967399.0,
      "step": 59255
    },
    {
      "entropy": 5.695899534225464,
      "epoch": 5.3755442670537015,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025146178835157876,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22105879634618758,
      "num_tokens": 122977891.0,
      "step": 59260
    },
    {
      "entropy": 5.741484832763672,
      "epoch": 5.375997822931785,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000251429609719072,
      "loss": 4.8905,
      "mean_token_accuracy": 0.226633220911026,
      "num_tokens": 122988446.0,
      "step": 59265
    },
    {
      "entropy": 5.76721601486206,
      "epoch": 5.376451378809869,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000251397431574007,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2309999793767929,
      "num_tokens": 122998021.0,
      "step": 59270
    },
    {
      "entropy": 5.661801195144653,
      "epoch": 5.376904934687953,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025136525391704904,
      "loss": 4.8531,
      "mean_token_accuracy": 0.22679445445537566,
      "num_tokens": 123008599.0,
      "step": 59275
    },
    {
      "entropy": 5.705998039245605,
      "epoch": 5.377358490566038,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002513330767488637,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23288679122924805,
      "num_tokens": 123019586.0,
      "step": 59280
    },
    {
      "entropy": 5.679974555969238,
      "epoch": 5.377812046444122,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025130090007011635,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23104263693094254,
      "num_tokens": 123030222.0,
      "step": 59285
    },
    {
      "entropy": 5.661655712127685,
      "epoch": 5.378265602322206,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002512687238814724,
      "loss": 4.7925,
      "mean_token_accuracy": 0.230720354616642,
      "num_tokens": 123041035.0,
      "step": 59290
    },
    {
      "entropy": 5.697479295730591,
      "epoch": 5.3787191582002905,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025123654818359737,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2341500148177147,
      "num_tokens": 123052559.0,
      "step": 59295
    },
    {
      "entropy": 5.754133510589599,
      "epoch": 5.379172714078375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002512043729771565,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2252386227250099,
      "num_tokens": 123062106.0,
      "step": 59300
    },
    {
      "entropy": 5.7089919090271,
      "epoch": 5.379626269956459,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025117219826281526,
      "loss": 4.7907,
      "mean_token_accuracy": 0.23825348615646363,
      "num_tokens": 123071670.0,
      "step": 59305
    },
    {
      "entropy": 5.72225284576416,
      "epoch": 5.380079825834542,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025114002404123917,
      "loss": 4.7409,
      "mean_token_accuracy": 0.24964421838521958,
      "num_tokens": 123082023.0,
      "step": 59310
    },
    {
      "entropy": 5.701400852203369,
      "epoch": 5.380533381712627,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025110785031309325,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2205074056982994,
      "num_tokens": 123092522.0,
      "step": 59315
    },
    {
      "entropy": 5.631285238265991,
      "epoch": 5.380986937590711,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025107567707904324,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2276730179786682,
      "num_tokens": 123102537.0,
      "step": 59320
    },
    {
      "entropy": 5.707534837722778,
      "epoch": 5.381440493468795,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002510435043397542,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2300427198410034,
      "num_tokens": 123112544.0,
      "step": 59325
    },
    {
      "entropy": 5.757447385787964,
      "epoch": 5.381894049346879,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002510113320958917,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22584124356508256,
      "num_tokens": 123123089.0,
      "step": 59330
    },
    {
      "entropy": 5.697446632385254,
      "epoch": 5.382347605224964,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002509791603481209,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22821266502141951,
      "num_tokens": 123133452.0,
      "step": 59335
    },
    {
      "entropy": 5.658330345153809,
      "epoch": 5.382801161103048,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025094698909710716,
      "loss": 4.7754,
      "mean_token_accuracy": 0.23334754407405853,
      "num_tokens": 123143739.0,
      "step": 59340
    },
    {
      "entropy": 5.740772342681884,
      "epoch": 5.383254716981132,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002509148183435159,
      "loss": 4.8636,
      "mean_token_accuracy": 0.2355750486254692,
      "num_tokens": 123154948.0,
      "step": 59345
    },
    {
      "entropy": 5.7623786449432375,
      "epoch": 5.3837082728592165,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025088264808801225,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21935809403657913,
      "num_tokens": 123165375.0,
      "step": 59350
    },
    {
      "entropy": 5.687585926055908,
      "epoch": 5.384161828737301,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002508504783312617,
      "loss": 4.8507,
      "mean_token_accuracy": 0.229604072868824,
      "num_tokens": 123176129.0,
      "step": 59355
    },
    {
      "entropy": 5.737585258483887,
      "epoch": 5.384615384615385,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025081830907392936,
      "loss": 4.8601,
      "mean_token_accuracy": 0.23247312903404235,
      "num_tokens": 123186499.0,
      "step": 59360
    },
    {
      "entropy": 5.682219505310059,
      "epoch": 5.385068940493468,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025078614031668044,
      "loss": 4.7759,
      "mean_token_accuracy": 0.23303068429231644,
      "num_tokens": 123197486.0,
      "step": 59365
    },
    {
      "entropy": 5.748718643188477,
      "epoch": 5.385522496371553,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025075397206018044,
      "loss": 4.8751,
      "mean_token_accuracy": 0.22715696096420288,
      "num_tokens": 123208107.0,
      "step": 59370
    },
    {
      "entropy": 5.702201986312867,
      "epoch": 5.385976052249637,
      "grad_norm": 1.25,
      "learning_rate": 0.00025072180430509445,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22316994667053222,
      "num_tokens": 123218993.0,
      "step": 59375
    },
    {
      "entropy": 5.7265204906463625,
      "epoch": 5.386429608127721,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002506896370520877,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2335875153541565,
      "num_tokens": 123229728.0,
      "step": 59380
    },
    {
      "entropy": 5.756487655639648,
      "epoch": 5.3868831640058055,
      "grad_norm": 1.375,
      "learning_rate": 0.0002506574703018255,
      "loss": 4.8798,
      "mean_token_accuracy": 0.23229512721300125,
      "num_tokens": 123240009.0,
      "step": 59385
    },
    {
      "entropy": 5.747384929656983,
      "epoch": 5.38733671988389,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025062530405497297,
      "loss": 4.8622,
      "mean_token_accuracy": 0.2261661931872368,
      "num_tokens": 123249827.0,
      "step": 59390
    },
    {
      "entropy": 5.611868619918823,
      "epoch": 5.387790275761974,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002505931383121954,
      "loss": 4.7793,
      "mean_token_accuracy": 0.24028478264808656,
      "num_tokens": 123260032.0,
      "step": 59395
    },
    {
      "entropy": 5.602818155288697,
      "epoch": 5.388243831640058,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025056097307415786,
      "loss": 4.7791,
      "mean_token_accuracy": 0.2280777022242546,
      "num_tokens": 123269818.0,
      "step": 59400
    },
    {
      "entropy": 5.727565050125122,
      "epoch": 5.3886973875181425,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025052880834152574,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2280734360218048,
      "num_tokens": 123280478.0,
      "step": 59405
    },
    {
      "entropy": 5.735644197463989,
      "epoch": 5.389150943396227,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000250496644114964,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22245387434959413,
      "num_tokens": 123291621.0,
      "step": 59410
    },
    {
      "entropy": 5.657926034927368,
      "epoch": 5.389604499274311,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002504644803951381,
      "loss": 4.7479,
      "mean_token_accuracy": 0.2409937024116516,
      "num_tokens": 123300906.0,
      "step": 59415
    },
    {
      "entropy": 5.725187492370606,
      "epoch": 5.3900580551523944,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002504323171827128,
      "loss": 4.8726,
      "mean_token_accuracy": 0.23460128903388977,
      "num_tokens": 123310961.0,
      "step": 59420
    },
    {
      "entropy": 5.677012586593628,
      "epoch": 5.390511611030479,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002504001544783535,
      "loss": 4.7983,
      "mean_token_accuracy": 0.23543121069669723,
      "num_tokens": 123321570.0,
      "step": 59425
    },
    {
      "entropy": 5.704794216156006,
      "epoch": 5.390965166908563,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025036799228272526,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22728723287582397,
      "num_tokens": 123331736.0,
      "step": 59430
    },
    {
      "entropy": 5.707384777069092,
      "epoch": 5.391418722786647,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002503358305964933,
      "loss": 4.9087,
      "mean_token_accuracy": 0.2302492842078209,
      "num_tokens": 123341893.0,
      "step": 59435
    },
    {
      "entropy": 5.716143560409546,
      "epoch": 5.3918722786647315,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025030366942032257,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22613864988088608,
      "num_tokens": 123351844.0,
      "step": 59440
    },
    {
      "entropy": 5.741736221313476,
      "epoch": 5.392325834542816,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002502715087548782,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22676898539066315,
      "num_tokens": 123362376.0,
      "step": 59445
    },
    {
      "entropy": 5.73913164138794,
      "epoch": 5.3927793904209,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002502393486008254,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23811442106962205,
      "num_tokens": 123372781.0,
      "step": 59450
    },
    {
      "entropy": 5.735100841522216,
      "epoch": 5.393232946298984,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002502071889588292,
      "loss": 4.9692,
      "mean_token_accuracy": 0.22459543645381927,
      "num_tokens": 123383047.0,
      "step": 59455
    },
    {
      "entropy": 5.634616708755493,
      "epoch": 5.393686502177069,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025017502982955457,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23532880395650863,
      "num_tokens": 123393265.0,
      "step": 59460
    },
    {
      "entropy": 5.63350396156311,
      "epoch": 5.394140058055152,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002501428712136667,
      "loss": 4.7749,
      "mean_token_accuracy": 0.23689111322164536,
      "num_tokens": 123403781.0,
      "step": 59465
    },
    {
      "entropy": 5.759675598144531,
      "epoch": 5.394593613933236,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002501107131118306,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23008581548929213,
      "num_tokens": 123413260.0,
      "step": 59470
    },
    {
      "entropy": 5.709977865219116,
      "epoch": 5.3950471698113205,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025007855552471123,
      "loss": 4.8442,
      "mean_token_accuracy": 0.22729507684707642,
      "num_tokens": 123423578.0,
      "step": 59475
    },
    {
      "entropy": 5.648068189620972,
      "epoch": 5.395500725689405,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025004639845297377,
      "loss": 4.7749,
      "mean_token_accuracy": 0.237044957280159,
      "num_tokens": 123434425.0,
      "step": 59480
    },
    {
      "entropy": 5.6421479225158695,
      "epoch": 5.395954281567489,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025001424189728314,
      "loss": 4.7285,
      "mean_token_accuracy": 0.2400059536099434,
      "num_tokens": 123444096.0,
      "step": 59485
    },
    {
      "entropy": 5.756652402877807,
      "epoch": 5.396407837445573,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024998208585830435,
      "loss": 4.9362,
      "mean_token_accuracy": 0.23037019819021226,
      "num_tokens": 123454655.0,
      "step": 59490
    },
    {
      "entropy": 5.676408386230468,
      "epoch": 5.3968613933236576,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024994993033670233,
      "loss": 4.8075,
      "mean_token_accuracy": 0.2338923841714859,
      "num_tokens": 123465742.0,
      "step": 59495
    },
    {
      "entropy": 5.79536452293396,
      "epoch": 5.397314949201742,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002499177753331422,
      "loss": 4.9226,
      "mean_token_accuracy": 0.22769764214754104,
      "num_tokens": 123476110.0,
      "step": 59500
    },
    {
      "entropy": 5.704824495315552,
      "epoch": 5.397768505079826,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002498856208482889,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22690316438674926,
      "num_tokens": 123486572.0,
      "step": 59505
    },
    {
      "entropy": 5.67619218826294,
      "epoch": 5.39822206095791,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002498534668828072,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22584255784749985,
      "num_tokens": 123498096.0,
      "step": 59510
    },
    {
      "entropy": 5.687308073043823,
      "epoch": 5.398675616835995,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002498213134373624,
      "loss": 4.8061,
      "mean_token_accuracy": 0.23428291380405425,
      "num_tokens": 123508290.0,
      "step": 59515
    },
    {
      "entropy": 5.762889051437378,
      "epoch": 5.399129172714078,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024978916051261914,
      "loss": 4.8342,
      "mean_token_accuracy": 0.23057093918323518,
      "num_tokens": 123518040.0,
      "step": 59520
    },
    {
      "entropy": 5.764799308776856,
      "epoch": 5.399582728592162,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002497570081092425,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22152202874422072,
      "num_tokens": 123528799.0,
      "step": 59525
    },
    {
      "entropy": 5.6839292526245115,
      "epoch": 5.4000362844702465,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002497248562278974,
      "loss": 4.8262,
      "mean_token_accuracy": 0.22887375354766845,
      "num_tokens": 123539114.0,
      "step": 59530
    },
    {
      "entropy": 5.722823810577393,
      "epoch": 5.400489840348331,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024969270486924867,
      "loss": 4.814,
      "mean_token_accuracy": 0.231020587682724,
      "num_tokens": 123549914.0,
      "step": 59535
    },
    {
      "entropy": 5.810749530792236,
      "epoch": 5.400943396226415,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024966055403396125,
      "loss": 4.9553,
      "mean_token_accuracy": 0.2199004262685776,
      "num_tokens": 123560067.0,
      "step": 59540
    },
    {
      "entropy": 5.66645565032959,
      "epoch": 5.401396952104499,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002496284037227,
      "loss": 4.7842,
      "mean_token_accuracy": 0.22927416115999222,
      "num_tokens": 123569756.0,
      "step": 59545
    },
    {
      "entropy": 5.745407962799073,
      "epoch": 5.401850507982584,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024959625393612987,
      "loss": 4.9074,
      "mean_token_accuracy": 0.2266767516732216,
      "num_tokens": 123580482.0,
      "step": 59550
    },
    {
      "entropy": 5.716536235809326,
      "epoch": 5.402304063860668,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002495641046749157,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22910026460886002,
      "num_tokens": 123592212.0,
      "step": 59555
    },
    {
      "entropy": 5.675992441177368,
      "epoch": 5.402757619738752,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002495319559397223,
      "loss": 4.8349,
      "mean_token_accuracy": 0.23374451398849488,
      "num_tokens": 123602795.0,
      "step": 59560
    },
    {
      "entropy": 5.687697553634644,
      "epoch": 5.403211175616836,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002494998077312146,
      "loss": 4.8225,
      "mean_token_accuracy": 0.23280998021364213,
      "num_tokens": 123613856.0,
      "step": 59565
    },
    {
      "entropy": 5.678142404556274,
      "epoch": 5.40366473149492,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024946766005005727,
      "loss": 4.8285,
      "mean_token_accuracy": 0.23477188646793365,
      "num_tokens": 123623536.0,
      "step": 59570
    },
    {
      "entropy": 5.664327096939087,
      "epoch": 5.404118287373004,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002494355128969153,
      "loss": 4.8258,
      "mean_token_accuracy": 0.22816223204135894,
      "num_tokens": 123634011.0,
      "step": 59575
    },
    {
      "entropy": 5.686284875869751,
      "epoch": 5.404571843251088,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024940336627245343,
      "loss": 4.828,
      "mean_token_accuracy": 0.2322561964392662,
      "num_tokens": 123644064.0,
      "step": 59580
    },
    {
      "entropy": 5.644005203247071,
      "epoch": 5.405025399129173,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002493712201773365,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23314585387706757,
      "num_tokens": 123654231.0,
      "step": 59585
    },
    {
      "entropy": 5.724025917053223,
      "epoch": 5.405478955007257,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002493390746122292,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22606738209724425,
      "num_tokens": 123664587.0,
      "step": 59590
    },
    {
      "entropy": 5.728795862197876,
      "epoch": 5.405932510885341,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002493069295777964,
      "loss": 4.8627,
      "mean_token_accuracy": 0.234894822537899,
      "num_tokens": 123674474.0,
      "step": 59595
    },
    {
      "entropy": 5.717601680755616,
      "epoch": 5.406386066763425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024927478507470285,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23070017397403716,
      "num_tokens": 123685108.0,
      "step": 59600
    },
    {
      "entropy": 5.6881263732910154,
      "epoch": 5.40683962264151,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024924264110361334,
      "loss": 4.8593,
      "mean_token_accuracy": 0.23079022914171218,
      "num_tokens": 123695723.0,
      "step": 59605
    },
    {
      "entropy": 5.753277349472046,
      "epoch": 5.407293178519594,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024921049766519257,
      "loss": 4.9558,
      "mean_token_accuracy": 0.2266206383705139,
      "num_tokens": 123706448.0,
      "step": 59610
    },
    {
      "entropy": 5.796307039260864,
      "epoch": 5.407746734397678,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002491783547601052,
      "loss": 4.9519,
      "mean_token_accuracy": 0.2275138959288597,
      "num_tokens": 123717373.0,
      "step": 59615
    },
    {
      "entropy": 5.757608842849732,
      "epoch": 5.4082002902757615,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002491462123890161,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22871136665344238,
      "num_tokens": 123727174.0,
      "step": 59620
    },
    {
      "entropy": 5.7160351276397705,
      "epoch": 5.408653846153846,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002491140705525899,
      "loss": 4.8644,
      "mean_token_accuracy": 0.2314464494585991,
      "num_tokens": 123736962.0,
      "step": 59625
    },
    {
      "entropy": 5.724878931045533,
      "epoch": 5.40910740203193,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024908192925149125,
      "loss": 4.832,
      "mean_token_accuracy": 0.23162737786769866,
      "num_tokens": 123748808.0,
      "step": 59630
    },
    {
      "entropy": 5.652881622314453,
      "epoch": 5.409560957910014,
      "grad_norm": 1.421875,
      "learning_rate": 0.000249049788486385,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23394618481397628,
      "num_tokens": 123759211.0,
      "step": 59635
    },
    {
      "entropy": 5.763215208053589,
      "epoch": 5.410014513788099,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002490176482579357,
      "loss": 4.9625,
      "mean_token_accuracy": 0.2214633882045746,
      "num_tokens": 123769568.0,
      "step": 59640
    },
    {
      "entropy": 5.790954494476319,
      "epoch": 5.410468069666183,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024898550856680815,
      "loss": 4.9062,
      "mean_token_accuracy": 0.22586793303489686,
      "num_tokens": 123779503.0,
      "step": 59645
    },
    {
      "entropy": 5.724041843414307,
      "epoch": 5.410921625544267,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024895336941366676,
      "loss": 4.8648,
      "mean_token_accuracy": 0.22586733102798462,
      "num_tokens": 123790194.0,
      "step": 59650
    },
    {
      "entropy": 5.851100492477417,
      "epoch": 5.411375181422351,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024892123079917647,
      "loss": 5.0721,
      "mean_token_accuracy": 0.21546741873025893,
      "num_tokens": 123800055.0,
      "step": 59655
    },
    {
      "entropy": 5.777582883834839,
      "epoch": 5.411828737300436,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002488890927240017,
      "loss": 4.8357,
      "mean_token_accuracy": 0.23444948047399522,
      "num_tokens": 123809358.0,
      "step": 59660
    },
    {
      "entropy": 5.660604190826416,
      "epoch": 5.41228229317852,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024885695518880706,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23222322314977645,
      "num_tokens": 123820294.0,
      "step": 59665
    },
    {
      "entropy": 5.6885223388671875,
      "epoch": 5.412735849056604,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024882481819425733,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22146472632884978,
      "num_tokens": 123830580.0,
      "step": 59670
    },
    {
      "entropy": 5.730199003219605,
      "epoch": 5.413189404934688,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024879268174101706,
      "loss": 4.8605,
      "mean_token_accuracy": 0.2222564399242401,
      "num_tokens": 123842089.0,
      "step": 59675
    },
    {
      "entropy": 5.673899126052857,
      "epoch": 5.413642960812772,
      "grad_norm": 1.5,
      "learning_rate": 0.00024876054582975075,
      "loss": 4.7668,
      "mean_token_accuracy": 0.2432022899389267,
      "num_tokens": 123851852.0,
      "step": 59680
    },
    {
      "entropy": 5.709301280975342,
      "epoch": 5.414096516690856,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002487284104611231,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23413625210523606,
      "num_tokens": 123863193.0,
      "step": 59685
    },
    {
      "entropy": 5.7616565227508545,
      "epoch": 5.41455007256894,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002486962756357986,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21514952331781387,
      "num_tokens": 123874771.0,
      "step": 59690
    },
    {
      "entropy": 5.765362310409546,
      "epoch": 5.415003628447025,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002486641413544418,
      "loss": 4.8451,
      "mean_token_accuracy": 0.2371772661805153,
      "num_tokens": 123885613.0,
      "step": 59695
    },
    {
      "entropy": 5.68542857170105,
      "epoch": 5.415457184325109,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002486320076177173,
      "loss": 4.7857,
      "mean_token_accuracy": 0.23211355954408647,
      "num_tokens": 123896181.0,
      "step": 59700
    },
    {
      "entropy": 5.685616397857666,
      "epoch": 5.415910740203193,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024859987442628963,
      "loss": 4.8341,
      "mean_token_accuracy": 0.23184244334697723,
      "num_tokens": 123906245.0,
      "step": 59705
    },
    {
      "entropy": 5.697358274459839,
      "epoch": 5.4163642960812775,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002485677417808232,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23833716958761214,
      "num_tokens": 123916596.0,
      "step": 59710
    },
    {
      "entropy": 5.671369838714599,
      "epoch": 5.416817851959362,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002485356096819826,
      "loss": 4.8132,
      "mean_token_accuracy": 0.23992773592472078,
      "num_tokens": 123927125.0,
      "step": 59715
    },
    {
      "entropy": 5.736529541015625,
      "epoch": 5.417271407837446,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002485034781304324,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22450391352176666,
      "num_tokens": 123937457.0,
      "step": 59720
    },
    {
      "entropy": 5.707201719284058,
      "epoch": 5.417724963715529,
      "grad_norm": 1.609375,
      "learning_rate": 0.000248471347126837,
      "loss": 4.9089,
      "mean_token_accuracy": 0.23186276406049727,
      "num_tokens": 123947178.0,
      "step": 59725
    },
    {
      "entropy": 5.719423103332519,
      "epoch": 5.418178519593614,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024843921667186094,
      "loss": 4.8342,
      "mean_token_accuracy": 0.23010358959436417,
      "num_tokens": 123957103.0,
      "step": 59730
    },
    {
      "entropy": 5.684950923919677,
      "epoch": 5.418632075471698,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002484070867661686,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22699456959962844,
      "num_tokens": 123967311.0,
      "step": 59735
    },
    {
      "entropy": 5.753493547439575,
      "epoch": 5.419085631349782,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002483749574104245,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2310578465461731,
      "num_tokens": 123977964.0,
      "step": 59740
    },
    {
      "entropy": 5.700353479385376,
      "epoch": 5.419539187227866,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002483428286052931,
      "loss": 4.8349,
      "mean_token_accuracy": 0.22962892353534697,
      "num_tokens": 123988345.0,
      "step": 59745
    },
    {
      "entropy": 5.708023262023926,
      "epoch": 5.419992743105951,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002483107003514387,
      "loss": 4.8631,
      "mean_token_accuracy": 0.23078318387269975,
      "num_tokens": 123998056.0,
      "step": 59750
    },
    {
      "entropy": 5.7044251441955565,
      "epoch": 5.420446298984035,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024827857264952583,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22503589391708373,
      "num_tokens": 124009104.0,
      "step": 59755
    },
    {
      "entropy": 5.6947344779968265,
      "epoch": 5.420899854862119,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024824644550021894,
      "loss": 4.8265,
      "mean_token_accuracy": 0.22412582188844682,
      "num_tokens": 124018975.0,
      "step": 59760
    },
    {
      "entropy": 5.648674058914184,
      "epoch": 5.4213534107402035,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002482143189041824,
      "loss": 4.7836,
      "mean_token_accuracy": 0.23374448120594024,
      "num_tokens": 124028770.0,
      "step": 59765
    },
    {
      "entropy": 5.648376417160034,
      "epoch": 5.421806966618288,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024818219286208053,
      "loss": 4.816,
      "mean_token_accuracy": 0.23295440673828124,
      "num_tokens": 124038625.0,
      "step": 59770
    },
    {
      "entropy": 5.701836824417114,
      "epoch": 5.422260522496371,
      "grad_norm": 1.375,
      "learning_rate": 0.0002481500673745778,
      "loss": 4.838,
      "mean_token_accuracy": 0.23757988959550858,
      "num_tokens": 124049063.0,
      "step": 59775
    },
    {
      "entropy": 5.7725166320800785,
      "epoch": 5.422714078374455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002481179424423385,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2285594567656517,
      "num_tokens": 124059564.0,
      "step": 59780
    },
    {
      "entropy": 5.7236165523529055,
      "epoch": 5.42316763425254,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002480858180660269,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22875185012817384,
      "num_tokens": 124069397.0,
      "step": 59785
    },
    {
      "entropy": 5.770729970932007,
      "epoch": 5.423621190130624,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002480536942463075,
      "loss": 4.8561,
      "mean_token_accuracy": 0.22415607720613479,
      "num_tokens": 124080173.0,
      "step": 59790
    },
    {
      "entropy": 5.779225826263428,
      "epoch": 5.424074746008708,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024802157098384464,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2277009069919586,
      "num_tokens": 124091445.0,
      "step": 59795
    },
    {
      "entropy": 5.717015123367309,
      "epoch": 5.4245283018867925,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024798944827930246,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22602895498275757,
      "num_tokens": 124100849.0,
      "step": 59800
    },
    {
      "entropy": 5.707440900802612,
      "epoch": 5.424981857764877,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002479573261333454,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23611694872379302,
      "num_tokens": 124110536.0,
      "step": 59805
    },
    {
      "entropy": 5.672256374359131,
      "epoch": 5.425435413642961,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002479252045466378,
      "loss": 4.8616,
      "mean_token_accuracy": 0.23820329904556276,
      "num_tokens": 124121329.0,
      "step": 59810
    },
    {
      "entropy": 5.631294679641724,
      "epoch": 5.425888969521045,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002478930835198439,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2255028799176216,
      "num_tokens": 124131611.0,
      "step": 59815
    },
    {
      "entropy": 5.661938190460205,
      "epoch": 5.4263425253991295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024786096305362776,
      "loss": 4.7915,
      "mean_token_accuracy": 0.24441658705472946,
      "num_tokens": 124141757.0,
      "step": 59820
    },
    {
      "entropy": 5.719198703765869,
      "epoch": 5.426796081277214,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024782884314865394,
      "loss": 4.8769,
      "mean_token_accuracy": 0.23181697875261306,
      "num_tokens": 124152410.0,
      "step": 59825
    },
    {
      "entropy": 5.716597032546997,
      "epoch": 5.427249637155297,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024779672380558654,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2425068959593773,
      "num_tokens": 124163206.0,
      "step": 59830
    },
    {
      "entropy": 5.730745410919189,
      "epoch": 5.4277031930333814,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002477646050250898,
      "loss": 4.8575,
      "mean_token_accuracy": 0.23176708072423935,
      "num_tokens": 124173208.0,
      "step": 59835
    },
    {
      "entropy": 5.66458215713501,
      "epoch": 5.428156748911466,
      "grad_norm": 1.375,
      "learning_rate": 0.000247732486807828,
      "loss": 4.7726,
      "mean_token_accuracy": 0.23408134132623673,
      "num_tokens": 124183759.0,
      "step": 59840
    },
    {
      "entropy": 5.812407827377319,
      "epoch": 5.42861030478955,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002477003691544654,
      "loss": 5.011,
      "mean_token_accuracy": 0.21841100454330445,
      "num_tokens": 124194517.0,
      "step": 59845
    },
    {
      "entropy": 5.665551090240479,
      "epoch": 5.429063860667634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002476682520656661,
      "loss": 4.7914,
      "mean_token_accuracy": 0.229350945353508,
      "num_tokens": 124204752.0,
      "step": 59850
    },
    {
      "entropy": 5.7206690311431885,
      "epoch": 5.4295174165457185,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024763613554209423,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22480456233024598,
      "num_tokens": 124214231.0,
      "step": 59855
    },
    {
      "entropy": 5.748619794845581,
      "epoch": 5.429970972423803,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002476040195844141,
      "loss": 4.8768,
      "mean_token_accuracy": 0.23301654160022736,
      "num_tokens": 124223984.0,
      "step": 59860
    },
    {
      "entropy": 5.6857470035552975,
      "epoch": 5.430424528301887,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024757190419328977,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2286092072725296,
      "num_tokens": 124234170.0,
      "step": 59865
    },
    {
      "entropy": 5.67919511795044,
      "epoch": 5.430878084179971,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002475397893693855,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23514575362205506,
      "num_tokens": 124244756.0,
      "step": 59870
    },
    {
      "entropy": 5.770411586761474,
      "epoch": 5.431331640058055,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024750767511336544,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2241489142179489,
      "num_tokens": 124256148.0,
      "step": 59875
    },
    {
      "entropy": 5.684573936462402,
      "epoch": 5.431785195936139,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002474755614258935,
      "loss": 4.807,
      "mean_token_accuracy": 0.23490413874387742,
      "num_tokens": 124266774.0,
      "step": 59880
    },
    {
      "entropy": 5.70974383354187,
      "epoch": 5.432238751814223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002474434483076341,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22833812683820726,
      "num_tokens": 124276747.0,
      "step": 59885
    },
    {
      "entropy": 5.707683277130127,
      "epoch": 5.4326923076923075,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002474113357592512,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22699266523122788,
      "num_tokens": 124286610.0,
      "step": 59890
    },
    {
      "entropy": 5.810425806045532,
      "epoch": 5.433145863570392,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002473792237814088,
      "loss": 4.921,
      "mean_token_accuracy": 0.22722531855106354,
      "num_tokens": 124296438.0,
      "step": 59895
    },
    {
      "entropy": 5.7903975486755375,
      "epoch": 5.433599419448476,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002473471123747712,
      "loss": 4.9533,
      "mean_token_accuracy": 0.2190373197197914,
      "num_tokens": 124305509.0,
      "step": 59900
    },
    {
      "entropy": 5.735112142562866,
      "epoch": 5.43405297532656,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002473150015400023,
      "loss": 4.8381,
      "mean_token_accuracy": 0.23276158422231674,
      "num_tokens": 124315911.0,
      "step": 59905
    },
    {
      "entropy": 5.682092571258545,
      "epoch": 5.4345065312046446,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002472828912777663,
      "loss": 4.8056,
      "mean_token_accuracy": 0.23754850775003433,
      "num_tokens": 124326670.0,
      "step": 59910
    },
    {
      "entropy": 5.67007851600647,
      "epoch": 5.434960087082729,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002472507815887271,
      "loss": 4.8348,
      "mean_token_accuracy": 0.23277391195297242,
      "num_tokens": 124336851.0,
      "step": 59915
    },
    {
      "entropy": 5.730233144760132,
      "epoch": 5.435413642960813,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024721867247354886,
      "loss": 4.8236,
      "mean_token_accuracy": 0.2333059325814247,
      "num_tokens": 124347056.0,
      "step": 59920
    },
    {
      "entropy": 5.720964002609253,
      "epoch": 5.435867198838897,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024718656393289554,
      "loss": 4.8442,
      "mean_token_accuracy": 0.22782956957817077,
      "num_tokens": 124357684.0,
      "step": 59925
    },
    {
      "entropy": 5.6352331161499025,
      "epoch": 5.436320754716981,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002471544559674311,
      "loss": 4.7104,
      "mean_token_accuracy": 0.24968432486057282,
      "num_tokens": 124368100.0,
      "step": 59930
    },
    {
      "entropy": 5.724936199188233,
      "epoch": 5.436774310595065,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002471223485778196,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2185131624341011,
      "num_tokens": 124378434.0,
      "step": 59935
    },
    {
      "entropy": 5.745794868469238,
      "epoch": 5.437227866473149,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024709024176472515,
      "loss": 4.8952,
      "mean_token_accuracy": 0.22772188484668732,
      "num_tokens": 124389730.0,
      "step": 59940
    },
    {
      "entropy": 5.80607647895813,
      "epoch": 5.4376814223512335,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024705813552881157,
      "loss": 4.9857,
      "mean_token_accuracy": 0.21483721137046813,
      "num_tokens": 124401566.0,
      "step": 59945
    },
    {
      "entropy": 5.713967561721802,
      "epoch": 5.438134978229318,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024702602987074283,
      "loss": 4.7979,
      "mean_token_accuracy": 0.23879697173833847,
      "num_tokens": 124411365.0,
      "step": 59950
    },
    {
      "entropy": 5.6538361549377445,
      "epoch": 5.438588534107402,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002469939247911829,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23280794471502303,
      "num_tokens": 124422193.0,
      "step": 59955
    },
    {
      "entropy": 5.693041801452637,
      "epoch": 5.439042089985486,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002469618202907958,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22304028421640396,
      "num_tokens": 124432564.0,
      "step": 59960
    },
    {
      "entropy": 5.7345054149627686,
      "epoch": 5.439495645863571,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002469297163702454,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2257700428366661,
      "num_tokens": 124442975.0,
      "step": 59965
    },
    {
      "entropy": 5.681476497650147,
      "epoch": 5.439949201741655,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002468976130301956,
      "loss": 4.8018,
      "mean_token_accuracy": 0.2303531989455223,
      "num_tokens": 124453084.0,
      "step": 59970
    },
    {
      "entropy": 5.806116724014283,
      "epoch": 5.440402757619739,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024686551027131037,
      "loss": 4.9845,
      "mean_token_accuracy": 0.21658553779125214,
      "num_tokens": 124462921.0,
      "step": 59975
    },
    {
      "entropy": 5.76521258354187,
      "epoch": 5.440856313497823,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002468334080942535,
      "loss": 4.828,
      "mean_token_accuracy": 0.22976527810096742,
      "num_tokens": 124473558.0,
      "step": 59980
    },
    {
      "entropy": 5.722191524505615,
      "epoch": 5.441309869375907,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002468013064996889,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2326936885714531,
      "num_tokens": 124483543.0,
      "step": 59985
    },
    {
      "entropy": 5.82045168876648,
      "epoch": 5.441763425253991,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024676920548828056,
      "loss": 4.9942,
      "mean_token_accuracy": 0.2187960758805275,
      "num_tokens": 124494278.0,
      "step": 59990
    },
    {
      "entropy": 5.833430004119873,
      "epoch": 5.442216981132075,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002467371050606922,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21803678572177887,
      "num_tokens": 124507007.0,
      "step": 59995
    },
    {
      "entropy": 5.743976688385009,
      "epoch": 5.44267053701016,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024670500521758765,
      "loss": 4.8042,
      "mean_token_accuracy": 0.23643605709075927,
      "num_tokens": 124517288.0,
      "step": 60000
    },
    {
      "epoch": 5.44267053701016,
      "eval_entropy": 5.425237088505499,
      "eval_loss": 4.944544792175293,
      "eval_mean_token_accuracy": 0.2332236254410776,
      "eval_num_tokens": 124517288.0,
      "eval_runtime": 20.6391,
      "eval_samples_per_second": 1713.256,
      "eval_steps_per_second": 214.157,
      "step": 60000
    },
    {
      "entropy": 5.6261743068695065,
      "epoch": 5.443124092888244,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024667290595963085,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2338068351149559,
      "num_tokens": 124527079.0,
      "step": 60005
    },
    {
      "entropy": 5.660961151123047,
      "epoch": 5.443577648766328,
      "grad_norm": 1.5,
      "learning_rate": 0.0002466408072874856,
      "loss": 4.8211,
      "mean_token_accuracy": 0.2319677487015724,
      "num_tokens": 124536997.0,
      "step": 60010
    },
    {
      "entropy": 5.735938835144043,
      "epoch": 5.444031204644412,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024660870920181565,
      "loss": 4.838,
      "mean_token_accuracy": 0.2323254808783531,
      "num_tokens": 124547368.0,
      "step": 60015
    },
    {
      "entropy": 5.78380937576294,
      "epoch": 5.444484760522497,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002465766117032849,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22189629822969437,
      "num_tokens": 124557888.0,
      "step": 60020
    },
    {
      "entropy": 5.717818260192871,
      "epoch": 5.444938316400581,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000246544514792557,
      "loss": 4.8057,
      "mean_token_accuracy": 0.23079380691051482,
      "num_tokens": 124567495.0,
      "step": 60025
    },
    {
      "entropy": 5.738111782073974,
      "epoch": 5.445391872278664,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002465124184702958,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22541180849075318,
      "num_tokens": 124577893.0,
      "step": 60030
    },
    {
      "entropy": 5.7150129795074465,
      "epoch": 5.4458454281567485,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024648032273716505,
      "loss": 4.8563,
      "mean_token_accuracy": 0.23110387176275254,
      "num_tokens": 124588453.0,
      "step": 60035
    },
    {
      "entropy": 5.654474687576294,
      "epoch": 5.446298984034833,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024644822759382846,
      "loss": 4.7857,
      "mean_token_accuracy": 0.23940775245428086,
      "num_tokens": 124598726.0,
      "step": 60040
    },
    {
      "entropy": 5.763272523880005,
      "epoch": 5.446752539912917,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002464161330409498,
      "loss": 4.898,
      "mean_token_accuracy": 0.23207311630249022,
      "num_tokens": 124610218.0,
      "step": 60045
    },
    {
      "entropy": 5.752389001846313,
      "epoch": 5.447206095791001,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024638403907919284,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2224534422159195,
      "num_tokens": 124619920.0,
      "step": 60050
    },
    {
      "entropy": 5.808909511566162,
      "epoch": 5.447659651669086,
      "grad_norm": 1.5,
      "learning_rate": 0.0002463519457092212,
      "loss": 4.9897,
      "mean_token_accuracy": 0.2251580238342285,
      "num_tokens": 124630815.0,
      "step": 60055
    },
    {
      "entropy": 5.771755027770996,
      "epoch": 5.44811320754717,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024631985293169863,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2225772425532341,
      "num_tokens": 124641783.0,
      "step": 60060
    },
    {
      "entropy": 5.760356521606445,
      "epoch": 5.448566763425254,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002462877607472888,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22783650010824202,
      "num_tokens": 124650943.0,
      "step": 60065
    },
    {
      "entropy": 5.757129096984864,
      "epoch": 5.449020319303338,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024625566915665545,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22864413410425186,
      "num_tokens": 124662065.0,
      "step": 60070
    },
    {
      "entropy": 5.737003231048584,
      "epoch": 5.449473875181423,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024622357816046216,
      "loss": 4.8246,
      "mean_token_accuracy": 0.2265040695667267,
      "num_tokens": 124672071.0,
      "step": 60075
    },
    {
      "entropy": 5.779267168045044,
      "epoch": 5.449927431059507,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024619148775937254,
      "loss": 4.9251,
      "mean_token_accuracy": 0.22235020101070405,
      "num_tokens": 124683085.0,
      "step": 60080
    },
    {
      "entropy": 5.741007614135742,
      "epoch": 5.45038098693759,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024615939795405033,
      "loss": 4.9374,
      "mean_token_accuracy": 0.22275055199861526,
      "num_tokens": 124693026.0,
      "step": 60085
    },
    {
      "entropy": 5.736038064956665,
      "epoch": 5.450834542815675,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002461273087451592,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2184591442346573,
      "num_tokens": 124704307.0,
      "step": 60090
    },
    {
      "entropy": 5.674058437347412,
      "epoch": 5.451288098693759,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024609522013336265,
      "loss": 4.8182,
      "mean_token_accuracy": 0.22804437577724457,
      "num_tokens": 124715659.0,
      "step": 60095
    },
    {
      "entropy": 5.696075057983398,
      "epoch": 5.451741654571843,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024606313211932425,
      "loss": 4.8449,
      "mean_token_accuracy": 0.23306350260972977,
      "num_tokens": 124727064.0,
      "step": 60100
    },
    {
      "entropy": 5.71390814781189,
      "epoch": 5.452195210449927,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002460310447037077,
      "loss": 4.8674,
      "mean_token_accuracy": 0.23954787850379944,
      "num_tokens": 124737896.0,
      "step": 60105
    },
    {
      "entropy": 5.655920600891113,
      "epoch": 5.452648766328012,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002459989578871765,
      "loss": 4.7201,
      "mean_token_accuracy": 0.23699544817209245,
      "num_tokens": 124748785.0,
      "step": 60110
    },
    {
      "entropy": 5.761770486831665,
      "epoch": 5.453102322206096,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024596687167039425,
      "loss": 4.9057,
      "mean_token_accuracy": 0.228071491420269,
      "num_tokens": 124760372.0,
      "step": 60115
    },
    {
      "entropy": 5.742548656463623,
      "epoch": 5.45355587808418,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024593478605402445,
      "loss": 4.8722,
      "mean_token_accuracy": 0.2260774090886116,
      "num_tokens": 124771372.0,
      "step": 60120
    },
    {
      "entropy": 5.719311761856079,
      "epoch": 5.4540094339622645,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002459027010387308,
      "loss": 4.9093,
      "mean_token_accuracy": 0.23187429308891297,
      "num_tokens": 124781734.0,
      "step": 60125
    },
    {
      "entropy": 5.701080846786499,
      "epoch": 5.454462989840349,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002458706166251766,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2326062798500061,
      "num_tokens": 124792141.0,
      "step": 60130
    },
    {
      "entropy": 5.690635633468628,
      "epoch": 5.454916545718432,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002458385328140255,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22913851737976074,
      "num_tokens": 124803154.0,
      "step": 60135
    },
    {
      "entropy": 5.677285861968994,
      "epoch": 5.455370101596516,
      "grad_norm": 1.421875,
      "learning_rate": 0.000245806449605941,
      "loss": 4.8055,
      "mean_token_accuracy": 0.23012640327215195,
      "num_tokens": 124813160.0,
      "step": 60140
    },
    {
      "entropy": 5.771250247955322,
      "epoch": 5.455823657474601,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024577436700158655,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22764389514923095,
      "num_tokens": 124823445.0,
      "step": 60145
    },
    {
      "entropy": 5.704864454269409,
      "epoch": 5.456277213352685,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002457422850016257,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2292759120464325,
      "num_tokens": 124834598.0,
      "step": 60150
    },
    {
      "entropy": 5.693383646011353,
      "epoch": 5.456730769230769,
      "grad_norm": 1.5,
      "learning_rate": 0.0002457102036067217,
      "loss": 4.7981,
      "mean_token_accuracy": 0.2345191478729248,
      "num_tokens": 124844292.0,
      "step": 60155
    },
    {
      "entropy": 5.662764263153076,
      "epoch": 5.457184325108853,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024567812281753825,
      "loss": 4.843,
      "mean_token_accuracy": 0.22857985496520997,
      "num_tokens": 124854119.0,
      "step": 60160
    },
    {
      "entropy": 5.634636259078979,
      "epoch": 5.457637880986938,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002456460426347387,
      "loss": 4.726,
      "mean_token_accuracy": 0.24492726475000381,
      "num_tokens": 124864441.0,
      "step": 60165
    },
    {
      "entropy": 5.696662664413452,
      "epoch": 5.458091436865022,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024561396305898645,
      "loss": 4.9266,
      "mean_token_accuracy": 0.22269272953271865,
      "num_tokens": 124874825.0,
      "step": 60170
    },
    {
      "entropy": 5.713459253311157,
      "epoch": 5.458544992743106,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024558188409094496,
      "loss": 4.822,
      "mean_token_accuracy": 0.2308145970106125,
      "num_tokens": 124883929.0,
      "step": 60175
    },
    {
      "entropy": 5.6939061164855955,
      "epoch": 5.4589985486211905,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024554980573127756,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23098620921373367,
      "num_tokens": 124894829.0,
      "step": 60180
    },
    {
      "entropy": 5.704517602920532,
      "epoch": 5.459452104499274,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024551772798064766,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2340299591422081,
      "num_tokens": 124905403.0,
      "step": 60185
    },
    {
      "entropy": 5.711206102371216,
      "epoch": 5.459905660377358,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024548565083971875,
      "loss": 4.8988,
      "mean_token_accuracy": 0.2248219668865204,
      "num_tokens": 124915309.0,
      "step": 60190
    },
    {
      "entropy": 5.631240940093994,
      "epoch": 5.460359216255442,
      "grad_norm": 1.421875,
      "learning_rate": 0.000245453574309154,
      "loss": 4.7313,
      "mean_token_accuracy": 0.2453741803765297,
      "num_tokens": 124924341.0,
      "step": 60195
    },
    {
      "entropy": 5.72589225769043,
      "epoch": 5.460812772133527,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002454214983896169,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22152200490236282,
      "num_tokens": 124934031.0,
      "step": 60200
    },
    {
      "entropy": 5.8094391345977785,
      "epoch": 5.461266328011611,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002453894230817708,
      "loss": 4.9855,
      "mean_token_accuracy": 0.22258255779743194,
      "num_tokens": 124945059.0,
      "step": 60205
    },
    {
      "entropy": 5.824787616729736,
      "epoch": 5.461719883889695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024535734838627895,
      "loss": 4.9535,
      "mean_token_accuracy": 0.21945765912532805,
      "num_tokens": 124956182.0,
      "step": 60210
    },
    {
      "entropy": 5.77811017036438,
      "epoch": 5.4621734397677795,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002453252743038046,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22574836760759354,
      "num_tokens": 124967067.0,
      "step": 60215
    },
    {
      "entropy": 5.674193716049194,
      "epoch": 5.462626995645864,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002452932008350112,
      "loss": 4.7979,
      "mean_token_accuracy": 0.23256291151046754,
      "num_tokens": 124978270.0,
      "step": 60220
    },
    {
      "entropy": 5.748060655593872,
      "epoch": 5.463080551523948,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024526112798056206,
      "loss": 4.914,
      "mean_token_accuracy": 0.2217566579580307,
      "num_tokens": 124990406.0,
      "step": 60225
    },
    {
      "entropy": 5.703971099853516,
      "epoch": 5.463534107402032,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002452290557411203,
      "loss": 4.845,
      "mean_token_accuracy": 0.23166223764419555,
      "num_tokens": 125000165.0,
      "step": 60230
    },
    {
      "entropy": 5.812888145446777,
      "epoch": 5.4639876632801165,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024519698411734923,
      "loss": 5.0269,
      "mean_token_accuracy": 0.22105565369129182,
      "num_tokens": 125010815.0,
      "step": 60235
    },
    {
      "entropy": 5.814827728271484,
      "epoch": 5.4644412191582,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024516491310991214,
      "loss": 4.9433,
      "mean_token_accuracy": 0.22153593748807907,
      "num_tokens": 125020125.0,
      "step": 60240
    },
    {
      "entropy": 5.748838901519775,
      "epoch": 5.464894775036284,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002451328427194723,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22312618792057037,
      "num_tokens": 125030253.0,
      "step": 60245
    },
    {
      "entropy": 5.690258979797363,
      "epoch": 5.4653483309143684,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024510077294669286,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2286802649497986,
      "num_tokens": 125040949.0,
      "step": 60250
    },
    {
      "entropy": 5.60951828956604,
      "epoch": 5.465801886792453,
      "grad_norm": 1.46875,
      "learning_rate": 0.000245068703792237,
      "loss": 4.7741,
      "mean_token_accuracy": 0.23858009427785873,
      "num_tokens": 125050549.0,
      "step": 60255
    },
    {
      "entropy": 5.67435040473938,
      "epoch": 5.466255442670537,
      "grad_norm": 1.296875,
      "learning_rate": 0.000245036635256768,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22893085032701493,
      "num_tokens": 125061291.0,
      "step": 60260
    },
    {
      "entropy": 5.726225185394287,
      "epoch": 5.466708998548621,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024500456734094907,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22297041714191437,
      "num_tokens": 125071381.0,
      "step": 60265
    },
    {
      "entropy": 5.777706909179687,
      "epoch": 5.4671625544267055,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024497250004544335,
      "loss": 4.8427,
      "mean_token_accuracy": 0.23264690935611726,
      "num_tokens": 125081995.0,
      "step": 60270
    },
    {
      "entropy": 5.752083587646484,
      "epoch": 5.46761611030479,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024494043337091394,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22941183745861055,
      "num_tokens": 125092436.0,
      "step": 60275
    },
    {
      "entropy": 5.709315586090088,
      "epoch": 5.468069666182874,
      "grad_norm": 1.40625,
      "learning_rate": 0.000244908367318024,
      "loss": 4.8546,
      "mean_token_accuracy": 0.22857104390859603,
      "num_tokens": 125103228.0,
      "step": 60280
    },
    {
      "entropy": 5.735686922073365,
      "epoch": 5.468523222060958,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002448763018874367,
      "loss": 4.8481,
      "mean_token_accuracy": 0.23278786689043046,
      "num_tokens": 125113616.0,
      "step": 60285
    },
    {
      "entropy": 5.775871610641479,
      "epoch": 5.468976777939042,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002448442370798151,
      "loss": 4.9096,
      "mean_token_accuracy": 0.22145676016807556,
      "num_tokens": 125124586.0,
      "step": 60290
    },
    {
      "entropy": 5.71171236038208,
      "epoch": 5.469430333817126,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024481217289582246,
      "loss": 4.838,
      "mean_token_accuracy": 0.23616165816783904,
      "num_tokens": 125134457.0,
      "step": 60295
    },
    {
      "entropy": 5.668897247314453,
      "epoch": 5.46988388969521,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002447801093361217,
      "loss": 4.8264,
      "mean_token_accuracy": 0.23388934433460234,
      "num_tokens": 125145116.0,
      "step": 60300
    },
    {
      "entropy": 5.750236177444458,
      "epoch": 5.4703374455732945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024474804640137605,
      "loss": 4.923,
      "mean_token_accuracy": 0.22885446548461913,
      "num_tokens": 125156164.0,
      "step": 60305
    },
    {
      "entropy": 5.663012075424194,
      "epoch": 5.470791001451379,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002447159840922485,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23298720866441727,
      "num_tokens": 125165955.0,
      "step": 60310
    },
    {
      "entropy": 5.635988330841064,
      "epoch": 5.471244557329463,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024468392240940205,
      "loss": 4.8028,
      "mean_token_accuracy": 0.23358301669359208,
      "num_tokens": 125176801.0,
      "step": 60315
    },
    {
      "entropy": 5.787669372558594,
      "epoch": 5.471698113207547,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002446518613534999,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22044066339731216,
      "num_tokens": 125186192.0,
      "step": 60320
    },
    {
      "entropy": 5.722063875198364,
      "epoch": 5.4721516690856316,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002446198009252049,
      "loss": 4.8498,
      "mean_token_accuracy": 0.22788583487272263,
      "num_tokens": 125197542.0,
      "step": 60325
    },
    {
      "entropy": 5.718541193008423,
      "epoch": 5.472605224963716,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002445877411251802,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22891288846731186,
      "num_tokens": 125207066.0,
      "step": 60330
    },
    {
      "entropy": 5.669090986251831,
      "epoch": 5.4730587808418,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002445556819540887,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24304095804691314,
      "num_tokens": 125215745.0,
      "step": 60335
    },
    {
      "entropy": 5.7373494625091555,
      "epoch": 5.4735123367198835,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002445236234125936,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2246638536453247,
      "num_tokens": 125226976.0,
      "step": 60340
    },
    {
      "entropy": 5.767173719406128,
      "epoch": 5.473965892597968,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024449156550135766,
      "loss": 4.9418,
      "mean_token_accuracy": 0.22226358503103255,
      "num_tokens": 125239223.0,
      "step": 60345
    },
    {
      "entropy": 5.6636920928955075,
      "epoch": 5.474419448476052,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024445950822104385,
      "loss": 4.7582,
      "mean_token_accuracy": 0.2336656153202057,
      "num_tokens": 125248962.0,
      "step": 60350
    },
    {
      "entropy": 5.726318120956421,
      "epoch": 5.474873004354136,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002444274515723153,
      "loss": 4.809,
      "mean_token_accuracy": 0.23631831854581833,
      "num_tokens": 125258326.0,
      "step": 60355
    },
    {
      "entropy": 5.701107215881348,
      "epoch": 5.4753265602322205,
      "grad_norm": 1.375,
      "learning_rate": 0.0002443953955558348,
      "loss": 4.8404,
      "mean_token_accuracy": 0.2241917744278908,
      "num_tokens": 125267711.0,
      "step": 60360
    },
    {
      "entropy": 5.716433429718018,
      "epoch": 5.475780116110305,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024436334017226535,
      "loss": 4.902,
      "mean_token_accuracy": 0.2297538325190544,
      "num_tokens": 125279267.0,
      "step": 60365
    },
    {
      "entropy": 5.680195999145508,
      "epoch": 5.476233671988389,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002443312854222697,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22962617576122285,
      "num_tokens": 125290671.0,
      "step": 60370
    },
    {
      "entropy": 5.701076602935791,
      "epoch": 5.476687227866473,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024429923130651107,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23194500207901,
      "num_tokens": 125300387.0,
      "step": 60375
    },
    {
      "entropy": 5.779751968383789,
      "epoch": 5.477140783744558,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024426717782565204,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2177848145365715,
      "num_tokens": 125310940.0,
      "step": 60380
    },
    {
      "entropy": 5.730344343185425,
      "epoch": 5.477594339622642,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002442351249803556,
      "loss": 4.8165,
      "mean_token_accuracy": 0.2320250988006592,
      "num_tokens": 125320716.0,
      "step": 60385
    },
    {
      "entropy": 5.735587072372437,
      "epoch": 5.478047895500726,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002442030727712847,
      "loss": 4.8771,
      "mean_token_accuracy": 0.2217429205775261,
      "num_tokens": 125331175.0,
      "step": 60390
    },
    {
      "entropy": 5.7621994972229,
      "epoch": 5.4785014513788095,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024417102119910205,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22555911242961885,
      "num_tokens": 125341396.0,
      "step": 60395
    },
    {
      "entropy": 5.663523483276367,
      "epoch": 5.478955007256894,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024413897026447052,
      "loss": 4.7724,
      "mean_token_accuracy": 0.23635817766189576,
      "num_tokens": 125352032.0,
      "step": 60400
    },
    {
      "entropy": 5.629005432128906,
      "epoch": 5.479408563134978,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000244106919968053,
      "loss": 4.7855,
      "mean_token_accuracy": 0.23510270565748215,
      "num_tokens": 125362909.0,
      "step": 60405
    },
    {
      "entropy": 5.793703699111939,
      "epoch": 5.479862119013062,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024407487031051224,
      "loss": 5.0023,
      "mean_token_accuracy": 0.21861820071935653,
      "num_tokens": 125373880.0,
      "step": 60410
    },
    {
      "entropy": 5.677214097976685,
      "epoch": 5.480315674891147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024404282129251097,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23707489520311356,
      "num_tokens": 125384364.0,
      "step": 60415
    },
    {
      "entropy": 5.682159519195556,
      "epoch": 5.480769230769231,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024401077291471212,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23558493852615356,
      "num_tokens": 125394985.0,
      "step": 60420
    },
    {
      "entropy": 5.776448965072632,
      "epoch": 5.481222786647315,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024397872517777837,
      "loss": 4.9271,
      "mean_token_accuracy": 0.22915688902139664,
      "num_tokens": 125405734.0,
      "step": 60425
    },
    {
      "entropy": 5.740930223464966,
      "epoch": 5.481676342525399,
      "grad_norm": 1.375,
      "learning_rate": 0.00024394667808237246,
      "loss": 4.9289,
      "mean_token_accuracy": 0.22548158317804337,
      "num_tokens": 125417143.0,
      "step": 60430
    },
    {
      "entropy": 5.683354282379151,
      "epoch": 5.482129898403484,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024391463162915718,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23406743109226227,
      "num_tokens": 125426830.0,
      "step": 60435
    },
    {
      "entropy": 5.756804990768432,
      "epoch": 5.482583454281568,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024388258581879525,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22018461674451828,
      "num_tokens": 125438332.0,
      "step": 60440
    },
    {
      "entropy": 5.742878580093384,
      "epoch": 5.483037010159651,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024385054065194934,
      "loss": 4.8448,
      "mean_token_accuracy": 0.23613244742155076,
      "num_tokens": 125447313.0,
      "step": 60445
    },
    {
      "entropy": 5.74176664352417,
      "epoch": 5.4834905660377355,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024381849612928224,
      "loss": 4.8289,
      "mean_token_accuracy": 0.23474746495485305,
      "num_tokens": 125457758.0,
      "step": 60450
    },
    {
      "entropy": 5.681018543243408,
      "epoch": 5.48394412191582,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024378645225145657,
      "loss": 4.8377,
      "mean_token_accuracy": 0.23235289007425308,
      "num_tokens": 125467748.0,
      "step": 60455
    },
    {
      "entropy": 5.659612417221069,
      "epoch": 5.484397677793904,
      "grad_norm": 1.484375,
      "learning_rate": 0.000243754409019135,
      "loss": 4.7295,
      "mean_token_accuracy": 0.23686396181583405,
      "num_tokens": 125477968.0,
      "step": 60460
    },
    {
      "entropy": 5.766493511199951,
      "epoch": 5.484851233671988,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024372236643298024,
      "loss": 4.9724,
      "mean_token_accuracy": 0.2279788926243782,
      "num_tokens": 125487651.0,
      "step": 60465
    },
    {
      "entropy": 5.719802570343018,
      "epoch": 5.485304789550073,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002436903244936548,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22971915155649186,
      "num_tokens": 125498093.0,
      "step": 60470
    },
    {
      "entropy": 5.6885573387146,
      "epoch": 5.485758345428157,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024365828320182154,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23538705855607986,
      "num_tokens": 125508333.0,
      "step": 60475
    },
    {
      "entropy": 5.7452850341796875,
      "epoch": 5.486211901306241,
      "grad_norm": 1.5,
      "learning_rate": 0.00024362624255814302,
      "loss": 4.8803,
      "mean_token_accuracy": 0.22923081368207932,
      "num_tokens": 125518872.0,
      "step": 60480
    },
    {
      "entropy": 5.7087384223937985,
      "epoch": 5.486665457184325,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024359420256328169,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22360996752977372,
      "num_tokens": 125530944.0,
      "step": 60485
    },
    {
      "entropy": 5.690398168563843,
      "epoch": 5.48711901306241,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002435621632179003,
      "loss": 4.8473,
      "mean_token_accuracy": 0.2315176710486412,
      "num_tokens": 125541324.0,
      "step": 60490
    },
    {
      "entropy": 5.659850692749023,
      "epoch": 5.487572568940493,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024353012452266138,
      "loss": 4.8086,
      "mean_token_accuracy": 0.23110146671533585,
      "num_tokens": 125550489.0,
      "step": 60495
    },
    {
      "entropy": 5.749499893188476,
      "epoch": 5.488026124818577,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024349808647822752,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2249292567372322,
      "num_tokens": 125560602.0,
      "step": 60500
    },
    {
      "entropy": 5.699468946456909,
      "epoch": 5.488479680696662,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002434660490852612,
      "loss": 4.766,
      "mean_token_accuracy": 0.241663658618927,
      "num_tokens": 125570054.0,
      "step": 60505
    },
    {
      "entropy": 5.7052477359771725,
      "epoch": 5.488933236574746,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024343401234442507,
      "loss": 4.7879,
      "mean_token_accuracy": 0.23329288512468338,
      "num_tokens": 125579626.0,
      "step": 60510
    },
    {
      "entropy": 5.681002044677735,
      "epoch": 5.48938679245283,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024340197625638167,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23585010766983033,
      "num_tokens": 125590352.0,
      "step": 60515
    },
    {
      "entropy": 5.7035054683685305,
      "epoch": 5.489840348330914,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024336994082179336,
      "loss": 4.8581,
      "mean_token_accuracy": 0.2276200383901596,
      "num_tokens": 125600897.0,
      "step": 60520
    },
    {
      "entropy": 5.679373550415039,
      "epoch": 5.490293904208999,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024333790604132277,
      "loss": 4.7629,
      "mean_token_accuracy": 0.2316649943590164,
      "num_tokens": 125612223.0,
      "step": 60525
    },
    {
      "entropy": 5.785789203643799,
      "epoch": 5.490747460087083,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024330587191563236,
      "loss": 4.9957,
      "mean_token_accuracy": 0.2175307586789131,
      "num_tokens": 125623289.0,
      "step": 60530
    },
    {
      "entropy": 5.789287757873535,
      "epoch": 5.491201015965167,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002432738384453846,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2321199432015419,
      "num_tokens": 125633712.0,
      "step": 60535
    },
    {
      "entropy": 5.682990646362304,
      "epoch": 5.4916545718432515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024324180563124186,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22714804708957673,
      "num_tokens": 125644089.0,
      "step": 60540
    },
    {
      "entropy": 5.721901607513428,
      "epoch": 5.492108127721336,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024320977347386675,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22467835545539855,
      "num_tokens": 125653363.0,
      "step": 60545
    },
    {
      "entropy": 5.826946020126343,
      "epoch": 5.492561683599419,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002431777419739216,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2178540363907814,
      "num_tokens": 125663774.0,
      "step": 60550
    },
    {
      "entropy": 5.604766321182251,
      "epoch": 5.493015239477503,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024314571113206884,
      "loss": 4.7115,
      "mean_token_accuracy": 0.23514879494905472,
      "num_tokens": 125674225.0,
      "step": 60555
    },
    {
      "entropy": 5.7029636859893795,
      "epoch": 5.493468795355588,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024311368094897097,
      "loss": 4.8093,
      "mean_token_accuracy": 0.2308464840054512,
      "num_tokens": 125684128.0,
      "step": 60560
    },
    {
      "entropy": 5.732748079299927,
      "epoch": 5.493922351233672,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024308165142529026,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22054243534803392,
      "num_tokens": 125694704.0,
      "step": 60565
    },
    {
      "entropy": 5.738520383834839,
      "epoch": 5.494375907111756,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024304962256168911,
      "loss": 4.8953,
      "mean_token_accuracy": 0.23187794983386995,
      "num_tokens": 125705216.0,
      "step": 60570
    },
    {
      "entropy": 5.757718849182129,
      "epoch": 5.49482946298984,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024301759435882992,
      "loss": 4.8948,
      "mean_token_accuracy": 0.22771986871957778,
      "num_tokens": 125715690.0,
      "step": 60575
    },
    {
      "entropy": 5.720482587814331,
      "epoch": 5.495283018867925,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024298556681737505,
      "loss": 4.8962,
      "mean_token_accuracy": 0.23047335296869279,
      "num_tokens": 125726179.0,
      "step": 60580
    },
    {
      "entropy": 5.724341249465942,
      "epoch": 5.495736574746009,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002429535399379868,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22965803146362304,
      "num_tokens": 125736498.0,
      "step": 60585
    },
    {
      "entropy": 5.6575850486755375,
      "epoch": 5.496190130624093,
      "grad_norm": 1.375,
      "learning_rate": 0.00024292151372132747,
      "loss": 4.771,
      "mean_token_accuracy": 0.2282101735472679,
      "num_tokens": 125746560.0,
      "step": 60590
    },
    {
      "entropy": 5.6618577003479,
      "epoch": 5.496643686502177,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024288948816805945,
      "loss": 4.7553,
      "mean_token_accuracy": 0.23694079518318176,
      "num_tokens": 125756973.0,
      "step": 60595
    },
    {
      "entropy": 5.708988237380981,
      "epoch": 5.497097242380261,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000242857463278845,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2332185313105583,
      "num_tokens": 125767889.0,
      "step": 60600
    },
    {
      "entropy": 5.729518413543701,
      "epoch": 5.497550798258345,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002428254390543464,
      "loss": 4.897,
      "mean_token_accuracy": 0.2329436346888542,
      "num_tokens": 125778879.0,
      "step": 60605
    },
    {
      "entropy": 5.7576884746551515,
      "epoch": 5.498004354136429,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024279341549522594,
      "loss": 4.9116,
      "mean_token_accuracy": 0.22694217711687087,
      "num_tokens": 125788654.0,
      "step": 60610
    },
    {
      "entropy": 5.699247789382935,
      "epoch": 5.498457910014514,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002427613926021458,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22697532027959824,
      "num_tokens": 125799343.0,
      "step": 60615
    },
    {
      "entropy": 5.72918438911438,
      "epoch": 5.498911465892598,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024272937037576832,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2205032840371132,
      "num_tokens": 125810415.0,
      "step": 60620
    },
    {
      "entropy": 5.7152392864227295,
      "epoch": 5.499365021770682,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024269734881675564,
      "loss": 4.8243,
      "mean_token_accuracy": 0.2257908284664154,
      "num_tokens": 125821016.0,
      "step": 60625
    },
    {
      "entropy": 5.741867160797119,
      "epoch": 5.4998185776487665,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024266532792577007,
      "loss": 4.8472,
      "mean_token_accuracy": 0.22880442589521408,
      "num_tokens": 125830881.0,
      "step": 60630
    },
    {
      "entropy": 5.687870502471924,
      "epoch": 5.500272133526851,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024263330770347375,
      "loss": 4.7811,
      "mean_token_accuracy": 0.23588261306285857,
      "num_tokens": 125839896.0,
      "step": 60635
    },
    {
      "entropy": 5.755599164962769,
      "epoch": 5.500725689404935,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024260128815052883,
      "loss": 4.9128,
      "mean_token_accuracy": 0.22972602993249894,
      "num_tokens": 125850434.0,
      "step": 60640
    },
    {
      "entropy": 5.65266227722168,
      "epoch": 5.501179245283019,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002425692692675976,
      "loss": 4.7734,
      "mean_token_accuracy": 0.22908254265785216,
      "num_tokens": 125860806.0,
      "step": 60645
    },
    {
      "entropy": 5.7576299667358395,
      "epoch": 5.501632801161103,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024253725105534207,
      "loss": 4.9025,
      "mean_token_accuracy": 0.23388800621032715,
      "num_tokens": 125871649.0,
      "step": 60650
    },
    {
      "entropy": 5.691917514801025,
      "epoch": 5.502086357039187,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024250523351442455,
      "loss": 4.8695,
      "mean_token_accuracy": 0.21965485960245132,
      "num_tokens": 125881855.0,
      "step": 60655
    },
    {
      "entropy": 5.728518533706665,
      "epoch": 5.502539912917271,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024247321664550698,
      "loss": 4.8974,
      "mean_token_accuracy": 0.23016136437654494,
      "num_tokens": 125891147.0,
      "step": 60660
    },
    {
      "entropy": 5.722679233551025,
      "epoch": 5.5029934687953554,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024244120044925166,
      "loss": 4.8418,
      "mean_token_accuracy": 0.2319852441549301,
      "num_tokens": 125900165.0,
      "step": 60665
    },
    {
      "entropy": 5.727681589126587,
      "epoch": 5.50344702467344,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024240918492632058,
      "loss": 4.8057,
      "mean_token_accuracy": 0.22560434490442277,
      "num_tokens": 125910540.0,
      "step": 60670
    },
    {
      "entropy": 5.757113742828369,
      "epoch": 5.503900580551524,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024237717007737582,
      "loss": 4.8241,
      "mean_token_accuracy": 0.23062733560800552,
      "num_tokens": 125920704.0,
      "step": 60675
    },
    {
      "entropy": 5.774916267395019,
      "epoch": 5.504354136429608,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024234515590307955,
      "loss": 4.9253,
      "mean_token_accuracy": 0.22505100518465043,
      "num_tokens": 125931417.0,
      "step": 60680
    },
    {
      "entropy": 5.64520845413208,
      "epoch": 5.5048076923076925,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024231314240409375,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23856865465641022,
      "num_tokens": 125942429.0,
      "step": 60685
    },
    {
      "entropy": 5.706238603591919,
      "epoch": 5.505261248185777,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024228112958108052,
      "loss": 4.8153,
      "mean_token_accuracy": 0.2415058821439743,
      "num_tokens": 125952611.0,
      "step": 60690
    },
    {
      "entropy": 5.680508708953857,
      "epoch": 5.505714804063861,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024224911743470195,
      "loss": 4.8257,
      "mean_token_accuracy": 0.22830231636762618,
      "num_tokens": 125963156.0,
      "step": 60695
    },
    {
      "entropy": 5.693733263015747,
      "epoch": 5.506168359941945,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024221710596561985,
      "loss": 4.8348,
      "mean_token_accuracy": 0.22750174403190612,
      "num_tokens": 125974207.0,
      "step": 60700
    },
    {
      "entropy": 5.706805229187012,
      "epoch": 5.506621915820029,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002421850951744964,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2354757159948349,
      "num_tokens": 125984340.0,
      "step": 60705
    },
    {
      "entropy": 5.7625932693481445,
      "epoch": 5.507075471698113,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024215308506199352,
      "loss": 4.8978,
      "mean_token_accuracy": 0.23050085604190826,
      "num_tokens": 125994211.0,
      "step": 60710
    },
    {
      "entropy": 5.810981273651123,
      "epoch": 5.507529027576197,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024212107562877322,
      "loss": 4.9605,
      "mean_token_accuracy": 0.2233694761991501,
      "num_tokens": 126005352.0,
      "step": 60715
    },
    {
      "entropy": 5.738763618469238,
      "epoch": 5.5079825834542815,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024208906687549743,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22531809210777282,
      "num_tokens": 126015833.0,
      "step": 60720
    },
    {
      "entropy": 5.7529644012451175,
      "epoch": 5.508436139332366,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002420570588028282,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23117870539426805,
      "num_tokens": 126026556.0,
      "step": 60725
    },
    {
      "entropy": 5.719589471817017,
      "epoch": 5.50888969521045,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024202505141142732,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22907328307628633,
      "num_tokens": 126037561.0,
      "step": 60730
    },
    {
      "entropy": 5.75527925491333,
      "epoch": 5.509343251088534,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024199304470195676,
      "loss": 4.9743,
      "mean_token_accuracy": 0.22360544502735138,
      "num_tokens": 126048008.0,
      "step": 60735
    },
    {
      "entropy": 5.737393426895141,
      "epoch": 5.5097968069666186,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002419610386750785,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22499755918979644,
      "num_tokens": 126058622.0,
      "step": 60740
    },
    {
      "entropy": 5.793855905532837,
      "epoch": 5.510250362844703,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024192903333145437,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22949065119028092,
      "num_tokens": 126069235.0,
      "step": 60745
    },
    {
      "entropy": 5.682419681549073,
      "epoch": 5.510703918722786,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002418970286717463,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23077600002288817,
      "num_tokens": 126079247.0,
      "step": 60750
    },
    {
      "entropy": 5.7281652927398685,
      "epoch": 5.5111574746008705,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024186502469661602,
      "loss": 4.8836,
      "mean_token_accuracy": 0.21841977089643477,
      "num_tokens": 126089060.0,
      "step": 60755
    },
    {
      "entropy": 5.69375376701355,
      "epoch": 5.511611030478955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024183302140672554,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22474341988563537,
      "num_tokens": 126099988.0,
      "step": 60760
    },
    {
      "entropy": 5.722544860839844,
      "epoch": 5.512064586357039,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024180101880273664,
      "loss": 4.863,
      "mean_token_accuracy": 0.22972475588321686,
      "num_tokens": 126110102.0,
      "step": 60765
    },
    {
      "entropy": 5.786341428756714,
      "epoch": 5.512518142235123,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024176901688531104,
      "loss": 4.8669,
      "mean_token_accuracy": 0.23351388722658156,
      "num_tokens": 126120342.0,
      "step": 60770
    },
    {
      "entropy": 5.741907978057862,
      "epoch": 5.5129716981132075,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002417370156551107,
      "loss": 4.9094,
      "mean_token_accuracy": 0.2254019245505333,
      "num_tokens": 126131121.0,
      "step": 60775
    },
    {
      "entropy": 5.762955999374389,
      "epoch": 5.513425253991292,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024170501511279734,
      "loss": 4.8916,
      "mean_token_accuracy": 0.22982196807861327,
      "num_tokens": 126141004.0,
      "step": 60780
    },
    {
      "entropy": 5.714068365097046,
      "epoch": 5.513878809869376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024167301525903273,
      "loss": 4.8549,
      "mean_token_accuracy": 0.2355598881840706,
      "num_tokens": 126150692.0,
      "step": 60785
    },
    {
      "entropy": 5.768978929519653,
      "epoch": 5.51433236574746,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024164101609447868,
      "loss": 4.929,
      "mean_token_accuracy": 0.2230737715959549,
      "num_tokens": 126161659.0,
      "step": 60790
    },
    {
      "entropy": 5.651105499267578,
      "epoch": 5.514785921625545,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024160901761979698,
      "loss": 4.8343,
      "mean_token_accuracy": 0.22865371257066727,
      "num_tokens": 126171958.0,
      "step": 60795
    },
    {
      "entropy": 5.723346948623657,
      "epoch": 5.515239477503629,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024157701983564927,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22877622097730638,
      "num_tokens": 126181491.0,
      "step": 60800
    },
    {
      "entropy": 5.7395836353302006,
      "epoch": 5.515693033381712,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024154502274269723,
      "loss": 4.882,
      "mean_token_accuracy": 0.22712644189596176,
      "num_tokens": 126192318.0,
      "step": 60805
    },
    {
      "entropy": 5.706558275222778,
      "epoch": 5.5161465892597965,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024151302634160276,
      "loss": 4.8245,
      "mean_token_accuracy": 0.23146522492170335,
      "num_tokens": 126202375.0,
      "step": 60810
    },
    {
      "entropy": 5.6870279788970945,
      "epoch": 5.516600145137881,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002414810306330274,
      "loss": 4.8247,
      "mean_token_accuracy": 0.23350509703159333,
      "num_tokens": 126211172.0,
      "step": 60815
    },
    {
      "entropy": 5.758696508407593,
      "epoch": 5.517053701015965,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002414490356176328,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22680942267179488,
      "num_tokens": 126221040.0,
      "step": 60820
    },
    {
      "entropy": 5.645207071304322,
      "epoch": 5.517507256894049,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024141704129608077,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23313572406768798,
      "num_tokens": 126230016.0,
      "step": 60825
    },
    {
      "entropy": 5.73769702911377,
      "epoch": 5.517960812772134,
      "grad_norm": 1.375,
      "learning_rate": 0.00024138504766903286,
      "loss": 4.9115,
      "mean_token_accuracy": 0.22844984382390976,
      "num_tokens": 126240343.0,
      "step": 60830
    },
    {
      "entropy": 5.742397308349609,
      "epoch": 5.518414368650218,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024135305473715074,
      "loss": 4.8954,
      "mean_token_accuracy": 0.2243331179022789,
      "num_tokens": 126251063.0,
      "step": 60835
    },
    {
      "entropy": 5.688250541687012,
      "epoch": 5.518867924528302,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024132106250109598,
      "loss": 4.7692,
      "mean_token_accuracy": 0.23476049602031707,
      "num_tokens": 126260312.0,
      "step": 60840
    },
    {
      "entropy": 5.663962173461914,
      "epoch": 5.519321480406386,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024128907096153024,
      "loss": 4.7751,
      "mean_token_accuracy": 0.23591236174106597,
      "num_tokens": 126270607.0,
      "step": 60845
    },
    {
      "entropy": 5.606388282775879,
      "epoch": 5.51977503628447,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024125708011911518,
      "loss": 4.7289,
      "mean_token_accuracy": 0.23867609351873398,
      "num_tokens": 126280885.0,
      "step": 60850
    },
    {
      "entropy": 5.69071774482727,
      "epoch": 5.520228592162555,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002412250899745122,
      "loss": 4.9954,
      "mean_token_accuracy": 0.21653251647949218,
      "num_tokens": 126291206.0,
      "step": 60855
    },
    {
      "entropy": 5.642111921310425,
      "epoch": 5.520682148040638,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024119310052838301,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23067246079444886,
      "num_tokens": 126302434.0,
      "step": 60860
    },
    {
      "entropy": 5.75415825843811,
      "epoch": 5.5211357039187225,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024116111178138918,
      "loss": 4.9035,
      "mean_token_accuracy": 0.22440622746944427,
      "num_tokens": 126313114.0,
      "step": 60865
    },
    {
      "entropy": 5.72152886390686,
      "epoch": 5.521589259796807,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002411291237341921,
      "loss": 4.7956,
      "mean_token_accuracy": 0.2371226206421852,
      "num_tokens": 126323172.0,
      "step": 60870
    },
    {
      "entropy": 5.679473972320556,
      "epoch": 5.522042815674891,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024109713638745335,
      "loss": 4.7836,
      "mean_token_accuracy": 0.23235729485750198,
      "num_tokens": 126334690.0,
      "step": 60875
    },
    {
      "entropy": 5.692976284027099,
      "epoch": 5.522496371552975,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024106514974183452,
      "loss": 4.8076,
      "mean_token_accuracy": 0.2404639795422554,
      "num_tokens": 126345097.0,
      "step": 60880
    },
    {
      "entropy": 5.710638618469238,
      "epoch": 5.52294992743106,
      "grad_norm": 1.5,
      "learning_rate": 0.000241033163797997,
      "loss": 4.853,
      "mean_token_accuracy": 0.22837281823158265,
      "num_tokens": 126356130.0,
      "step": 60885
    },
    {
      "entropy": 5.688620281219483,
      "epoch": 5.523403483309144,
      "grad_norm": 1.765625,
      "learning_rate": 0.0002410011785566023,
      "loss": 4.8455,
      "mean_token_accuracy": 0.22933405339717866,
      "num_tokens": 126366640.0,
      "step": 60890
    },
    {
      "entropy": 5.780199193954468,
      "epoch": 5.523857039187228,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024096919401831197,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2162567511200905,
      "num_tokens": 126378465.0,
      "step": 60895
    },
    {
      "entropy": 5.701440382003784,
      "epoch": 5.524310595065312,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024093721018378733,
      "loss": 4.8175,
      "mean_token_accuracy": 0.23009692877531052,
      "num_tokens": 126388506.0,
      "step": 60900
    },
    {
      "entropy": 5.748069381713867,
      "epoch": 5.524764150943396,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024090522705368984,
      "loss": 4.9025,
      "mean_token_accuracy": 0.21922058165073394,
      "num_tokens": 126398674.0,
      "step": 60905
    },
    {
      "entropy": 5.625267791748047,
      "epoch": 5.52521770682148,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024087324462868098,
      "loss": 4.7237,
      "mean_token_accuracy": 0.24445029199123383,
      "num_tokens": 126409671.0,
      "step": 60910
    },
    {
      "entropy": 5.721931838989258,
      "epoch": 5.525671262699564,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024084126290942215,
      "loss": 4.9413,
      "mean_token_accuracy": 0.22845844626426698,
      "num_tokens": 126419797.0,
      "step": 60915
    },
    {
      "entropy": 5.730300617218018,
      "epoch": 5.526124818577649,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024080928189657464,
      "loss": 4.9401,
      "mean_token_accuracy": 0.22168278247117995,
      "num_tokens": 126430964.0,
      "step": 60920
    },
    {
      "entropy": 5.740819835662842,
      "epoch": 5.526578374455733,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002407773015907999,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2272664025425911,
      "num_tokens": 126440909.0,
      "step": 60925
    },
    {
      "entropy": 5.749761486053467,
      "epoch": 5.527031930333817,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024074532199275928,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2301769509911537,
      "num_tokens": 126450492.0,
      "step": 60930
    },
    {
      "entropy": 5.681548738479615,
      "epoch": 5.527485486211901,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024071334310311415,
      "loss": 4.841,
      "mean_token_accuracy": 0.23002849221229554,
      "num_tokens": 126462209.0,
      "step": 60935
    },
    {
      "entropy": 5.753965568542481,
      "epoch": 5.527939042089986,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024068136492252584,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22541719824075698,
      "num_tokens": 126473226.0,
      "step": 60940
    },
    {
      "entropy": 5.7603813171386715,
      "epoch": 5.52839259796807,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024064938745165564,
      "loss": 4.8442,
      "mean_token_accuracy": 0.23630797117948532,
      "num_tokens": 126483419.0,
      "step": 60945
    },
    {
      "entropy": 5.715794038772583,
      "epoch": 5.528846153846154,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024061741069116488,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2350405991077423,
      "num_tokens": 126494076.0,
      "step": 60950
    },
    {
      "entropy": 5.741288661956787,
      "epoch": 5.5292997097242385,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002405854346417148,
      "loss": 4.8954,
      "mean_token_accuracy": 0.22792763859033585,
      "num_tokens": 126504339.0,
      "step": 60955
    },
    {
      "entropy": 5.613232135772705,
      "epoch": 5.529753265602322,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024055345930396667,
      "loss": 4.714,
      "mean_token_accuracy": 0.2388572335243225,
      "num_tokens": 126514110.0,
      "step": 60960
    },
    {
      "entropy": 5.649919509887695,
      "epoch": 5.530206821480406,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024052148467858183,
      "loss": 4.7841,
      "mean_token_accuracy": 0.23458351641893388,
      "num_tokens": 126524520.0,
      "step": 60965
    },
    {
      "entropy": 5.69794716835022,
      "epoch": 5.53066037735849,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024048951076622145,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22699105143547058,
      "num_tokens": 126536069.0,
      "step": 60970
    },
    {
      "entropy": 5.658146858215332,
      "epoch": 5.531113933236575,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024045753756754673,
      "loss": 4.747,
      "mean_token_accuracy": 0.24211256504058837,
      "num_tokens": 126545374.0,
      "step": 60975
    },
    {
      "entropy": 5.682596206665039,
      "epoch": 5.531567489114659,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024042556508321904,
      "loss": 4.8659,
      "mean_token_accuracy": 0.23006605952978135,
      "num_tokens": 126555995.0,
      "step": 60980
    },
    {
      "entropy": 5.681816864013672,
      "epoch": 5.532021044992743,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024039359331389944,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2305373191833496,
      "num_tokens": 126566281.0,
      "step": 60985
    },
    {
      "entropy": 5.716967821121216,
      "epoch": 5.532474600870827,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0002403616222602491,
      "loss": 4.8705,
      "mean_token_accuracy": 0.23679172694683076,
      "num_tokens": 126575333.0,
      "step": 60990
    },
    {
      "entropy": 5.7135130882263185,
      "epoch": 5.532928156748912,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024032965192292928,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23075680583715438,
      "num_tokens": 126586071.0,
      "step": 60995
    },
    {
      "entropy": 5.652644872665405,
      "epoch": 5.533381712626996,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024029768230260113,
      "loss": 4.8402,
      "mean_token_accuracy": 0.2249441549181938,
      "num_tokens": 126596641.0,
      "step": 61000
    },
    {
      "entropy": 5.701812601089477,
      "epoch": 5.533835268505079,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002402657133999257,
      "loss": 4.8864,
      "mean_token_accuracy": 0.2321029856801033,
      "num_tokens": 126607521.0,
      "step": 61005
    },
    {
      "entropy": 5.701250505447388,
      "epoch": 5.5342888243831645,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024023374521556413,
      "loss": 4.7997,
      "mean_token_accuracy": 0.23630696088075637,
      "num_tokens": 126617513.0,
      "step": 61010
    },
    {
      "entropy": 5.693952417373657,
      "epoch": 5.534742380261248,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024020177775017766,
      "loss": 4.8133,
      "mean_token_accuracy": 0.23605897277593613,
      "num_tokens": 126627552.0,
      "step": 61015
    },
    {
      "entropy": 5.673954677581787,
      "epoch": 5.535195936139332,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024016981100442725,
      "loss": 4.8316,
      "mean_token_accuracy": 0.2342689484357834,
      "num_tokens": 126637587.0,
      "step": 61020
    },
    {
      "entropy": 5.628733730316162,
      "epoch": 5.535649492017416,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024013784497897402,
      "loss": 4.7152,
      "mean_token_accuracy": 0.24162406027317046,
      "num_tokens": 126647123.0,
      "step": 61025
    },
    {
      "entropy": 5.736734056472779,
      "epoch": 5.536103047895501,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024010587967447904,
      "loss": 4.8945,
      "mean_token_accuracy": 0.22766339927911758,
      "num_tokens": 126657489.0,
      "step": 61030
    },
    {
      "entropy": 5.764311647415161,
      "epoch": 5.536556603773585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024007391509160337,
      "loss": 4.858,
      "mean_token_accuracy": 0.2302749291062355,
      "num_tokens": 126667691.0,
      "step": 61035
    },
    {
      "entropy": 5.767987442016602,
      "epoch": 5.537010159651669,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024004195123100808,
      "loss": 4.9141,
      "mean_token_accuracy": 0.22086220234632492,
      "num_tokens": 126678614.0,
      "step": 61040
    },
    {
      "entropy": 5.7172043800354,
      "epoch": 5.5374637155297535,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024000998809335406,
      "loss": 4.8565,
      "mean_token_accuracy": 0.23122409582138062,
      "num_tokens": 126688241.0,
      "step": 61045
    },
    {
      "entropy": 5.711239528656006,
      "epoch": 5.537917271407838,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023997802567930245,
      "loss": 4.8333,
      "mean_token_accuracy": 0.23144749104976653,
      "num_tokens": 126698056.0,
      "step": 61050
    },
    {
      "entropy": 5.679987859725952,
      "epoch": 5.538370827285922,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023994606398951418,
      "loss": 4.7903,
      "mean_token_accuracy": 0.2424955725669861,
      "num_tokens": 126709336.0,
      "step": 61055
    },
    {
      "entropy": 5.713873720169067,
      "epoch": 5.538824383164005,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023991410302465023,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2308898389339447,
      "num_tokens": 126720917.0,
      "step": 61060
    },
    {
      "entropy": 5.686146688461304,
      "epoch": 5.53927793904209,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023988214278537157,
      "loss": 4.8379,
      "mean_token_accuracy": 0.22043481916189195,
      "num_tokens": 126732059.0,
      "step": 61065
    },
    {
      "entropy": 5.742831993103027,
      "epoch": 5.539731494920174,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002398501832723392,
      "loss": 4.8732,
      "mean_token_accuracy": 0.22956515848636627,
      "num_tokens": 126743354.0,
      "step": 61070
    },
    {
      "entropy": 5.701641607284546,
      "epoch": 5.540185050798258,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023981822448621397,
      "loss": 4.8224,
      "mean_token_accuracy": 0.2374078407883644,
      "num_tokens": 126753468.0,
      "step": 61075
    },
    {
      "entropy": 5.721153259277344,
      "epoch": 5.5406386066763424,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00023978626642765676,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2347380429506302,
      "num_tokens": 126763229.0,
      "step": 61080
    },
    {
      "entropy": 5.71700119972229,
      "epoch": 5.541092162554427,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002397543090973286,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23238044530153273,
      "num_tokens": 126773997.0,
      "step": 61085
    },
    {
      "entropy": 5.720271587371826,
      "epoch": 5.541545718432511,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023972235249589035,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23149219453334807,
      "num_tokens": 126784124.0,
      "step": 61090
    },
    {
      "entropy": 5.789032316207885,
      "epoch": 5.541999274310595,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023969039662400271,
      "loss": 4.9308,
      "mean_token_accuracy": 0.23060414046049119,
      "num_tokens": 126795206.0,
      "step": 61095
    },
    {
      "entropy": 5.743089914321899,
      "epoch": 5.5424528301886795,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023965844148232673,
      "loss": 4.9479,
      "mean_token_accuracy": 0.22326236814260483,
      "num_tokens": 126805798.0,
      "step": 61100
    },
    {
      "entropy": 5.702437686920166,
      "epoch": 5.542906386066764,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023962648707152322,
      "loss": 4.812,
      "mean_token_accuracy": 0.24463436156511306,
      "num_tokens": 126816189.0,
      "step": 61105
    },
    {
      "entropy": 5.7386534214019775,
      "epoch": 5.543359941944848,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023959453339225296,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22458382844924926,
      "num_tokens": 126827219.0,
      "step": 61110
    },
    {
      "entropy": 5.802410840988159,
      "epoch": 5.543813497822931,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023956258044517675,
      "loss": 4.935,
      "mean_token_accuracy": 0.2251037433743477,
      "num_tokens": 126836930.0,
      "step": 61115
    },
    {
      "entropy": 5.7191712856292725,
      "epoch": 5.544267053701016,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023953062823095546,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2302938312292099,
      "num_tokens": 126846163.0,
      "step": 61120
    },
    {
      "entropy": 5.693466663360596,
      "epoch": 5.5447206095791,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023949867675024982,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22965019345283508,
      "num_tokens": 126857141.0,
      "step": 61125
    },
    {
      "entropy": 5.765407228469849,
      "epoch": 5.545174165457184,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002394667260037206,
      "loss": 4.9555,
      "mean_token_accuracy": 0.23211514502763747,
      "num_tokens": 126866872.0,
      "step": 61130
    },
    {
      "entropy": 5.627369689941406,
      "epoch": 5.5456277213352685,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023943477599202846,
      "loss": 4.694,
      "mean_token_accuracy": 0.2448469579219818,
      "num_tokens": 126877367.0,
      "step": 61135
    },
    {
      "entropy": 5.703247594833374,
      "epoch": 5.546081277213353,
      "grad_norm": 1.375,
      "learning_rate": 0.0002394028267158342,
      "loss": 4.8872,
      "mean_token_accuracy": 0.2254680171608925,
      "num_tokens": 126888161.0,
      "step": 61140
    },
    {
      "entropy": 5.747354936599732,
      "epoch": 5.546534833091437,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002393708781757986,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2304744690656662,
      "num_tokens": 126897981.0,
      "step": 61145
    },
    {
      "entropy": 5.734670209884643,
      "epoch": 5.546988388969521,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023933893037258235,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2302619159221649,
      "num_tokens": 126908696.0,
      "step": 61150
    },
    {
      "entropy": 5.815563917160034,
      "epoch": 5.5474419448476056,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023930698330684602,
      "loss": 4.924,
      "mean_token_accuracy": 0.21962177008390427,
      "num_tokens": 126919336.0,
      "step": 61155
    },
    {
      "entropy": 5.766284608840943,
      "epoch": 5.547895500725689,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023927503697925045,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2312251329421997,
      "num_tokens": 126929975.0,
      "step": 61160
    },
    {
      "entropy": 5.737932586669922,
      "epoch": 5.548349056603773,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023924309139045625,
      "loss": 4.8604,
      "mean_token_accuracy": 0.23116905987262726,
      "num_tokens": 126940489.0,
      "step": 61165
    },
    {
      "entropy": 5.666020917892456,
      "epoch": 5.5488026124818575,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023921114654112396,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22817158252000808,
      "num_tokens": 126950681.0,
      "step": 61170
    },
    {
      "entropy": 5.729557514190674,
      "epoch": 5.549256168359942,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023917920243191426,
      "loss": 4.8629,
      "mean_token_accuracy": 0.2256164252758026,
      "num_tokens": 126961267.0,
      "step": 61175
    },
    {
      "entropy": 5.796693992614746,
      "epoch": 5.549709724238026,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023914725906348783,
      "loss": 4.9462,
      "mean_token_accuracy": 0.22018204182386397,
      "num_tokens": 126972100.0,
      "step": 61180
    },
    {
      "entropy": 5.698697662353515,
      "epoch": 5.55016328011611,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023911531643650526,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2312246486544609,
      "num_tokens": 126982806.0,
      "step": 61185
    },
    {
      "entropy": 5.686219024658203,
      "epoch": 5.5506168359941945,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023908337455162698,
      "loss": 4.7656,
      "mean_token_accuracy": 0.24144561886787413,
      "num_tokens": 126992218.0,
      "step": 61190
    },
    {
      "entropy": 5.666381740570069,
      "epoch": 5.551070391872279,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023905143340951373,
      "loss": 4.7639,
      "mean_token_accuracy": 0.24183388650417328,
      "num_tokens": 127002783.0,
      "step": 61195
    },
    {
      "entropy": 5.724459314346314,
      "epoch": 5.551523947750363,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000239019493010826,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2232982873916626,
      "num_tokens": 127013720.0,
      "step": 61200
    },
    {
      "entropy": 5.771685409545898,
      "epoch": 5.551977503628447,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002389875533562243,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22996153235435485,
      "num_tokens": 127024443.0,
      "step": 61205
    },
    {
      "entropy": 5.627539110183716,
      "epoch": 5.552431059506532,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023895561444636914,
      "loss": 4.7078,
      "mean_token_accuracy": 0.23731009811162948,
      "num_tokens": 127035565.0,
      "step": 61210
    },
    {
      "entropy": 5.651168346405029,
      "epoch": 5.552884615384615,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023892367628192119,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23574999570846558,
      "num_tokens": 127044762.0,
      "step": 61215
    },
    {
      "entropy": 5.651391220092774,
      "epoch": 5.553338171262699,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023889173886354072,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23182796090841293,
      "num_tokens": 127054893.0,
      "step": 61220
    },
    {
      "entropy": 5.699566268920899,
      "epoch": 5.5537917271407835,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023885980219188829,
      "loss": 4.88,
      "mean_token_accuracy": 0.23242794275283812,
      "num_tokens": 127065063.0,
      "step": 61225
    },
    {
      "entropy": 5.68003888130188,
      "epoch": 5.554245283018868,
      "grad_norm": 1.625,
      "learning_rate": 0.00023882786626762436,
      "loss": 4.8299,
      "mean_token_accuracy": 0.23100514113903045,
      "num_tokens": 127075894.0,
      "step": 61230
    },
    {
      "entropy": 5.765844345092773,
      "epoch": 5.554698838896952,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002387959310914094,
      "loss": 4.8852,
      "mean_token_accuracy": 0.23355148434638978,
      "num_tokens": 127086299.0,
      "step": 61235
    },
    {
      "entropy": 5.719778633117675,
      "epoch": 5.555152394775036,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023876399666390374,
      "loss": 4.8683,
      "mean_token_accuracy": 0.22506873160600663,
      "num_tokens": 127096913.0,
      "step": 61240
    },
    {
      "entropy": 5.782762813568115,
      "epoch": 5.555605950653121,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023873206298576793,
      "loss": 4.9484,
      "mean_token_accuracy": 0.21988336443901063,
      "num_tokens": 127107467.0,
      "step": 61245
    },
    {
      "entropy": 5.765958595275879,
      "epoch": 5.556059506531205,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002387001300576623,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21128684729337693,
      "num_tokens": 127119121.0,
      "step": 61250
    },
    {
      "entropy": 5.713018131256104,
      "epoch": 5.556513062409289,
      "grad_norm": 1.375,
      "learning_rate": 0.00023866819788024725,
      "loss": 4.8435,
      "mean_token_accuracy": 0.2265893816947937,
      "num_tokens": 127129305.0,
      "step": 61255
    },
    {
      "entropy": 5.711279392242432,
      "epoch": 5.556966618287373,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023863626645418301,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22727397829294205,
      "num_tokens": 127138867.0,
      "step": 61260
    },
    {
      "entropy": 5.752798175811767,
      "epoch": 5.557420174165458,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023860433578013014,
      "loss": 4.895,
      "mean_token_accuracy": 0.22807956784963607,
      "num_tokens": 127148702.0,
      "step": 61265
    },
    {
      "entropy": 5.719301223754883,
      "epoch": 5.557873730043541,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002385724058587489,
      "loss": 4.8177,
      "mean_token_accuracy": 0.24155167043209075,
      "num_tokens": 127158533.0,
      "step": 61270
    },
    {
      "entropy": 5.680095052719116,
      "epoch": 5.558327285921625,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002385404766906995,
      "loss": 4.747,
      "mean_token_accuracy": 0.24127518832683564,
      "num_tokens": 127169115.0,
      "step": 61275
    },
    {
      "entropy": 5.67265625,
      "epoch": 5.5587808417997095,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023850854827664242,
      "loss": 4.8404,
      "mean_token_accuracy": 0.23173908591270448,
      "num_tokens": 127180396.0,
      "step": 61280
    },
    {
      "entropy": 5.677108669281006,
      "epoch": 5.559234397677794,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023847662061723786,
      "loss": 4.8355,
      "mean_token_accuracy": 0.23343948274850845,
      "num_tokens": 127191341.0,
      "step": 61285
    },
    {
      "entropy": 5.599442052841186,
      "epoch": 5.559687953555878,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002384446937131461,
      "loss": 4.7448,
      "mean_token_accuracy": 0.24159968495368958,
      "num_tokens": 127201321.0,
      "step": 61290
    },
    {
      "entropy": 5.773750448226929,
      "epoch": 5.560141509433962,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023841276756502735,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22737145572900772,
      "num_tokens": 127211490.0,
      "step": 61295
    },
    {
      "entropy": 5.745524501800537,
      "epoch": 5.560595065312047,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002383808421735419,
      "loss": 4.9055,
      "mean_token_accuracy": 0.22131581157445906,
      "num_tokens": 127221719.0,
      "step": 61300
    },
    {
      "entropy": 5.719996547698974,
      "epoch": 5.561048621190131,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023834891753934996,
      "loss": 4.8802,
      "mean_token_accuracy": 0.2249760314822197,
      "num_tokens": 127232543.0,
      "step": 61305
    },
    {
      "entropy": 5.724157524108887,
      "epoch": 5.561502177068215,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023831699366311173,
      "loss": 4.8988,
      "mean_token_accuracy": 0.2289251998066902,
      "num_tokens": 127242549.0,
      "step": 61310
    },
    {
      "entropy": 5.758150815963745,
      "epoch": 5.5619557329462985,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023828507054548737,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2327059969305992,
      "num_tokens": 127252539.0,
      "step": 61315
    },
    {
      "entropy": 5.762049770355224,
      "epoch": 5.562409288824383,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023825314818713721,
      "loss": 4.8667,
      "mean_token_accuracy": 0.23178091943264006,
      "num_tokens": 127262639.0,
      "step": 61320
    },
    {
      "entropy": 5.7273002624511715,
      "epoch": 5.562862844702467,
      "grad_norm": 1.5,
      "learning_rate": 0.00023822122658872125,
      "loss": 4.828,
      "mean_token_accuracy": 0.2305402636528015,
      "num_tokens": 127272609.0,
      "step": 61325
    },
    {
      "entropy": 5.70020604133606,
      "epoch": 5.563316400580551,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002381893057508997,
      "loss": 4.9359,
      "mean_token_accuracy": 0.22984673976898193,
      "num_tokens": 127283539.0,
      "step": 61330
    },
    {
      "entropy": 5.725558233261109,
      "epoch": 5.563769956458636,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023815738567433277,
      "loss": 4.955,
      "mean_token_accuracy": 0.21957223117351532,
      "num_tokens": 127294566.0,
      "step": 61335
    },
    {
      "entropy": 5.673550319671631,
      "epoch": 5.56422351233672,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023812546635968042,
      "loss": 4.7952,
      "mean_token_accuracy": 0.2346173033118248,
      "num_tokens": 127304197.0,
      "step": 61340
    },
    {
      "entropy": 5.681006193161011,
      "epoch": 5.564677068214804,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023809354780760279,
      "loss": 4.7356,
      "mean_token_accuracy": 0.239576855301857,
      "num_tokens": 127313834.0,
      "step": 61345
    },
    {
      "entropy": 5.667253828048706,
      "epoch": 5.565130624092888,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023806163001876009,
      "loss": 4.8225,
      "mean_token_accuracy": 0.23047062307596206,
      "num_tokens": 127324880.0,
      "step": 61350
    },
    {
      "entropy": 5.650677967071533,
      "epoch": 5.565584179970973,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023802971299381222,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22508554309606552,
      "num_tokens": 127334732.0,
      "step": 61355
    },
    {
      "entropy": 5.6561178207397464,
      "epoch": 5.566037735849057,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002379977967334193,
      "loss": 4.7897,
      "mean_token_accuracy": 0.242897866666317,
      "num_tokens": 127345224.0,
      "step": 61360
    },
    {
      "entropy": 5.719110441207886,
      "epoch": 5.566491291727141,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023796588123824143,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22191160619258882,
      "num_tokens": 127356041.0,
      "step": 61365
    },
    {
      "entropy": 5.708858633041382,
      "epoch": 5.566944847605225,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023793396650893852,
      "loss": 4.8248,
      "mean_token_accuracy": 0.2343386173248291,
      "num_tokens": 127365576.0,
      "step": 61370
    },
    {
      "entropy": 5.762853574752808,
      "epoch": 5.567398403483309,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023790205254617063,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21783490628004074,
      "num_tokens": 127375933.0,
      "step": 61375
    },
    {
      "entropy": 5.683960628509522,
      "epoch": 5.567851959361393,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023787013935059776,
      "loss": 4.888,
      "mean_token_accuracy": 0.22080762088298797,
      "num_tokens": 127386187.0,
      "step": 61380
    },
    {
      "entropy": 5.759936237335205,
      "epoch": 5.568305515239477,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023783822692287988,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2278902754187584,
      "num_tokens": 127395537.0,
      "step": 61385
    },
    {
      "entropy": 5.7492725372314455,
      "epoch": 5.568759071117562,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023780631526367697,
      "loss": 4.8487,
      "mean_token_accuracy": 0.22202683240175247,
      "num_tokens": 127405745.0,
      "step": 61390
    },
    {
      "entropy": 5.675144338607788,
      "epoch": 5.569212626995646,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023777440437364888,
      "loss": 4.8516,
      "mean_token_accuracy": 0.23118832111358642,
      "num_tokens": 127415430.0,
      "step": 61395
    },
    {
      "entropy": 5.733362627029419,
      "epoch": 5.56966618287373,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023774249425345562,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22491608262062074,
      "num_tokens": 127426058.0,
      "step": 61400
    },
    {
      "entropy": 5.7560488224029545,
      "epoch": 5.570119738751814,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023771058490375707,
      "loss": 4.8545,
      "mean_token_accuracy": 0.22934226989746093,
      "num_tokens": 127437401.0,
      "step": 61405
    },
    {
      "entropy": 5.759156274795532,
      "epoch": 5.570573294629899,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023767867632521313,
      "loss": 4.9625,
      "mean_token_accuracy": 0.2161521077156067,
      "num_tokens": 127446927.0,
      "step": 61410
    },
    {
      "entropy": 5.682780647277832,
      "epoch": 5.571026850507983,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023764676851848365,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22643545418977737,
      "num_tokens": 127457468.0,
      "step": 61415
    },
    {
      "entropy": 5.725375366210938,
      "epoch": 5.571480406386067,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023761486148422855,
      "loss": 4.8444,
      "mean_token_accuracy": 0.22605564445257187,
      "num_tokens": 127467722.0,
      "step": 61420
    },
    {
      "entropy": 5.703731966018677,
      "epoch": 5.571933962264151,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002375829552231077,
      "loss": 4.8178,
      "mean_token_accuracy": 0.233597294986248,
      "num_tokens": 127478735.0,
      "step": 61425
    },
    {
      "entropy": 5.700780200958252,
      "epoch": 5.572387518142235,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023755104973578072,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23173812925815582,
      "num_tokens": 127489500.0,
      "step": 61430
    },
    {
      "entropy": 5.6809240818023685,
      "epoch": 5.572841074020319,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00023751914502290772,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2283536672592163,
      "num_tokens": 127499515.0,
      "step": 61435
    },
    {
      "entropy": 5.768828058242798,
      "epoch": 5.573294629898403,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023748724108514825,
      "loss": 4.9072,
      "mean_token_accuracy": 0.2275032341480255,
      "num_tokens": 127509310.0,
      "step": 61440
    },
    {
      "entropy": 5.814745378494263,
      "epoch": 5.573748185776488,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002374553379231622,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22654060870409012,
      "num_tokens": 127519413.0,
      "step": 61445
    },
    {
      "entropy": 5.763829374313355,
      "epoch": 5.574201741654572,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023742343553760938,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22525244653224946,
      "num_tokens": 127529100.0,
      "step": 61450
    },
    {
      "entropy": 5.706058359146118,
      "epoch": 5.574655297532656,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023739153392914947,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2326437920331955,
      "num_tokens": 127539538.0,
      "step": 61455
    },
    {
      "entropy": 5.742603397369384,
      "epoch": 5.5751088534107405,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023735963309844222,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23110800832509995,
      "num_tokens": 127549351.0,
      "step": 61460
    },
    {
      "entropy": 5.770081090927124,
      "epoch": 5.575562409288825,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023732773304614734,
      "loss": 4.956,
      "mean_token_accuracy": 0.21804733723402023,
      "num_tokens": 127559676.0,
      "step": 61465
    },
    {
      "entropy": 5.741412210464477,
      "epoch": 5.576015965166908,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002372958337729246,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22807824462652207,
      "num_tokens": 127569448.0,
      "step": 61470
    },
    {
      "entropy": 5.752772951126099,
      "epoch": 5.576469521044992,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023726393527943358,
      "loss": 4.8289,
      "mean_token_accuracy": 0.230268594622612,
      "num_tokens": 127580466.0,
      "step": 61475
    },
    {
      "entropy": 5.701215076446533,
      "epoch": 5.576923076923077,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023723203756633395,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23307819217443465,
      "num_tokens": 127590955.0,
      "step": 61480
    },
    {
      "entropy": 5.650945234298706,
      "epoch": 5.577376632801161,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023720014063428546,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23200378566980362,
      "num_tokens": 127601410.0,
      "step": 61485
    },
    {
      "entropy": 5.67879548072815,
      "epoch": 5.577830188679245,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023716824448394768,
      "loss": 4.7719,
      "mean_token_accuracy": 0.24405987709760665,
      "num_tokens": 127611275.0,
      "step": 61490
    },
    {
      "entropy": 5.651198291778565,
      "epoch": 5.5782837445573294,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002371363491159803,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2313280299305916,
      "num_tokens": 127621660.0,
      "step": 61495
    },
    {
      "entropy": 5.727491950988769,
      "epoch": 5.578737300435414,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023710445453104274,
      "loss": 4.9331,
      "mean_token_accuracy": 0.21992328613996506,
      "num_tokens": 127631534.0,
      "step": 61500
    },
    {
      "entropy": 5.6335042953491214,
      "epoch": 5.579190856313498,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023707256072979483,
      "loss": 4.8297,
      "mean_token_accuracy": 0.22820919156074523,
      "num_tokens": 127641742.0,
      "step": 61505
    },
    {
      "entropy": 5.692401885986328,
      "epoch": 5.579644412191582,
      "grad_norm": 1.390625,
      "learning_rate": 0.000237040667712896,
      "loss": 4.8794,
      "mean_token_accuracy": 0.2311510294675827,
      "num_tokens": 127653832.0,
      "step": 61510
    },
    {
      "entropy": 5.70913667678833,
      "epoch": 5.5800979680696665,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002370087754810058,
      "loss": 4.8057,
      "mean_token_accuracy": 0.2306920811533928,
      "num_tokens": 127664072.0,
      "step": 61515
    },
    {
      "entropy": 5.688839292526245,
      "epoch": 5.580551523947751,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023697688403478384,
      "loss": 4.7739,
      "mean_token_accuracy": 0.24163439571857454,
      "num_tokens": 127673725.0,
      "step": 61520
    },
    {
      "entropy": 5.659263372421265,
      "epoch": 5.581005079825834,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023694499337488962,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23520078063011168,
      "num_tokens": 127684616.0,
      "step": 61525
    },
    {
      "entropy": 5.711013507843018,
      "epoch": 5.581458635703918,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002369131035019826,
      "loss": 4.9049,
      "mean_token_accuracy": 0.22797464728355407,
      "num_tokens": 127695395.0,
      "step": 61530
    },
    {
      "entropy": 5.680735921859741,
      "epoch": 5.581912191582003,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023688121441672234,
      "loss": 4.8178,
      "mean_token_accuracy": 0.23974889516830444,
      "num_tokens": 127704712.0,
      "step": 61535
    },
    {
      "entropy": 5.672369718551636,
      "epoch": 5.582365747460087,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023684932611976823,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22459943145513533,
      "num_tokens": 127715913.0,
      "step": 61540
    },
    {
      "entropy": 5.713598394393921,
      "epoch": 5.582819303338171,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023681743861177985,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2282384976744652,
      "num_tokens": 127725639.0,
      "step": 61545
    },
    {
      "entropy": 5.659256315231323,
      "epoch": 5.5832728592162555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023678555189341656,
      "loss": 4.8383,
      "mean_token_accuracy": 0.22559576183557511,
      "num_tokens": 127737285.0,
      "step": 61550
    },
    {
      "entropy": 5.697662115097046,
      "epoch": 5.58372641509434,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002367536659653378,
      "loss": 4.7831,
      "mean_token_accuracy": 0.23591790497303008,
      "num_tokens": 127746492.0,
      "step": 61555
    },
    {
      "entropy": 5.675088167190552,
      "epoch": 5.584179970972424,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002367217808282029,
      "loss": 4.8179,
      "mean_token_accuracy": 0.2377396047115326,
      "num_tokens": 127756491.0,
      "step": 61560
    },
    {
      "entropy": 5.730838966369629,
      "epoch": 5.584633526850508,
      "grad_norm": 1.5,
      "learning_rate": 0.00023668989648267135,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2283640056848526,
      "num_tokens": 127767934.0,
      "step": 61565
    },
    {
      "entropy": 5.7693257331848145,
      "epoch": 5.585087082728592,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023665801292940253,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2367238938808441,
      "num_tokens": 127778219.0,
      "step": 61570
    },
    {
      "entropy": 5.761605501174927,
      "epoch": 5.585540638606677,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023662613016905577,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2323038950562477,
      "num_tokens": 127787484.0,
      "step": 61575
    },
    {
      "entropy": 5.663483619689941,
      "epoch": 5.58599419448476,
      "grad_norm": 1.375,
      "learning_rate": 0.00023659424820229042,
      "loss": 4.7378,
      "mean_token_accuracy": 0.23844195455312728,
      "num_tokens": 127797409.0,
      "step": 61580
    },
    {
      "entropy": 5.678190803527832,
      "epoch": 5.5864477503628445,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002365623670297658,
      "loss": 4.7673,
      "mean_token_accuracy": 0.2350816771388054,
      "num_tokens": 127807919.0,
      "step": 61585
    },
    {
      "entropy": 5.713662624359131,
      "epoch": 5.586901306240929,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002365304866521412,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2224103793501854,
      "num_tokens": 127817225.0,
      "step": 61590
    },
    {
      "entropy": 5.659422922134399,
      "epoch": 5.587354862119013,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023649860707007593,
      "loss": 4.7911,
      "mean_token_accuracy": 0.2386000633239746,
      "num_tokens": 127827287.0,
      "step": 61595
    },
    {
      "entropy": 5.678902101516724,
      "epoch": 5.587808417997097,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023646672828422932,
      "loss": 4.8293,
      "mean_token_accuracy": 0.23240037262439728,
      "num_tokens": 127838264.0,
      "step": 61600
    },
    {
      "entropy": 5.727884960174561,
      "epoch": 5.5882619738751815,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023643485029526057,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22589536309242247,
      "num_tokens": 127848214.0,
      "step": 61605
    },
    {
      "entropy": 5.740785455703735,
      "epoch": 5.588715529753266,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002364029731038289,
      "loss": 4.8309,
      "mean_token_accuracy": 0.23147848546504973,
      "num_tokens": 127857912.0,
      "step": 61610
    },
    {
      "entropy": 5.7763991355896,
      "epoch": 5.58916908563135,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002363710967105937,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22785936892032624,
      "num_tokens": 127867749.0,
      "step": 61615
    },
    {
      "entropy": 5.73713698387146,
      "epoch": 5.589622641509434,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023633922111621398,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22562996596097945,
      "num_tokens": 127878410.0,
      "step": 61620
    },
    {
      "entropy": 5.754213428497314,
      "epoch": 5.590076197387518,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023630734632134898,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22216808944940566,
      "num_tokens": 127888375.0,
      "step": 61625
    },
    {
      "entropy": 5.71306881904602,
      "epoch": 5.590529753265602,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023627547232665802,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22840454280376435,
      "num_tokens": 127898010.0,
      "step": 61630
    },
    {
      "entropy": 5.683782577514648,
      "epoch": 5.590983309143686,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002362435991328001,
      "loss": 4.7549,
      "mean_token_accuracy": 0.23939770609140396,
      "num_tokens": 127908780.0,
      "step": 61635
    },
    {
      "entropy": 5.7092832088470455,
      "epoch": 5.5914368650217705,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023621172674043444,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22838462740182877,
      "num_tokens": 127918897.0,
      "step": 61640
    },
    {
      "entropy": 5.697630739212036,
      "epoch": 5.591890420899855,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002361798551502201,
      "loss": 4.8272,
      "mean_token_accuracy": 0.23232369720935822,
      "num_tokens": 127929687.0,
      "step": 61645
    },
    {
      "entropy": 5.732925844192505,
      "epoch": 5.592343976777939,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002361479843628163,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2290133774280548,
      "num_tokens": 127940676.0,
      "step": 61650
    },
    {
      "entropy": 5.7027050971984865,
      "epoch": 5.592797532656023,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002361161143788821,
      "loss": 4.828,
      "mean_token_accuracy": 0.2306281104683876,
      "num_tokens": 127950891.0,
      "step": 61655
    },
    {
      "entropy": 5.689784860610962,
      "epoch": 5.593251088534108,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002360842451990765,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23921673446893693,
      "num_tokens": 127961871.0,
      "step": 61660
    },
    {
      "entropy": 5.716077661514282,
      "epoch": 5.593704644412192,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023605237682405871,
      "loss": 4.8214,
      "mean_token_accuracy": 0.23170038908720017,
      "num_tokens": 127970602.0,
      "step": 61665
    },
    {
      "entropy": 5.750817680358887,
      "epoch": 5.594158200290276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023602050925448765,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21163566410541534,
      "num_tokens": 127981672.0,
      "step": 61670
    },
    {
      "entropy": 5.702906656265259,
      "epoch": 5.59461175616836,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023598864249102243,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23143409341573715,
      "num_tokens": 127990995.0,
      "step": 61675
    },
    {
      "entropy": 5.717066812515259,
      "epoch": 5.595065312046444,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023595677653432195,
      "loss": 4.8179,
      "mean_token_accuracy": 0.23188737779855728,
      "num_tokens": 128000490.0,
      "step": 61680
    },
    {
      "entropy": 5.729816436767578,
      "epoch": 5.595518867924528,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023592491138504535,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2229366973042488,
      "num_tokens": 128011257.0,
      "step": 61685
    },
    {
      "entropy": 5.7207043170928955,
      "epoch": 5.595972423802612,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002358930470438515,
      "loss": 4.8973,
      "mean_token_accuracy": 0.2245500311255455,
      "num_tokens": 128021297.0,
      "step": 61690
    },
    {
      "entropy": 5.7373046875,
      "epoch": 5.5964259796806965,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002358611835113994,
      "loss": 4.9172,
      "mean_token_accuracy": 0.22668554931879042,
      "num_tokens": 128031170.0,
      "step": 61695
    },
    {
      "entropy": 5.668882274627686,
      "epoch": 5.596879535558781,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023582932078834801,
      "loss": 4.7521,
      "mean_token_accuracy": 0.23814094364643096,
      "num_tokens": 128040845.0,
      "step": 61700
    },
    {
      "entropy": 5.6624860763549805,
      "epoch": 5.597333091436865,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023579745887535624,
      "loss": 4.8465,
      "mean_token_accuracy": 0.23343331664800643,
      "num_tokens": 128051082.0,
      "step": 61705
    },
    {
      "entropy": 5.761852645874024,
      "epoch": 5.597786647314949,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023576559777308304,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22882286310195923,
      "num_tokens": 128062611.0,
      "step": 61710
    },
    {
      "entropy": 5.743003797531128,
      "epoch": 5.598240203193034,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023573373748218723,
      "loss": 4.8431,
      "mean_token_accuracy": 0.2375922754406929,
      "num_tokens": 128071608.0,
      "step": 61715
    },
    {
      "entropy": 5.681014728546143,
      "epoch": 5.598693759071118,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002357018780033278,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23237644135951996,
      "num_tokens": 128081912.0,
      "step": 61720
    },
    {
      "entropy": 5.748305368423462,
      "epoch": 5.599147314949201,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023567001933716354,
      "loss": 4.8581,
      "mean_token_accuracy": 0.23110074996948243,
      "num_tokens": 128092015.0,
      "step": 61725
    },
    {
      "entropy": 5.709585523605346,
      "epoch": 5.599600870827286,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023563816148435324,
      "loss": 4.7766,
      "mean_token_accuracy": 0.24055940359830857,
      "num_tokens": 128102148.0,
      "step": 61730
    },
    {
      "entropy": 5.707119989395141,
      "epoch": 5.60005442670537,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002356063044455558,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22995828241109847,
      "num_tokens": 128112163.0,
      "step": 61735
    },
    {
      "entropy": 5.720515537261963,
      "epoch": 5.600507982583454,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023557444822143008,
      "loss": 4.845,
      "mean_token_accuracy": 0.23439643383026124,
      "num_tokens": 128123059.0,
      "step": 61740
    },
    {
      "entropy": 5.677988529205322,
      "epoch": 5.600961538461538,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023554259281263474,
      "loss": 4.8259,
      "mean_token_accuracy": 0.22496495842933656,
      "num_tokens": 128133171.0,
      "step": 61745
    },
    {
      "entropy": 5.683382654190064,
      "epoch": 5.601415094339623,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023551073821982872,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2318251520395279,
      "num_tokens": 128143240.0,
      "step": 61750
    },
    {
      "entropy": 5.75920147895813,
      "epoch": 5.601868650217707,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023547888444367066,
      "loss": 4.871,
      "mean_token_accuracy": 0.23042369335889817,
      "num_tokens": 128154434.0,
      "step": 61755
    },
    {
      "entropy": 5.719506072998047,
      "epoch": 5.602322206095791,
      "grad_norm": 1.375,
      "learning_rate": 0.00023544703148481932,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23564208745956422,
      "num_tokens": 128164684.0,
      "step": 61760
    },
    {
      "entropy": 5.715956783294677,
      "epoch": 5.602775761973875,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023541517934393347,
      "loss": 4.8487,
      "mean_token_accuracy": 0.238513545691967,
      "num_tokens": 128174118.0,
      "step": 61765
    },
    {
      "entropy": 5.614346218109131,
      "epoch": 5.60322931785196,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002353833280216718,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23202347457408906,
      "num_tokens": 128184780.0,
      "step": 61770
    },
    {
      "entropy": 5.630583763122559,
      "epoch": 5.603682873730044,
      "grad_norm": 1.359375,
      "learning_rate": 0.000235351477518693,
      "loss": 4.7259,
      "mean_token_accuracy": 0.2351904645562172,
      "num_tokens": 128194178.0,
      "step": 61775
    },
    {
      "entropy": 5.595767307281494,
      "epoch": 5.604136429608127,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002353196278356557,
      "loss": 4.7354,
      "mean_token_accuracy": 0.23944087326526642,
      "num_tokens": 128205817.0,
      "step": 61780
    },
    {
      "entropy": 5.686732149124145,
      "epoch": 5.604589985486212,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023528777897321868,
      "loss": 4.7894,
      "mean_token_accuracy": 0.2364732578396797,
      "num_tokens": 128215753.0,
      "step": 61785
    },
    {
      "entropy": 5.671860265731811,
      "epoch": 5.605043541364296,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023525593093204052,
      "loss": 4.8049,
      "mean_token_accuracy": 0.23986871242523194,
      "num_tokens": 128225763.0,
      "step": 61790
    },
    {
      "entropy": 5.791018533706665,
      "epoch": 5.60549709724238,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023522408371277976,
      "loss": 4.9975,
      "mean_token_accuracy": 0.22140823304653168,
      "num_tokens": 128237185.0,
      "step": 61795
    },
    {
      "entropy": 5.655772113800049,
      "epoch": 5.605950653120464,
      "grad_norm": 1.375,
      "learning_rate": 0.00023519223731609511,
      "loss": 4.8226,
      "mean_token_accuracy": 0.24225003272294998,
      "num_tokens": 128247578.0,
      "step": 61800
    },
    {
      "entropy": 5.650177621841431,
      "epoch": 5.606404208998549,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023516039174264515,
      "loss": 4.7294,
      "mean_token_accuracy": 0.23663100153207778,
      "num_tokens": 128258391.0,
      "step": 61805
    },
    {
      "entropy": 5.684117841720581,
      "epoch": 5.606857764876633,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023512854699308844,
      "loss": 4.8362,
      "mean_token_accuracy": 0.23049745410680772,
      "num_tokens": 128269498.0,
      "step": 61810
    },
    {
      "entropy": 5.732565402984619,
      "epoch": 5.607311320754717,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023509670306808356,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23161069303750992,
      "num_tokens": 128279544.0,
      "step": 61815
    },
    {
      "entropy": 5.6488487243652346,
      "epoch": 5.607764876632801,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023506485996828901,
      "loss": 4.74,
      "mean_token_accuracy": 0.24429285973310472,
      "num_tokens": 128290376.0,
      "step": 61820
    },
    {
      "entropy": 5.633038902282715,
      "epoch": 5.608218432510886,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023503301769436336,
      "loss": 4.7848,
      "mean_token_accuracy": 0.23443714827299117,
      "num_tokens": 128300598.0,
      "step": 61825
    },
    {
      "entropy": 5.691554069519043,
      "epoch": 5.60867198838897,
      "grad_norm": 1.375,
      "learning_rate": 0.00023500117624696497,
      "loss": 4.8252,
      "mean_token_accuracy": 0.23659974485635757,
      "num_tokens": 128310866.0,
      "step": 61830
    },
    {
      "entropy": 5.690186548233032,
      "epoch": 5.609125544267053,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002349693356267526,
      "loss": 4.861,
      "mean_token_accuracy": 0.22934631407260894,
      "num_tokens": 128321666.0,
      "step": 61835
    },
    {
      "entropy": 5.702060890197754,
      "epoch": 5.609579100145138,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023493749583438451,
      "loss": 4.784,
      "mean_token_accuracy": 0.2401848167181015,
      "num_tokens": 128331457.0,
      "step": 61840
    },
    {
      "entropy": 5.717726707458496,
      "epoch": 5.610032656023222,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023490565687051923,
      "loss": 4.826,
      "mean_token_accuracy": 0.22881388813257217,
      "num_tokens": 128342796.0,
      "step": 61845
    },
    {
      "entropy": 5.67452392578125,
      "epoch": 5.610486211901306,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002348738187358151,
      "loss": 4.8745,
      "mean_token_accuracy": 0.23206167668104172,
      "num_tokens": 128353758.0,
      "step": 61850
    },
    {
      "entropy": 5.729216861724853,
      "epoch": 5.61093976777939,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023484198143093073,
      "loss": 4.8479,
      "mean_token_accuracy": 0.23403186202049256,
      "num_tokens": 128364205.0,
      "step": 61855
    },
    {
      "entropy": 5.753251028060913,
      "epoch": 5.611393323657475,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023481014495652441,
      "loss": 4.939,
      "mean_token_accuracy": 0.22173262536525726,
      "num_tokens": 128375632.0,
      "step": 61860
    },
    {
      "entropy": 5.7818153381347654,
      "epoch": 5.611846879535559,
      "grad_norm": 1.671875,
      "learning_rate": 0.00023477830931325445,
      "loss": 4.8615,
      "mean_token_accuracy": 0.2348748579621315,
      "num_tokens": 128384605.0,
      "step": 61865
    },
    {
      "entropy": 5.687752342224121,
      "epoch": 5.612300435413643,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023474647450177937,
      "loss": 4.7468,
      "mean_token_accuracy": 0.24213570952415467,
      "num_tokens": 128394026.0,
      "step": 61870
    },
    {
      "entropy": 5.628443431854248,
      "epoch": 5.6127539912917275,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002347146405227575,
      "loss": 4.7643,
      "mean_token_accuracy": 0.24388414323329927,
      "num_tokens": 128403791.0,
      "step": 61875
    },
    {
      "entropy": 5.671854257583618,
      "epoch": 5.613207547169811,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023468280737684705,
      "loss": 4.8564,
      "mean_token_accuracy": 0.231588414311409,
      "num_tokens": 128414313.0,
      "step": 61880
    },
    {
      "entropy": 5.712482357025147,
      "epoch": 5.613661103047895,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002346509750647064,
      "loss": 4.8717,
      "mean_token_accuracy": 0.23885704725980758,
      "num_tokens": 128424619.0,
      "step": 61885
    },
    {
      "entropy": 5.655836343765259,
      "epoch": 5.614114658925979,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023461914358699387,
      "loss": 4.7993,
      "mean_token_accuracy": 0.24111268818378448,
      "num_tokens": 128433691.0,
      "step": 61890
    },
    {
      "entropy": 5.762513971328735,
      "epoch": 5.614568214804064,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023458731294436787,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2272263541817665,
      "num_tokens": 128443730.0,
      "step": 61895
    },
    {
      "entropy": 5.671576499938965,
      "epoch": 5.615021770682148,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023455548313748638,
      "loss": 4.7665,
      "mean_token_accuracy": 0.2360390082001686,
      "num_tokens": 128454293.0,
      "step": 61900
    },
    {
      "entropy": 5.695105981826782,
      "epoch": 5.615475326560232,
      "grad_norm": 1.375,
      "learning_rate": 0.00023452365416700788,
      "loss": 4.8612,
      "mean_token_accuracy": 0.2281895473599434,
      "num_tokens": 128465025.0,
      "step": 61905
    },
    {
      "entropy": 5.70420823097229,
      "epoch": 5.6159288824383164,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023449182603359054,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22258401811122894,
      "num_tokens": 128476763.0,
      "step": 61910
    },
    {
      "entropy": 5.662468671798706,
      "epoch": 5.616382438316401,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023445999873789248,
      "loss": 4.7847,
      "mean_token_accuracy": 0.22935588508844376,
      "num_tokens": 128486593.0,
      "step": 61915
    },
    {
      "entropy": 5.691277599334716,
      "epoch": 5.616835994194485,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023442817228057206,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22790638655424117,
      "num_tokens": 128497733.0,
      "step": 61920
    },
    {
      "entropy": 5.732460975646973,
      "epoch": 5.617289550072569,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002343963466622874,
      "loss": 4.8286,
      "mean_token_accuracy": 0.2384435385465622,
      "num_tokens": 128507657.0,
      "step": 61925
    },
    {
      "entropy": 5.701535987854004,
      "epoch": 5.6177431059506535,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023436452188369656,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23988644033670425,
      "num_tokens": 128517803.0,
      "step": 61930
    },
    {
      "entropy": 5.705656814575195,
      "epoch": 5.618196661828737,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023433269794545774,
      "loss": 4.8721,
      "mean_token_accuracy": 0.23124347776174545,
      "num_tokens": 128527612.0,
      "step": 61935
    },
    {
      "entropy": 5.748022508621216,
      "epoch": 5.618650217706821,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023430087484822915,
      "loss": 4.8731,
      "mean_token_accuracy": 0.23606279939413072,
      "num_tokens": 128538527.0,
      "step": 61940
    },
    {
      "entropy": 5.743792104721069,
      "epoch": 5.619103773584905,
      "grad_norm": 1.375,
      "learning_rate": 0.00023426905259266883,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22781775444746016,
      "num_tokens": 128549659.0,
      "step": 61945
    },
    {
      "entropy": 5.769913196563721,
      "epoch": 5.61955732946299,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023423723117943485,
      "loss": 4.932,
      "mean_token_accuracy": 0.22496574819087983,
      "num_tokens": 128560605.0,
      "step": 61950
    },
    {
      "entropy": 5.749667501449585,
      "epoch": 5.620010885341074,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023420541060918538,
      "loss": 4.8209,
      "mean_token_accuracy": 0.24466813057661058,
      "num_tokens": 128571092.0,
      "step": 61955
    },
    {
      "entropy": 5.664315271377563,
      "epoch": 5.620464441219158,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002341735908825784,
      "loss": 4.8232,
      "mean_token_accuracy": 0.237323197722435,
      "num_tokens": 128582183.0,
      "step": 61960
    },
    {
      "entropy": 5.703614807128906,
      "epoch": 5.6209179970972425,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023414177200027197,
      "loss": 4.912,
      "mean_token_accuracy": 0.2246032878756523,
      "num_tokens": 128593704.0,
      "step": 61965
    },
    {
      "entropy": 5.695023679733277,
      "epoch": 5.621371552975327,
      "grad_norm": 1.375,
      "learning_rate": 0.00023410995396292407,
      "loss": 4.8339,
      "mean_token_accuracy": 0.22547393143177033,
      "num_tokens": 128603878.0,
      "step": 61970
    },
    {
      "entropy": 5.655227279663086,
      "epoch": 5.621825108853411,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023407813677119282,
      "loss": 4.7871,
      "mean_token_accuracy": 0.2358610823750496,
      "num_tokens": 128614609.0,
      "step": 61975
    },
    {
      "entropy": 5.715513086318969,
      "epoch": 5.622278664731495,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023404632042573604,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23555189967155457,
      "num_tokens": 128624806.0,
      "step": 61980
    },
    {
      "entropy": 5.705550765991211,
      "epoch": 5.62273222060958,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023401450492721183,
      "loss": 4.8138,
      "mean_token_accuracy": 0.2358323037624359,
      "num_tokens": 128635939.0,
      "step": 61985
    },
    {
      "entropy": 5.804441499710083,
      "epoch": 5.623185776487663,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002339826902762781,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22199366092681885,
      "num_tokens": 128646404.0,
      "step": 61990
    },
    {
      "entropy": 5.673663711547851,
      "epoch": 5.623639332365747,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023395087647359275,
      "loss": 4.79,
      "mean_token_accuracy": 0.23647813349962235,
      "num_tokens": 128656426.0,
      "step": 61995
    },
    {
      "entropy": 5.7186198234558105,
      "epoch": 5.6240928882438315,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023391906351981378,
      "loss": 4.9243,
      "mean_token_accuracy": 0.22360572814941407,
      "num_tokens": 128665854.0,
      "step": 62000
    },
    {
      "entropy": 5.607046365737915,
      "epoch": 5.624546444121916,
      "grad_norm": 1.375,
      "learning_rate": 0.000233887251415599,
      "loss": 4.747,
      "mean_token_accuracy": 0.2373987004160881,
      "num_tokens": 128676006.0,
      "step": 62005
    },
    {
      "entropy": 5.653600788116455,
      "epoch": 5.625,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023385544016160637,
      "loss": 4.7512,
      "mean_token_accuracy": 0.23105500191450118,
      "num_tokens": 128686108.0,
      "step": 62010
    },
    {
      "entropy": 5.709413242340088,
      "epoch": 5.625453555878084,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002338236297584937,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2299770325422287,
      "num_tokens": 128695656.0,
      "step": 62015
    },
    {
      "entropy": 5.692032718658448,
      "epoch": 5.6259071117561685,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023379182020691892,
      "loss": 4.878,
      "mean_token_accuracy": 0.22721488624811173,
      "num_tokens": 128705414.0,
      "step": 62020
    },
    {
      "entropy": 5.714777040481567,
      "epoch": 5.626360667634253,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023376001150753972,
      "loss": 4.8405,
      "mean_token_accuracy": 0.23092198818922044,
      "num_tokens": 128716532.0,
      "step": 62025
    },
    {
      "entropy": 5.677502679824829,
      "epoch": 5.626814223512337,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000233728203661014,
      "loss": 4.8583,
      "mean_token_accuracy": 0.22708972692489623,
      "num_tokens": 128727156.0,
      "step": 62030
    },
    {
      "entropy": 5.753334856033325,
      "epoch": 5.62726777939042,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002336963966679996,
      "loss": 4.8469,
      "mean_token_accuracy": 0.2268613815307617,
      "num_tokens": 128737160.0,
      "step": 62035
    },
    {
      "entropy": 5.671037006378174,
      "epoch": 5.627721335268505,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023366459052915424,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22879908680915834,
      "num_tokens": 128748657.0,
      "step": 62040
    },
    {
      "entropy": 5.685969495773316,
      "epoch": 5.628174891146589,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002336327852451356,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23440443277359008,
      "num_tokens": 128759200.0,
      "step": 62045
    },
    {
      "entropy": 5.664090871810913,
      "epoch": 5.628628447024673,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002336009808166016,
      "loss": 4.8341,
      "mean_token_accuracy": 0.22511786520481109,
      "num_tokens": 128769351.0,
      "step": 62050
    },
    {
      "entropy": 5.748419904708863,
      "epoch": 5.6290820029027575,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023356917724420983,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2273329496383667,
      "num_tokens": 128779540.0,
      "step": 62055
    },
    {
      "entropy": 5.77805528640747,
      "epoch": 5.629535558780842,
      "grad_norm": 1.421875,
      "learning_rate": 0.000233537374528618,
      "loss": 4.9292,
      "mean_token_accuracy": 0.21694613993167877,
      "num_tokens": 128790159.0,
      "step": 62060
    },
    {
      "entropy": 5.762298965454102,
      "epoch": 5.629989114658926,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023350557267048382,
      "loss": 4.911,
      "mean_token_accuracy": 0.21990331411361694,
      "num_tokens": 128801679.0,
      "step": 62065
    },
    {
      "entropy": 5.750947427749634,
      "epoch": 5.63044267053701,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023347377167046501,
      "loss": 4.8685,
      "mean_token_accuracy": 0.23088972866535187,
      "num_tokens": 128812511.0,
      "step": 62070
    },
    {
      "entropy": 5.683723974227905,
      "epoch": 5.630896226415095,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023344197152921925,
      "loss": 4.7895,
      "mean_token_accuracy": 0.2391177996993065,
      "num_tokens": 128822703.0,
      "step": 62075
    },
    {
      "entropy": 5.707107973098755,
      "epoch": 5.631349782293179,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000233410172247404,
      "loss": 4.8225,
      "mean_token_accuracy": 0.22888254523277282,
      "num_tokens": 128832429.0,
      "step": 62080
    },
    {
      "entropy": 5.717544984817505,
      "epoch": 5.631803338171263,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023337837382567705,
      "loss": 4.8699,
      "mean_token_accuracy": 0.2297389566898346,
      "num_tokens": 128843429.0,
      "step": 62085
    },
    {
      "entropy": 5.711619091033936,
      "epoch": 5.6322568940493465,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002333465762646959,
      "loss": 4.85,
      "mean_token_accuracy": 0.23868427574634551,
      "num_tokens": 128852967.0,
      "step": 62090
    },
    {
      "entropy": 5.7712846279144285,
      "epoch": 5.632710449927431,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002333147795651182,
      "loss": 4.9346,
      "mean_token_accuracy": 0.22562339305877685,
      "num_tokens": 128863811.0,
      "step": 62095
    },
    {
      "entropy": 5.694263219833374,
      "epoch": 5.633164005805515,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023328298372760143,
      "loss": 4.7976,
      "mean_token_accuracy": 0.23267922103404998,
      "num_tokens": 128874377.0,
      "step": 62100
    },
    {
      "entropy": 5.719774675369263,
      "epoch": 5.633617561683599,
      "grad_norm": 1.375,
      "learning_rate": 0.0002332511887528032,
      "loss": 4.7972,
      "mean_token_accuracy": 0.22861261516809464,
      "num_tokens": 128884355.0,
      "step": 62105
    },
    {
      "entropy": 5.692353439331055,
      "epoch": 5.6340711175616836,
      "grad_norm": 1.375,
      "learning_rate": 0.0002332193946413811,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23450472205877304,
      "num_tokens": 128894385.0,
      "step": 62110
    },
    {
      "entropy": 5.780172395706177,
      "epoch": 5.634524673439768,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023318760139399248,
      "loss": 4.957,
      "mean_token_accuracy": 0.22021305859088897,
      "num_tokens": 128906414.0,
      "step": 62115
    },
    {
      "entropy": 5.770833730697632,
      "epoch": 5.634978229317852,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023315580901129497,
      "loss": 4.868,
      "mean_token_accuracy": 0.231300088763237,
      "num_tokens": 128916302.0,
      "step": 62120
    },
    {
      "entropy": 5.741064214706421,
      "epoch": 5.635431785195936,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023312401749394596,
      "loss": 4.8666,
      "mean_token_accuracy": 0.22474328130483628,
      "num_tokens": 128926366.0,
      "step": 62125
    },
    {
      "entropy": 5.6479414939880375,
      "epoch": 5.635885341074021,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002330922268426029,
      "loss": 4.7911,
      "mean_token_accuracy": 0.23558294922113418,
      "num_tokens": 128936716.0,
      "step": 62130
    },
    {
      "entropy": 5.712172269821167,
      "epoch": 5.636338896952105,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023306043705792335,
      "loss": 4.8239,
      "mean_token_accuracy": 0.2288702741265297,
      "num_tokens": 128948802.0,
      "step": 62135
    },
    {
      "entropy": 5.725664281845093,
      "epoch": 5.636792452830189,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023302864814056463,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22780560404062272,
      "num_tokens": 128959914.0,
      "step": 62140
    },
    {
      "entropy": 5.69707465171814,
      "epoch": 5.6372460087082725,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023299686009118414,
      "loss": 4.8365,
      "mean_token_accuracy": 0.23312680125236512,
      "num_tokens": 128970666.0,
      "step": 62145
    },
    {
      "entropy": 5.6913330078125,
      "epoch": 5.637699564586357,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023296507291043929,
      "loss": 4.827,
      "mean_token_accuracy": 0.22831233739852905,
      "num_tokens": 128981302.0,
      "step": 62150
    },
    {
      "entropy": 5.716566467285157,
      "epoch": 5.638153120464441,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002329332865989875,
      "loss": 4.8785,
      "mean_token_accuracy": 0.22638676762580873,
      "num_tokens": 128991242.0,
      "step": 62155
    },
    {
      "entropy": 5.752957391738891,
      "epoch": 5.638606676342525,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000232901501157486,
      "loss": 4.8581,
      "mean_token_accuracy": 0.23421460390090942,
      "num_tokens": 129000778.0,
      "step": 62160
    },
    {
      "entropy": 5.718532657623291,
      "epoch": 5.63906023222061,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023286971658659217,
      "loss": 4.8732,
      "mean_token_accuracy": 0.2280690774321556,
      "num_tokens": 129011799.0,
      "step": 62165
    },
    {
      "entropy": 5.709915208816528,
      "epoch": 5.639513788098694,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023283793288696338,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2348971411585808,
      "num_tokens": 129022038.0,
      "step": 62170
    },
    {
      "entropy": 5.708548545837402,
      "epoch": 5.639967343976778,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023280615005925688,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22686359733343125,
      "num_tokens": 129032337.0,
      "step": 62175
    },
    {
      "entropy": 5.754452276229858,
      "epoch": 5.640420899854862,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023277436810412993,
      "loss": 4.892,
      "mean_token_accuracy": 0.23262661695480347,
      "num_tokens": 129042391.0,
      "step": 62180
    },
    {
      "entropy": 5.6474450588226315,
      "epoch": 5.640874455732947,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023274258702223977,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2265176445245743,
      "num_tokens": 129052872.0,
      "step": 62185
    },
    {
      "entropy": 5.713178253173828,
      "epoch": 5.64132801161103,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002327108068142438,
      "loss": 4.871,
      "mean_token_accuracy": 0.22406278252601625,
      "num_tokens": 129063483.0,
      "step": 62190
    },
    {
      "entropy": 5.755850791931152,
      "epoch": 5.641781567489114,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000232679027480799,
      "loss": 4.8012,
      "mean_token_accuracy": 0.22808083593845369,
      "num_tokens": 129075357.0,
      "step": 62195
    },
    {
      "entropy": 5.7634612083435055,
      "epoch": 5.642235123367199,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023264724902256274,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22762440592050553,
      "num_tokens": 129086424.0,
      "step": 62200
    },
    {
      "entropy": 5.657752561569214,
      "epoch": 5.642688679245283,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023261547144019212,
      "loss": 4.7602,
      "mean_token_accuracy": 0.23175329864025115,
      "num_tokens": 129096350.0,
      "step": 62205
    },
    {
      "entropy": 5.663941192626953,
      "epoch": 5.643142235123367,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023258369473434443,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23483803123235703,
      "num_tokens": 129106858.0,
      "step": 62210
    },
    {
      "entropy": 5.772548484802246,
      "epoch": 5.643595791001451,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023255191890567668,
      "loss": 4.908,
      "mean_token_accuracy": 0.22645144760608674,
      "num_tokens": 129116545.0,
      "step": 62215
    },
    {
      "entropy": 5.705436897277832,
      "epoch": 5.644049346879536,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023252014395484612,
      "loss": 4.8409,
      "mean_token_accuracy": 0.2325573295354843,
      "num_tokens": 129126053.0,
      "step": 62220
    },
    {
      "entropy": 5.71001296043396,
      "epoch": 5.64450290275762,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023248836988250975,
      "loss": 4.8971,
      "mean_token_accuracy": 0.219446824491024,
      "num_tokens": 129136297.0,
      "step": 62225
    },
    {
      "entropy": 5.68692364692688,
      "epoch": 5.644956458635704,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002324565966893248,
      "loss": 4.8077,
      "mean_token_accuracy": 0.23309404104948045,
      "num_tokens": 129146809.0,
      "step": 62230
    },
    {
      "entropy": 5.717669105529785,
      "epoch": 5.645410014513788,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023242482437594818,
      "loss": 4.913,
      "mean_token_accuracy": 0.22534934133291246,
      "num_tokens": 129157682.0,
      "step": 62235
    },
    {
      "entropy": 5.7060919284820555,
      "epoch": 5.645863570391873,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023239305294303714,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22301640957593918,
      "num_tokens": 129167637.0,
      "step": 62240
    },
    {
      "entropy": 5.673026990890503,
      "epoch": 5.646317126269956,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023236128239124853,
      "loss": 4.7217,
      "mean_token_accuracy": 0.2356308162212372,
      "num_tokens": 129177449.0,
      "step": 62245
    },
    {
      "entropy": 5.642708826065063,
      "epoch": 5.64677068214804,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023232951272123952,
      "loss": 4.819,
      "mean_token_accuracy": 0.2340735450387001,
      "num_tokens": 129186740.0,
      "step": 62250
    },
    {
      "entropy": 5.725921154022217,
      "epoch": 5.647224238026125,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023229774393366704,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22838261723518372,
      "num_tokens": 129197371.0,
      "step": 62255
    },
    {
      "entropy": 5.746433544158935,
      "epoch": 5.647677793904209,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023226597602918813,
      "loss": 4.9175,
      "mean_token_accuracy": 0.22593754529953003,
      "num_tokens": 129208326.0,
      "step": 62260
    },
    {
      "entropy": 5.6979598045349125,
      "epoch": 5.648131349782293,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002322342090084597,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22361129522323608,
      "num_tokens": 129219434.0,
      "step": 62265
    },
    {
      "entropy": 5.63729076385498,
      "epoch": 5.648584905660377,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023220244287213866,
      "loss": 4.78,
      "mean_token_accuracy": 0.2344679594039917,
      "num_tokens": 129230582.0,
      "step": 62270
    },
    {
      "entropy": 5.654370260238648,
      "epoch": 5.649038461538462,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023217067762088212,
      "loss": 4.7423,
      "mean_token_accuracy": 0.2399286612868309,
      "num_tokens": 129240476.0,
      "step": 62275
    },
    {
      "entropy": 5.664772319793701,
      "epoch": 5.649492017416546,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023213891325534687,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23383836299180985,
      "num_tokens": 129250566.0,
      "step": 62280
    },
    {
      "entropy": 5.698317432403565,
      "epoch": 5.64994557329463,
      "grad_norm": 1.375,
      "learning_rate": 0.0002321071497761897,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22979585826396942,
      "num_tokens": 129261173.0,
      "step": 62285
    },
    {
      "entropy": 5.766669607162475,
      "epoch": 5.650399129172714,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023207538718406773,
      "loss": 4.9571,
      "mean_token_accuracy": 0.2301727741956711,
      "num_tokens": 129271775.0,
      "step": 62290
    },
    {
      "entropy": 5.78709568977356,
      "epoch": 5.650852685050799,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023204362547963765,
      "loss": 4.9383,
      "mean_token_accuracy": 0.22813550084829332,
      "num_tokens": 129282203.0,
      "step": 62295
    },
    {
      "entropy": 5.79157133102417,
      "epoch": 5.651306240928882,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023201186466355634,
      "loss": 4.8978,
      "mean_token_accuracy": 0.22365251928567886,
      "num_tokens": 129292584.0,
      "step": 62300
    },
    {
      "entropy": 5.756330680847168,
      "epoch": 5.651759796806966,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023198010473648057,
      "loss": 4.9028,
      "mean_token_accuracy": 0.22564199268817903,
      "num_tokens": 129303301.0,
      "step": 62305
    },
    {
      "entropy": 5.726636505126953,
      "epoch": 5.652213352685051,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023194834569906724,
      "loss": 4.8257,
      "mean_token_accuracy": 0.23108285367488862,
      "num_tokens": 129314229.0,
      "step": 62310
    },
    {
      "entropy": 5.68978443145752,
      "epoch": 5.652666908563135,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023191658755197315,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23420827388763427,
      "num_tokens": 129324998.0,
      "step": 62315
    },
    {
      "entropy": 5.573440361022949,
      "epoch": 5.653120464441219,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023188483029585487,
      "loss": 4.7096,
      "mean_token_accuracy": 0.23708305656909942,
      "num_tokens": 129336133.0,
      "step": 62320
    },
    {
      "entropy": 5.6467584609985355,
      "epoch": 5.6535740203193035,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002318530739313694,
      "loss": 4.8264,
      "mean_token_accuracy": 0.23566127121448516,
      "num_tokens": 129346416.0,
      "step": 62325
    },
    {
      "entropy": 5.706980037689209,
      "epoch": 5.654027576197388,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023182131845917337,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2277364581823349,
      "num_tokens": 129357214.0,
      "step": 62330
    },
    {
      "entropy": 5.669563961029053,
      "epoch": 5.654481132075472,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002317895638799234,
      "loss": 4.7367,
      "mean_token_accuracy": 0.23591073900461196,
      "num_tokens": 129366796.0,
      "step": 62335
    },
    {
      "entropy": 5.704401731491089,
      "epoch": 5.654934687953556,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023175781019427633,
      "loss": 4.8356,
      "mean_token_accuracy": 0.23546013832092286,
      "num_tokens": 129376791.0,
      "step": 62340
    },
    {
      "entropy": 5.658244800567627,
      "epoch": 5.65538824383164,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002317260574028887,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22835089713335038,
      "num_tokens": 129387589.0,
      "step": 62345
    },
    {
      "entropy": 5.768177938461304,
      "epoch": 5.655841799709724,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002316943055064173,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2281471610069275,
      "num_tokens": 129397041.0,
      "step": 62350
    },
    {
      "entropy": 5.6309412002563475,
      "epoch": 5.656295355587808,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002316625545055186,
      "loss": 4.7282,
      "mean_token_accuracy": 0.24469517171382904,
      "num_tokens": 129406984.0,
      "step": 62355
    },
    {
      "entropy": 5.697713804244995,
      "epoch": 5.656748911465892,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023163080440084943,
      "loss": 4.786,
      "mean_token_accuracy": 0.2324086531996727,
      "num_tokens": 129417809.0,
      "step": 62360
    },
    {
      "entropy": 5.594459009170532,
      "epoch": 5.657202467343977,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023159905519306624,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23691221326589584,
      "num_tokens": 129427838.0,
      "step": 62365
    },
    {
      "entropy": 5.704296827316284,
      "epoch": 5.657656023222061,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023156730688282556,
      "loss": 4.8966,
      "mean_token_accuracy": 0.23179901987314225,
      "num_tokens": 129437536.0,
      "step": 62370
    },
    {
      "entropy": 5.733917093276977,
      "epoch": 5.658109579100145,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023153555947078415,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23659985661506652,
      "num_tokens": 129447930.0,
      "step": 62375
    },
    {
      "entropy": 5.707721710205078,
      "epoch": 5.6585631349782295,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023150381295759844,
      "loss": 4.8364,
      "mean_token_accuracy": 0.23677935153245927,
      "num_tokens": 129457740.0,
      "step": 62380
    },
    {
      "entropy": 5.774998044967651,
      "epoch": 5.659016690856314,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002314720673439249,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21839645355939866,
      "num_tokens": 129468340.0,
      "step": 62385
    },
    {
      "entropy": 5.737592315673828,
      "epoch": 5.659470246734398,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023144032263042015,
      "loss": 4.886,
      "mean_token_accuracy": 0.2256598502397537,
      "num_tokens": 129478880.0,
      "step": 62390
    },
    {
      "entropy": 5.713773918151856,
      "epoch": 5.659923802612482,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023140857881774057,
      "loss": 4.8143,
      "mean_token_accuracy": 0.23245948553085327,
      "num_tokens": 129489362.0,
      "step": 62395
    },
    {
      "entropy": 5.762184381484985,
      "epoch": 5.660377358490566,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023137683590654273,
      "loss": 4.9063,
      "mean_token_accuracy": 0.22381252646446229,
      "num_tokens": 129500294.0,
      "step": 62400
    },
    {
      "entropy": 5.706961727142334,
      "epoch": 5.66083091436865,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023134509389748315,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23314416855573655,
      "num_tokens": 129510258.0,
      "step": 62405
    },
    {
      "entropy": 5.692757654190063,
      "epoch": 5.661284470246734,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000231313352791218,
      "loss": 4.8608,
      "mean_token_accuracy": 0.23565135151147842,
      "num_tokens": 129520390.0,
      "step": 62410
    },
    {
      "entropy": 5.681824111938477,
      "epoch": 5.6617380261248185,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023128161258840387,
      "loss": 4.8149,
      "mean_token_accuracy": 0.23193133473396302,
      "num_tokens": 129531165.0,
      "step": 62415
    },
    {
      "entropy": 5.738641595840454,
      "epoch": 5.662191582002903,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023124987328969714,
      "loss": 4.9181,
      "mean_token_accuracy": 0.2245556443929672,
      "num_tokens": 129541468.0,
      "step": 62420
    },
    {
      "entropy": 5.754482173919678,
      "epoch": 5.662645137880987,
      "grad_norm": 1.375,
      "learning_rate": 0.00023121813489575428,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22632600963115693,
      "num_tokens": 129551536.0,
      "step": 62425
    },
    {
      "entropy": 5.702982664108276,
      "epoch": 5.663098693759071,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023118639740723134,
      "loss": 4.8458,
      "mean_token_accuracy": 0.23314953446388245,
      "num_tokens": 129561101.0,
      "step": 62430
    },
    {
      "entropy": 5.802996683120727,
      "epoch": 5.6635522496371555,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023115466082478505,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2148928537964821,
      "num_tokens": 129572040.0,
      "step": 62435
    },
    {
      "entropy": 5.687436199188232,
      "epoch": 5.66400580551524,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023112292514907147,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23048067092895508,
      "num_tokens": 129582243.0,
      "step": 62440
    },
    {
      "entropy": 5.672156763076782,
      "epoch": 5.664459361393323,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000231091190380747,
      "loss": 4.7929,
      "mean_token_accuracy": 0.2322392076253891,
      "num_tokens": 129592950.0,
      "step": 62445
    },
    {
      "entropy": 5.676977205276489,
      "epoch": 5.664912917271408,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002310594565204679,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23535776287317275,
      "num_tokens": 129603936.0,
      "step": 62450
    },
    {
      "entropy": 5.735214138031006,
      "epoch": 5.665366473149492,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002310277235688904,
      "loss": 4.8824,
      "mean_token_accuracy": 0.22753021121025085,
      "num_tokens": 129613132.0,
      "step": 62455
    },
    {
      "entropy": 5.7502501010894775,
      "epoch": 5.665820029027576,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023099599152667083,
      "loss": 4.9364,
      "mean_token_accuracy": 0.2261253148317337,
      "num_tokens": 129624016.0,
      "step": 62460
    },
    {
      "entropy": 5.770652103424072,
      "epoch": 5.66627358490566,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023096426039446533,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22816359698772432,
      "num_tokens": 129634423.0,
      "step": 62465
    },
    {
      "entropy": 5.714783573150635,
      "epoch": 5.6667271407837445,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023093253017293015,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22794755101203917,
      "num_tokens": 129644133.0,
      "step": 62470
    },
    {
      "entropy": 5.646713829040527,
      "epoch": 5.667180696661829,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023090080086272152,
      "loss": 4.7916,
      "mean_token_accuracy": 0.23369920551776885,
      "num_tokens": 129653761.0,
      "step": 62475
    },
    {
      "entropy": 5.654772758483887,
      "epoch": 5.667634252539913,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023086907246449556,
      "loss": 4.8501,
      "mean_token_accuracy": 0.2270941987633705,
      "num_tokens": 129664628.0,
      "step": 62480
    },
    {
      "entropy": 5.6589995384216305,
      "epoch": 5.668087808417997,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023083734497890835,
      "loss": 4.8104,
      "mean_token_accuracy": 0.24180588573217393,
      "num_tokens": 129675301.0,
      "step": 62485
    },
    {
      "entropy": 5.757870292663574,
      "epoch": 5.668541364296082,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023080561840661617,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21789691746234893,
      "num_tokens": 129685599.0,
      "step": 62490
    },
    {
      "entropy": 5.747638845443726,
      "epoch": 5.668994920174166,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002307738927482751,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2181578353047371,
      "num_tokens": 129695927.0,
      "step": 62495
    },
    {
      "entropy": 5.714643955230713,
      "epoch": 5.669448476052249,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023074216800454106,
      "loss": 4.873,
      "mean_token_accuracy": 0.22620508968830108,
      "num_tokens": 129706422.0,
      "step": 62500
    },
    {
      "entropy": 5.696066808700562,
      "epoch": 5.6699020319303335,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002307104441760704,
      "loss": 4.8651,
      "mean_token_accuracy": 0.21950265616178513,
      "num_tokens": 129716125.0,
      "step": 62505
    },
    {
      "entropy": 5.705029058456421,
      "epoch": 5.670355587808418,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023067872126351902,
      "loss": 4.8312,
      "mean_token_accuracy": 0.23361015617847442,
      "num_tokens": 129727017.0,
      "step": 62510
    },
    {
      "entropy": 5.703234434127808,
      "epoch": 5.670809143686502,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023064699926754297,
      "loss": 4.828,
      "mean_token_accuracy": 0.23364530950784684,
      "num_tokens": 129737420.0,
      "step": 62515
    },
    {
      "entropy": 5.697514533996582,
      "epoch": 5.671262699564586,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023061527818879818,
      "loss": 4.7848,
      "mean_token_accuracy": 0.2297925606369972,
      "num_tokens": 129747090.0,
      "step": 62520
    },
    {
      "entropy": 5.740170335769653,
      "epoch": 5.6717162554426706,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023058355802794083,
      "loss": 4.862,
      "mean_token_accuracy": 0.23608438968658446,
      "num_tokens": 129757600.0,
      "step": 62525
    },
    {
      "entropy": 5.699066972732544,
      "epoch": 5.672169811320755,
      "grad_norm": 1.25,
      "learning_rate": 0.00023055183878562685,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2272272452712059,
      "num_tokens": 129767815.0,
      "step": 62530
    },
    {
      "entropy": 5.685151767730713,
      "epoch": 5.672623367198839,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023052012046251207,
      "loss": 4.8056,
      "mean_token_accuracy": 0.23249715119600295,
      "num_tokens": 129779222.0,
      "step": 62535
    },
    {
      "entropy": 5.729890060424805,
      "epoch": 5.673076923076923,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023048840305925266,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22731157392263412,
      "num_tokens": 129790985.0,
      "step": 62540
    },
    {
      "entropy": 5.745668029785156,
      "epoch": 5.673530478955008,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023045668657650437,
      "loss": 4.7673,
      "mean_token_accuracy": 0.2440663143992424,
      "num_tokens": 129802386.0,
      "step": 62545
    },
    {
      "entropy": 5.673512649536133,
      "epoch": 5.673984034833092,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023042497101492305,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23462175130844115,
      "num_tokens": 129812345.0,
      "step": 62550
    },
    {
      "entropy": 5.735498571395874,
      "epoch": 5.674437590711175,
      "grad_norm": 1.5,
      "learning_rate": 0.00023039325637516485,
      "loss": 4.8753,
      "mean_token_accuracy": 0.22779848873615266,
      "num_tokens": 129823430.0,
      "step": 62555
    },
    {
      "entropy": 5.7728437900543215,
      "epoch": 5.6748911465892595,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002303615426578854,
      "loss": 4.9361,
      "mean_token_accuracy": 0.22631009221076964,
      "num_tokens": 129833715.0,
      "step": 62560
    },
    {
      "entropy": 5.737469291687011,
      "epoch": 5.675344702467344,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023032982986374062,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22946788966655732,
      "num_tokens": 129845561.0,
      "step": 62565
    },
    {
      "entropy": 5.680110025405884,
      "epoch": 5.675798258345428,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023029811799338628,
      "loss": 4.8467,
      "mean_token_accuracy": 0.2396669939160347,
      "num_tokens": 129855228.0,
      "step": 62570
    },
    {
      "entropy": 5.7124340534210205,
      "epoch": 5.676251814223512,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023026640704747837,
      "loss": 4.8259,
      "mean_token_accuracy": 0.23413684368133544,
      "num_tokens": 129865788.0,
      "step": 62575
    },
    {
      "entropy": 5.801585912704468,
      "epoch": 5.676705370101597,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002302346970266725,
      "loss": 4.9237,
      "mean_token_accuracy": 0.22622643858194352,
      "num_tokens": 129875788.0,
      "step": 62580
    },
    {
      "entropy": 5.711309957504272,
      "epoch": 5.677158925979681,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023020298793162452,
      "loss": 4.8372,
      "mean_token_accuracy": 0.2328108847141266,
      "num_tokens": 129886892.0,
      "step": 62585
    },
    {
      "entropy": 5.713310050964355,
      "epoch": 5.677612481857765,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023017127976299013,
      "loss": 4.9505,
      "mean_token_accuracy": 0.22130731791257857,
      "num_tokens": 129897834.0,
      "step": 62590
    },
    {
      "entropy": 5.708551836013794,
      "epoch": 5.678066037735849,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023013957252142515,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23231416344642639,
      "num_tokens": 129908038.0,
      "step": 62595
    },
    {
      "entropy": 5.757876539230347,
      "epoch": 5.678519593613933,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023010786620758522,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2230415940284729,
      "num_tokens": 129918386.0,
      "step": 62600
    },
    {
      "entropy": 5.740434074401856,
      "epoch": 5.678973149492017,
      "grad_norm": 1.421875,
      "learning_rate": 0.000230076160822126,
      "loss": 4.8978,
      "mean_token_accuracy": 0.2185784175992012,
      "num_tokens": 129928776.0,
      "step": 62605
    },
    {
      "entropy": 5.680328035354615,
      "epoch": 5.679426705370101,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002300444563657033,
      "loss": 4.7926,
      "mean_token_accuracy": 0.238804329931736,
      "num_tokens": 129939234.0,
      "step": 62610
    },
    {
      "entropy": 5.623304796218872,
      "epoch": 5.679880261248186,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023001275283897273,
      "loss": 4.7598,
      "mean_token_accuracy": 0.23833031803369523,
      "num_tokens": 129949506.0,
      "step": 62615
    },
    {
      "entropy": 5.702211141586304,
      "epoch": 5.68033381712627,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002299810502425898,
      "loss": 4.819,
      "mean_token_accuracy": 0.23719388544559478,
      "num_tokens": 129959815.0,
      "step": 62620
    },
    {
      "entropy": 5.695876026153565,
      "epoch": 5.680787373004354,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022994934857721027,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22557591646909714,
      "num_tokens": 129969598.0,
      "step": 62625
    },
    {
      "entropy": 5.791470527648926,
      "epoch": 5.681240928882438,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022991764784348974,
      "loss": 4.8765,
      "mean_token_accuracy": 0.23300676941871643,
      "num_tokens": 129979497.0,
      "step": 62630
    },
    {
      "entropy": 5.699743032455444,
      "epoch": 5.681694484760523,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022988594804208362,
      "loss": 4.8292,
      "mean_token_accuracy": 0.2346063584089279,
      "num_tokens": 129989716.0,
      "step": 62635
    },
    {
      "entropy": 5.684230089187622,
      "epoch": 5.682148040638607,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022985424917364766,
      "loss": 4.8242,
      "mean_token_accuracy": 0.23369326889514924,
      "num_tokens": 129999676.0,
      "step": 62640
    },
    {
      "entropy": 5.712541151046753,
      "epoch": 5.682601596516691,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022982255123883734,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23450972884893417,
      "num_tokens": 130009346.0,
      "step": 62645
    },
    {
      "entropy": 5.695904636383057,
      "epoch": 5.683055152394775,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002297908542383082,
      "loss": 4.8793,
      "mean_token_accuracy": 0.21998976320028304,
      "num_tokens": 130021072.0,
      "step": 62650
    },
    {
      "entropy": 5.654925584793091,
      "epoch": 5.683508708272859,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022975915817271565,
      "loss": 4.8409,
      "mean_token_accuracy": 0.23152619302272798,
      "num_tokens": 130030899.0,
      "step": 62655
    },
    {
      "entropy": 5.6585663795471195,
      "epoch": 5.683962264150943,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022972746304271525,
      "loss": 4.8143,
      "mean_token_accuracy": 0.23983338922262193,
      "num_tokens": 130040345.0,
      "step": 62660
    },
    {
      "entropy": 5.709490442276001,
      "epoch": 5.684415820029027,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022969576884896247,
      "loss": 4.8893,
      "mean_token_accuracy": 0.22308728098869324,
      "num_tokens": 130050627.0,
      "step": 62665
    },
    {
      "entropy": 5.6778051376342775,
      "epoch": 5.684869375907112,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022966407559211273,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23699894100427626,
      "num_tokens": 130060592.0,
      "step": 62670
    },
    {
      "entropy": 5.700448417663575,
      "epoch": 5.685322931785196,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022963238327282143,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22715283781290055,
      "num_tokens": 130070285.0,
      "step": 62675
    },
    {
      "entropy": 5.745352077484131,
      "epoch": 5.68577648766328,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002296006918917441,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2258310243487358,
      "num_tokens": 130080356.0,
      "step": 62680
    },
    {
      "entropy": 5.733156061172485,
      "epoch": 5.686230043541364,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022956900144953596,
      "loss": 4.918,
      "mean_token_accuracy": 0.2223566398024559,
      "num_tokens": 130091532.0,
      "step": 62685
    },
    {
      "entropy": 5.651312160491943,
      "epoch": 5.686683599419449,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002295373119468524,
      "loss": 4.7686,
      "mean_token_accuracy": 0.23276668339967727,
      "num_tokens": 130101853.0,
      "step": 62690
    },
    {
      "entropy": 5.709974145889282,
      "epoch": 5.687137155297533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022950562338434888,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22525174021720887,
      "num_tokens": 130111801.0,
      "step": 62695
    },
    {
      "entropy": 5.707341289520263,
      "epoch": 5.687590711175617,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022947393576268067,
      "loss": 4.854,
      "mean_token_accuracy": 0.2348189875483513,
      "num_tokens": 130123007.0,
      "step": 62700
    },
    {
      "entropy": 5.732383966445923,
      "epoch": 5.6880442670537015,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000229442249082503,
      "loss": 4.8663,
      "mean_token_accuracy": 0.22617467045783995,
      "num_tokens": 130133699.0,
      "step": 62705
    },
    {
      "entropy": 5.794244194030762,
      "epoch": 5.688497822931785,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002294105633444713,
      "loss": 4.918,
      "mean_token_accuracy": 0.2252281442284584,
      "num_tokens": 130144178.0,
      "step": 62710
    },
    {
      "entropy": 5.6423381805419925,
      "epoch": 5.688951378809869,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002293788785492408,
      "loss": 4.7898,
      "mean_token_accuracy": 0.2287445366382599,
      "num_tokens": 130155218.0,
      "step": 62715
    },
    {
      "entropy": 5.648793649673462,
      "epoch": 5.689404934687953,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022934719469746662,
      "loss": 4.86,
      "mean_token_accuracy": 0.23462009131908418,
      "num_tokens": 130166346.0,
      "step": 62720
    },
    {
      "entropy": 5.692072200775146,
      "epoch": 5.689858490566038,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022931551178980415,
      "loss": 4.8437,
      "mean_token_accuracy": 0.23292098939418793,
      "num_tokens": 130176354.0,
      "step": 62725
    },
    {
      "entropy": 5.689882707595825,
      "epoch": 5.690312046444122,
      "grad_norm": 1.375,
      "learning_rate": 0.0002292838298269085,
      "loss": 4.7725,
      "mean_token_accuracy": 0.2390355497598648,
      "num_tokens": 130185920.0,
      "step": 62730
    },
    {
      "entropy": 5.726787900924682,
      "epoch": 5.690765602322206,
      "grad_norm": 1.375,
      "learning_rate": 0.00022925214880943496,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22370177656412124,
      "num_tokens": 130196670.0,
      "step": 62735
    },
    {
      "entropy": 5.755134439468383,
      "epoch": 5.6912191582002905,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022922046873803854,
      "loss": 4.8687,
      "mean_token_accuracy": 0.23360845297574998,
      "num_tokens": 130206567.0,
      "step": 62740
    },
    {
      "entropy": 5.717805290222168,
      "epoch": 5.691672714078375,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002291887896133746,
      "loss": 4.8675,
      "mean_token_accuracy": 0.23027462214231492,
      "num_tokens": 130217835.0,
      "step": 62745
    },
    {
      "entropy": 5.644756507873535,
      "epoch": 5.692126269956459,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022915711143609808,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2391272246837616,
      "num_tokens": 130227968.0,
      "step": 62750
    },
    {
      "entropy": 5.706371164321899,
      "epoch": 5.692579825834542,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002291254342068642,
      "loss": 4.8656,
      "mean_token_accuracy": 0.23287387639284135,
      "num_tokens": 130237863.0,
      "step": 62755
    },
    {
      "entropy": 5.713207340240478,
      "epoch": 5.693033381712627,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022909375792632802,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2283276617527008,
      "num_tokens": 130249296.0,
      "step": 62760
    },
    {
      "entropy": 5.68371353149414,
      "epoch": 5.693486937590711,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002290620825951447,
      "loss": 4.7803,
      "mean_token_accuracy": 0.2354516103863716,
      "num_tokens": 130259117.0,
      "step": 62765
    },
    {
      "entropy": 5.755583047866821,
      "epoch": 5.693940493468795,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022903040821396913,
      "loss": 4.9088,
      "mean_token_accuracy": 0.22456689029932023,
      "num_tokens": 130269678.0,
      "step": 62770
    },
    {
      "entropy": 5.741656923294068,
      "epoch": 5.694394049346879,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002289987347834565,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22576333284378053,
      "num_tokens": 130280238.0,
      "step": 62775
    },
    {
      "entropy": 5.718924283981323,
      "epoch": 5.694847605224964,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022896706230426166,
      "loss": 4.8522,
      "mean_token_accuracy": 0.227787521481514,
      "num_tokens": 130290586.0,
      "step": 62780
    },
    {
      "entropy": 5.727214002609253,
      "epoch": 5.695301161103048,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022893539077703984,
      "loss": 4.8233,
      "mean_token_accuracy": 0.23773084282875062,
      "num_tokens": 130300921.0,
      "step": 62785
    },
    {
      "entropy": 5.68786096572876,
      "epoch": 5.695754716981132,
      "grad_norm": 1.375,
      "learning_rate": 0.00022890372020244578,
      "loss": 4.8632,
      "mean_token_accuracy": 0.23040202409029006,
      "num_tokens": 130311420.0,
      "step": 62790
    },
    {
      "entropy": 5.744601106643676,
      "epoch": 5.6962082728592165,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022887205058113458,
      "loss": 4.9217,
      "mean_token_accuracy": 0.22650371491909027,
      "num_tokens": 130322411.0,
      "step": 62795
    },
    {
      "entropy": 5.74550724029541,
      "epoch": 5.696661828737301,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002288403819137611,
      "loss": 4.8778,
      "mean_token_accuracy": 0.23398361802101136,
      "num_tokens": 130332422.0,
      "step": 62800
    },
    {
      "entropy": 5.69567461013794,
      "epoch": 5.697115384615385,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002288087142009803,
      "loss": 4.8157,
      "mean_token_accuracy": 0.23364341408014297,
      "num_tokens": 130342221.0,
      "step": 62805
    },
    {
      "entropy": 5.735183811187744,
      "epoch": 5.697568940493468,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000228777047443447,
      "loss": 4.8385,
      "mean_token_accuracy": 0.23240811824798585,
      "num_tokens": 130352882.0,
      "step": 62810
    },
    {
      "entropy": 5.715462112426758,
      "epoch": 5.698022496371553,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022874538164181625,
      "loss": 4.8616,
      "mean_token_accuracy": 0.22306873500347138,
      "num_tokens": 130363521.0,
      "step": 62815
    },
    {
      "entropy": 5.681339979171753,
      "epoch": 5.698476052249637,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022871371679674275,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22841319292783738,
      "num_tokens": 130374737.0,
      "step": 62820
    },
    {
      "entropy": 5.788036727905274,
      "epoch": 5.698929608127721,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022868205290888134,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2267611011862755,
      "num_tokens": 130385677.0,
      "step": 62825
    },
    {
      "entropy": 5.750956916809082,
      "epoch": 5.6993831640058055,
      "grad_norm": 1.40625,
      "learning_rate": 0.000228650389978887,
      "loss": 4.8647,
      "mean_token_accuracy": 0.23269987106323242,
      "num_tokens": 130396708.0,
      "step": 62830
    },
    {
      "entropy": 5.685616111755371,
      "epoch": 5.69983671988389,
      "grad_norm": 1.5,
      "learning_rate": 0.0002286187280074143,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2342742145061493,
      "num_tokens": 130406789.0,
      "step": 62835
    },
    {
      "entropy": 5.671590566635132,
      "epoch": 5.700290275761974,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002285870669951181,
      "loss": 4.7885,
      "mean_token_accuracy": 0.24718116074800492,
      "num_tokens": 130416876.0,
      "step": 62840
    },
    {
      "entropy": 5.729632902145386,
      "epoch": 5.700743831640058,
      "grad_norm": 1.375,
      "learning_rate": 0.00022855540694265326,
      "loss": 4.8254,
      "mean_token_accuracy": 0.2341738075017929,
      "num_tokens": 130427580.0,
      "step": 62845
    },
    {
      "entropy": 5.719961071014405,
      "epoch": 5.7011973875181425,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022852374785067438,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2295947328209877,
      "num_tokens": 130437870.0,
      "step": 62850
    },
    {
      "entropy": 5.666269540786743,
      "epoch": 5.701650943396227,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022849208971983625,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23261915594339372,
      "num_tokens": 130447852.0,
      "step": 62855
    },
    {
      "entropy": 5.702665138244629,
      "epoch": 5.702104499274311,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022846043255079357,
      "loss": 4.8744,
      "mean_token_accuracy": 0.23225070387125016,
      "num_tokens": 130458322.0,
      "step": 62860
    },
    {
      "entropy": 5.7153972625732425,
      "epoch": 5.7025580551523944,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000228428776344201,
      "loss": 4.7707,
      "mean_token_accuracy": 0.23792447596788407,
      "num_tokens": 130468344.0,
      "step": 62865
    },
    {
      "entropy": 5.674361085891723,
      "epoch": 5.703011611030479,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022839712110071315,
      "loss": 4.7633,
      "mean_token_accuracy": 0.23437548726797103,
      "num_tokens": 130478288.0,
      "step": 62870
    },
    {
      "entropy": 5.6497876167297365,
      "epoch": 5.703465166908563,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002283654668209848,
      "loss": 4.846,
      "mean_token_accuracy": 0.23170033246278762,
      "num_tokens": 130488223.0,
      "step": 62875
    },
    {
      "entropy": 5.583140087127686,
      "epoch": 5.703918722786647,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022833381350567046,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2447405606508255,
      "num_tokens": 130498602.0,
      "step": 62880
    },
    {
      "entropy": 5.663861465454102,
      "epoch": 5.7043722786647315,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022830216115542467,
      "loss": 4.7908,
      "mean_token_accuracy": 0.24123494923114777,
      "num_tokens": 130509479.0,
      "step": 62885
    },
    {
      "entropy": 5.672888469696045,
      "epoch": 5.704825834542816,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002282705097709022,
      "loss": 4.8593,
      "mean_token_accuracy": 0.22845116257667542,
      "num_tokens": 130520606.0,
      "step": 62890
    },
    {
      "entropy": 5.732314586639404,
      "epoch": 5.7052793904209,
      "grad_norm": 1.25,
      "learning_rate": 0.00022823885935275746,
      "loss": 4.9291,
      "mean_token_accuracy": 0.2244260609149933,
      "num_tokens": 130532447.0,
      "step": 62895
    },
    {
      "entropy": 5.689919281005859,
      "epoch": 5.705732946298984,
      "grad_norm": 1.5,
      "learning_rate": 0.00022820720990164502,
      "loss": 4.8143,
      "mean_token_accuracy": 0.22928913980722426,
      "num_tokens": 130542047.0,
      "step": 62900
    },
    {
      "entropy": 5.662133073806762,
      "epoch": 5.706186502177069,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022817556141821932,
      "loss": 4.7936,
      "mean_token_accuracy": 0.23312143981456757,
      "num_tokens": 130552631.0,
      "step": 62905
    },
    {
      "entropy": 5.666987895965576,
      "epoch": 5.706640058055152,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022814391390313504,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23213072568178178,
      "num_tokens": 130562323.0,
      "step": 62910
    },
    {
      "entropy": 5.6874110221862795,
      "epoch": 5.707093613933236,
      "grad_norm": 1.5,
      "learning_rate": 0.00022811226735704654,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23270241618156434,
      "num_tokens": 130573310.0,
      "step": 62915
    },
    {
      "entropy": 5.7375883102417,
      "epoch": 5.7075471698113205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022808062178060828,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22886516004800797,
      "num_tokens": 130584059.0,
      "step": 62920
    },
    {
      "entropy": 5.682730865478516,
      "epoch": 5.708000725689405,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022804897717447476,
      "loss": 4.8038,
      "mean_token_accuracy": 0.22870425879955292,
      "num_tokens": 130594101.0,
      "step": 62925
    },
    {
      "entropy": 5.640892362594604,
      "epoch": 5.708454281567489,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022801733353930038,
      "loss": 4.7944,
      "mean_token_accuracy": 0.2343431904911995,
      "num_tokens": 130604531.0,
      "step": 62930
    },
    {
      "entropy": 5.658610153198242,
      "epoch": 5.708907837445573,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002279856908757394,
      "loss": 4.7619,
      "mean_token_accuracy": 0.24331524223089218,
      "num_tokens": 130614213.0,
      "step": 62935
    },
    {
      "entropy": 5.6416689395904545,
      "epoch": 5.7093613933236576,
      "grad_norm": 1.375,
      "learning_rate": 0.0002279540491844464,
      "loss": 4.784,
      "mean_token_accuracy": 0.23386640846729279,
      "num_tokens": 130624809.0,
      "step": 62940
    },
    {
      "entropy": 5.678088092803955,
      "epoch": 5.709814949201742,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022792240846607558,
      "loss": 4.8583,
      "mean_token_accuracy": 0.2302267462015152,
      "num_tokens": 130633874.0,
      "step": 62945
    },
    {
      "entropy": 5.626293468475342,
      "epoch": 5.710268505079826,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022789076872128144,
      "loss": 4.8317,
      "mean_token_accuracy": 0.2370610147714615,
      "num_tokens": 130644256.0,
      "step": 62950
    },
    {
      "entropy": 5.6638116359710695,
      "epoch": 5.71072206095791,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022785912995071817,
      "loss": 4.8076,
      "mean_token_accuracy": 0.2361317366361618,
      "num_tokens": 130654318.0,
      "step": 62955
    },
    {
      "entropy": 5.703345060348511,
      "epoch": 5.711175616835995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002278274921550401,
      "loss": 4.8376,
      "mean_token_accuracy": 0.22421566694974898,
      "num_tokens": 130664256.0,
      "step": 62960
    },
    {
      "entropy": 5.798236560821533,
      "epoch": 5.711629172714078,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022779585533490155,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2226380631327629,
      "num_tokens": 130674238.0,
      "step": 62965
    },
    {
      "entropy": 5.723135375976563,
      "epoch": 5.712082728592162,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022776421949095667,
      "loss": 4.843,
      "mean_token_accuracy": 0.22149204909801484,
      "num_tokens": 130685308.0,
      "step": 62970
    },
    {
      "entropy": 5.732873868942261,
      "epoch": 5.7125362844702465,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022773258462385981,
      "loss": 4.8318,
      "mean_token_accuracy": 0.23698252141475679,
      "num_tokens": 130695773.0,
      "step": 62975
    },
    {
      "entropy": 5.773996353149414,
      "epoch": 5.712989840348331,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002277009507342651,
      "loss": 4.8638,
      "mean_token_accuracy": 0.23249631077051164,
      "num_tokens": 130705730.0,
      "step": 62980
    },
    {
      "entropy": 5.725247573852539,
      "epoch": 5.713443396226415,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022766931782282685,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22640817761421203,
      "num_tokens": 130717154.0,
      "step": 62985
    },
    {
      "entropy": 5.6692214012146,
      "epoch": 5.713896952104499,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022763768589019912,
      "loss": 4.8586,
      "mean_token_accuracy": 0.2278900235891342,
      "num_tokens": 130726963.0,
      "step": 62990
    },
    {
      "entropy": 5.682397651672363,
      "epoch": 5.714350507982584,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022760605493703616,
      "loss": 4.91,
      "mean_token_accuracy": 0.2243878185749054,
      "num_tokens": 130736736.0,
      "step": 62995
    },
    {
      "entropy": 5.66461820602417,
      "epoch": 5.714804063860668,
      "grad_norm": 1.375,
      "learning_rate": 0.00022757442496399206,
      "loss": 4.7704,
      "mean_token_accuracy": 0.23821602016687393,
      "num_tokens": 130746758.0,
      "step": 63000
    },
    {
      "epoch": 5.714804063860668,
      "eval_entropy": 5.415451145927291,
      "eval_loss": 4.939684867858887,
      "eval_mean_token_accuracy": 0.23387119694701416,
      "eval_num_tokens": 130746758.0,
      "eval_runtime": 20.6651,
      "eval_samples_per_second": 1711.096,
      "eval_steps_per_second": 213.887,
      "step": 63000
    },
    {
      "entropy": 5.686853694915771,
      "epoch": 5.715257619738752,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022754279597172084,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23511806428432463,
      "num_tokens": 130756962.0,
      "step": 63005
    },
    {
      "entropy": 5.699848556518555,
      "epoch": 5.7157111756168355,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002275111679608768,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2320326790213585,
      "num_tokens": 130768553.0,
      "step": 63010
    },
    {
      "entropy": 5.757534170150757,
      "epoch": 5.716164731494921,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022747954093211388,
      "loss": 4.9123,
      "mean_token_accuracy": 0.2244052991271019,
      "num_tokens": 130778683.0,
      "step": 63015
    },
    {
      "entropy": 5.659536266326905,
      "epoch": 5.716618287373004,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002274479148860862,
      "loss": 4.7892,
      "mean_token_accuracy": 0.2404308319091797,
      "num_tokens": 130789109.0,
      "step": 63020
    },
    {
      "entropy": 5.643398666381836,
      "epoch": 5.717071843251088,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002274162898234477,
      "loss": 4.7994,
      "mean_token_accuracy": 0.23691997677087784,
      "num_tokens": 130800376.0,
      "step": 63025
    },
    {
      "entropy": 5.764399147033691,
      "epoch": 5.717525399129173,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022738466574485244,
      "loss": 4.913,
      "mean_token_accuracy": 0.2267802283167839,
      "num_tokens": 130811688.0,
      "step": 63030
    },
    {
      "entropy": 5.711898899078369,
      "epoch": 5.717978955007257,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022735304265095445,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22930900752544403,
      "num_tokens": 130822137.0,
      "step": 63035
    },
    {
      "entropy": 5.666014623641968,
      "epoch": 5.718432510885341,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022732142054240767,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23397473245859146,
      "num_tokens": 130831635.0,
      "step": 63040
    },
    {
      "entropy": 5.687845277786255,
      "epoch": 5.718886066763425,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022728979941986612,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2340848445892334,
      "num_tokens": 130841277.0,
      "step": 63045
    },
    {
      "entropy": 5.592782068252563,
      "epoch": 5.71933962264151,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022725817928398367,
      "loss": 4.7271,
      "mean_token_accuracy": 0.23751054257154464,
      "num_tokens": 130851252.0,
      "step": 63050
    },
    {
      "entropy": 5.718686580657959,
      "epoch": 5.719793178519594,
      "grad_norm": 1.375,
      "learning_rate": 0.00022722656013541415,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23347547948360442,
      "num_tokens": 130861756.0,
      "step": 63055
    },
    {
      "entropy": 5.713470506668091,
      "epoch": 5.720246734397678,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022719494197481155,
      "loss": 4.7823,
      "mean_token_accuracy": 0.23758601099252702,
      "num_tokens": 130871284.0,
      "step": 63060
    },
    {
      "entropy": 5.710532379150391,
      "epoch": 5.7207002902757615,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002271633248028298,
      "loss": 4.8041,
      "mean_token_accuracy": 0.2355360895395279,
      "num_tokens": 130882320.0,
      "step": 63065
    },
    {
      "entropy": 5.710180473327637,
      "epoch": 5.721153846153846,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022713170862012267,
      "loss": 4.8467,
      "mean_token_accuracy": 0.21973631978034974,
      "num_tokens": 130893711.0,
      "step": 63070
    },
    {
      "entropy": 5.657866382598877,
      "epoch": 5.72160740203193,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000227100093427344,
      "loss": 4.83,
      "mean_token_accuracy": 0.23055712282657623,
      "num_tokens": 130903543.0,
      "step": 63075
    },
    {
      "entropy": 5.65033917427063,
      "epoch": 5.722060957910014,
      "grad_norm": 1.5,
      "learning_rate": 0.0002270684792251476,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24237634837627411,
      "num_tokens": 130913819.0,
      "step": 63080
    },
    {
      "entropy": 5.68483338356018,
      "epoch": 5.722514513788099,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022703686601418727,
      "loss": 4.837,
      "mean_token_accuracy": 0.24108044058084488,
      "num_tokens": 130924628.0,
      "step": 63085
    },
    {
      "entropy": 5.688007259368897,
      "epoch": 5.722968069666183,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022700525379511672,
      "loss": 4.83,
      "mean_token_accuracy": 0.23169467747211456,
      "num_tokens": 130934163.0,
      "step": 63090
    },
    {
      "entropy": 5.708333110809326,
      "epoch": 5.723421625544267,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022697364256858983,
      "loss": 4.7822,
      "mean_token_accuracy": 0.23416255563497543,
      "num_tokens": 130944449.0,
      "step": 63095
    },
    {
      "entropy": 5.754124307632447,
      "epoch": 5.723875181422351,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022694203233526023,
      "loss": 4.8494,
      "mean_token_accuracy": 0.23744977414608,
      "num_tokens": 130954595.0,
      "step": 63100
    },
    {
      "entropy": 5.714716529846191,
      "epoch": 5.724328737300436,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022691042309578164,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22687697410583496,
      "num_tokens": 130965405.0,
      "step": 63105
    },
    {
      "entropy": 5.698559761047363,
      "epoch": 5.72478229317852,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022687881485080775,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22261785119771957,
      "num_tokens": 130975984.0,
      "step": 63110
    },
    {
      "entropy": 5.703161907196045,
      "epoch": 5.725235849056604,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002268472076009922,
      "loss": 4.8313,
      "mean_token_accuracy": 0.2341375544667244,
      "num_tokens": 130987038.0,
      "step": 63115
    },
    {
      "entropy": 5.68642520904541,
      "epoch": 5.725689404934688,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002268156013469887,
      "loss": 4.7663,
      "mean_token_accuracy": 0.2383050188422203,
      "num_tokens": 130996872.0,
      "step": 63120
    },
    {
      "entropy": 5.719441986083984,
      "epoch": 5.726142960812772,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022678399608945082,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23222773224115373,
      "num_tokens": 131008128.0,
      "step": 63125
    },
    {
      "entropy": 5.74805269241333,
      "epoch": 5.726596516690856,
      "grad_norm": 1.375,
      "learning_rate": 0.00022675239182903223,
      "loss": 4.9535,
      "mean_token_accuracy": 0.22380514442920685,
      "num_tokens": 131019014.0,
      "step": 63130
    },
    {
      "entropy": 5.736402463912964,
      "epoch": 5.72705007256894,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022672078856638645,
      "loss": 4.8734,
      "mean_token_accuracy": 0.23975236862897872,
      "num_tokens": 131029148.0,
      "step": 63135
    },
    {
      "entropy": 5.670983600616455,
      "epoch": 5.727503628447025,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022668918630216705,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23411910682916642,
      "num_tokens": 131039290.0,
      "step": 63140
    },
    {
      "entropy": 5.686109352111816,
      "epoch": 5.727957184325109,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002266575850370276,
      "loss": 4.7995,
      "mean_token_accuracy": 0.2338786616921425,
      "num_tokens": 131050264.0,
      "step": 63145
    },
    {
      "entropy": 5.646191787719727,
      "epoch": 5.728410740203193,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022662598477162166,
      "loss": 4.7927,
      "mean_token_accuracy": 0.22836600095033646,
      "num_tokens": 131060089.0,
      "step": 63150
    },
    {
      "entropy": 5.743436288833618,
      "epoch": 5.7288642960812775,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002265943855066026,
      "loss": 4.9213,
      "mean_token_accuracy": 0.2220624193549156,
      "num_tokens": 131072336.0,
      "step": 63155
    },
    {
      "entropy": 5.697641801834107,
      "epoch": 5.729317851959362,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000226562787242624,
      "loss": 4.841,
      "mean_token_accuracy": 0.22818912714719772,
      "num_tokens": 131082028.0,
      "step": 63160
    },
    {
      "entropy": 5.755266284942627,
      "epoch": 5.729771407837445,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022653118998033932,
      "loss": 4.8813,
      "mean_token_accuracy": 0.23088353872299194,
      "num_tokens": 131091899.0,
      "step": 63165
    },
    {
      "entropy": 5.674013328552246,
      "epoch": 5.73022496371553,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022649959372040207,
      "loss": 4.769,
      "mean_token_accuracy": 0.23614854961633683,
      "num_tokens": 131102385.0,
      "step": 63170
    },
    {
      "entropy": 5.719228363037109,
      "epoch": 5.730678519593614,
      "grad_norm": 1.375,
      "learning_rate": 0.00022646799846346544,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2247608095407486,
      "num_tokens": 131112407.0,
      "step": 63175
    },
    {
      "entropy": 5.6935076236724855,
      "epoch": 5.731132075471698,
      "grad_norm": 1.5,
      "learning_rate": 0.000226436404210183,
      "loss": 4.7891,
      "mean_token_accuracy": 0.2314327225089073,
      "num_tokens": 131122450.0,
      "step": 63180
    },
    {
      "entropy": 5.681711292266845,
      "epoch": 5.731585631349782,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002264048109612082,
      "loss": 4.7731,
      "mean_token_accuracy": 0.2332440957427025,
      "num_tokens": 131133409.0,
      "step": 63185
    },
    {
      "entropy": 5.73307318687439,
      "epoch": 5.732039187227866,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022637321871719419,
      "loss": 4.8831,
      "mean_token_accuracy": 0.23753801584243775,
      "num_tokens": 131142878.0,
      "step": 63190
    },
    {
      "entropy": 5.716517639160156,
      "epoch": 5.732492743105951,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002263416274787944,
      "loss": 4.8053,
      "mean_token_accuracy": 0.23467761725187303,
      "num_tokens": 131152442.0,
      "step": 63195
    },
    {
      "entropy": 5.784765386581421,
      "epoch": 5.732946298984035,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002263100372466622,
      "loss": 4.8636,
      "mean_token_accuracy": 0.23366738259792327,
      "num_tokens": 131163581.0,
      "step": 63200
    },
    {
      "entropy": 5.670667600631714,
      "epoch": 5.733399854862119,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022627844802145086,
      "loss": 4.7523,
      "mean_token_accuracy": 0.23678569495677948,
      "num_tokens": 131173198.0,
      "step": 63205
    },
    {
      "entropy": 5.696275043487549,
      "epoch": 5.7338534107402035,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022624685980381353,
      "loss": 4.8898,
      "mean_token_accuracy": 0.22279966473579407,
      "num_tokens": 131183073.0,
      "step": 63210
    },
    {
      "entropy": 5.689581775665284,
      "epoch": 5.734306966618288,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022621527259440366,
      "loss": 4.8414,
      "mean_token_accuracy": 0.2329220399260521,
      "num_tokens": 131194222.0,
      "step": 63215
    },
    {
      "entropy": 5.77886734008789,
      "epoch": 5.734760522496371,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022618368639387438,
      "loss": 4.9169,
      "mean_token_accuracy": 0.23275212198495865,
      "num_tokens": 131204724.0,
      "step": 63220
    },
    {
      "entropy": 5.785861539840698,
      "epoch": 5.735214078374455,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022615210120287887,
      "loss": 4.9624,
      "mean_token_accuracy": 0.22598420828580856,
      "num_tokens": 131214079.0,
      "step": 63225
    },
    {
      "entropy": 5.732250356674195,
      "epoch": 5.73566763425254,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022612051702207032,
      "loss": 4.866,
      "mean_token_accuracy": 0.22469810396432877,
      "num_tokens": 131224494.0,
      "step": 63230
    },
    {
      "entropy": 5.672440767288208,
      "epoch": 5.736121190130624,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000226088933852102,
      "loss": 4.8157,
      "mean_token_accuracy": 0.23359074890613557,
      "num_tokens": 131235604.0,
      "step": 63235
    },
    {
      "entropy": 5.642015600204468,
      "epoch": 5.736574746008708,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022605735169362701,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23179123252630235,
      "num_tokens": 131246857.0,
      "step": 63240
    },
    {
      "entropy": 5.731518316268921,
      "epoch": 5.7370283018867925,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002260257705472984,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22232674509286882,
      "num_tokens": 131257508.0,
      "step": 63245
    },
    {
      "entropy": 5.725886678695678,
      "epoch": 5.737481857764877,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022599419041376934,
      "loss": 4.8838,
      "mean_token_accuracy": 0.23048363178968428,
      "num_tokens": 131267820.0,
      "step": 63250
    },
    {
      "entropy": 5.779127168655395,
      "epoch": 5.737935413642961,
      "grad_norm": 1.34375,
      "learning_rate": 0.000225962611293693,
      "loss": 4.9557,
      "mean_token_accuracy": 0.22196493446826934,
      "num_tokens": 131278307.0,
      "step": 63255
    },
    {
      "entropy": 5.717890024185181,
      "epoch": 5.738388969521045,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022593103318772228,
      "loss": 4.821,
      "mean_token_accuracy": 0.23404375910758973,
      "num_tokens": 131288576.0,
      "step": 63260
    },
    {
      "entropy": 5.718027925491333,
      "epoch": 5.7388425253991295,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002258994560965103,
      "loss": 4.8112,
      "mean_token_accuracy": 0.23255764842033386,
      "num_tokens": 131299419.0,
      "step": 63265
    },
    {
      "entropy": 5.703716230392456,
      "epoch": 5.739296081277214,
      "grad_norm": 1.375,
      "learning_rate": 0.00022586788002071007,
      "loss": 4.857,
      "mean_token_accuracy": 0.22847480028867723,
      "num_tokens": 131309722.0,
      "step": 63270
    },
    {
      "entropy": 5.726007747650146,
      "epoch": 5.739749637155297,
      "grad_norm": 1.5,
      "learning_rate": 0.00022583630496097456,
      "loss": 4.798,
      "mean_token_accuracy": 0.23745923340320588,
      "num_tokens": 131318955.0,
      "step": 63275
    },
    {
      "entropy": 5.734606790542602,
      "epoch": 5.7402031930333814,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022580473091795688,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22497989982366562,
      "num_tokens": 131329113.0,
      "step": 63280
    },
    {
      "entropy": 5.731538200378418,
      "epoch": 5.740656748911466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002257731578923099,
      "loss": 4.89,
      "mean_token_accuracy": 0.22249395847320558,
      "num_tokens": 131339473.0,
      "step": 63285
    },
    {
      "entropy": 5.753970432281494,
      "epoch": 5.74111030478955,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022574158588468647,
      "loss": 4.9972,
      "mean_token_accuracy": 0.22218993306159973,
      "num_tokens": 131349654.0,
      "step": 63290
    },
    {
      "entropy": 5.802382087707519,
      "epoch": 5.741563860667634,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022571001489573968,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2303476467728615,
      "num_tokens": 131359870.0,
      "step": 63295
    },
    {
      "entropy": 5.696162271499634,
      "epoch": 5.7420174165457185,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022567844492612233,
      "loss": 4.7531,
      "mean_token_accuracy": 0.23567847907543182,
      "num_tokens": 131369568.0,
      "step": 63300
    },
    {
      "entropy": 5.849718284606934,
      "epoch": 5.742470972423803,
      "grad_norm": 1.375,
      "learning_rate": 0.00022564687597648726,
      "loss": 5.0562,
      "mean_token_accuracy": 0.21899161040782927,
      "num_tokens": 131381254.0,
      "step": 63305
    },
    {
      "entropy": 5.65283899307251,
      "epoch": 5.742924528301887,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022561530804748737,
      "loss": 4.7579,
      "mean_token_accuracy": 0.241523440182209,
      "num_tokens": 131391182.0,
      "step": 63310
    },
    {
      "entropy": 5.76723165512085,
      "epoch": 5.743378084179971,
      "grad_norm": 1.375,
      "learning_rate": 0.0002255837411397756,
      "loss": 4.9796,
      "mean_token_accuracy": 0.2197033330798149,
      "num_tokens": 131402125.0,
      "step": 63315
    },
    {
      "entropy": 5.7264015674591064,
      "epoch": 5.743831640058055,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022555217525400457,
      "loss": 4.8362,
      "mean_token_accuracy": 0.21928201913833617,
      "num_tokens": 131411994.0,
      "step": 63320
    },
    {
      "entropy": 5.738026523590088,
      "epoch": 5.744285195936139,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002255206103908271,
      "loss": 4.8892,
      "mean_token_accuracy": 0.22544224113225936,
      "num_tokens": 131423864.0,
      "step": 63325
    },
    {
      "entropy": 5.6593324661254885,
      "epoch": 5.744738751814223,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022548904655089608,
      "loss": 4.7725,
      "mean_token_accuracy": 0.22993759214878082,
      "num_tokens": 131434320.0,
      "step": 63330
    },
    {
      "entropy": 5.694047355651856,
      "epoch": 5.7451923076923075,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022545748373486424,
      "loss": 4.844,
      "mean_token_accuracy": 0.22731305062770843,
      "num_tokens": 131444628.0,
      "step": 63335
    },
    {
      "entropy": 5.700935888290405,
      "epoch": 5.745645863570392,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022542592194338418,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23699092566967012,
      "num_tokens": 131455644.0,
      "step": 63340
    },
    {
      "entropy": 5.830265235900879,
      "epoch": 5.746099419448476,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022539436117710877,
      "loss": 4.9216,
      "mean_token_accuracy": 0.23563510924577713,
      "num_tokens": 131466138.0,
      "step": 63345
    },
    {
      "entropy": 5.784493255615234,
      "epoch": 5.74655297532656,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022536280143669053,
      "loss": 4.9572,
      "mean_token_accuracy": 0.22323852628469468,
      "num_tokens": 131476153.0,
      "step": 63350
    },
    {
      "entropy": 5.712936782836914,
      "epoch": 5.7470065312046446,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022533124272278222,
      "loss": 4.8468,
      "mean_token_accuracy": 0.22877589911222457,
      "num_tokens": 131485973.0,
      "step": 63355
    },
    {
      "entropy": 5.70575098991394,
      "epoch": 5.747460087082729,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022529968503603654,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2273195743560791,
      "num_tokens": 131496428.0,
      "step": 63360
    },
    {
      "entropy": 5.754439640045166,
      "epoch": 5.747913642960813,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022526812837710597,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22434744387865066,
      "num_tokens": 131507836.0,
      "step": 63365
    },
    {
      "entropy": 5.712545871734619,
      "epoch": 5.748367198838897,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002252365727466432,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22143404334783554,
      "num_tokens": 131518548.0,
      "step": 63370
    },
    {
      "entropy": 5.6591424465179445,
      "epoch": 5.748820754716981,
      "grad_norm": 1.375,
      "learning_rate": 0.00022520501814530074,
      "loss": 4.7491,
      "mean_token_accuracy": 0.23576533496379853,
      "num_tokens": 131528784.0,
      "step": 63375
    },
    {
      "entropy": 5.669259643554687,
      "epoch": 5.749274310595065,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022517346457373127,
      "loss": 4.8245,
      "mean_token_accuracy": 0.23776365369558333,
      "num_tokens": 131539617.0,
      "step": 63380
    },
    {
      "entropy": 5.716305112838745,
      "epoch": 5.749727866473149,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022514191203258722,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22639358043670654,
      "num_tokens": 131550279.0,
      "step": 63385
    },
    {
      "entropy": 5.710794353485108,
      "epoch": 5.7501814223512335,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002251103605225211,
      "loss": 4.838,
      "mean_token_accuracy": 0.23499393761157988,
      "num_tokens": 131561494.0,
      "step": 63390
    },
    {
      "entropy": 5.714746618270874,
      "epoch": 5.750634978229318,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022507881004418546,
      "loss": 4.8248,
      "mean_token_accuracy": 0.23626979291439057,
      "num_tokens": 131572166.0,
      "step": 63395
    },
    {
      "entropy": 5.7467200756073,
      "epoch": 5.751088534107402,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022504726059823277,
      "loss": 4.9138,
      "mean_token_accuracy": 0.22410165667533874,
      "num_tokens": 131582224.0,
      "step": 63400
    },
    {
      "entropy": 5.657828235626221,
      "epoch": 5.751542089985486,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022501571218531554,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2313920259475708,
      "num_tokens": 131592612.0,
      "step": 63405
    },
    {
      "entropy": 5.675454902648926,
      "epoch": 5.751995645863571,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022498416480608597,
      "loss": 4.7977,
      "mean_token_accuracy": 0.2312883660197258,
      "num_tokens": 131603002.0,
      "step": 63410
    },
    {
      "entropy": 5.694905996322632,
      "epoch": 5.752449201741655,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002249526184611967,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22645934373140336,
      "num_tokens": 131614045.0,
      "step": 63415
    },
    {
      "entropy": 5.72091760635376,
      "epoch": 5.752902757619739,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002249210731513,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23558998852968216,
      "num_tokens": 131623395.0,
      "step": 63420
    },
    {
      "entropy": 5.699629592895508,
      "epoch": 5.753356313497823,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022488952887704823,
      "loss": 4.8636,
      "mean_token_accuracy": 0.23203110843896865,
      "num_tokens": 131633985.0,
      "step": 63425
    },
    {
      "entropy": 5.683747673034668,
      "epoch": 5.753809869375907,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022485798563909383,
      "loss": 4.7503,
      "mean_token_accuracy": 0.23852866888046265,
      "num_tokens": 131644136.0,
      "step": 63430
    },
    {
      "entropy": 5.666155815124512,
      "epoch": 5.754263425253991,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002248264434380891,
      "loss": 4.8064,
      "mean_token_accuracy": 0.22998733073472977,
      "num_tokens": 131654197.0,
      "step": 63435
    },
    {
      "entropy": 5.749076795578003,
      "epoch": 5.754716981132075,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022479490227468623,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22965109050273896,
      "num_tokens": 131664697.0,
      "step": 63440
    },
    {
      "entropy": 5.6820472240447994,
      "epoch": 5.75517053701016,
      "grad_norm": 1.375,
      "learning_rate": 0.00022476336214953757,
      "loss": 4.801,
      "mean_token_accuracy": 0.23977261632680893,
      "num_tokens": 131674822.0,
      "step": 63445
    },
    {
      "entropy": 5.693502712249756,
      "epoch": 5.755624092888244,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022473182306329548,
      "loss": 4.8121,
      "mean_token_accuracy": 0.2306712970137596,
      "num_tokens": 131684716.0,
      "step": 63450
    },
    {
      "entropy": 5.726043319702148,
      "epoch": 5.756077648766328,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022470028501661205,
      "loss": 4.9343,
      "mean_token_accuracy": 0.21777404844760895,
      "num_tokens": 131696608.0,
      "step": 63455
    },
    {
      "entropy": 5.704763269424438,
      "epoch": 5.756531204644412,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022466874801013948,
      "loss": 4.8405,
      "mean_token_accuracy": 0.23036837726831436,
      "num_tokens": 131707247.0,
      "step": 63460
    },
    {
      "entropy": 5.672975540161133,
      "epoch": 5.756984760522497,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022463721204453007,
      "loss": 4.7977,
      "mean_token_accuracy": 0.2429825782775879,
      "num_tokens": 131716877.0,
      "step": 63465
    },
    {
      "entropy": 5.643925428390503,
      "epoch": 5.757438316400581,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000224605677120436,
      "loss": 4.7646,
      "mean_token_accuracy": 0.2313605010509491,
      "num_tokens": 131728061.0,
      "step": 63470
    },
    {
      "entropy": 5.621161413192749,
      "epoch": 5.757891872278664,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022457414323850928,
      "loss": 4.7159,
      "mean_token_accuracy": 0.23923666328191756,
      "num_tokens": 131737679.0,
      "step": 63475
    },
    {
      "entropy": 5.758856773376465,
      "epoch": 5.7583454281567485,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002245426103994022,
      "loss": 4.9451,
      "mean_token_accuracy": 0.22139064967632294,
      "num_tokens": 131747570.0,
      "step": 63480
    },
    {
      "entropy": 5.623011922836303,
      "epoch": 5.758798984034833,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002245110786037668,
      "loss": 4.6884,
      "mean_token_accuracy": 0.23950936049222946,
      "num_tokens": 131757216.0,
      "step": 63485
    },
    {
      "entropy": 5.618896532058716,
      "epoch": 5.759252539912917,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022447954785225516,
      "loss": 4.7573,
      "mean_token_accuracy": 0.23708366155624389,
      "num_tokens": 131767964.0,
      "step": 63490
    },
    {
      "entropy": 5.664538764953614,
      "epoch": 5.759706095791001,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022444801814551924,
      "loss": 4.9182,
      "mean_token_accuracy": 0.22437411695718765,
      "num_tokens": 131779155.0,
      "step": 63495
    },
    {
      "entropy": 5.7186281204223635,
      "epoch": 5.760159651669086,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002244164894842113,
      "loss": 4.8422,
      "mean_token_accuracy": 0.23568241596221923,
      "num_tokens": 131789091.0,
      "step": 63500
    },
    {
      "entropy": 5.681352043151856,
      "epoch": 5.76061320754717,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022438496186898322,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22722960263490677,
      "num_tokens": 131799205.0,
      "step": 63505
    },
    {
      "entropy": 5.668429327011109,
      "epoch": 5.761066763425254,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022435343530048695,
      "loss": 4.8345,
      "mean_token_accuracy": 0.22536790370941162,
      "num_tokens": 131808803.0,
      "step": 63510
    },
    {
      "entropy": 5.7067484855651855,
      "epoch": 5.761520319303338,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022432190977937462,
      "loss": 4.8456,
      "mean_token_accuracy": 0.2290523588657379,
      "num_tokens": 131818682.0,
      "step": 63515
    },
    {
      "entropy": 5.710638046264648,
      "epoch": 5.761973875181423,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022429038530629813,
      "loss": 4.8545,
      "mean_token_accuracy": 0.2174447551369667,
      "num_tokens": 131829705.0,
      "step": 63520
    },
    {
      "entropy": 5.686687803268432,
      "epoch": 5.762427431059507,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022425886188190937,
      "loss": 4.806,
      "mean_token_accuracy": 0.23263431787490846,
      "num_tokens": 131839206.0,
      "step": 63525
    },
    {
      "entropy": 5.662824487686157,
      "epoch": 5.76288098693759,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022422733950686025,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2304491549730301,
      "num_tokens": 131848695.0,
      "step": 63530
    },
    {
      "entropy": 5.724411487579346,
      "epoch": 5.763334542815675,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022419581818180275,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22479831129312516,
      "num_tokens": 131858443.0,
      "step": 63535
    },
    {
      "entropy": 5.752908229827881,
      "epoch": 5.763788098693759,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022416429790738862,
      "loss": 4.8901,
      "mean_token_accuracy": 0.2304064929485321,
      "num_tokens": 131868280.0,
      "step": 63540
    },
    {
      "entropy": 5.785575294494629,
      "epoch": 5.764241654571843,
      "grad_norm": 1.375,
      "learning_rate": 0.00022413277868426972,
      "loss": 4.9152,
      "mean_token_accuracy": 0.22138692438602448,
      "num_tokens": 131880013.0,
      "step": 63545
    },
    {
      "entropy": 5.7676756381988525,
      "epoch": 5.764695210449927,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022410126051309803,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22621283531188965,
      "num_tokens": 131890828.0,
      "step": 63550
    },
    {
      "entropy": 5.717794799804688,
      "epoch": 5.765148766328012,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002240697433945252,
      "loss": 4.8168,
      "mean_token_accuracy": 0.2332173079252243,
      "num_tokens": 131900817.0,
      "step": 63555
    },
    {
      "entropy": 5.74787974357605,
      "epoch": 5.765602322206096,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000224038227329203,
      "loss": 4.8481,
      "mean_token_accuracy": 0.22847286611795425,
      "num_tokens": 131911237.0,
      "step": 63560
    },
    {
      "entropy": 5.713052082061767,
      "epoch": 5.76605587808418,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002240067123177833,
      "loss": 4.8767,
      "mean_token_accuracy": 0.2293548434972763,
      "num_tokens": 131921429.0,
      "step": 63565
    },
    {
      "entropy": 5.6516258239746096,
      "epoch": 5.7665094339622645,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002239751983609178,
      "loss": 4.8158,
      "mean_token_accuracy": 0.23042524605989456,
      "num_tokens": 131932812.0,
      "step": 63570
    },
    {
      "entropy": 5.780234003067017,
      "epoch": 5.766962989840349,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002239436854592582,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22497458159923553,
      "num_tokens": 131943360.0,
      "step": 63575
    },
    {
      "entropy": 5.743438386917115,
      "epoch": 5.767416545718433,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022391217361345612,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22683090418577195,
      "num_tokens": 131953713.0,
      "step": 63580
    },
    {
      "entropy": 5.686519050598145,
      "epoch": 5.767870101596516,
      "grad_norm": 1.375,
      "learning_rate": 0.00022388066282416337,
      "loss": 4.7981,
      "mean_token_accuracy": 0.2327667087316513,
      "num_tokens": 131964959.0,
      "step": 63585
    },
    {
      "entropy": 5.7109198570251465,
      "epoch": 5.768323657474601,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022384915309203151,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22709384560585022,
      "num_tokens": 131975127.0,
      "step": 63590
    },
    {
      "entropy": 5.663993406295776,
      "epoch": 5.768777213352685,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022381764441771224,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2346167579293251,
      "num_tokens": 131985125.0,
      "step": 63595
    },
    {
      "entropy": 5.67778582572937,
      "epoch": 5.769230769230769,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002237861368018571,
      "loss": 4.836,
      "mean_token_accuracy": 0.22666365206241607,
      "num_tokens": 131995745.0,
      "step": 63600
    },
    {
      "entropy": 5.689883184432984,
      "epoch": 5.769684325108853,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022375463024511773,
      "loss": 4.8312,
      "mean_token_accuracy": 0.23898562490940095,
      "num_tokens": 132006335.0,
      "step": 63605
    },
    {
      "entropy": 5.7314657211303714,
      "epoch": 5.770137880986938,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022372312474814565,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2267378881573677,
      "num_tokens": 132016589.0,
      "step": 63610
    },
    {
      "entropy": 5.765860605239868,
      "epoch": 5.770591436865022,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00022369162031159236,
      "loss": 4.8293,
      "mean_token_accuracy": 0.23277963548898697,
      "num_tokens": 132026430.0,
      "step": 63615
    },
    {
      "entropy": 5.731919527053833,
      "epoch": 5.771044992743106,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002236601169361095,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22326403856277466,
      "num_tokens": 132036790.0,
      "step": 63620
    },
    {
      "entropy": 5.7052216053009035,
      "epoch": 5.7714985486211905,
      "grad_norm": 1.25,
      "learning_rate": 0.0002236286146223485,
      "loss": 4.843,
      "mean_token_accuracy": 0.2333589166402817,
      "num_tokens": 132047379.0,
      "step": 63625
    },
    {
      "entropy": 5.70922966003418,
      "epoch": 5.771952104499274,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022359711337096083,
      "loss": 4.9176,
      "mean_token_accuracy": 0.2281218484044075,
      "num_tokens": 132058628.0,
      "step": 63630
    },
    {
      "entropy": 5.706657981872558,
      "epoch": 5.772405660377358,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022356561318259794,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22784441858530044,
      "num_tokens": 132068133.0,
      "step": 63635
    },
    {
      "entropy": 5.691334962844849,
      "epoch": 5.772859216255442,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002235341140579113,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23380258232355117,
      "num_tokens": 132077813.0,
      "step": 63640
    },
    {
      "entropy": 5.709412240982056,
      "epoch": 5.773312772133527,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022350261599755227,
      "loss": 4.8916,
      "mean_token_accuracy": 0.23015026599168778,
      "num_tokens": 132087823.0,
      "step": 63645
    },
    {
      "entropy": 5.782384586334229,
      "epoch": 5.773766328011611,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002234711190021723,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22848503291606903,
      "num_tokens": 132098393.0,
      "step": 63650
    },
    {
      "entropy": 5.709535360336304,
      "epoch": 5.774219883889695,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002234396230724227,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2270656034350395,
      "num_tokens": 132109146.0,
      "step": 63655
    },
    {
      "entropy": 5.782077836990356,
      "epoch": 5.7746734397677795,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002234081282089549,
      "loss": 4.9662,
      "mean_token_accuracy": 0.21931857764720916,
      "num_tokens": 132119286.0,
      "step": 63660
    },
    {
      "entropy": 5.703988122940063,
      "epoch": 5.775126995645864,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022337663441242007,
      "loss": 4.7585,
      "mean_token_accuracy": 0.23756499737501144,
      "num_tokens": 132129123.0,
      "step": 63665
    },
    {
      "entropy": 5.709490871429443,
      "epoch": 5.775580551523948,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002233451416834696,
      "loss": 4.8766,
      "mean_token_accuracy": 0.22410326153039933,
      "num_tokens": 132139935.0,
      "step": 63670
    },
    {
      "entropy": 5.636438703536987,
      "epoch": 5.776034107402032,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022331365002275477,
      "loss": 4.8024,
      "mean_token_accuracy": 0.23417148888111114,
      "num_tokens": 132150407.0,
      "step": 63675
    },
    {
      "entropy": 5.731329250335693,
      "epoch": 5.7764876632801165,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022328215943092684,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2299938455224037,
      "num_tokens": 132160487.0,
      "step": 63680
    },
    {
      "entropy": 5.824496173858643,
      "epoch": 5.7769412191582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022325066990863714,
      "loss": 5.0333,
      "mean_token_accuracy": 0.21468248069286347,
      "num_tokens": 132170868.0,
      "step": 63685
    },
    {
      "entropy": 5.747001218795776,
      "epoch": 5.777394775036284,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022321918145653662,
      "loss": 4.8459,
      "mean_token_accuracy": 0.23581941723823546,
      "num_tokens": 132181774.0,
      "step": 63690
    },
    {
      "entropy": 5.792680358886718,
      "epoch": 5.7778483309143684,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022318769407527668,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2313005730509758,
      "num_tokens": 132191862.0,
      "step": 63695
    },
    {
      "entropy": 5.810448741912841,
      "epoch": 5.778301886792453,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022315620776550843,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21706912219524382,
      "num_tokens": 132202385.0,
      "step": 63700
    },
    {
      "entropy": 5.666048336029053,
      "epoch": 5.778755442670537,
      "grad_norm": 1.3125,
      "learning_rate": 0.000223124722527883,
      "loss": 4.8045,
      "mean_token_accuracy": 0.23389448672533036,
      "num_tokens": 132213654.0,
      "step": 63705
    },
    {
      "entropy": 5.7846152782440186,
      "epoch": 5.779208998548621,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022309323836305152,
      "loss": 4.9304,
      "mean_token_accuracy": 0.22568233460187911,
      "num_tokens": 132224816.0,
      "step": 63710
    },
    {
      "entropy": 5.71210618019104,
      "epoch": 5.7796625544267055,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002230617552716651,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23488956540822983,
      "num_tokens": 132234347.0,
      "step": 63715
    },
    {
      "entropy": 5.709124422073364,
      "epoch": 5.78011611030479,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022303027325437487,
      "loss": 4.7669,
      "mean_token_accuracy": 0.24214558750391008,
      "num_tokens": 132244413.0,
      "step": 63720
    },
    {
      "entropy": 5.731365966796875,
      "epoch": 5.780569666182874,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022299879231183173,
      "loss": 4.8523,
      "mean_token_accuracy": 0.22870507091283798,
      "num_tokens": 132254093.0,
      "step": 63725
    },
    {
      "entropy": 5.681227493286133,
      "epoch": 5.781023222060957,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022296731244468693,
      "loss": 4.8025,
      "mean_token_accuracy": 0.2321344196796417,
      "num_tokens": 132264449.0,
      "step": 63730
    },
    {
      "entropy": 5.697452878952026,
      "epoch": 5.781476777939043,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022293583365359131,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23591717034578324,
      "num_tokens": 132274784.0,
      "step": 63735
    },
    {
      "entropy": 5.7185760021209715,
      "epoch": 5.781930333817126,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002229043559391959,
      "loss": 4.794,
      "mean_token_accuracy": 0.23195090740919114,
      "num_tokens": 132284289.0,
      "step": 63740
    },
    {
      "entropy": 5.683293008804322,
      "epoch": 5.78238388969521,
      "grad_norm": 1.625,
      "learning_rate": 0.00022287287930215167,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23342871516942978,
      "num_tokens": 132293749.0,
      "step": 63745
    },
    {
      "entropy": 5.653101205825806,
      "epoch": 5.7828374455732945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022284140374310963,
      "loss": 4.7632,
      "mean_token_accuracy": 0.23878976255655288,
      "num_tokens": 132304129.0,
      "step": 63750
    },
    {
      "entropy": 5.687844514846802,
      "epoch": 5.783291001451379,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022280992926272065,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23659754395484925,
      "num_tokens": 132316102.0,
      "step": 63755
    },
    {
      "entropy": 5.6932008266448975,
      "epoch": 5.783744557329463,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022277845586163558,
      "loss": 4.8791,
      "mean_token_accuracy": 0.23118950724601744,
      "num_tokens": 132327421.0,
      "step": 63760
    },
    {
      "entropy": 5.758880376815796,
      "epoch": 5.784198113207547,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022274698354050543,
      "loss": 4.8967,
      "mean_token_accuracy": 0.22567750066518782,
      "num_tokens": 132338285.0,
      "step": 63765
    },
    {
      "entropy": 5.6797239780426025,
      "epoch": 5.7846516690856316,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022271551229998093,
      "loss": 4.8195,
      "mean_token_accuracy": 0.22967805117368698,
      "num_tokens": 132348288.0,
      "step": 63770
    },
    {
      "entropy": 5.725140714645386,
      "epoch": 5.785105224963716,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022268404214071297,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23324329107999803,
      "num_tokens": 132358225.0,
      "step": 63775
    },
    {
      "entropy": 5.7775050640106205,
      "epoch": 5.7855587808418,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022265257306335235,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21804296523332595,
      "num_tokens": 132369000.0,
      "step": 63780
    },
    {
      "entropy": 5.60500054359436,
      "epoch": 5.7860123367198835,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022262110506854988,
      "loss": 4.7757,
      "mean_token_accuracy": 0.23960756361484528,
      "num_tokens": 132379461.0,
      "step": 63785
    },
    {
      "entropy": 5.691548204421997,
      "epoch": 5.786465892597968,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022258963815695627,
      "loss": 4.8373,
      "mean_token_accuracy": 0.23102234452962875,
      "num_tokens": 132389869.0,
      "step": 63790
    },
    {
      "entropy": 5.697595119476318,
      "epoch": 5.786919448476052,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022255817232922227,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23627406358718872,
      "num_tokens": 132399463.0,
      "step": 63795
    },
    {
      "entropy": 5.712697744369507,
      "epoch": 5.787373004354136,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022252670758599868,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22518020421266555,
      "num_tokens": 132410633.0,
      "step": 63800
    },
    {
      "entropy": 5.716022872924805,
      "epoch": 5.7878265602322205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002224952439279362,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23225726187229156,
      "num_tokens": 132421340.0,
      "step": 63805
    },
    {
      "entropy": 5.692303419113159,
      "epoch": 5.788280116110305,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002224637813556853,
      "loss": 4.8434,
      "mean_token_accuracy": 0.2273004174232483,
      "num_tokens": 132430870.0,
      "step": 63810
    },
    {
      "entropy": 5.695868253707886,
      "epoch": 5.788733671988389,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022243231986989693,
      "loss": 4.7343,
      "mean_token_accuracy": 0.2388512060046196,
      "num_tokens": 132441120.0,
      "step": 63815
    },
    {
      "entropy": 5.687720727920532,
      "epoch": 5.789187227866473,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022240085947122153,
      "loss": 4.7433,
      "mean_token_accuracy": 0.24314164966344834,
      "num_tokens": 132451149.0,
      "step": 63820
    },
    {
      "entropy": 5.677066135406494,
      "epoch": 5.789640783744558,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022236940016030977,
      "loss": 4.8344,
      "mean_token_accuracy": 0.23416468352079392,
      "num_tokens": 132461566.0,
      "step": 63825
    },
    {
      "entropy": 5.675561904907227,
      "epoch": 5.790094339622642,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002223379419378122,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23612527400255204,
      "num_tokens": 132471486.0,
      "step": 63830
    },
    {
      "entropy": 5.6846692085266115,
      "epoch": 5.790547895500726,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022230648480437942,
      "loss": 4.8305,
      "mean_token_accuracy": 0.2325798213481903,
      "num_tokens": 132482391.0,
      "step": 63835
    },
    {
      "entropy": 5.683035516738892,
      "epoch": 5.7910014513788095,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022227502876066203,
      "loss": 4.8335,
      "mean_token_accuracy": 0.2310647800564766,
      "num_tokens": 132492235.0,
      "step": 63840
    },
    {
      "entropy": 5.718653678894043,
      "epoch": 5.791455007256894,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022224357380731035,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22446997910737992,
      "num_tokens": 132501984.0,
      "step": 63845
    },
    {
      "entropy": 5.696605777740478,
      "epoch": 5.791908563134978,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002222121199449751,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23300075381994248,
      "num_tokens": 132512608.0,
      "step": 63850
    },
    {
      "entropy": 5.696591854095459,
      "epoch": 5.792362119013062,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022218066717430662,
      "loss": 4.7373,
      "mean_token_accuracy": 0.2406098648905754,
      "num_tokens": 132522682.0,
      "step": 63855
    },
    {
      "entropy": 5.743563604354859,
      "epoch": 5.792815674891147,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002221492154959554,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2334130510687828,
      "num_tokens": 132531680.0,
      "step": 63860
    },
    {
      "entropy": 5.710639190673828,
      "epoch": 5.793269230769231,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002221177649105719,
      "loss": 4.8132,
      "mean_token_accuracy": 0.23221060037612914,
      "num_tokens": 132542149.0,
      "step": 63865
    },
    {
      "entropy": 5.670086860656738,
      "epoch": 5.793722786647315,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022208631541880653,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2276678428053856,
      "num_tokens": 132552249.0,
      "step": 63870
    },
    {
      "entropy": 5.722401332855225,
      "epoch": 5.794176342525399,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002220548670213096,
      "loss": 4.8681,
      "mean_token_accuracy": 0.23591484874486923,
      "num_tokens": 132562471.0,
      "step": 63875
    },
    {
      "entropy": 5.717200040817261,
      "epoch": 5.794629898403484,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022202341971873152,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22975470423698424,
      "num_tokens": 132572357.0,
      "step": 63880
    },
    {
      "entropy": 5.684429883956909,
      "epoch": 5.795083454281567,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022199197351172262,
      "loss": 4.8966,
      "mean_token_accuracy": 0.23000154197216033,
      "num_tokens": 132582020.0,
      "step": 63885
    },
    {
      "entropy": 5.762054061889648,
      "epoch": 5.795537010159651,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022196052840093322,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22321600317955018,
      "num_tokens": 132592018.0,
      "step": 63890
    },
    {
      "entropy": 5.704104328155518,
      "epoch": 5.7959905660377355,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002219290843870136,
      "loss": 4.8259,
      "mean_token_accuracy": 0.23158313035964967,
      "num_tokens": 132602433.0,
      "step": 63895
    },
    {
      "entropy": 5.692022705078125,
      "epoch": 5.79644412191582,
      "grad_norm": 1.765625,
      "learning_rate": 0.00022189764147061402,
      "loss": 4.7683,
      "mean_token_accuracy": 0.2454630345106125,
      "num_tokens": 132612633.0,
      "step": 63900
    },
    {
      "entropy": 5.711274719238281,
      "epoch": 5.796897677793904,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002218661996523848,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2277887463569641,
      "num_tokens": 132622432.0,
      "step": 63905
    },
    {
      "entropy": 5.703212833404541,
      "epoch": 5.797351233671988,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002218347589329761,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22847386747598647,
      "num_tokens": 132632770.0,
      "step": 63910
    },
    {
      "entropy": 5.6290504932403564,
      "epoch": 5.797804789550073,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022180331931303805,
      "loss": 4.77,
      "mean_token_accuracy": 0.23153109103441238,
      "num_tokens": 132642745.0,
      "step": 63915
    },
    {
      "entropy": 5.630648612976074,
      "epoch": 5.798258345428157,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022177188079322102,
      "loss": 4.7986,
      "mean_token_accuracy": 0.2280721202492714,
      "num_tokens": 132653023.0,
      "step": 63920
    },
    {
      "entropy": 5.744391393661499,
      "epoch": 5.798711901306241,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022174044337417505,
      "loss": 4.9428,
      "mean_token_accuracy": 0.22624078392982483,
      "num_tokens": 132662143.0,
      "step": 63925
    },
    {
      "entropy": 5.7665934562683105,
      "epoch": 5.799165457184325,
      "grad_norm": 1.375,
      "learning_rate": 0.00022170900705655017,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22667589038610458,
      "num_tokens": 132671541.0,
      "step": 63930
    },
    {
      "entropy": 5.675007009506226,
      "epoch": 5.79961901306241,
      "grad_norm": 1.375,
      "learning_rate": 0.00022167757184099669,
      "loss": 4.8127,
      "mean_token_accuracy": 0.23186843991279601,
      "num_tokens": 132681660.0,
      "step": 63935
    },
    {
      "entropy": 5.693827676773071,
      "epoch": 5.800072568940493,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002216461377281646,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2376233533024788,
      "num_tokens": 132692625.0,
      "step": 63940
    },
    {
      "entropy": 5.655860424041748,
      "epoch": 5.800526124818577,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022161470471870404,
      "loss": 4.7798,
      "mean_token_accuracy": 0.23996298760175705,
      "num_tokens": 132703082.0,
      "step": 63945
    },
    {
      "entropy": 5.7192689895629885,
      "epoch": 5.800979680696662,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022158327281326484,
      "loss": 4.848,
      "mean_token_accuracy": 0.22902771830558777,
      "num_tokens": 132713434.0,
      "step": 63950
    },
    {
      "entropy": 5.706898927688599,
      "epoch": 5.801433236574746,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022155184201249728,
      "loss": 4.8395,
      "mean_token_accuracy": 0.22824840545654296,
      "num_tokens": 132724266.0,
      "step": 63955
    },
    {
      "entropy": 5.72227988243103,
      "epoch": 5.80188679245283,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022152041231705128,
      "loss": 4.839,
      "mean_token_accuracy": 0.23621970117092134,
      "num_tokens": 132734417.0,
      "step": 63960
    },
    {
      "entropy": 5.692641019821167,
      "epoch": 5.802340348330914,
      "grad_norm": 1.375,
      "learning_rate": 0.0002214889837275767,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23266859203577042,
      "num_tokens": 132745596.0,
      "step": 63965
    },
    {
      "entropy": 5.6876002788543705,
      "epoch": 5.802793904208999,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022145755624472364,
      "loss": 4.8054,
      "mean_token_accuracy": 0.23073718696832657,
      "num_tokens": 132754852.0,
      "step": 63970
    },
    {
      "entropy": 5.711327648162841,
      "epoch": 5.803247460087083,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002214261298691419,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22981164753437042,
      "num_tokens": 132767341.0,
      "step": 63975
    },
    {
      "entropy": 5.613332796096802,
      "epoch": 5.803701015965167,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002213947046014815,
      "loss": 4.6963,
      "mean_token_accuracy": 0.2497832342982292,
      "num_tokens": 132777372.0,
      "step": 63980
    },
    {
      "entropy": 5.678045177459717,
      "epoch": 5.8041545718432515,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022136328044239218,
      "loss": 4.7676,
      "mean_token_accuracy": 0.23420565724372863,
      "num_tokens": 132786845.0,
      "step": 63985
    },
    {
      "entropy": 5.7365867614746096,
      "epoch": 5.804608127721336,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000221331857392524,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22795893400907516,
      "num_tokens": 132796699.0,
      "step": 63990
    },
    {
      "entropy": 5.700076055526734,
      "epoch": 5.805061683599419,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022130043545252664,
      "loss": 4.82,
      "mean_token_accuracy": 0.23929930627346038,
      "num_tokens": 132806448.0,
      "step": 63995
    },
    {
      "entropy": 5.708661460876465,
      "epoch": 5.805515239477503,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022126901462304993,
      "loss": 4.802,
      "mean_token_accuracy": 0.23162735551595687,
      "num_tokens": 132815570.0,
      "step": 64000
    },
    {
      "entropy": 5.594670295715332,
      "epoch": 5.805968795355588,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002212375949047437,
      "loss": 4.7565,
      "mean_token_accuracy": 0.23595983237028123,
      "num_tokens": 132825524.0,
      "step": 64005
    },
    {
      "entropy": 5.683086967468261,
      "epoch": 5.806422351233672,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022120617629825773,
      "loss": 4.8335,
      "mean_token_accuracy": 0.22977345883846284,
      "num_tokens": 132836312.0,
      "step": 64010
    },
    {
      "entropy": 5.612058305740357,
      "epoch": 5.806875907111756,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002211747588042417,
      "loss": 4.7202,
      "mean_token_accuracy": 0.23707927763462067,
      "num_tokens": 132845897.0,
      "step": 64015
    },
    {
      "entropy": 5.68422532081604,
      "epoch": 5.80732946298984,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022114334242334544,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2379963591694832,
      "num_tokens": 132855882.0,
      "step": 64020
    },
    {
      "entropy": 5.678504228591919,
      "epoch": 5.807783018867925,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022111192715621857,
      "loss": 4.8157,
      "mean_token_accuracy": 0.23374857306480407,
      "num_tokens": 132865989.0,
      "step": 64025
    },
    {
      "entropy": 5.717402696609497,
      "epoch": 5.808236574746009,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022108051300351078,
      "loss": 4.8723,
      "mean_token_accuracy": 0.23436804115772247,
      "num_tokens": 132876314.0,
      "step": 64030
    },
    {
      "entropy": 5.708573055267334,
      "epoch": 5.808690130624093,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002210490999658717,
      "loss": 4.7885,
      "mean_token_accuracy": 0.23921446800231932,
      "num_tokens": 132885498.0,
      "step": 64035
    },
    {
      "entropy": 5.646019744873047,
      "epoch": 5.809143686502177,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022101768804395094,
      "loss": 4.7653,
      "mean_token_accuracy": 0.23076444566249849,
      "num_tokens": 132895389.0,
      "step": 64040
    },
    {
      "entropy": 5.682502412796021,
      "epoch": 5.809597242380261,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022098627723839826,
      "loss": 4.813,
      "mean_token_accuracy": 0.23453418761491776,
      "num_tokens": 132905332.0,
      "step": 64045
    },
    {
      "entropy": 5.675707530975342,
      "epoch": 5.810050798258345,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000220954867549863,
      "loss": 4.865,
      "mean_token_accuracy": 0.21896573156118393,
      "num_tokens": 132915402.0,
      "step": 64050
    },
    {
      "entropy": 5.699703788757324,
      "epoch": 5.810504354136429,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002209234589789949,
      "loss": 4.8713,
      "mean_token_accuracy": 0.22996346205472945,
      "num_tokens": 132925117.0,
      "step": 64055
    },
    {
      "entropy": 5.7215064525604244,
      "epoch": 5.810957910014514,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022089205152644347,
      "loss": 4.8406,
      "mean_token_accuracy": 0.22522860914468765,
      "num_tokens": 132935163.0,
      "step": 64060
    },
    {
      "entropy": 5.771512174606324,
      "epoch": 5.811411465892598,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002208606451928582,
      "loss": 4.9223,
      "mean_token_accuracy": 0.22888771891593934,
      "num_tokens": 132944971.0,
      "step": 64065
    },
    {
      "entropy": 5.77207989692688,
      "epoch": 5.811865021770682,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002208292399788886,
      "loss": 4.9012,
      "mean_token_accuracy": 0.2227646604180336,
      "num_tokens": 132955199.0,
      "step": 64070
    },
    {
      "entropy": 5.767969179153442,
      "epoch": 5.8123185776487665,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002207978358851841,
      "loss": 4.9128,
      "mean_token_accuracy": 0.22005248814821243,
      "num_tokens": 132966144.0,
      "step": 64075
    },
    {
      "entropy": 5.804065942764282,
      "epoch": 5.812772133526851,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022076643291239417,
      "loss": 4.9041,
      "mean_token_accuracy": 0.2258058950304985,
      "num_tokens": 132976866.0,
      "step": 64080
    },
    {
      "entropy": 5.730074119567871,
      "epoch": 5.813225689404935,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002207350310611682,
      "loss": 4.891,
      "mean_token_accuracy": 0.22520802766084672,
      "num_tokens": 132988178.0,
      "step": 64085
    },
    {
      "entropy": 5.713765954971313,
      "epoch": 5.813679245283019,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022070363033215563,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2276656448841095,
      "num_tokens": 132998631.0,
      "step": 64090
    },
    {
      "entropy": 5.70262451171875,
      "epoch": 5.814132801161103,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022067223072600584,
      "loss": 4.8547,
      "mean_token_accuracy": 0.2267395168542862,
      "num_tokens": 133008864.0,
      "step": 64095
    },
    {
      "entropy": 5.661644458770752,
      "epoch": 5.814586357039187,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002206408322433681,
      "loss": 4.8467,
      "mean_token_accuracy": 0.22563537806272507,
      "num_tokens": 133019771.0,
      "step": 64100
    },
    {
      "entropy": 5.760131025314331,
      "epoch": 5.815039912917271,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022060943488489188,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22612787187099456,
      "num_tokens": 133030123.0,
      "step": 64105
    },
    {
      "entropy": 5.686515283584595,
      "epoch": 5.8154934687953554,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022057803865122645,
      "loss": 4.798,
      "mean_token_accuracy": 0.2307887777686119,
      "num_tokens": 133040104.0,
      "step": 64110
    },
    {
      "entropy": 5.6988130569458,
      "epoch": 5.81594702467344,
      "grad_norm": 1.5,
      "learning_rate": 0.000220546643543021,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23425949960947037,
      "num_tokens": 133049594.0,
      "step": 64115
    },
    {
      "entropy": 5.632229423522949,
      "epoch": 5.816400580551524,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022051524956092483,
      "loss": 4.7438,
      "mean_token_accuracy": 0.24535926729440688,
      "num_tokens": 133060233.0,
      "step": 64120
    },
    {
      "entropy": 5.733109951019287,
      "epoch": 5.816854136429608,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002204838567055872,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22233099192380906,
      "num_tokens": 133070486.0,
      "step": 64125
    },
    {
      "entropy": 5.730430841445923,
      "epoch": 5.8173076923076925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022045246497765726,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2285383403301239,
      "num_tokens": 133080432.0,
      "step": 64130
    },
    {
      "entropy": 5.709640455245972,
      "epoch": 5.817761248185777,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022042107437778426,
      "loss": 4.8363,
      "mean_token_accuracy": 0.23434161990880967,
      "num_tokens": 133090349.0,
      "step": 64135
    },
    {
      "entropy": 5.699872303009033,
      "epoch": 5.818214804063861,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022038968490661738,
      "loss": 4.7669,
      "mean_token_accuracy": 0.2407387986779213,
      "num_tokens": 133101055.0,
      "step": 64140
    },
    {
      "entropy": 5.706423902511597,
      "epoch": 5.818668359941945,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022035829656480565,
      "loss": 4.8248,
      "mean_token_accuracy": 0.22402965277433395,
      "num_tokens": 133111185.0,
      "step": 64145
    },
    {
      "entropy": 5.706963205337525,
      "epoch": 5.819121915820029,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022032690935299838,
      "loss": 4.8423,
      "mean_token_accuracy": 0.22511114031076432,
      "num_tokens": 133121567.0,
      "step": 64150
    },
    {
      "entropy": 5.662488317489624,
      "epoch": 5.819575471698113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002202955232718445,
      "loss": 4.7419,
      "mean_token_accuracy": 0.2475334420800209,
      "num_tokens": 133132306.0,
      "step": 64155
    },
    {
      "entropy": 5.681830739974975,
      "epoch": 5.820029027576197,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002202641383219931,
      "loss": 4.8214,
      "mean_token_accuracy": 0.2358947604894638,
      "num_tokens": 133143418.0,
      "step": 64160
    },
    {
      "entropy": 5.7157539367675785,
      "epoch": 5.8204825834542815,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022023275450409335,
      "loss": 4.8766,
      "mean_token_accuracy": 0.2255534753203392,
      "num_tokens": 133154420.0,
      "step": 64165
    },
    {
      "entropy": 5.693794441223145,
      "epoch": 5.820936139332366,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022020137181879413,
      "loss": 4.7972,
      "mean_token_accuracy": 0.23326994925737382,
      "num_tokens": 133163557.0,
      "step": 64170
    },
    {
      "entropy": 5.720128917694092,
      "epoch": 5.82138969521045,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022016999026674457,
      "loss": 4.758,
      "mean_token_accuracy": 0.23495329320430755,
      "num_tokens": 133173902.0,
      "step": 64175
    },
    {
      "entropy": 5.701705503463745,
      "epoch": 5.821843251088534,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022013860984859346,
      "loss": 4.7639,
      "mean_token_accuracy": 0.243974669277668,
      "num_tokens": 133183488.0,
      "step": 64180
    },
    {
      "entropy": 5.731578683853149,
      "epoch": 5.8222968069666186,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022010723056499,
      "loss": 4.8398,
      "mean_token_accuracy": 0.23226927369832992,
      "num_tokens": 133194014.0,
      "step": 64185
    },
    {
      "entropy": 5.774499034881591,
      "epoch": 5.822750362844703,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000220075852416583,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2285216197371483,
      "num_tokens": 133204019.0,
      "step": 64190
    },
    {
      "entropy": 5.675313329696655,
      "epoch": 5.823203918722786,
      "grad_norm": 1.25,
      "learning_rate": 0.0002200444754040213,
      "loss": 4.8549,
      "mean_token_accuracy": 0.23879691809415818,
      "num_tokens": 133214290.0,
      "step": 64195
    },
    {
      "entropy": 5.669415712356567,
      "epoch": 5.8236574746008705,
      "grad_norm": 1.375,
      "learning_rate": 0.0002200130995279539,
      "loss": 4.8057,
      "mean_token_accuracy": 0.23468693792819978,
      "num_tokens": 133225102.0,
      "step": 64200
    },
    {
      "entropy": 5.680475807189941,
      "epoch": 5.824111030478955,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002199817247890296,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23575611412525177,
      "num_tokens": 133235683.0,
      "step": 64205
    },
    {
      "entropy": 5.704711151123047,
      "epoch": 5.824564586357039,
      "grad_norm": 1.25,
      "learning_rate": 0.00021995035118789724,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2251909226179123,
      "num_tokens": 133247664.0,
      "step": 64210
    },
    {
      "entropy": 5.668287944793701,
      "epoch": 5.825018142235123,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021991897872520566,
      "loss": 4.7828,
      "mean_token_accuracy": 0.24355259984731675,
      "num_tokens": 133257686.0,
      "step": 64215
    },
    {
      "entropy": 5.725055265426636,
      "epoch": 5.8254716981132075,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021988760740160364,
      "loss": 4.825,
      "mean_token_accuracy": 0.23194470256567,
      "num_tokens": 133267295.0,
      "step": 64220
    },
    {
      "entropy": 5.664690113067627,
      "epoch": 5.825925253991292,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021985623721773996,
      "loss": 4.7071,
      "mean_token_accuracy": 0.23817880749702453,
      "num_tokens": 133277540.0,
      "step": 64225
    },
    {
      "entropy": 5.662933254241944,
      "epoch": 5.826378809869376,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021982486817426328,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23242394477128983,
      "num_tokens": 133288160.0,
      "step": 64230
    },
    {
      "entropy": 5.686497402191162,
      "epoch": 5.82683236574746,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021979350027182242,
      "loss": 4.8078,
      "mean_token_accuracy": 0.24215407222509383,
      "num_tokens": 133298648.0,
      "step": 64235
    },
    {
      "entropy": 5.7292064189910885,
      "epoch": 5.827285921625545,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021976213351106605,
      "loss": 4.8418,
      "mean_token_accuracy": 0.2358512043952942,
      "num_tokens": 133309320.0,
      "step": 64240
    },
    {
      "entropy": 5.740872240066528,
      "epoch": 5.827739477503629,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021973076789264285,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2252182424068451,
      "num_tokens": 133318948.0,
      "step": 64245
    },
    {
      "entropy": 5.6959539413452145,
      "epoch": 5.828193033381712,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002196994034172014,
      "loss": 4.8112,
      "mean_token_accuracy": 0.22920002937316894,
      "num_tokens": 133328573.0,
      "step": 64250
    },
    {
      "entropy": 5.658938789367676,
      "epoch": 5.8286465892597965,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021966804008539042,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23243148773908615,
      "num_tokens": 133338503.0,
      "step": 64255
    },
    {
      "entropy": 5.703886461257935,
      "epoch": 5.829100145137881,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021963667789785853,
      "loss": 4.8402,
      "mean_token_accuracy": 0.2318193644285202,
      "num_tokens": 133349733.0,
      "step": 64260
    },
    {
      "entropy": 5.710337352752686,
      "epoch": 5.829553701015965,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021960531685525413,
      "loss": 4.8468,
      "mean_token_accuracy": 0.23084611296653748,
      "num_tokens": 133360639.0,
      "step": 64265
    },
    {
      "entropy": 5.745991802215576,
      "epoch": 5.830007256894049,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021957395695822597,
      "loss": 4.8554,
      "mean_token_accuracy": 0.23134303092956543,
      "num_tokens": 133370678.0,
      "step": 64270
    },
    {
      "entropy": 5.760033130645752,
      "epoch": 5.830460812772134,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021954259820742251,
      "loss": 4.883,
      "mean_token_accuracy": 0.2351992756128311,
      "num_tokens": 133382140.0,
      "step": 64275
    },
    {
      "entropy": 5.69479398727417,
      "epoch": 5.830914368650218,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021951124060349221,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23260274827480315,
      "num_tokens": 133392152.0,
      "step": 64280
    },
    {
      "entropy": 5.760529899597168,
      "epoch": 5.831367924528302,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021947988414708358,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22147225886583327,
      "num_tokens": 133402543.0,
      "step": 64285
    },
    {
      "entropy": 5.775099420547486,
      "epoch": 5.831821480406386,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002194485288388452,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22739844173192977,
      "num_tokens": 133413418.0,
      "step": 64290
    },
    {
      "entropy": 5.693542003631592,
      "epoch": 5.83227503628447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002194171746794253,
      "loss": 4.868,
      "mean_token_accuracy": 0.22950057834386825,
      "num_tokens": 133424342.0,
      "step": 64295
    },
    {
      "entropy": 5.799184370040893,
      "epoch": 5.832728592162555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002193858216694724,
      "loss": 5.002,
      "mean_token_accuracy": 0.21661666184663772,
      "num_tokens": 133435403.0,
      "step": 64300
    },
    {
      "entropy": 5.740971040725708,
      "epoch": 5.833182148040638,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002193544698096349,
      "loss": 4.9536,
      "mean_token_accuracy": 0.22935347110033036,
      "num_tokens": 133446267.0,
      "step": 64305
    },
    {
      "entropy": 5.707323741912842,
      "epoch": 5.8336357039187225,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021932311910056117,
      "loss": 4.8553,
      "mean_token_accuracy": 0.23070345520973207,
      "num_tokens": 133457156.0,
      "step": 64310
    },
    {
      "entropy": 5.691985988616944,
      "epoch": 5.834089259796807,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021929176954289947,
      "loss": 4.8223,
      "mean_token_accuracy": 0.23197563588619233,
      "num_tokens": 133467583.0,
      "step": 64315
    },
    {
      "entropy": 5.75888295173645,
      "epoch": 5.834542815674891,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021926042113729822,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2173672005534172,
      "num_tokens": 133478019.0,
      "step": 64320
    },
    {
      "entropy": 5.659050559997558,
      "epoch": 5.834996371552975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021922907388440566,
      "loss": 4.7755,
      "mean_token_accuracy": 0.24159383326768874,
      "num_tokens": 133487521.0,
      "step": 64325
    },
    {
      "entropy": 5.71269006729126,
      "epoch": 5.83544992743106,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021919772778487007,
      "loss": 4.8771,
      "mean_token_accuracy": 0.2255738392472267,
      "num_tokens": 133498453.0,
      "step": 64330
    },
    {
      "entropy": 5.7249494075775145,
      "epoch": 5.835903483309144,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002191663828393396,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22894511967897416,
      "num_tokens": 133509234.0,
      "step": 64335
    },
    {
      "entropy": 5.713944005966186,
      "epoch": 5.836357039187228,
      "grad_norm": 1.375,
      "learning_rate": 0.00021913503904846265,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22784450203180312,
      "num_tokens": 133519983.0,
      "step": 64340
    },
    {
      "entropy": 5.726438379287719,
      "epoch": 5.836810595065312,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021910369641288735,
      "loss": 4.7956,
      "mean_token_accuracy": 0.22858690321445466,
      "num_tokens": 133530294.0,
      "step": 64345
    },
    {
      "entropy": 5.704834413528443,
      "epoch": 5.837264150943396,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021907235493326177,
      "loss": 4.8451,
      "mean_token_accuracy": 0.23907008469104768,
      "num_tokens": 133539979.0,
      "step": 64350
    },
    {
      "entropy": 5.683802270889283,
      "epoch": 5.83771770682148,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021904101461023417,
      "loss": 4.8598,
      "mean_token_accuracy": 0.23512554466724395,
      "num_tokens": 133550729.0,
      "step": 64355
    },
    {
      "entropy": 5.6977952003479,
      "epoch": 5.838171262699564,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021900967544445272,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23637585937976838,
      "num_tokens": 133561143.0,
      "step": 64360
    },
    {
      "entropy": 5.734751796722412,
      "epoch": 5.838624818577649,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021897833743656536,
      "loss": 4.8704,
      "mean_token_accuracy": 0.23615441620349883,
      "num_tokens": 133572039.0,
      "step": 64365
    },
    {
      "entropy": 5.6879559516906735,
      "epoch": 5.839078374455733,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021894700058722028,
      "loss": 4.7954,
      "mean_token_accuracy": 0.23487950563430787,
      "num_tokens": 133581206.0,
      "step": 64370
    },
    {
      "entropy": 5.666264438629151,
      "epoch": 5.839531930333817,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021891566489706555,
      "loss": 4.8402,
      "mean_token_accuracy": 0.2296026259660721,
      "num_tokens": 133591724.0,
      "step": 64375
    },
    {
      "entropy": 5.725771951675415,
      "epoch": 5.839985486211901,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021888433036674915,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2237562969326973,
      "num_tokens": 133602282.0,
      "step": 64380
    },
    {
      "entropy": 5.676604843139648,
      "epoch": 5.840439042089986,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000218852996996919,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23795992136001587,
      "num_tokens": 133612389.0,
      "step": 64385
    },
    {
      "entropy": 5.7072591304779055,
      "epoch": 5.84089259796807,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002188216647882233,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22544897496700286,
      "num_tokens": 133622059.0,
      "step": 64390
    },
    {
      "entropy": 5.73408613204956,
      "epoch": 5.841346153846154,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021879033374130985,
      "loss": 4.8487,
      "mean_token_accuracy": 0.22732639759778978,
      "num_tokens": 133632423.0,
      "step": 64395
    },
    {
      "entropy": 5.71442437171936,
      "epoch": 5.8417997097242385,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021875900385682663,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23668632954359053,
      "num_tokens": 133643197.0,
      "step": 64400
    },
    {
      "entropy": 5.705300045013428,
      "epoch": 5.842253265602322,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021872767513542152,
      "loss": 4.8523,
      "mean_token_accuracy": 0.23048931658267974,
      "num_tokens": 133653313.0,
      "step": 64405
    },
    {
      "entropy": 5.750592947006226,
      "epoch": 5.842706821480406,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021869634757774243,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22703205794095993,
      "num_tokens": 133663082.0,
      "step": 64410
    },
    {
      "entropy": 5.6714620113372805,
      "epoch": 5.84316037735849,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021866502118443726,
      "loss": 4.7764,
      "mean_token_accuracy": 0.23143004775047302,
      "num_tokens": 133673213.0,
      "step": 64415
    },
    {
      "entropy": 5.67293381690979,
      "epoch": 5.843613933236575,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021863369595615374,
      "loss": 4.7434,
      "mean_token_accuracy": 0.23926090747117995,
      "num_tokens": 133682779.0,
      "step": 64420
    },
    {
      "entropy": 5.669696950912476,
      "epoch": 5.844067489114659,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002186023718935398,
      "loss": 4.8157,
      "mean_token_accuracy": 0.22643478810787201,
      "num_tokens": 133693404.0,
      "step": 64425
    },
    {
      "entropy": 5.650432586669922,
      "epoch": 5.844521044992743,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021857104899724323,
      "loss": 4.7815,
      "mean_token_accuracy": 0.23652153164148332,
      "num_tokens": 133703621.0,
      "step": 64430
    },
    {
      "entropy": 5.755836915969849,
      "epoch": 5.844974600870827,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002185397272679116,
      "loss": 4.9375,
      "mean_token_accuracy": 0.22712422758340836,
      "num_tokens": 133714944.0,
      "step": 64435
    },
    {
      "entropy": 5.7252672672271725,
      "epoch": 5.845428156748912,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021850840670619294,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22437018901109695,
      "num_tokens": 133725434.0,
      "step": 64440
    },
    {
      "entropy": 5.588098955154419,
      "epoch": 5.845881712626996,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021847708731273475,
      "loss": 4.6746,
      "mean_token_accuracy": 0.24744361788034439,
      "num_tokens": 133736629.0,
      "step": 64445
    },
    {
      "entropy": 5.734464931488037,
      "epoch": 5.846335268505079,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021844576908818487,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23663600236177446,
      "num_tokens": 133746967.0,
      "step": 64450
    },
    {
      "entropy": 5.7360738754272464,
      "epoch": 5.8467888243831645,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021841445203319083,
      "loss": 4.82,
      "mean_token_accuracy": 0.2308531552553177,
      "num_tokens": 133757723.0,
      "step": 64455
    },
    {
      "entropy": 5.6967182636260985,
      "epoch": 5.847242380261248,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002183831361484004,
      "loss": 4.826,
      "mean_token_accuracy": 0.2240427926182747,
      "num_tokens": 133768718.0,
      "step": 64460
    },
    {
      "entropy": 5.710841655731201,
      "epoch": 5.847695936139332,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021835182143446102,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2259274512529373,
      "num_tokens": 133779206.0,
      "step": 64465
    },
    {
      "entropy": 5.600605058670044,
      "epoch": 5.848149492017416,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021832050789202047,
      "loss": 4.7521,
      "mean_token_accuracy": 0.2432217538356781,
      "num_tokens": 133789668.0,
      "step": 64470
    },
    {
      "entropy": 5.687100601196289,
      "epoch": 5.848603047895501,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021828919552172632,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23292637914419173,
      "num_tokens": 133800525.0,
      "step": 64475
    },
    {
      "entropy": 5.713022136688233,
      "epoch": 5.849056603773585,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021825788432422606,
      "loss": 4.7698,
      "mean_token_accuracy": 0.23814596831798554,
      "num_tokens": 133811887.0,
      "step": 64480
    },
    {
      "entropy": 5.755997133255005,
      "epoch": 5.849510159651669,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021822657430016712,
      "loss": 4.972,
      "mean_token_accuracy": 0.22518094033002853,
      "num_tokens": 133822278.0,
      "step": 64485
    },
    {
      "entropy": 5.66053524017334,
      "epoch": 5.8499637155297535,
      "grad_norm": 1.671875,
      "learning_rate": 0.00021819526545019714,
      "loss": 4.8168,
      "mean_token_accuracy": 0.2345369800925255,
      "num_tokens": 133831705.0,
      "step": 64490
    },
    {
      "entropy": 5.735858917236328,
      "epoch": 5.850417271407838,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021816395777496357,
      "loss": 4.8256,
      "mean_token_accuracy": 0.23551902025938035,
      "num_tokens": 133842673.0,
      "step": 64495
    },
    {
      "entropy": 5.697670984268188,
      "epoch": 5.850870827285922,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021813265127511384,
      "loss": 4.8262,
      "mean_token_accuracy": 0.22885168045759202,
      "num_tokens": 133852792.0,
      "step": 64500
    },
    {
      "entropy": 5.743348217010498,
      "epoch": 5.851324383164005,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021810134595129528,
      "loss": 4.807,
      "mean_token_accuracy": 0.23063733130693437,
      "num_tokens": 133862733.0,
      "step": 64505
    },
    {
      "entropy": 5.72846417427063,
      "epoch": 5.85177793904209,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021807004180415552,
      "loss": 4.9497,
      "mean_token_accuracy": 0.22012560367584227,
      "num_tokens": 133873757.0,
      "step": 64510
    },
    {
      "entropy": 5.742285537719726,
      "epoch": 5.852231494920174,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021803873883434173,
      "loss": 4.8232,
      "mean_token_accuracy": 0.22749418914318084,
      "num_tokens": 133883829.0,
      "step": 64515
    },
    {
      "entropy": 5.737408685684204,
      "epoch": 5.852685050798258,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021800743704250124,
      "loss": 4.9572,
      "mean_token_accuracy": 0.22443864941596986,
      "num_tokens": 133894323.0,
      "step": 64520
    },
    {
      "entropy": 5.655583333969116,
      "epoch": 5.8531386066763424,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021797613642928165,
      "loss": 4.7742,
      "mean_token_accuracy": 0.23735732585191727,
      "num_tokens": 133904990.0,
      "step": 64525
    },
    {
      "entropy": 5.749254465103149,
      "epoch": 5.853592162554427,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021794483699533003,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22938340455293654,
      "num_tokens": 133914668.0,
      "step": 64530
    },
    {
      "entropy": 5.671967601776123,
      "epoch": 5.854045718432511,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021791353874129376,
      "loss": 4.7956,
      "mean_token_accuracy": 0.23278047889471054,
      "num_tokens": 133924611.0,
      "step": 64535
    },
    {
      "entropy": 5.623067903518677,
      "epoch": 5.854499274310595,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021788224166782,
      "loss": 4.7491,
      "mean_token_accuracy": 0.2354249119758606,
      "num_tokens": 133935707.0,
      "step": 64540
    },
    {
      "entropy": 5.711665153503418,
      "epoch": 5.8549528301886795,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021785094577555603,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23614072054624557,
      "num_tokens": 133945744.0,
      "step": 64545
    },
    {
      "entropy": 5.802614498138428,
      "epoch": 5.855406386066764,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021781965106514907,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2277312234044075,
      "num_tokens": 133956073.0,
      "step": 64550
    },
    {
      "entropy": 5.6751195907592775,
      "epoch": 5.855859941944848,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002177883575372463,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2341318443417549,
      "num_tokens": 133966009.0,
      "step": 64555
    },
    {
      "entropy": 5.765043926239014,
      "epoch": 5.856313497822931,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021775706519249488,
      "loss": 4.9522,
      "mean_token_accuracy": 0.22647254168987274,
      "num_tokens": 133975759.0,
      "step": 64560
    },
    {
      "entropy": 5.72235631942749,
      "epoch": 5.856767053701016,
      "grad_norm": 1.65625,
      "learning_rate": 0.00021772577403154193,
      "loss": 4.9281,
      "mean_token_accuracy": 0.22166973501443862,
      "num_tokens": 133986537.0,
      "step": 64565
    },
    {
      "entropy": 5.63547420501709,
      "epoch": 5.8572206095791,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002176944840550346,
      "loss": 4.78,
      "mean_token_accuracy": 0.2368450105190277,
      "num_tokens": 133997069.0,
      "step": 64570
    },
    {
      "entropy": 5.79290542602539,
      "epoch": 5.857674165457184,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021766319526361988,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2194149449467659,
      "num_tokens": 134008266.0,
      "step": 64575
    },
    {
      "entropy": 5.7092184066772464,
      "epoch": 5.8581277213352685,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021763190765794493,
      "loss": 4.7435,
      "mean_token_accuracy": 0.24050482362508774,
      "num_tokens": 134017848.0,
      "step": 64580
    },
    {
      "entropy": 5.7046526908874515,
      "epoch": 5.858581277213353,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021760062123865682,
      "loss": 4.7827,
      "mean_token_accuracy": 0.2310303971171379,
      "num_tokens": 134027720.0,
      "step": 64585
    },
    {
      "entropy": 5.755787754058838,
      "epoch": 5.859034833091437,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002175693360064024,
      "loss": 4.9317,
      "mean_token_accuracy": 0.22554749697446824,
      "num_tokens": 134037473.0,
      "step": 64590
    },
    {
      "entropy": 5.683589029312134,
      "epoch": 5.859488388969521,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021753805196182875,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2250048980116844,
      "num_tokens": 134048373.0,
      "step": 64595
    },
    {
      "entropy": 5.736261081695557,
      "epoch": 5.8599419448476056,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002175067691055828,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22666025459766387,
      "num_tokens": 134058631.0,
      "step": 64600
    },
    {
      "entropy": 5.653164196014404,
      "epoch": 5.860395500725689,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021747548743831162,
      "loss": 4.7682,
      "mean_token_accuracy": 0.24018004089593886,
      "num_tokens": 134069826.0,
      "step": 64605
    },
    {
      "entropy": 5.733295440673828,
      "epoch": 5.860849056603773,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021744420696066197,
      "loss": 4.879,
      "mean_token_accuracy": 0.2292304366827011,
      "num_tokens": 134080248.0,
      "step": 64610
    },
    {
      "entropy": 5.74349422454834,
      "epoch": 5.8613026124818575,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021741292767328075,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22405824065208435,
      "num_tokens": 134091549.0,
      "step": 64615
    },
    {
      "entropy": 5.747530174255371,
      "epoch": 5.861756168359942,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002173816495768149,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2204025939106941,
      "num_tokens": 134101892.0,
      "step": 64620
    },
    {
      "entropy": 5.753232479095459,
      "epoch": 5.862209724238026,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002173503726719112,
      "loss": 4.8855,
      "mean_token_accuracy": 0.23269327878952026,
      "num_tokens": 134112073.0,
      "step": 64625
    },
    {
      "entropy": 5.661357450485229,
      "epoch": 5.86266328011611,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021731909695921653,
      "loss": 4.768,
      "mean_token_accuracy": 0.2353372424840927,
      "num_tokens": 134123015.0,
      "step": 64630
    },
    {
      "entropy": 5.679569292068481,
      "epoch": 5.8631168359941945,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021728782243937751,
      "loss": 4.8341,
      "mean_token_accuracy": 0.23501363545656204,
      "num_tokens": 134134742.0,
      "step": 64635
    },
    {
      "entropy": 5.711278009414673,
      "epoch": 5.863570391872279,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021725654911304115,
      "loss": 4.7525,
      "mean_token_accuracy": 0.23217216730117798,
      "num_tokens": 134145477.0,
      "step": 64640
    },
    {
      "entropy": 5.6674741268157955,
      "epoch": 5.864023947750363,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002172252769808541,
      "loss": 4.8258,
      "mean_token_accuracy": 0.2343647763133049,
      "num_tokens": 134155987.0,
      "step": 64645
    },
    {
      "entropy": 5.750072431564331,
      "epoch": 5.864477503628447,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002171940060434629,
      "loss": 4.8935,
      "mean_token_accuracy": 0.228919617831707,
      "num_tokens": 134165044.0,
      "step": 64650
    },
    {
      "entropy": 5.72748908996582,
      "epoch": 5.864931059506532,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021716273630151446,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22114880084991456,
      "num_tokens": 134174786.0,
      "step": 64655
    },
    {
      "entropy": 5.7193090438842775,
      "epoch": 5.865384615384615,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002171314677556554,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23266323953866958,
      "num_tokens": 134184769.0,
      "step": 64660
    },
    {
      "entropy": 5.727167749404908,
      "epoch": 5.865838171262699,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021710020040653234,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23142008036375045,
      "num_tokens": 134194288.0,
      "step": 64665
    },
    {
      "entropy": 5.786126661300659,
      "epoch": 5.8662917271407835,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002170689342547918,
      "loss": 4.9943,
      "mean_token_accuracy": 0.21952083706855774,
      "num_tokens": 134205150.0,
      "step": 64670
    },
    {
      "entropy": 5.64331464767456,
      "epoch": 5.866745283018868,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021703766930108054,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23402672708034516,
      "num_tokens": 134216221.0,
      "step": 64675
    },
    {
      "entropy": 5.743455982208252,
      "epoch": 5.867198838896952,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021700640554604507,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22943958044052123,
      "num_tokens": 134226396.0,
      "step": 64680
    },
    {
      "entropy": 5.715174388885498,
      "epoch": 5.867652394775036,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021697514299033176,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22979769110679626,
      "num_tokens": 134237564.0,
      "step": 64685
    },
    {
      "entropy": 5.714475631713867,
      "epoch": 5.868105950653121,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021694388163458745,
      "loss": 4.802,
      "mean_token_accuracy": 0.2330876260995865,
      "num_tokens": 134247840.0,
      "step": 64690
    },
    {
      "entropy": 5.766328191757202,
      "epoch": 5.868559506531205,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002169126214794584,
      "loss": 4.8752,
      "mean_token_accuracy": 0.23280754387378694,
      "num_tokens": 134258366.0,
      "step": 64695
    },
    {
      "entropy": 5.702075386047364,
      "epoch": 5.869013062409289,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021688136252559116,
      "loss": 4.7741,
      "mean_token_accuracy": 0.2392452374100685,
      "num_tokens": 134268551.0,
      "step": 64700
    },
    {
      "entropy": 5.796337270736695,
      "epoch": 5.869466618287373,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021685010477363214,
      "loss": 4.9228,
      "mean_token_accuracy": 0.22333958148956298,
      "num_tokens": 134279954.0,
      "step": 64705
    },
    {
      "entropy": 5.721205472946167,
      "epoch": 5.869920174165458,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021681884822422775,
      "loss": 4.8542,
      "mean_token_accuracy": 0.23122881799936296,
      "num_tokens": 134290719.0,
      "step": 64710
    },
    {
      "entropy": 5.718817710876465,
      "epoch": 5.870373730043541,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021678759287802448,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22654211819171904,
      "num_tokens": 134300556.0,
      "step": 64715
    },
    {
      "entropy": 5.714639377593994,
      "epoch": 5.870827285921625,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021675633873566853,
      "loss": 4.9238,
      "mean_token_accuracy": 0.22221433222293854,
      "num_tokens": 134312265.0,
      "step": 64720
    },
    {
      "entropy": 5.717086124420166,
      "epoch": 5.8712808417997095,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021672508579780636,
      "loss": 4.8839,
      "mean_token_accuracy": 0.2246651381254196,
      "num_tokens": 134323124.0,
      "step": 64725
    },
    {
      "entropy": 5.727870893478394,
      "epoch": 5.871734397677794,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002166938340650843,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2356557622551918,
      "num_tokens": 134331775.0,
      "step": 64730
    },
    {
      "entropy": 5.718722152709961,
      "epoch": 5.872187953555878,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021666258353814855,
      "loss": 4.8496,
      "mean_token_accuracy": 0.2325512483716011,
      "num_tokens": 134342471.0,
      "step": 64735
    },
    {
      "entropy": 5.68462176322937,
      "epoch": 5.872641509433962,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021663133421764554,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23270603120326996,
      "num_tokens": 134353380.0,
      "step": 64740
    },
    {
      "entropy": 5.709676218032837,
      "epoch": 5.873095065312047,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021660008610422133,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22414373755455017,
      "num_tokens": 134364060.0,
      "step": 64745
    },
    {
      "entropy": 5.700548696517944,
      "epoch": 5.873548621190131,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021656883919852227,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2168250098824501,
      "num_tokens": 134374337.0,
      "step": 64750
    },
    {
      "entropy": 5.703072786331177,
      "epoch": 5.874002177068215,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021653759350119444,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23427161127328872,
      "num_tokens": 134385868.0,
      "step": 64755
    },
    {
      "entropy": 5.779260063171387,
      "epoch": 5.8744557329462985,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021650634901288412,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22348724901676179,
      "num_tokens": 134397324.0,
      "step": 64760
    },
    {
      "entropy": 5.771374130249024,
      "epoch": 5.874909288824383,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002164751057342374,
      "loss": 4.8887,
      "mean_token_accuracy": 0.22075001299381256,
      "num_tokens": 134407408.0,
      "step": 64765
    },
    {
      "entropy": 5.702367973327637,
      "epoch": 5.875362844702467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002164438636659004,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22891307175159453,
      "num_tokens": 134416948.0,
      "step": 64770
    },
    {
      "entropy": 5.664295673370361,
      "epoch": 5.875816400580551,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021641262280851925,
      "loss": 4.7976,
      "mean_token_accuracy": 0.23703758269548417,
      "num_tokens": 134428274.0,
      "step": 64775
    },
    {
      "entropy": 5.641127204895019,
      "epoch": 5.876269956458636,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021638138316273998,
      "loss": 4.7589,
      "mean_token_accuracy": 0.23737382888793945,
      "num_tokens": 134437968.0,
      "step": 64780
    },
    {
      "entropy": 5.638536405563355,
      "epoch": 5.87672351233672,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002163501447292086,
      "loss": 4.7796,
      "mean_token_accuracy": 0.2309774488210678,
      "num_tokens": 134448349.0,
      "step": 64785
    },
    {
      "entropy": 5.69555630683899,
      "epoch": 5.877177068214804,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002163189075085712,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23588410317897796,
      "num_tokens": 134458867.0,
      "step": 64790
    },
    {
      "entropy": 5.767416334152221,
      "epoch": 5.877630624092888,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021628767150147377,
      "loss": 4.9241,
      "mean_token_accuracy": 0.22212524861097335,
      "num_tokens": 134469588.0,
      "step": 64795
    },
    {
      "entropy": 5.737659358978272,
      "epoch": 5.878084179970973,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021625643670856225,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23954017013311385,
      "num_tokens": 134479698.0,
      "step": 64800
    },
    {
      "entropy": 5.699388742446899,
      "epoch": 5.878537735849057,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021622520313048255,
      "loss": 4.7412,
      "mean_token_accuracy": 0.23723701387643814,
      "num_tokens": 134490015.0,
      "step": 64805
    },
    {
      "entropy": 5.716898632049561,
      "epoch": 5.878991291727141,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021619397076788065,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22950996905565263,
      "num_tokens": 134499536.0,
      "step": 64810
    },
    {
      "entropy": 5.660066938400268,
      "epoch": 5.879444847605225,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021616273962140242,
      "loss": 4.7732,
      "mean_token_accuracy": 0.2366795390844345,
      "num_tokens": 134510191.0,
      "step": 64815
    },
    {
      "entropy": 5.676793575286865,
      "epoch": 5.879898403483309,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021613150969169377,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2344587817788124,
      "num_tokens": 134520387.0,
      "step": 64820
    },
    {
      "entropy": 5.696463680267334,
      "epoch": 5.880351959361393,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021610028097940044,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22914999574422837,
      "num_tokens": 134530368.0,
      "step": 64825
    },
    {
      "entropy": 5.752265310287475,
      "epoch": 5.880805515239477,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002160690534851683,
      "loss": 4.904,
      "mean_token_accuracy": 0.22250230759382247,
      "num_tokens": 134540275.0,
      "step": 64830
    },
    {
      "entropy": 5.751452207565308,
      "epoch": 5.881259071117562,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002160378272096432,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22713685780763626,
      "num_tokens": 134551564.0,
      "step": 64835
    },
    {
      "entropy": 5.651205348968506,
      "epoch": 5.881712626995646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021600660215347084,
      "loss": 4.8055,
      "mean_token_accuracy": 0.23687406331300737,
      "num_tokens": 134562929.0,
      "step": 64840
    },
    {
      "entropy": 5.688595581054687,
      "epoch": 5.88216618287373,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000215975378317297,
      "loss": 4.8227,
      "mean_token_accuracy": 0.23596263229846953,
      "num_tokens": 134573171.0,
      "step": 64845
    },
    {
      "entropy": 5.670881032943726,
      "epoch": 5.882619738751814,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021594415570176735,
      "loss": 4.7779,
      "mean_token_accuracy": 0.23235397040843964,
      "num_tokens": 134583132.0,
      "step": 64850
    },
    {
      "entropy": 5.675733518600464,
      "epoch": 5.883073294629899,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021591293430752763,
      "loss": 4.8153,
      "mean_token_accuracy": 0.23706115186214446,
      "num_tokens": 134594070.0,
      "step": 64855
    },
    {
      "entropy": 5.6701374530792235,
      "epoch": 5.883526850507983,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021588171413522344,
      "loss": 4.8336,
      "mean_token_accuracy": 0.22844426333904266,
      "num_tokens": 134603964.0,
      "step": 64860
    },
    {
      "entropy": 5.66662015914917,
      "epoch": 5.883980406386067,
      "grad_norm": 1.25,
      "learning_rate": 0.00021585049518550053,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23414244651794433,
      "num_tokens": 134615951.0,
      "step": 64865
    },
    {
      "entropy": 5.672884178161621,
      "epoch": 5.884433962264151,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002158192774590045,
      "loss": 4.8461,
      "mean_token_accuracy": 0.22847237437963486,
      "num_tokens": 134626397.0,
      "step": 64870
    },
    {
      "entropy": 5.674179649353027,
      "epoch": 5.884887518142235,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021578806095638077,
      "loss": 4.8026,
      "mean_token_accuracy": 0.23085835725069045,
      "num_tokens": 134637264.0,
      "step": 64875
    },
    {
      "entropy": 5.741304063796997,
      "epoch": 5.885341074020319,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021575684567827502,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22768388241529464,
      "num_tokens": 134648170.0,
      "step": 64880
    },
    {
      "entropy": 5.713961267471314,
      "epoch": 5.885794629898403,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021572563162533293,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23057930618524553,
      "num_tokens": 134659174.0,
      "step": 64885
    },
    {
      "entropy": 5.710559606552124,
      "epoch": 5.886248185776488,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021569441879819973,
      "loss": 4.8682,
      "mean_token_accuracy": 0.23287445157766343,
      "num_tokens": 134669810.0,
      "step": 64890
    },
    {
      "entropy": 5.7550863265991214,
      "epoch": 5.886701741654572,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021566320719752113,
      "loss": 4.9525,
      "mean_token_accuracy": 0.22257194370031358,
      "num_tokens": 134679777.0,
      "step": 64895
    },
    {
      "entropy": 5.628110647201538,
      "epoch": 5.887155297532656,
      "grad_norm": 1.5,
      "learning_rate": 0.00021563199682394257,
      "loss": 4.76,
      "mean_token_accuracy": 0.23747376054525376,
      "num_tokens": 134690052.0,
      "step": 64900
    },
    {
      "entropy": 5.718479490280151,
      "epoch": 5.8876088534107405,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021560078767810943,
      "loss": 4.9407,
      "mean_token_accuracy": 0.22336386144161224,
      "num_tokens": 134701191.0,
      "step": 64905
    },
    {
      "entropy": 5.709280872344971,
      "epoch": 5.888062409288825,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002155695797606671,
      "loss": 4.9097,
      "mean_token_accuracy": 0.2322255089879036,
      "num_tokens": 134711852.0,
      "step": 64910
    },
    {
      "entropy": 5.630353879928589,
      "epoch": 5.888515965166908,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021553837307226103,
      "loss": 4.7034,
      "mean_token_accuracy": 0.25170817971229553,
      "num_tokens": 134722391.0,
      "step": 64915
    },
    {
      "entropy": 5.71423077583313,
      "epoch": 5.888969521044992,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021550716761353656,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2264970526099205,
      "num_tokens": 134732274.0,
      "step": 64920
    },
    {
      "entropy": 5.689841461181641,
      "epoch": 5.889423076923077,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021547596338513896,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2406940460205078,
      "num_tokens": 134742877.0,
      "step": 64925
    },
    {
      "entropy": 5.74317946434021,
      "epoch": 5.889876632801161,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021544476038771367,
      "loss": 4.897,
      "mean_token_accuracy": 0.22056645154953003,
      "num_tokens": 134752821.0,
      "step": 64930
    },
    {
      "entropy": 5.698602533340454,
      "epoch": 5.890330188679245,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021541355862190591,
      "loss": 4.8662,
      "mean_token_accuracy": 0.22912800163030625,
      "num_tokens": 134763175.0,
      "step": 64935
    },
    {
      "entropy": 5.735682392120362,
      "epoch": 5.8907837445573294,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021538235808836093,
      "loss": 4.843,
      "mean_token_accuracy": 0.23462677597999573,
      "num_tokens": 134773037.0,
      "step": 64940
    },
    {
      "entropy": 5.667714023590088,
      "epoch": 5.891237300435414,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021535115878772398,
      "loss": 4.7647,
      "mean_token_accuracy": 0.23964281231164933,
      "num_tokens": 134783306.0,
      "step": 64945
    },
    {
      "entropy": 5.696981573104859,
      "epoch": 5.891690856313498,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002153199607206403,
      "loss": 4.8327,
      "mean_token_accuracy": 0.23326566368341445,
      "num_tokens": 134793216.0,
      "step": 64950
    },
    {
      "entropy": 5.782320261001587,
      "epoch": 5.892144412191582,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021528876388775503,
      "loss": 4.9307,
      "mean_token_accuracy": 0.22453901767730713,
      "num_tokens": 134802705.0,
      "step": 64955
    },
    {
      "entropy": 5.641586971282959,
      "epoch": 5.8925979680696665,
      "grad_norm": 1.375,
      "learning_rate": 0.00021525756828971328,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23187139630317688,
      "num_tokens": 134813151.0,
      "step": 64960
    },
    {
      "entropy": 5.706297445297241,
      "epoch": 5.893051523947751,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021522637392716037,
      "loss": 4.7943,
      "mean_token_accuracy": 0.24066072255373,
      "num_tokens": 134823281.0,
      "step": 64965
    },
    {
      "entropy": 5.725685691833496,
      "epoch": 5.893505079825834,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002151951808007412,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22719634920358658,
      "num_tokens": 134833965.0,
      "step": 64970
    },
    {
      "entropy": 5.711892938613891,
      "epoch": 5.893958635703918,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002151639889111009,
      "loss": 4.8533,
      "mean_token_accuracy": 0.22199054807424545,
      "num_tokens": 134844043.0,
      "step": 64975
    },
    {
      "entropy": 5.742860794067383,
      "epoch": 5.894412191582003,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002151327982588846,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22906060367822648,
      "num_tokens": 134853011.0,
      "step": 64980
    },
    {
      "entropy": 5.734624671936035,
      "epoch": 5.894865747460087,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021510160884473723,
      "loss": 4.845,
      "mean_token_accuracy": 0.2255838245153427,
      "num_tokens": 134863660.0,
      "step": 64985
    },
    {
      "entropy": 5.721452474594116,
      "epoch": 5.895319303338171,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021507042066930387,
      "loss": 4.8512,
      "mean_token_accuracy": 0.23456015437841415,
      "num_tokens": 134874026.0,
      "step": 64990
    },
    {
      "entropy": 5.793324041366577,
      "epoch": 5.8957728592162555,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021503923373322953,
      "loss": 4.9079,
      "mean_token_accuracy": 0.22527338415384293,
      "num_tokens": 134883381.0,
      "step": 64995
    },
    {
      "entropy": 5.689508724212646,
      "epoch": 5.89622641509434,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021500804803715906,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23396133333444596,
      "num_tokens": 134893782.0,
      "step": 65000
    },
    {
      "entropy": 5.699884080886841,
      "epoch": 5.896679970972424,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002149768635817375,
      "loss": 4.8357,
      "mean_token_accuracy": 0.23244285583496094,
      "num_tokens": 134904698.0,
      "step": 65005
    },
    {
      "entropy": 5.6692615985870365,
      "epoch": 5.897133526850508,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021494568036760963,
      "loss": 4.7166,
      "mean_token_accuracy": 0.2452177330851555,
      "num_tokens": 134914764.0,
      "step": 65010
    },
    {
      "entropy": 5.7309643745422365,
      "epoch": 5.897587082728592,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002149144983954205,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22733092159032822,
      "num_tokens": 134925679.0,
      "step": 65015
    },
    {
      "entropy": 5.738805437088013,
      "epoch": 5.898040638606677,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021488331766581476,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22272309064865112,
      "num_tokens": 134935233.0,
      "step": 65020
    },
    {
      "entropy": 5.669378662109375,
      "epoch": 5.89849419448476,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021485213817943732,
      "loss": 4.8143,
      "mean_token_accuracy": 0.23060944080352783,
      "num_tokens": 134945014.0,
      "step": 65025
    },
    {
      "entropy": 5.7024119853973385,
      "epoch": 5.8989477503628445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021482095993693306,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23787232637405395,
      "num_tokens": 134955225.0,
      "step": 65030
    },
    {
      "entropy": 5.608464622497559,
      "epoch": 5.899401306240929,
      "grad_norm": 1.5,
      "learning_rate": 0.00021478978293894662,
      "loss": 4.7842,
      "mean_token_accuracy": 0.24830244481563568,
      "num_tokens": 134965457.0,
      "step": 65035
    },
    {
      "entropy": 5.726892518997192,
      "epoch": 5.899854862119013,
      "grad_norm": 1.5,
      "learning_rate": 0.0002147586071861229,
      "loss": 4.8759,
      "mean_token_accuracy": 0.23186870366334916,
      "num_tokens": 134976819.0,
      "step": 65040
    },
    {
      "entropy": 5.71324348449707,
      "epoch": 5.900308417997097,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021472743267910644,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23207805603742598,
      "num_tokens": 134987179.0,
      "step": 65045
    },
    {
      "entropy": 5.7195446491241455,
      "epoch": 5.9007619738751815,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002146962594185421,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23499255180358886,
      "num_tokens": 134997673.0,
      "step": 65050
    },
    {
      "entropy": 5.73437385559082,
      "epoch": 5.901215529753266,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002146650874050744,
      "loss": 4.9402,
      "mean_token_accuracy": 0.22193656861782074,
      "num_tokens": 135008512.0,
      "step": 65055
    },
    {
      "entropy": 5.660632228851318,
      "epoch": 5.90166908563135,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021463391663934818,
      "loss": 4.7554,
      "mean_token_accuracy": 0.23766975849866867,
      "num_tokens": 135018600.0,
      "step": 65060
    },
    {
      "entropy": 5.648023319244385,
      "epoch": 5.902122641509434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021460274712200789,
      "loss": 4.7482,
      "mean_token_accuracy": 0.23536430150270463,
      "num_tokens": 135029248.0,
      "step": 65065
    },
    {
      "entropy": 5.65427885055542,
      "epoch": 5.902576197387518,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002145715788536981,
      "loss": 4.7925,
      "mean_token_accuracy": 0.2308007687330246,
      "num_tokens": 135040223.0,
      "step": 65070
    },
    {
      "entropy": 5.645760202407837,
      "epoch": 5.903029753265602,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021454041183506357,
      "loss": 4.7674,
      "mean_token_accuracy": 0.23435329496860505,
      "num_tokens": 135049759.0,
      "step": 65075
    },
    {
      "entropy": 5.712612104415894,
      "epoch": 5.903483309143686,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002145092460667487,
      "loss": 4.9371,
      "mean_token_accuracy": 0.22299854904413224,
      "num_tokens": 135059485.0,
      "step": 65080
    },
    {
      "entropy": 5.589196872711182,
      "epoch": 5.9039368650217705,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021447808154939797,
      "loss": 4.6803,
      "mean_token_accuracy": 0.24733156710863113,
      "num_tokens": 135069536.0,
      "step": 65085
    },
    {
      "entropy": 5.7180359840393065,
      "epoch": 5.904390420899855,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021444691828365604,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22579895853996276,
      "num_tokens": 135080020.0,
      "step": 65090
    },
    {
      "entropy": 5.735708808898925,
      "epoch": 5.904843976777939,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021441575627016723,
      "loss": 4.8557,
      "mean_token_accuracy": 0.2344529524445534,
      "num_tokens": 135090374.0,
      "step": 65095
    },
    {
      "entropy": 5.762402963638306,
      "epoch": 5.905297532656023,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021438459550957606,
      "loss": 4.9119,
      "mean_token_accuracy": 0.22915777564048767,
      "num_tokens": 135101437.0,
      "step": 65100
    },
    {
      "entropy": 5.726598882675171,
      "epoch": 5.905751088534108,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021435343600252683,
      "loss": 4.9322,
      "mean_token_accuracy": 0.22318273037672043,
      "num_tokens": 135112113.0,
      "step": 65105
    },
    {
      "entropy": 5.58823447227478,
      "epoch": 5.906204644412192,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000214322277749664,
      "loss": 4.7504,
      "mean_token_accuracy": 0.23672262877225875,
      "num_tokens": 135123718.0,
      "step": 65110
    },
    {
      "entropy": 5.745626544952392,
      "epoch": 5.906658200290276,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021429112075163197,
      "loss": 4.8508,
      "mean_token_accuracy": 0.23014121800661086,
      "num_tokens": 135134693.0,
      "step": 65115
    },
    {
      "entropy": 5.731426429748535,
      "epoch": 5.90711175616836,
      "grad_norm": 1.375,
      "learning_rate": 0.00021425996500907502,
      "loss": 4.8455,
      "mean_token_accuracy": 0.23540338277816772,
      "num_tokens": 135145057.0,
      "step": 65120
    },
    {
      "entropy": 5.705480575561523,
      "epoch": 5.907565312046444,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002142288105226375,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23111762404441832,
      "num_tokens": 135155667.0,
      "step": 65125
    },
    {
      "entropy": 5.781964921951294,
      "epoch": 5.908018867924528,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021419765729296364,
      "loss": 4.9579,
      "mean_token_accuracy": 0.21528838872909545,
      "num_tokens": 135166677.0,
      "step": 65130
    },
    {
      "entropy": 5.7356836795806885,
      "epoch": 5.908472423802612,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002141665053206977,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22963547259569167,
      "num_tokens": 135177029.0,
      "step": 65135
    },
    {
      "entropy": 5.699616479873657,
      "epoch": 5.9089259796806965,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021413535460648392,
      "loss": 4.8667,
      "mean_token_accuracy": 0.2253642350435257,
      "num_tokens": 135187798.0,
      "step": 65140
    },
    {
      "entropy": 5.640416860580444,
      "epoch": 5.909379535558781,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021410420515096656,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23034027963876724,
      "num_tokens": 135198805.0,
      "step": 65145
    },
    {
      "entropy": 5.777015447616577,
      "epoch": 5.909833091436865,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021407305695478974,
      "loss": 4.8793,
      "mean_token_accuracy": 0.23001266568899154,
      "num_tokens": 135209781.0,
      "step": 65150
    },
    {
      "entropy": 5.683942794799805,
      "epoch": 5.910286647314949,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021404191001859763,
      "loss": 4.7671,
      "mean_token_accuracy": 0.23463771790266036,
      "num_tokens": 135219683.0,
      "step": 65155
    },
    {
      "entropy": 5.6349273204803465,
      "epoch": 5.910740203193034,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021401076434303436,
      "loss": 4.7508,
      "mean_token_accuracy": 0.235362246632576,
      "num_tokens": 135230085.0,
      "step": 65160
    },
    {
      "entropy": 5.707024049758911,
      "epoch": 5.911193759071118,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021397961992874405,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22978705763816834,
      "num_tokens": 135239601.0,
      "step": 65165
    },
    {
      "entropy": 5.73832745552063,
      "epoch": 5.911647314949201,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021394847677637068,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2277977228164673,
      "num_tokens": 135249649.0,
      "step": 65170
    },
    {
      "entropy": 5.697419786453247,
      "epoch": 5.912100870827286,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002139173348865584,
      "loss": 4.7953,
      "mean_token_accuracy": 0.23803581148386002,
      "num_tokens": 135259948.0,
      "step": 65175
    },
    {
      "entropy": 5.650695323944092,
      "epoch": 5.91255442670537,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021388619425995127,
      "loss": 4.8069,
      "mean_token_accuracy": 0.23489006757736205,
      "num_tokens": 135270650.0,
      "step": 65180
    },
    {
      "entropy": 5.692415618896485,
      "epoch": 5.913007982583454,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002138550548971931,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22781420946121217,
      "num_tokens": 135281547.0,
      "step": 65185
    },
    {
      "entropy": 5.682812976837158,
      "epoch": 5.913461538461538,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021382391679892804,
      "loss": 4.7831,
      "mean_token_accuracy": 0.23837026804685593,
      "num_tokens": 135290256.0,
      "step": 65190
    },
    {
      "entropy": 5.699838066101075,
      "epoch": 5.913915094339623,
      "grad_norm": 1.375,
      "learning_rate": 0.00021379277996579993,
      "loss": 4.8581,
      "mean_token_accuracy": 0.23004069924354553,
      "num_tokens": 135300542.0,
      "step": 65195
    },
    {
      "entropy": 5.678812074661255,
      "epoch": 5.914368650217707,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002137616443984528,
      "loss": 4.7673,
      "mean_token_accuracy": 0.23421405851840973,
      "num_tokens": 135311354.0,
      "step": 65200
    },
    {
      "entropy": 5.691715908050537,
      "epoch": 5.914822206095791,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021373051009753035,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2275543689727783,
      "num_tokens": 135322620.0,
      "step": 65205
    },
    {
      "entropy": 5.701027154922485,
      "epoch": 5.915275761973875,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002136993770636767,
      "loss": 4.8572,
      "mean_token_accuracy": 0.2333531692624092,
      "num_tokens": 135333199.0,
      "step": 65210
    },
    {
      "entropy": 5.711991310119629,
      "epoch": 5.91572931785196,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021366824529753554,
      "loss": 4.8708,
      "mean_token_accuracy": 0.23229458630084993,
      "num_tokens": 135343964.0,
      "step": 65215
    },
    {
      "entropy": 5.720633172988892,
      "epoch": 5.916182873730044,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021363711479975063,
      "loss": 4.8305,
      "mean_token_accuracy": 0.22948067635297775,
      "num_tokens": 135354598.0,
      "step": 65220
    },
    {
      "entropy": 5.6685709953308105,
      "epoch": 5.916636429608127,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021360598557096583,
      "loss": 4.8527,
      "mean_token_accuracy": 0.22756435871124267,
      "num_tokens": 135365228.0,
      "step": 65225
    },
    {
      "entropy": 5.676816844940186,
      "epoch": 5.917089985486212,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021357485761182488,
      "loss": 4.8377,
      "mean_token_accuracy": 0.23408458530902862,
      "num_tokens": 135375612.0,
      "step": 65230
    },
    {
      "entropy": 5.706959676742554,
      "epoch": 5.917543541364296,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021354373092297157,
      "loss": 4.8596,
      "mean_token_accuracy": 0.23186856508255005,
      "num_tokens": 135386333.0,
      "step": 65235
    },
    {
      "entropy": 5.699850416183471,
      "epoch": 5.91799709724238,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021351260550504954,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23044949918985366,
      "num_tokens": 135396563.0,
      "step": 65240
    },
    {
      "entropy": 5.684767723083496,
      "epoch": 5.918450653120464,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002134814813587025,
      "loss": 4.8005,
      "mean_token_accuracy": 0.23572303354740143,
      "num_tokens": 135406895.0,
      "step": 65245
    },
    {
      "entropy": 5.7507976531982425,
      "epoch": 5.918904208998549,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00021345035848457412,
      "loss": 4.8531,
      "mean_token_accuracy": 0.23778323084115982,
      "num_tokens": 135416946.0,
      "step": 65250
    },
    {
      "entropy": 5.744440364837646,
      "epoch": 5.919357764876633,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021341923688330806,
      "loss": 4.8891,
      "mean_token_accuracy": 0.23353171199560166,
      "num_tokens": 135427662.0,
      "step": 65255
    },
    {
      "entropy": 5.681900405883789,
      "epoch": 5.919811320754717,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002133881165555478,
      "loss": 4.8128,
      "mean_token_accuracy": 0.2271723583340645,
      "num_tokens": 135437416.0,
      "step": 65260
    },
    {
      "entropy": 5.679203796386719,
      "epoch": 5.920264876632801,
      "grad_norm": 1.484375,
      "learning_rate": 0.000213356997501937,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23794140964746474,
      "num_tokens": 135447643.0,
      "step": 65265
    },
    {
      "entropy": 5.676841402053833,
      "epoch": 5.920718432510886,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021332587972311924,
      "loss": 4.8613,
      "mean_token_accuracy": 0.2334899440407753,
      "num_tokens": 135458604.0,
      "step": 65270
    },
    {
      "entropy": 5.677072620391845,
      "epoch": 5.92117198838897,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002132947632197379,
      "loss": 4.7803,
      "mean_token_accuracy": 0.23348648548126222,
      "num_tokens": 135469591.0,
      "step": 65275
    },
    {
      "entropy": 5.612297201156617,
      "epoch": 5.921625544267053,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002132636479924367,
      "loss": 4.7718,
      "mean_token_accuracy": 0.2364058867096901,
      "num_tokens": 135479656.0,
      "step": 65280
    },
    {
      "entropy": 5.73899245262146,
      "epoch": 5.922079100145138,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021323253404185893,
      "loss": 4.8927,
      "mean_token_accuracy": 0.21755610406398773,
      "num_tokens": 135490838.0,
      "step": 65285
    },
    {
      "entropy": 5.656654262542725,
      "epoch": 5.922532656023222,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021320142136864806,
      "loss": 4.7412,
      "mean_token_accuracy": 0.24251447468996049,
      "num_tokens": 135500862.0,
      "step": 65290
    },
    {
      "entropy": 5.7606712818145756,
      "epoch": 5.922986211901306,
      "grad_norm": 1.65625,
      "learning_rate": 0.00021317030997344767,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22255772054195405,
      "num_tokens": 135510521.0,
      "step": 65295
    },
    {
      "entropy": 5.692900705337524,
      "epoch": 5.92343976777939,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021313919985690088,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23333031088113784,
      "num_tokens": 135520348.0,
      "step": 65300
    },
    {
      "entropy": 5.711297416687012,
      "epoch": 5.923893323657475,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021310809101965134,
      "loss": 4.8465,
      "mean_token_accuracy": 0.22689120918512345,
      "num_tokens": 135531482.0,
      "step": 65305
    },
    {
      "entropy": 5.664238023757934,
      "epoch": 5.924346879535559,
      "grad_norm": 1.375,
      "learning_rate": 0.0002130769834623421,
      "loss": 4.8076,
      "mean_token_accuracy": 0.22964251935482025,
      "num_tokens": 135541555.0,
      "step": 65310
    },
    {
      "entropy": 5.698928260803223,
      "epoch": 5.924800435413643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021304587718561674,
      "loss": 4.8219,
      "mean_token_accuracy": 0.2299882099032402,
      "num_tokens": 135553459.0,
      "step": 65315
    },
    {
      "entropy": 5.805053329467773,
      "epoch": 5.9252539912917275,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021301477219011838,
      "loss": 4.9357,
      "mean_token_accuracy": 0.22440853714942932,
      "num_tokens": 135563580.0,
      "step": 65320
    },
    {
      "entropy": 5.71978530883789,
      "epoch": 5.925707547169811,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002129836684764903,
      "loss": 4.8695,
      "mean_token_accuracy": 0.22719806134700776,
      "num_tokens": 135575131.0,
      "step": 65325
    },
    {
      "entropy": 5.781659936904907,
      "epoch": 5.926161103047895,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021295256604537583,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2235329657793045,
      "num_tokens": 135585427.0,
      "step": 65330
    },
    {
      "entropy": 5.681419420242309,
      "epoch": 5.926614658925979,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002129214648974181,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2311760514974594,
      "num_tokens": 135595118.0,
      "step": 65335
    },
    {
      "entropy": 5.779699373245239,
      "epoch": 5.927068214804064,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021289036503326027,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22080785483121873,
      "num_tokens": 135606449.0,
      "step": 65340
    },
    {
      "entropy": 5.651597261428833,
      "epoch": 5.927521770682148,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021285926645354548,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2344757840037346,
      "num_tokens": 135616250.0,
      "step": 65345
    },
    {
      "entropy": 5.6861825466156,
      "epoch": 5.927975326560232,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021282816915891694,
      "loss": 4.8506,
      "mean_token_accuracy": 0.2314467504620552,
      "num_tokens": 135626882.0,
      "step": 65350
    },
    {
      "entropy": 5.650212574005127,
      "epoch": 5.9284288824383164,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021279707315001773,
      "loss": 4.7826,
      "mean_token_accuracy": 0.2420904204249382,
      "num_tokens": 135636824.0,
      "step": 65355
    },
    {
      "entropy": 5.771042728424073,
      "epoch": 5.928882438316401,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021276597842749084,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2235725775361061,
      "num_tokens": 135647310.0,
      "step": 65360
    },
    {
      "entropy": 5.753791761398316,
      "epoch": 5.929335994194485,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002127348849919794,
      "loss": 4.867,
      "mean_token_accuracy": 0.22946480810642242,
      "num_tokens": 135657192.0,
      "step": 65365
    },
    {
      "entropy": 5.634783983230591,
      "epoch": 5.929789550072569,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021270379284412644,
      "loss": 4.7303,
      "mean_token_accuracy": 0.24482865482568741,
      "num_tokens": 135667780.0,
      "step": 65370
    },
    {
      "entropy": 5.652531433105469,
      "epoch": 5.9302431059506535,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002126727019845749,
      "loss": 4.7181,
      "mean_token_accuracy": 0.24845789670944213,
      "num_tokens": 135677891.0,
      "step": 65375
    },
    {
      "entropy": 5.675779342651367,
      "epoch": 5.930696661828737,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021264161241396767,
      "loss": 4.7877,
      "mean_token_accuracy": 0.24108323603868484,
      "num_tokens": 135687575.0,
      "step": 65380
    },
    {
      "entropy": 5.722821569442749,
      "epoch": 5.931150217706821,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002126105241329479,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2276616632938385,
      "num_tokens": 135698592.0,
      "step": 65385
    },
    {
      "entropy": 5.596069383621216,
      "epoch": 5.931603773584905,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021257943714215845,
      "loss": 4.7044,
      "mean_token_accuracy": 0.24014799147844315,
      "num_tokens": 135709159.0,
      "step": 65390
    },
    {
      "entropy": 5.760043907165527,
      "epoch": 5.93205732946299,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021254835144224205,
      "loss": 4.9443,
      "mean_token_accuracy": 0.22515711486339568,
      "num_tokens": 135718639.0,
      "step": 65395
    },
    {
      "entropy": 5.710653829574585,
      "epoch": 5.932510885341074,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021251726703384167,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22428881078958512,
      "num_tokens": 135729494.0,
      "step": 65400
    },
    {
      "entropy": 5.711664009094238,
      "epoch": 5.932964441219158,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021248618391760005,
      "loss": 4.8497,
      "mean_token_accuracy": 0.231598162651062,
      "num_tokens": 135740456.0,
      "step": 65405
    },
    {
      "entropy": 5.633752632141113,
      "epoch": 5.9334179970972425,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021245510209416018,
      "loss": 4.7601,
      "mean_token_accuracy": 0.24266624748706817,
      "num_tokens": 135750363.0,
      "step": 65410
    },
    {
      "entropy": 5.640265369415284,
      "epoch": 5.933871552975327,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021242402156416474,
      "loss": 4.7968,
      "mean_token_accuracy": 0.23679712563753127,
      "num_tokens": 135760505.0,
      "step": 65415
    },
    {
      "entropy": 5.720248985290527,
      "epoch": 5.934325108853411,
      "grad_norm": 1.5,
      "learning_rate": 0.0002123929423282564,
      "loss": 4.789,
      "mean_token_accuracy": 0.23586370795965195,
      "num_tokens": 135770708.0,
      "step": 65420
    },
    {
      "entropy": 5.736480093002319,
      "epoch": 5.934778664731495,
      "grad_norm": 1.375,
      "learning_rate": 0.000212361864387078,
      "loss": 4.8957,
      "mean_token_accuracy": 0.22733516693115235,
      "num_tokens": 135780761.0,
      "step": 65425
    },
    {
      "entropy": 5.740129804611206,
      "epoch": 5.93523222060958,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021233078774127225,
      "loss": 4.9203,
      "mean_token_accuracy": 0.2302001193165779,
      "num_tokens": 135790427.0,
      "step": 65430
    },
    {
      "entropy": 5.749075984954834,
      "epoch": 5.935685776487663,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021229971239148176,
      "loss": 4.919,
      "mean_token_accuracy": 0.2326383873820305,
      "num_tokens": 135801962.0,
      "step": 65435
    },
    {
      "entropy": 5.682842874526978,
      "epoch": 5.936139332365747,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021226863833834913,
      "loss": 4.869,
      "mean_token_accuracy": 0.22693073600530625,
      "num_tokens": 135811942.0,
      "step": 65440
    },
    {
      "entropy": 5.762048053741455,
      "epoch": 5.9365928882438315,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021223756558251712,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22512658089399337,
      "num_tokens": 135822102.0,
      "step": 65445
    },
    {
      "entropy": 5.732373380661011,
      "epoch": 5.937046444121916,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021220649412462823,
      "loss": 4.8638,
      "mean_token_accuracy": 0.23527825027704238,
      "num_tokens": 135832804.0,
      "step": 65450
    },
    {
      "entropy": 5.706991386413574,
      "epoch": 5.9375,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021217542396532503,
      "loss": 4.8186,
      "mean_token_accuracy": 0.233304525911808,
      "num_tokens": 135842415.0,
      "step": 65455
    },
    {
      "entropy": 5.7051492691040036,
      "epoch": 5.937953555878084,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002121443551052501,
      "loss": 4.841,
      "mean_token_accuracy": 0.23461349904537201,
      "num_tokens": 135853084.0,
      "step": 65460
    },
    {
      "entropy": 5.7483419418334964,
      "epoch": 5.9384071117561685,
      "grad_norm": 1.640625,
      "learning_rate": 0.00021211328754504587,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2313359186053276,
      "num_tokens": 135863215.0,
      "step": 65465
    },
    {
      "entropy": 5.694288682937622,
      "epoch": 5.938860667634253,
      "grad_norm": 1.5,
      "learning_rate": 0.00021208222128535491,
      "loss": 4.8354,
      "mean_token_accuracy": 0.22531788051128387,
      "num_tokens": 135874167.0,
      "step": 65470
    },
    {
      "entropy": 5.670178651809692,
      "epoch": 5.939314223512337,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021205115632681965,
      "loss": 4.8146,
      "mean_token_accuracy": 0.2346041053533554,
      "num_tokens": 135883309.0,
      "step": 65475
    },
    {
      "entropy": 5.7348034381866455,
      "epoch": 5.93976777939042,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021202009267008254,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22596477270126342,
      "num_tokens": 135893750.0,
      "step": 65480
    },
    {
      "entropy": 5.750947046279907,
      "epoch": 5.940221335268505,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021198903031578598,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23440378308296203,
      "num_tokens": 135904244.0,
      "step": 65485
    },
    {
      "entropy": 5.671503305435181,
      "epoch": 5.940674891146589,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021195796926457223,
      "loss": 4.8071,
      "mean_token_accuracy": 0.23022500872612,
      "num_tokens": 135913538.0,
      "step": 65490
    },
    {
      "entropy": 5.69728045463562,
      "epoch": 5.941128447024673,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021192690951708383,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2395867258310318,
      "num_tokens": 135924711.0,
      "step": 65495
    },
    {
      "entropy": 5.678213262557984,
      "epoch": 5.9415820029027575,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021189585107396296,
      "loss": 4.8244,
      "mean_token_accuracy": 0.23176230639219284,
      "num_tokens": 135935389.0,
      "step": 65500
    },
    {
      "entropy": 5.625677394866943,
      "epoch": 5.942035558780842,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021186479393585196,
      "loss": 4.7623,
      "mean_token_accuracy": 0.2465125858783722,
      "num_tokens": 135945311.0,
      "step": 65505
    },
    {
      "entropy": 5.685950803756714,
      "epoch": 5.942489114658926,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002118337381033932,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23745209574699402,
      "num_tokens": 135954782.0,
      "step": 65510
    },
    {
      "entropy": 5.696381616592407,
      "epoch": 5.94294267053701,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021180268357722876,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23207719027996063,
      "num_tokens": 135965060.0,
      "step": 65515
    },
    {
      "entropy": 5.694208383560181,
      "epoch": 5.943396226415095,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021177163035800102,
      "loss": 4.9169,
      "mean_token_accuracy": 0.22568323761224746,
      "num_tokens": 135977373.0,
      "step": 65520
    },
    {
      "entropy": 5.662731742858886,
      "epoch": 5.943849782293179,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002117405784463519,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2263045623898506,
      "num_tokens": 135988426.0,
      "step": 65525
    },
    {
      "entropy": 5.6850262641906735,
      "epoch": 5.944303338171263,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002117095278429239,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23105384558439254,
      "num_tokens": 135998636.0,
      "step": 65530
    },
    {
      "entropy": 5.756299066543579,
      "epoch": 5.9447568940493465,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021167847854835893,
      "loss": 4.8733,
      "mean_token_accuracy": 0.23385589718818664,
      "num_tokens": 136009665.0,
      "step": 65535
    },
    {
      "entropy": 5.681488847732544,
      "epoch": 5.945210449927431,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021164743056329915,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22804685980081557,
      "num_tokens": 136020024.0,
      "step": 65540
    },
    {
      "entropy": 5.626901865005493,
      "epoch": 5.945664005805515,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021161638388838666,
      "loss": 4.732,
      "mean_token_accuracy": 0.24493592232465744,
      "num_tokens": 136030563.0,
      "step": 65545
    },
    {
      "entropy": 5.638283205032349,
      "epoch": 5.946117561683599,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021158533852426353,
      "loss": 4.7458,
      "mean_token_accuracy": 0.2381819322705269,
      "num_tokens": 136042000.0,
      "step": 65550
    },
    {
      "entropy": 5.6803360939025875,
      "epoch": 5.9465711175616836,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021155429447157178,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23132775574922562,
      "num_tokens": 136051570.0,
      "step": 65555
    },
    {
      "entropy": 5.708084774017334,
      "epoch": 5.947024673439768,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021152325173095328,
      "loss": 4.864,
      "mean_token_accuracy": 0.23349713683128356,
      "num_tokens": 136062170.0,
      "step": 65560
    },
    {
      "entropy": 5.631809139251709,
      "epoch": 5.947478229317852,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002114922103030502,
      "loss": 4.8008,
      "mean_token_accuracy": 0.23207793533802032,
      "num_tokens": 136071216.0,
      "step": 65565
    },
    {
      "entropy": 5.7031115055084225,
      "epoch": 5.947931785195936,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021146117018850435,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22445889413356782,
      "num_tokens": 136081359.0,
      "step": 65570
    },
    {
      "entropy": 5.694078874588013,
      "epoch": 5.948385341074021,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002114301313879577,
      "loss": 4.8359,
      "mean_token_accuracy": 0.2335870623588562,
      "num_tokens": 136092157.0,
      "step": 65575
    },
    {
      "entropy": 5.643405294418335,
      "epoch": 5.948838896952105,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002113990939020522,
      "loss": 4.7024,
      "mean_token_accuracy": 0.23991701006889343,
      "num_tokens": 136102077.0,
      "step": 65580
    },
    {
      "entropy": 5.722413063049316,
      "epoch": 5.949292452830189,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021136805773142958,
      "loss": 4.8495,
      "mean_token_accuracy": 0.23065562099218367,
      "num_tokens": 136113736.0,
      "step": 65585
    },
    {
      "entropy": 5.780789041519165,
      "epoch": 5.9497460087082725,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021133702287673173,
      "loss": 4.9525,
      "mean_token_accuracy": 0.2209150582551956,
      "num_tokens": 136123818.0,
      "step": 65590
    },
    {
      "entropy": 5.712390518188476,
      "epoch": 5.950199564586357,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002113059893386005,
      "loss": 4.8188,
      "mean_token_accuracy": 0.2272142842411995,
      "num_tokens": 136133340.0,
      "step": 65595
    },
    {
      "entropy": 5.7641833305358885,
      "epoch": 5.950653120464441,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00021127495711767763,
      "loss": 4.9924,
      "mean_token_accuracy": 0.22451099902391433,
      "num_tokens": 136144342.0,
      "step": 65600
    },
    {
      "entropy": 5.769424200057983,
      "epoch": 5.951106676342525,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021124392621460491,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21875570416450502,
      "num_tokens": 136155246.0,
      "step": 65605
    },
    {
      "entropy": 5.670628881454467,
      "epoch": 5.95156023222061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021121289663002395,
      "loss": 4.8388,
      "mean_token_accuracy": 0.24060185849666596,
      "num_tokens": 136165654.0,
      "step": 65610
    },
    {
      "entropy": 5.706529760360718,
      "epoch": 5.952013788098694,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002111818683645767,
      "loss": 4.818,
      "mean_token_accuracy": 0.23690466433763505,
      "num_tokens": 136176301.0,
      "step": 65615
    },
    {
      "entropy": 5.6994977474212645,
      "epoch": 5.952467343976778,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021115084141890456,
      "loss": 4.8122,
      "mean_token_accuracy": 0.23582924157381058,
      "num_tokens": 136187241.0,
      "step": 65620
    },
    {
      "entropy": 5.703855752944946,
      "epoch": 5.952920899854862,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002111198157936493,
      "loss": 4.8521,
      "mean_token_accuracy": 0.2286621630191803,
      "num_tokens": 136197436.0,
      "step": 65625
    },
    {
      "entropy": 5.594985103607177,
      "epoch": 5.953374455732947,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021108879148945255,
      "loss": 4.7282,
      "mean_token_accuracy": 0.2380169838666916,
      "num_tokens": 136207739.0,
      "step": 65630
    },
    {
      "entropy": 5.671165323257446,
      "epoch": 5.95382801161103,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021105776850695592,
      "loss": 4.8238,
      "mean_token_accuracy": 0.2284463807940483,
      "num_tokens": 136217230.0,
      "step": 65635
    },
    {
      "entropy": 5.703950023651123,
      "epoch": 5.954281567489114,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002110267468468009,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22954328060150148,
      "num_tokens": 136226626.0,
      "step": 65640
    },
    {
      "entropy": 5.700776243209839,
      "epoch": 5.954735123367199,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000210995726509629,
      "loss": 4.8049,
      "mean_token_accuracy": 0.2307190477848053,
      "num_tokens": 136236404.0,
      "step": 65645
    },
    {
      "entropy": 5.767683458328247,
      "epoch": 5.955188679245283,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021096470749608183,
      "loss": 4.898,
      "mean_token_accuracy": 0.233057402074337,
      "num_tokens": 136246735.0,
      "step": 65650
    },
    {
      "entropy": 5.738810300827026,
      "epoch": 5.955642235123367,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021093368980680083,
      "loss": 4.8732,
      "mean_token_accuracy": 0.223726287484169,
      "num_tokens": 136257286.0,
      "step": 65655
    },
    {
      "entropy": 5.63177375793457,
      "epoch": 5.956095791001451,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021090267344242743,
      "loss": 4.8047,
      "mean_token_accuracy": 0.22789078801870347,
      "num_tokens": 136268245.0,
      "step": 65660
    },
    {
      "entropy": 5.746881341934204,
      "epoch": 5.956549346879536,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021087165840360313,
      "loss": 4.939,
      "mean_token_accuracy": 0.22652660310268402,
      "num_tokens": 136279592.0,
      "step": 65665
    },
    {
      "entropy": 5.7114709377288815,
      "epoch": 5.95700290275762,
      "grad_norm": 1.375,
      "learning_rate": 0.00021084064469096926,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2380344420671463,
      "num_tokens": 136290128.0,
      "step": 65670
    },
    {
      "entropy": 5.74973692893982,
      "epoch": 5.957456458635704,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002108096323051672,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2344082236289978,
      "num_tokens": 136300506.0,
      "step": 65675
    },
    {
      "entropy": 5.719472742080688,
      "epoch": 5.957910014513788,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021077862124683827,
      "loss": 4.8577,
      "mean_token_accuracy": 0.22801639139652252,
      "num_tokens": 136310205.0,
      "step": 65680
    },
    {
      "entropy": 5.706651973724365,
      "epoch": 5.958363570391873,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002107476115166238,
      "loss": 4.8578,
      "mean_token_accuracy": 0.23116149455308915,
      "num_tokens": 136321870.0,
      "step": 65685
    },
    {
      "entropy": 5.682117366790772,
      "epoch": 5.958817126269956,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002107166031151652,
      "loss": 4.7799,
      "mean_token_accuracy": 0.2320043697953224,
      "num_tokens": 136332534.0,
      "step": 65690
    },
    {
      "entropy": 5.730310773849487,
      "epoch": 5.95927068214804,
      "grad_norm": 1.5,
      "learning_rate": 0.0002106855960431035,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22578218281269075,
      "num_tokens": 136341805.0,
      "step": 65695
    },
    {
      "entropy": 5.703595590591431,
      "epoch": 5.959724238026125,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021065459030108013,
      "loss": 4.8644,
      "mean_token_accuracy": 0.22308290600776673,
      "num_tokens": 136352664.0,
      "step": 65700
    },
    {
      "entropy": 5.7365867614746096,
      "epoch": 5.960177793904209,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021062358588973628,
      "loss": 4.8479,
      "mean_token_accuracy": 0.225959113240242,
      "num_tokens": 136362520.0,
      "step": 65705
    },
    {
      "entropy": 5.699816465377808,
      "epoch": 5.960631349782293,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021059258280971298,
      "loss": 4.8063,
      "mean_token_accuracy": 0.23584224432706832,
      "num_tokens": 136371930.0,
      "step": 65710
    },
    {
      "entropy": 5.728770160675049,
      "epoch": 5.961084905660377,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021056158106165157,
      "loss": 4.8632,
      "mean_token_accuracy": 0.23167552351951598,
      "num_tokens": 136381698.0,
      "step": 65715
    },
    {
      "entropy": 5.742983198165893,
      "epoch": 5.961538461538462,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021053058064619308,
      "loss": 4.781,
      "mean_token_accuracy": 0.2375450164079666,
      "num_tokens": 136390946.0,
      "step": 65720
    },
    {
      "entropy": 5.704337120056152,
      "epoch": 5.961992017416546,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021049958156397854,
      "loss": 4.8159,
      "mean_token_accuracy": 0.24166524410247803,
      "num_tokens": 136400028.0,
      "step": 65725
    },
    {
      "entropy": 5.710370206832886,
      "epoch": 5.96244557329463,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002104685838156491,
      "loss": 4.8584,
      "mean_token_accuracy": 0.231605863571167,
      "num_tokens": 136409626.0,
      "step": 65730
    },
    {
      "entropy": 5.702501344680786,
      "epoch": 5.962899129172714,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002104375874018458,
      "loss": 5.0061,
      "mean_token_accuracy": 0.21049256026744842,
      "num_tokens": 136420753.0,
      "step": 65735
    },
    {
      "entropy": 5.73600401878357,
      "epoch": 5.963352685050799,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002104065923232097,
      "loss": 4.8616,
      "mean_token_accuracy": 0.23356228023767472,
      "num_tokens": 136431342.0,
      "step": 65740
    },
    {
      "entropy": 5.730835866928101,
      "epoch": 5.963806240928882,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002103755985803817,
      "loss": 4.8569,
      "mean_token_accuracy": 0.23077307641506195,
      "num_tokens": 136441419.0,
      "step": 65745
    },
    {
      "entropy": 5.83523736000061,
      "epoch": 5.964259796806966,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021034460617400276,
      "loss": 5.0022,
      "mean_token_accuracy": 0.22291755825281143,
      "num_tokens": 136451685.0,
      "step": 65750
    },
    {
      "entropy": 5.709998273849488,
      "epoch": 5.964713352685051,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021031361510471386,
      "loss": 4.8153,
      "mean_token_accuracy": 0.22971617579460143,
      "num_tokens": 136461212.0,
      "step": 65755
    },
    {
      "entropy": 5.790847682952881,
      "epoch": 5.965166908563135,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002102826253731559,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22353093028068544,
      "num_tokens": 136471531.0,
      "step": 65760
    },
    {
      "entropy": 5.732756233215332,
      "epoch": 5.965620464441219,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002102516369799697,
      "loss": 4.8522,
      "mean_token_accuracy": 0.23320553749799727,
      "num_tokens": 136482109.0,
      "step": 65765
    },
    {
      "entropy": 5.721735143661499,
      "epoch": 5.9660740203193035,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002102206499257962,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23912979662418365,
      "num_tokens": 136492981.0,
      "step": 65770
    },
    {
      "entropy": 5.738364315032959,
      "epoch": 5.966527576197388,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002101896642112761,
      "loss": 4.8398,
      "mean_token_accuracy": 0.2287017822265625,
      "num_tokens": 136502333.0,
      "step": 65775
    },
    {
      "entropy": 5.711202144622803,
      "epoch": 5.966981132075472,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002101586798370503,
      "loss": 4.8068,
      "mean_token_accuracy": 0.237689870595932,
      "num_tokens": 136512940.0,
      "step": 65780
    },
    {
      "entropy": 5.65767879486084,
      "epoch": 5.967434687953556,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021012769680375955,
      "loss": 4.7957,
      "mean_token_accuracy": 0.23027997612953185,
      "num_tokens": 136523557.0,
      "step": 65785
    },
    {
      "entropy": 5.731646156311035,
      "epoch": 5.96788824383164,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021009671511204453,
      "loss": 4.8831,
      "mean_token_accuracy": 0.22745839208364488,
      "num_tokens": 136533352.0,
      "step": 65790
    },
    {
      "entropy": 5.774239778518677,
      "epoch": 5.968341799709724,
      "grad_norm": 1.375,
      "learning_rate": 0.00021006573476254597,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22702645063400267,
      "num_tokens": 136543377.0,
      "step": 65795
    },
    {
      "entropy": 5.658892440795898,
      "epoch": 5.968795355587808,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021003475575590458,
      "loss": 4.7644,
      "mean_token_accuracy": 0.2426096811890602,
      "num_tokens": 136553088.0,
      "step": 65800
    },
    {
      "entropy": 5.69241361618042,
      "epoch": 5.969248911465892,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000210003778092761,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22747244238853453,
      "num_tokens": 136563319.0,
      "step": 65805
    },
    {
      "entropy": 5.670722007751465,
      "epoch": 5.969702467343977,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002099728017737559,
      "loss": 4.8306,
      "mean_token_accuracy": 0.23336442708969116,
      "num_tokens": 136574278.0,
      "step": 65810
    },
    {
      "entropy": 5.701848030090332,
      "epoch": 5.970156023222061,
      "grad_norm": 1.5,
      "learning_rate": 0.00020994182679952978,
      "loss": 4.8747,
      "mean_token_accuracy": 0.2279978334903717,
      "num_tokens": 136583318.0,
      "step": 65815
    },
    {
      "entropy": 5.696918725967407,
      "epoch": 5.970609579100145,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020991085317072333,
      "loss": 4.8601,
      "mean_token_accuracy": 0.22686487436294556,
      "num_tokens": 136594226.0,
      "step": 65820
    },
    {
      "entropy": 5.737337636947632,
      "epoch": 5.9710631349782295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020987988088797687,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22961728870868683,
      "num_tokens": 136605540.0,
      "step": 65825
    },
    {
      "entropy": 5.717337036132813,
      "epoch": 5.971516690856314,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002098489099519312,
      "loss": 4.815,
      "mean_token_accuracy": 0.2346593976020813,
      "num_tokens": 136616154.0,
      "step": 65830
    },
    {
      "entropy": 5.6974524974823,
      "epoch": 5.971970246734398,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020981794036322666,
      "loss": 4.8395,
      "mean_token_accuracy": 0.2265813559293747,
      "num_tokens": 136626064.0,
      "step": 65835
    },
    {
      "entropy": 5.681202030181884,
      "epoch": 5.972423802612482,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002097869721225036,
      "loss": 4.8188,
      "mean_token_accuracy": 0.23605132550001146,
      "num_tokens": 136636326.0,
      "step": 65840
    },
    {
      "entropy": 5.674501609802246,
      "epoch": 5.972877358490566,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020975600523040272,
      "loss": 4.8506,
      "mean_token_accuracy": 0.23519093543291092,
      "num_tokens": 136646675.0,
      "step": 65845
    },
    {
      "entropy": 5.6578044414520265,
      "epoch": 5.97333091436865,
      "grad_norm": 1.375,
      "learning_rate": 0.00020972503968756418,
      "loss": 4.7753,
      "mean_token_accuracy": 0.23498208224773406,
      "num_tokens": 136657041.0,
      "step": 65850
    },
    {
      "entropy": 5.709125232696533,
      "epoch": 5.973784470246734,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020969407549462855,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23026481419801711,
      "num_tokens": 136666156.0,
      "step": 65855
    },
    {
      "entropy": 5.672545909881592,
      "epoch": 5.9742380261248185,
      "grad_norm": 1.375,
      "learning_rate": 0.00020966311265223597,
      "loss": 4.7749,
      "mean_token_accuracy": 0.23801990896463393,
      "num_tokens": 136675870.0,
      "step": 65860
    },
    {
      "entropy": 5.640111017227173,
      "epoch": 5.974691582002903,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002096321511610269,
      "loss": 4.7871,
      "mean_token_accuracy": 0.23574888855218887,
      "num_tokens": 136686635.0,
      "step": 65865
    },
    {
      "entropy": 5.804391002655029,
      "epoch": 5.975145137880987,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020960119102164166,
      "loss": 5.0228,
      "mean_token_accuracy": 0.2174474850296974,
      "num_tokens": 136698365.0,
      "step": 65870
    },
    {
      "entropy": 5.7050464153289795,
      "epoch": 5.975598693759071,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020957023223472037,
      "loss": 4.8509,
      "mean_token_accuracy": 0.22432173043489456,
      "num_tokens": 136709629.0,
      "step": 65875
    },
    {
      "entropy": 5.79775652885437,
      "epoch": 5.9760522496371555,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020953927480090346,
      "loss": 4.9185,
      "mean_token_accuracy": 0.22257318645715712,
      "num_tokens": 136718996.0,
      "step": 65880
    },
    {
      "entropy": 5.705549097061157,
      "epoch": 5.97650580551524,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000209508318720831,
      "loss": 4.8563,
      "mean_token_accuracy": 0.232795786857605,
      "num_tokens": 136729999.0,
      "step": 65885
    },
    {
      "entropy": 5.728742980957032,
      "epoch": 5.976959361393323,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020947736399514318,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23408622443675994,
      "num_tokens": 136739809.0,
      "step": 65890
    },
    {
      "entropy": 5.703338766098023,
      "epoch": 5.977412917271408,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020944641062448004,
      "loss": 4.8164,
      "mean_token_accuracy": 0.2311257764697075,
      "num_tokens": 136750423.0,
      "step": 65895
    },
    {
      "entropy": 5.69181227684021,
      "epoch": 5.977866473149492,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020941545860948203,
      "loss": 4.8263,
      "mean_token_accuracy": 0.22766685336828232,
      "num_tokens": 136761907.0,
      "step": 65900
    },
    {
      "entropy": 5.787675237655639,
      "epoch": 5.978320029027576,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000209384507950789,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2172337532043457,
      "num_tokens": 136772437.0,
      "step": 65905
    },
    {
      "entropy": 5.776616621017456,
      "epoch": 5.97877358490566,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000209353558649041,
      "loss": 4.8664,
      "mean_token_accuracy": 0.23476210087537766,
      "num_tokens": 136782768.0,
      "step": 65910
    },
    {
      "entropy": 5.739399003982544,
      "epoch": 5.9792271407837445,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020932261070487817,
      "loss": 4.8401,
      "mean_token_accuracy": 0.2343429371714592,
      "num_tokens": 136792339.0,
      "step": 65915
    },
    {
      "entropy": 5.660113906860351,
      "epoch": 5.979680696661829,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020929166411894046,
      "loss": 4.7731,
      "mean_token_accuracy": 0.2379525274038315,
      "num_tokens": 136803242.0,
      "step": 65920
    },
    {
      "entropy": 5.767369985580444,
      "epoch": 5.980134252539913,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020926071889186787,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22634360939264297,
      "num_tokens": 136812934.0,
      "step": 65925
    },
    {
      "entropy": 5.70105938911438,
      "epoch": 5.980587808417997,
      "grad_norm": 1.5,
      "learning_rate": 0.00020922977502430047,
      "loss": 4.7922,
      "mean_token_accuracy": 0.24266529828310013,
      "num_tokens": 136822749.0,
      "step": 65930
    },
    {
      "entropy": 5.574490070343018,
      "epoch": 5.981041364296082,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020919883251687804,
      "loss": 4.6729,
      "mean_token_accuracy": 0.2407844826579094,
      "num_tokens": 136832573.0,
      "step": 65935
    },
    {
      "entropy": 5.620280694961548,
      "epoch": 5.981494920174166,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020916789137024052,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23487131297588348,
      "num_tokens": 136842372.0,
      "step": 65940
    },
    {
      "entropy": 5.663778638839721,
      "epoch": 5.981948476052249,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020913695158502772,
      "loss": 4.7903,
      "mean_token_accuracy": 0.2353670120239258,
      "num_tokens": 136853626.0,
      "step": 65945
    },
    {
      "entropy": 5.736944675445557,
      "epoch": 5.9824020319303335,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020910601316187955,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22104260325431824,
      "num_tokens": 136864281.0,
      "step": 65950
    },
    {
      "entropy": 5.697096490859986,
      "epoch": 5.982855587808418,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002090750761014359,
      "loss": 4.8402,
      "mean_token_accuracy": 0.2292373239994049,
      "num_tokens": 136874451.0,
      "step": 65955
    },
    {
      "entropy": 5.701625919342041,
      "epoch": 5.983309143686502,
      "grad_norm": 1.5,
      "learning_rate": 0.0002090441404043364,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22053072303533555,
      "num_tokens": 136884125.0,
      "step": 65960
    },
    {
      "entropy": 5.688125848770142,
      "epoch": 5.983762699564586,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020901320607122093,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2302434340119362,
      "num_tokens": 136894278.0,
      "step": 65965
    },
    {
      "entropy": 5.696059989929199,
      "epoch": 5.9842162554426706,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020898227310272915,
      "loss": 4.8414,
      "mean_token_accuracy": 0.22879777252674102,
      "num_tokens": 136904212.0,
      "step": 65970
    },
    {
      "entropy": 5.687975358963013,
      "epoch": 5.984669811320755,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020895134149950085,
      "loss": 4.7593,
      "mean_token_accuracy": 0.2395613521337509,
      "num_tokens": 136914591.0,
      "step": 65975
    },
    {
      "entropy": 5.6619553565979,
      "epoch": 5.985123367198839,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020892041126217555,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23717708736658097,
      "num_tokens": 136925271.0,
      "step": 65980
    },
    {
      "entropy": 5.651717805862427,
      "epoch": 5.985576923076923,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000208889482391393,
      "loss": 4.8175,
      "mean_token_accuracy": 0.2331619530916214,
      "num_tokens": 136936575.0,
      "step": 65985
    },
    {
      "entropy": 5.700466060638428,
      "epoch": 5.986030478955008,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020885855488779288,
      "loss": 4.8587,
      "mean_token_accuracy": 0.23225611448287964,
      "num_tokens": 136946488.0,
      "step": 65990
    },
    {
      "entropy": 5.6811669826507565,
      "epoch": 5.986484034833092,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020882762875201456,
      "loss": 4.876,
      "mean_token_accuracy": 0.2313245564699173,
      "num_tokens": 136956592.0,
      "step": 65995
    },
    {
      "entropy": 5.774746751785278,
      "epoch": 5.986937590711175,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020879670398469785,
      "loss": 4.9047,
      "mean_token_accuracy": 0.21954434514045715,
      "num_tokens": 136966878.0,
      "step": 66000
    },
    {
      "epoch": 5.986937590711175,
      "eval_entropy": 5.415217533046843,
      "eval_loss": 4.935329914093018,
      "eval_mean_token_accuracy": 0.2348741578165762,
      "eval_num_tokens": 136966878.0,
      "eval_runtime": 20.6546,
      "eval_samples_per_second": 1711.966,
      "eval_steps_per_second": 213.996,
      "step": 66000
    },
    {
      "entropy": 5.7120434761047365,
      "epoch": 5.9873911465892595,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020876578058648217,
      "loss": 4.8871,
      "mean_token_accuracy": 0.22430450916290284,
      "num_tokens": 136977626.0,
      "step": 66005
    },
    {
      "entropy": 5.633639144897461,
      "epoch": 5.987844702467344,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020873485855800694,
      "loss": 4.8227,
      "mean_token_accuracy": 0.23515801429748534,
      "num_tokens": 136987584.0,
      "step": 66010
    },
    {
      "entropy": 5.6864964962005615,
      "epoch": 5.988298258345428,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002087039378999118,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22297294139862062,
      "num_tokens": 136996556.0,
      "step": 66015
    },
    {
      "entropy": 5.723286533355713,
      "epoch": 5.988751814223512,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002086730186128361,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22778346836566926,
      "num_tokens": 137006381.0,
      "step": 66020
    },
    {
      "entropy": 5.705532693862915,
      "epoch": 5.989205370101597,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002086421006974193,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2308787301182747,
      "num_tokens": 137016575.0,
      "step": 66025
    },
    {
      "entropy": 5.769444274902344,
      "epoch": 5.989658925979681,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020861118415430068,
      "loss": 4.923,
      "mean_token_accuracy": 0.22601603120565414,
      "num_tokens": 137026825.0,
      "step": 66030
    },
    {
      "entropy": 5.736752080917358,
      "epoch": 5.990112481857765,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002085802689841198,
      "loss": 4.8516,
      "mean_token_accuracy": 0.23132312893867493,
      "num_tokens": 137037064.0,
      "step": 66035
    },
    {
      "entropy": 5.665262603759766,
      "epoch": 5.990566037735849,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020854935518751583,
      "loss": 4.8253,
      "mean_token_accuracy": 0.23468652963638306,
      "num_tokens": 137048114.0,
      "step": 66040
    },
    {
      "entropy": 5.786039066314697,
      "epoch": 5.991019593613933,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020851844276512815,
      "loss": 4.986,
      "mean_token_accuracy": 0.21647345572710036,
      "num_tokens": 137057942.0,
      "step": 66045
    },
    {
      "entropy": 5.713873291015625,
      "epoch": 5.991473149492017,
      "grad_norm": 1.40625,
      "learning_rate": 0.000208487531717596,
      "loss": 4.7629,
      "mean_token_accuracy": 0.23971940726041793,
      "num_tokens": 137067422.0,
      "step": 66050
    },
    {
      "entropy": 5.708318281173706,
      "epoch": 5.991926705370101,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020845662204555866,
      "loss": 4.8516,
      "mean_token_accuracy": 0.23010155111551284,
      "num_tokens": 137077613.0,
      "step": 66055
    },
    {
      "entropy": 5.687849617004394,
      "epoch": 5.992380261248186,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002084257137496553,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2299363926053047,
      "num_tokens": 137086934.0,
      "step": 66060
    },
    {
      "entropy": 5.68806004524231,
      "epoch": 5.99283381712627,
      "grad_norm": 1.375,
      "learning_rate": 0.00020839480683052514,
      "loss": 4.8024,
      "mean_token_accuracy": 0.2436286523938179,
      "num_tokens": 137097548.0,
      "step": 66065
    },
    {
      "entropy": 5.6854085445404055,
      "epoch": 5.993287373004354,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020836390128880738,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22560672610998153,
      "num_tokens": 137107458.0,
      "step": 66070
    },
    {
      "entropy": 5.673308324813843,
      "epoch": 5.993740928882438,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020833299712514115,
      "loss": 4.7746,
      "mean_token_accuracy": 0.2296238899230957,
      "num_tokens": 137117666.0,
      "step": 66075
    },
    {
      "entropy": 5.736688899993896,
      "epoch": 5.994194484760523,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020830209434016545,
      "loss": 4.8715,
      "mean_token_accuracy": 0.21756862103939056,
      "num_tokens": 137127599.0,
      "step": 66080
    },
    {
      "entropy": 5.682034158706665,
      "epoch": 5.994648040638607,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020827119293451946,
      "loss": 4.8594,
      "mean_token_accuracy": 0.23564452230930327,
      "num_tokens": 137137925.0,
      "step": 66085
    },
    {
      "entropy": 5.607802534103394,
      "epoch": 5.995101596516691,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020824029290884223,
      "loss": 4.7094,
      "mean_token_accuracy": 0.2466371089220047,
      "num_tokens": 137149054.0,
      "step": 66090
    },
    {
      "entropy": 5.686489868164062,
      "epoch": 5.995555152394775,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020820939426377278,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22752151489257813,
      "num_tokens": 137159503.0,
      "step": 66095
    },
    {
      "entropy": 5.703406810760498,
      "epoch": 5.996008708272859,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020817849699995,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22722985595464706,
      "num_tokens": 137170140.0,
      "step": 66100
    },
    {
      "entropy": 5.651288890838623,
      "epoch": 5.996462264150943,
      "grad_norm": 1.328125,
      "learning_rate": 0.000208147601118013,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23402173221111297,
      "num_tokens": 137180591.0,
      "step": 66105
    },
    {
      "entropy": 5.6975480079650875,
      "epoch": 5.996915820029027,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002081167066186006,
      "loss": 4.8205,
      "mean_token_accuracy": 0.23163412660360336,
      "num_tokens": 137191928.0,
      "step": 66110
    },
    {
      "entropy": 5.756398630142212,
      "epoch": 5.997369375907112,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002080858135023518,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22371741831302644,
      "num_tokens": 137200909.0,
      "step": 66115
    },
    {
      "entropy": 5.707106447219848,
      "epoch": 5.997822931785196,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020805492176990538,
      "loss": 4.8523,
      "mean_token_accuracy": 0.227319398522377,
      "num_tokens": 137212277.0,
      "step": 66120
    },
    {
      "entropy": 5.726680088043213,
      "epoch": 5.99827648766328,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002080240314219003,
      "loss": 4.847,
      "mean_token_accuracy": 0.2264995977282524,
      "num_tokens": 137222761.0,
      "step": 66125
    },
    {
      "entropy": 5.7206401348114015,
      "epoch": 5.998730043541364,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002079931424589752,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23156281262636186,
      "num_tokens": 137233686.0,
      "step": 66130
    },
    {
      "entropy": 5.665046310424804,
      "epoch": 5.999183599419449,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020796225488176908,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23079515248537064,
      "num_tokens": 137244122.0,
      "step": 66135
    },
    {
      "entropy": 5.6283371925354,
      "epoch": 5.999637155297533,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002079313686909206,
      "loss": 4.7203,
      "mean_token_accuracy": 0.24085850417613983,
      "num_tokens": 137255089.0,
      "step": 66140
    },
    {
      "entropy": 5.744688606262207,
      "epoch": 6.000090711175617,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002079004838870685,
      "loss": 4.8686,
      "mean_token_accuracy": 0.23014982044696808,
      "num_tokens": 137266052.0,
      "step": 66145
    },
    {
      "entropy": 5.690227031707764,
      "epoch": 6.000544267053701,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020786960047085147,
      "loss": 4.8061,
      "mean_token_accuracy": 0.22908749282360077,
      "num_tokens": 137276761.0,
      "step": 66150
    },
    {
      "entropy": 5.688071393966675,
      "epoch": 6.000997822931785,
      "grad_norm": 1.375,
      "learning_rate": 0.00020783871844290826,
      "loss": 4.7967,
      "mean_token_accuracy": 0.24154350012540818,
      "num_tokens": 137286780.0,
      "step": 66155
    },
    {
      "entropy": 5.735907936096192,
      "epoch": 6.001451378809869,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020780783780387747,
      "loss": 4.8848,
      "mean_token_accuracy": 0.23042248338460922,
      "num_tokens": 137297211.0,
      "step": 66160
    },
    {
      "entropy": 5.6744283676147464,
      "epoch": 6.001904934687953,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002077769585543976,
      "loss": 4.8658,
      "mean_token_accuracy": 0.22859189808368682,
      "num_tokens": 137307295.0,
      "step": 66165
    },
    {
      "entropy": 5.7036608219146725,
      "epoch": 6.002358490566038,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020774608069510747,
      "loss": 4.7754,
      "mean_token_accuracy": 0.23708581179380417,
      "num_tokens": 137316677.0,
      "step": 66170
    },
    {
      "entropy": 5.699928045272827,
      "epoch": 6.002812046444122,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020771520422664552,
      "loss": 4.7613,
      "mean_token_accuracy": 0.24313358068466187,
      "num_tokens": 137326254.0,
      "step": 66175
    },
    {
      "entropy": 5.715155601501465,
      "epoch": 6.003265602322206,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002076843291496503,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2271281898021698,
      "num_tokens": 137336974.0,
      "step": 66180
    },
    {
      "entropy": 5.599990129470825,
      "epoch": 6.0037191582002905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020765345546476023,
      "loss": 4.7349,
      "mean_token_accuracy": 0.24299487471580505,
      "num_tokens": 137347170.0,
      "step": 66185
    },
    {
      "entropy": 5.685761594772339,
      "epoch": 6.004172714078375,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020762258317261396,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23638418316841125,
      "num_tokens": 137356908.0,
      "step": 66190
    },
    {
      "entropy": 5.633751010894775,
      "epoch": 6.004626269956459,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020759171227384982,
      "loss": 4.725,
      "mean_token_accuracy": 0.233607742190361,
      "num_tokens": 137367279.0,
      "step": 66195
    },
    {
      "entropy": 5.611336183547974,
      "epoch": 6.005079825834543,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020756084276910613,
      "loss": 4.7633,
      "mean_token_accuracy": 0.23796157240867616,
      "num_tokens": 137378296.0,
      "step": 66200
    },
    {
      "entropy": 5.638360023498535,
      "epoch": 6.005533381712627,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002075299746590215,
      "loss": 4.7986,
      "mean_token_accuracy": 0.24045598208904267,
      "num_tokens": 137389290.0,
      "step": 66205
    },
    {
      "entropy": 5.636170387268066,
      "epoch": 6.005986937590711,
      "grad_norm": 1.25,
      "learning_rate": 0.0002074991079442342,
      "loss": 4.7405,
      "mean_token_accuracy": 0.24274501353502273,
      "num_tokens": 137399803.0,
      "step": 66210
    },
    {
      "entropy": 5.730358123779297,
      "epoch": 6.006440493468795,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020746824262538255,
      "loss": 4.8234,
      "mean_token_accuracy": 0.23537158966064453,
      "num_tokens": 137410467.0,
      "step": 66215
    },
    {
      "entropy": 5.63616099357605,
      "epoch": 6.006894049346879,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002074373787031048,
      "loss": 4.7318,
      "mean_token_accuracy": 0.2408212959766388,
      "num_tokens": 137420830.0,
      "step": 66220
    },
    {
      "entropy": 5.650511932373047,
      "epoch": 6.007347605224964,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002074065161780393,
      "loss": 4.7846,
      "mean_token_accuracy": 0.2372092992067337,
      "num_tokens": 137431052.0,
      "step": 66225
    },
    {
      "entropy": 5.668917798995972,
      "epoch": 6.007801161103048,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020737565505082433,
      "loss": 4.8527,
      "mean_token_accuracy": 0.2301172137260437,
      "num_tokens": 137440270.0,
      "step": 66230
    },
    {
      "entropy": 5.682622241973877,
      "epoch": 6.008254716981132,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020734479532209793,
      "loss": 4.7748,
      "mean_token_accuracy": 0.2348934143781662,
      "num_tokens": 137450276.0,
      "step": 66235
    },
    {
      "entropy": 5.729228448867798,
      "epoch": 6.0087082728592165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020731393699249844,
      "loss": 4.7909,
      "mean_token_accuracy": 0.2422514259815216,
      "num_tokens": 137461202.0,
      "step": 66240
    },
    {
      "entropy": 5.717363405227661,
      "epoch": 6.009161828737301,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002072830800626641,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2472658485174179,
      "num_tokens": 137471202.0,
      "step": 66245
    },
    {
      "entropy": 5.743350172042847,
      "epoch": 6.009615384615385,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020725222453323283,
      "loss": 4.9078,
      "mean_token_accuracy": 0.22270871698856354,
      "num_tokens": 137481373.0,
      "step": 66250
    },
    {
      "entropy": 5.680462169647217,
      "epoch": 6.010068940493468,
      "grad_norm": 1.375,
      "learning_rate": 0.00020722137040484284,
      "loss": 4.8662,
      "mean_token_accuracy": 0.2312432497739792,
      "num_tokens": 137492045.0,
      "step": 66255
    },
    {
      "entropy": 5.6881390571594235,
      "epoch": 6.010522496371553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020719051767813213,
      "loss": 4.7954,
      "mean_token_accuracy": 0.23511325865983962,
      "num_tokens": 137501910.0,
      "step": 66260
    },
    {
      "entropy": 5.744770526885986,
      "epoch": 6.010976052249637,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020715966635373884,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2300735041499138,
      "num_tokens": 137512722.0,
      "step": 66265
    },
    {
      "entropy": 5.755110740661621,
      "epoch": 6.011429608127721,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020712881643230092,
      "loss": 4.8419,
      "mean_token_accuracy": 0.2446837231516838,
      "num_tokens": 137524245.0,
      "step": 66270
    },
    {
      "entropy": 5.74254584312439,
      "epoch": 6.0118831640058055,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002070979679144564,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22609800696372986,
      "num_tokens": 137533977.0,
      "step": 66275
    },
    {
      "entropy": 5.776634454727173,
      "epoch": 6.01233671988389,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020706712080084317,
      "loss": 4.8871,
      "mean_token_accuracy": 0.22914343029260636,
      "num_tokens": 137545801.0,
      "step": 66280
    },
    {
      "entropy": 5.7342822551727295,
      "epoch": 6.012790275761974,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020703627509209915,
      "loss": 4.87,
      "mean_token_accuracy": 0.22705447524785996,
      "num_tokens": 137556109.0,
      "step": 66285
    },
    {
      "entropy": 5.65423698425293,
      "epoch": 6.013243831640058,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020700543078886241,
      "loss": 4.721,
      "mean_token_accuracy": 0.24472247660160065,
      "num_tokens": 137568060.0,
      "step": 66290
    },
    {
      "entropy": 5.697555065155029,
      "epoch": 6.0136973875181425,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002069745878917706,
      "loss": 4.8,
      "mean_token_accuracy": 0.23531196117401124,
      "num_tokens": 137579167.0,
      "step": 66295
    },
    {
      "entropy": 5.711184549331665,
      "epoch": 6.014150943396227,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020694374640146162,
      "loss": 4.8077,
      "mean_token_accuracy": 0.2326766476035118,
      "num_tokens": 137589050.0,
      "step": 66300
    },
    {
      "entropy": 5.694968318939209,
      "epoch": 6.01460449927431,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020691290631857335,
      "loss": 4.8139,
      "mean_token_accuracy": 0.24122208952903748,
      "num_tokens": 137599496.0,
      "step": 66305
    },
    {
      "entropy": 5.693858194351196,
      "epoch": 6.0150580551523944,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002068820676437435,
      "loss": 4.8241,
      "mean_token_accuracy": 0.23364632576704025,
      "num_tokens": 137609272.0,
      "step": 66310
    },
    {
      "entropy": 5.650151491165161,
      "epoch": 6.015511611030479,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002068512303776099,
      "loss": 4.7348,
      "mean_token_accuracy": 0.2337775081396103,
      "num_tokens": 137619541.0,
      "step": 66315
    },
    {
      "entropy": 5.71425609588623,
      "epoch": 6.015965166908563,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002068203945208102,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22920305877923966,
      "num_tokens": 137629894.0,
      "step": 66320
    },
    {
      "entropy": 5.825317811965943,
      "epoch": 6.016418722786647,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020678956007398215,
      "loss": 4.9495,
      "mean_token_accuracy": 0.2230623722076416,
      "num_tokens": 137639892.0,
      "step": 66325
    },
    {
      "entropy": 5.7066504001617435,
      "epoch": 6.0168722786647315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020675872703776327,
      "loss": 4.8182,
      "mean_token_accuracy": 0.2280263841152191,
      "num_tokens": 137650166.0,
      "step": 66330
    },
    {
      "entropy": 5.580956029891968,
      "epoch": 6.017325834542816,
      "grad_norm": 1.5,
      "learning_rate": 0.0002067278954127914,
      "loss": 4.7003,
      "mean_token_accuracy": 0.2334149405360222,
      "num_tokens": 137659676.0,
      "step": 66335
    },
    {
      "entropy": 5.644603443145752,
      "epoch": 6.0177793904209,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020669706519970405,
      "loss": 4.8458,
      "mean_token_accuracy": 0.2241022616624832,
      "num_tokens": 137670257.0,
      "step": 66340
    },
    {
      "entropy": 5.649445390701294,
      "epoch": 6.018232946298984,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020666623639913873,
      "loss": 4.782,
      "mean_token_accuracy": 0.2346022456884384,
      "num_tokens": 137680797.0,
      "step": 66345
    },
    {
      "entropy": 5.732213687896729,
      "epoch": 6.018686502177069,
      "grad_norm": 1.375,
      "learning_rate": 0.0002066354090117331,
      "loss": 4.8898,
      "mean_token_accuracy": 0.22817437201738358,
      "num_tokens": 137692199.0,
      "step": 66350
    },
    {
      "entropy": 5.715999269485474,
      "epoch": 6.019140058055152,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020660458303812469,
      "loss": 4.875,
      "mean_token_accuracy": 0.21924983710050583,
      "num_tokens": 137703485.0,
      "step": 66355
    },
    {
      "entropy": 5.712945461273193,
      "epoch": 6.019593613933236,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020657375847895089,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2335523322224617,
      "num_tokens": 137714940.0,
      "step": 66360
    },
    {
      "entropy": 5.699818897247314,
      "epoch": 6.0200471698113205,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020654293533484914,
      "loss": 4.8386,
      "mean_token_accuracy": 0.22639009058475495,
      "num_tokens": 137724389.0,
      "step": 66365
    },
    {
      "entropy": 5.652038097381592,
      "epoch": 6.020500725689405,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020651211360645693,
      "loss": 4.7029,
      "mean_token_accuracy": 0.2493144914507866,
      "num_tokens": 137734360.0,
      "step": 66370
    },
    {
      "entropy": 5.76816840171814,
      "epoch": 6.020954281567489,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002064812932944118,
      "loss": 4.906,
      "mean_token_accuracy": 0.2304202139377594,
      "num_tokens": 137745479.0,
      "step": 66375
    },
    {
      "entropy": 5.678042602539063,
      "epoch": 6.021407837445573,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020645047439935083,
      "loss": 4.7343,
      "mean_token_accuracy": 0.23495284020900725,
      "num_tokens": 137755162.0,
      "step": 66380
    },
    {
      "entropy": 5.664081192016601,
      "epoch": 6.0218613933236576,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020641965692191157,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2420924946665764,
      "num_tokens": 137766090.0,
      "step": 66385
    },
    {
      "entropy": 5.681673383712768,
      "epoch": 6.022314949201742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020638884086273136,
      "loss": 4.8127,
      "mean_token_accuracy": 0.23515650779008865,
      "num_tokens": 137775526.0,
      "step": 66390
    },
    {
      "entropy": 5.7674051284790036,
      "epoch": 6.022768505079826,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020635802622244733,
      "loss": 4.9531,
      "mean_token_accuracy": 0.22328424453735352,
      "num_tokens": 137786005.0,
      "step": 66395
    },
    {
      "entropy": 5.713361120223999,
      "epoch": 6.02322206095791,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020632721300169676,
      "loss": 4.7887,
      "mean_token_accuracy": 0.22990315705537795,
      "num_tokens": 137796609.0,
      "step": 66400
    },
    {
      "entropy": 5.725145244598389,
      "epoch": 6.023675616835995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020629640120111702,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22440956979990007,
      "num_tokens": 137806698.0,
      "step": 66405
    },
    {
      "entropy": 5.736941957473755,
      "epoch": 6.024129172714078,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002062655908213452,
      "loss": 4.888,
      "mean_token_accuracy": 0.22774115651845933,
      "num_tokens": 137816457.0,
      "step": 66410
    },
    {
      "entropy": 5.698803091049195,
      "epoch": 6.024582728592162,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002062347818630184,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23287934213876724,
      "num_tokens": 137826191.0,
      "step": 66415
    },
    {
      "entropy": 5.731973266601562,
      "epoch": 6.0250362844702465,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002062039743267739,
      "loss": 4.8566,
      "mean_token_accuracy": 0.23113005459308625,
      "num_tokens": 137837713.0,
      "step": 66420
    },
    {
      "entropy": 5.7274738311767575,
      "epoch": 6.025489840348331,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020617316821324878,
      "loss": 4.7639,
      "mean_token_accuracy": 0.24735352545976638,
      "num_tokens": 137847960.0,
      "step": 66425
    },
    {
      "entropy": 5.754007244110108,
      "epoch": 6.025943396226415,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020614236352307997,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2258964642882347,
      "num_tokens": 137858931.0,
      "step": 66430
    },
    {
      "entropy": 5.640301704406738,
      "epoch": 6.026396952104499,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002061115602569047,
      "loss": 4.73,
      "mean_token_accuracy": 0.23930452913045883,
      "num_tokens": 137869402.0,
      "step": 66435
    },
    {
      "entropy": 5.683822679519653,
      "epoch": 6.026850507982584,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020608075841535995,
      "loss": 4.7892,
      "mean_token_accuracy": 0.2386031061410904,
      "num_tokens": 137879908.0,
      "step": 66440
    },
    {
      "entropy": 5.6613483905792235,
      "epoch": 6.027304063860668,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020604995799908266,
      "loss": 4.7721,
      "mean_token_accuracy": 0.23461044281721116,
      "num_tokens": 137888993.0,
      "step": 66445
    },
    {
      "entropy": 5.658319425582886,
      "epoch": 6.027757619738752,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020601915900870976,
      "loss": 4.8198,
      "mean_token_accuracy": 0.23244257420301437,
      "num_tokens": 137899560.0,
      "step": 66450
    },
    {
      "entropy": 5.665250396728515,
      "epoch": 6.028211175616836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020598836144487825,
      "loss": 4.7823,
      "mean_token_accuracy": 0.23749201744794846,
      "num_tokens": 137909988.0,
      "step": 66455
    },
    {
      "entropy": 5.7098174571990965,
      "epoch": 6.02866473149492,
      "grad_norm": 1.375,
      "learning_rate": 0.000205957565308225,
      "loss": 4.7945,
      "mean_token_accuracy": 0.2375682845711708,
      "num_tokens": 137919458.0,
      "step": 66460
    },
    {
      "entropy": 5.657073926925659,
      "epoch": 6.029118287373004,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020592677059938687,
      "loss": 4.7701,
      "mean_token_accuracy": 0.23576471954584122,
      "num_tokens": 137929489.0,
      "step": 66465
    },
    {
      "entropy": 5.685831737518311,
      "epoch": 6.029571843251088,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020589597731900079,
      "loss": 4.7935,
      "mean_token_accuracy": 0.23368506729602814,
      "num_tokens": 137939775.0,
      "step": 66470
    },
    {
      "entropy": 5.657836103439331,
      "epoch": 6.030025399129173,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020586518546770354,
      "loss": 4.7894,
      "mean_token_accuracy": 0.23525859415531158,
      "num_tokens": 137950580.0,
      "step": 66475
    },
    {
      "entropy": 5.680651521682739,
      "epoch": 6.030478955007257,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020583439504613183,
      "loss": 4.8099,
      "mean_token_accuracy": 0.23265518844127656,
      "num_tokens": 137960836.0,
      "step": 66480
    },
    {
      "entropy": 5.715020799636841,
      "epoch": 6.030932510885341,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020580360605492237,
      "loss": 4.8541,
      "mean_token_accuracy": 0.2348325029015541,
      "num_tokens": 137972195.0,
      "step": 66485
    },
    {
      "entropy": 5.651127624511719,
      "epoch": 6.031386066763425,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020577281849471206,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23096729964017867,
      "num_tokens": 137982314.0,
      "step": 66490
    },
    {
      "entropy": 5.624353122711182,
      "epoch": 6.03183962264151,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020574203236613754,
      "loss": 4.7572,
      "mean_token_accuracy": 0.229055055975914,
      "num_tokens": 137992865.0,
      "step": 66495
    },
    {
      "entropy": 5.800323343276977,
      "epoch": 6.032293178519594,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020571124766983535,
      "loss": 4.9012,
      "mean_token_accuracy": 0.22548136860132217,
      "num_tokens": 138003048.0,
      "step": 66500
    },
    {
      "entropy": 5.734719276428223,
      "epoch": 6.032746734397678,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020568046440644228,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22816185355186464,
      "num_tokens": 138013341.0,
      "step": 66505
    },
    {
      "entropy": 5.71274356842041,
      "epoch": 6.0332002902757615,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020564968257659482,
      "loss": 4.8095,
      "mean_token_accuracy": 0.2285443902015686,
      "num_tokens": 138023623.0,
      "step": 66510
    },
    {
      "entropy": 5.721702241897583,
      "epoch": 6.033653846153846,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002056189021809296,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2246854767203331,
      "num_tokens": 138035556.0,
      "step": 66515
    },
    {
      "entropy": 5.663994693756104,
      "epoch": 6.03410740203193,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020558812322008318,
      "loss": 4.7819,
      "mean_token_accuracy": 0.23936557322740554,
      "num_tokens": 138045849.0,
      "step": 66520
    },
    {
      "entropy": 5.682425594329834,
      "epoch": 6.034560957910014,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002055573456946921,
      "loss": 4.8101,
      "mean_token_accuracy": 0.233328241109848,
      "num_tokens": 138057033.0,
      "step": 66525
    },
    {
      "entropy": 5.679525089263916,
      "epoch": 6.035014513788099,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020552656960539278,
      "loss": 4.803,
      "mean_token_accuracy": 0.2333708867430687,
      "num_tokens": 138068638.0,
      "step": 66530
    },
    {
      "entropy": 5.707609510421753,
      "epoch": 6.035468069666183,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020549579495282168,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2303635522723198,
      "num_tokens": 138079638.0,
      "step": 66535
    },
    {
      "entropy": 5.750111865997314,
      "epoch": 6.035921625544267,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002054650217376153,
      "loss": 4.909,
      "mean_token_accuracy": 0.23269391655921937,
      "num_tokens": 138089753.0,
      "step": 66540
    },
    {
      "entropy": 5.766170263290405,
      "epoch": 6.036375181422351,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020543424996040993,
      "loss": 4.8955,
      "mean_token_accuracy": 0.23189998865127565,
      "num_tokens": 138099829.0,
      "step": 66545
    },
    {
      "entropy": 5.705867910385132,
      "epoch": 6.036828737300436,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020540347962184202,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23691103011369705,
      "num_tokens": 138110722.0,
      "step": 66550
    },
    {
      "entropy": 5.665182495117188,
      "epoch": 6.03728229317852,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020537271072254792,
      "loss": 4.8114,
      "mean_token_accuracy": 0.23336922228336335,
      "num_tokens": 138120753.0,
      "step": 66555
    },
    {
      "entropy": 5.643378877639771,
      "epoch": 6.037735849056604,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020534194326316386,
      "loss": 4.7945,
      "mean_token_accuracy": 0.23034674376249314,
      "num_tokens": 138131372.0,
      "step": 66560
    },
    {
      "entropy": 5.737073230743408,
      "epoch": 6.038189404934688,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020531117724432624,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22738378643989562,
      "num_tokens": 138141057.0,
      "step": 66565
    },
    {
      "entropy": 5.659091567993164,
      "epoch": 6.038642960812772,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020528041266667114,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2443648800253868,
      "num_tokens": 138150638.0,
      "step": 66570
    },
    {
      "entropy": 5.712225151062012,
      "epoch": 6.039096516690856,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020524964953083492,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23754813373088837,
      "num_tokens": 138160924.0,
      "step": 66575
    },
    {
      "entropy": 5.679871082305908,
      "epoch": 6.03955007256894,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002052188878374537,
      "loss": 4.8062,
      "mean_token_accuracy": 0.23660861998796462,
      "num_tokens": 138171865.0,
      "step": 66580
    },
    {
      "entropy": 5.623663759231567,
      "epoch": 6.040003628447025,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020518812758716365,
      "loss": 4.7577,
      "mean_token_accuracy": 0.2387099415063858,
      "num_tokens": 138181521.0,
      "step": 66585
    },
    {
      "entropy": 5.739480066299438,
      "epoch": 6.040457184325109,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020515736878060092,
      "loss": 4.9314,
      "mean_token_accuracy": 0.22605279386043547,
      "num_tokens": 138192265.0,
      "step": 66590
    },
    {
      "entropy": 5.674508857727051,
      "epoch": 6.040910740203193,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020512661141840167,
      "loss": 4.787,
      "mean_token_accuracy": 0.23600708991289138,
      "num_tokens": 138202912.0,
      "step": 66595
    },
    {
      "entropy": 5.718644142150879,
      "epoch": 6.0413642960812775,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020509585550120186,
      "loss": 4.8705,
      "mean_token_accuracy": 0.23043214082717894,
      "num_tokens": 138213376.0,
      "step": 66600
    },
    {
      "entropy": 5.718833494186401,
      "epoch": 6.041817851959362,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002050651010296375,
      "loss": 4.8141,
      "mean_token_accuracy": 0.24058228582143784,
      "num_tokens": 138224190.0,
      "step": 66605
    },
    {
      "entropy": 5.719875335693359,
      "epoch": 6.042271407837446,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020503434800434479,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2368616208434105,
      "num_tokens": 138235112.0,
      "step": 66610
    },
    {
      "entropy": 5.695919561386108,
      "epoch": 6.042724963715529,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002050035964259595,
      "loss": 4.8404,
      "mean_token_accuracy": 0.2282659590244293,
      "num_tokens": 138246534.0,
      "step": 66615
    },
    {
      "entropy": 5.640859603881836,
      "epoch": 6.043178519593614,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020497284629511769,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2299182504415512,
      "num_tokens": 138257478.0,
      "step": 66620
    },
    {
      "entropy": 5.655445957183838,
      "epoch": 6.043632075471698,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002049420976124553,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23504508286714554,
      "num_tokens": 138268099.0,
      "step": 66625
    },
    {
      "entropy": 5.754044675827027,
      "epoch": 6.044085631349782,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020491135037860818,
      "loss": 4.8613,
      "mean_token_accuracy": 0.23880242705345153,
      "num_tokens": 138278982.0,
      "step": 66630
    },
    {
      "entropy": 5.740449619293213,
      "epoch": 6.044539187227866,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020488060459421215,
      "loss": 4.8627,
      "mean_token_accuracy": 0.23334280550479888,
      "num_tokens": 138289078.0,
      "step": 66635
    },
    {
      "entropy": 5.674741649627686,
      "epoch": 6.044992743105951,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020484986025990317,
      "loss": 4.7484,
      "mean_token_accuracy": 0.23832366913557052,
      "num_tokens": 138299177.0,
      "step": 66640
    },
    {
      "entropy": 5.746947479248047,
      "epoch": 6.045446298984035,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020481911737631686,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22529525756835939,
      "num_tokens": 138309576.0,
      "step": 66645
    },
    {
      "entropy": 5.711149454116821,
      "epoch": 6.045899854862119,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020478837594408916,
      "loss": 4.8647,
      "mean_token_accuracy": 0.2322438269853592,
      "num_tokens": 138318576.0,
      "step": 66650
    },
    {
      "entropy": 5.749949312210083,
      "epoch": 6.0463534107402035,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002047576359638557,
      "loss": 4.901,
      "mean_token_accuracy": 0.2221514031291008,
      "num_tokens": 138328091.0,
      "step": 66655
    },
    {
      "entropy": 5.628675699234009,
      "epoch": 6.046806966618288,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020472689743625223,
      "loss": 4.8198,
      "mean_token_accuracy": 0.22953065484762192,
      "num_tokens": 138338925.0,
      "step": 66660
    },
    {
      "entropy": 5.708606243133545,
      "epoch": 6.047260522496371,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020469616036191446,
      "loss": 4.8319,
      "mean_token_accuracy": 0.23284886181354522,
      "num_tokens": 138349749.0,
      "step": 66665
    },
    {
      "entropy": 5.695165157318115,
      "epoch": 6.047714078374455,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020466542474147803,
      "loss": 4.7598,
      "mean_token_accuracy": 0.2327522188425064,
      "num_tokens": 138360224.0,
      "step": 66670
    },
    {
      "entropy": 5.722689247131347,
      "epoch": 6.04816763425254,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020463469057557855,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23417565673589708,
      "num_tokens": 138370445.0,
      "step": 66675
    },
    {
      "entropy": 5.702287483215332,
      "epoch": 6.048621190130624,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002046039578648515,
      "loss": 4.7826,
      "mean_token_accuracy": 0.2320740193128586,
      "num_tokens": 138381456.0,
      "step": 66680
    },
    {
      "entropy": 5.679199695587158,
      "epoch": 6.049074746008708,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002045732266099326,
      "loss": 4.8291,
      "mean_token_accuracy": 0.23368181884288788,
      "num_tokens": 138391403.0,
      "step": 66685
    },
    {
      "entropy": 5.687116909027099,
      "epoch": 6.0495283018867925,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002045424968114573,
      "loss": 4.8202,
      "mean_token_accuracy": 0.23053296953439711,
      "num_tokens": 138402274.0,
      "step": 66690
    },
    {
      "entropy": 5.694399023056031,
      "epoch": 6.049981857764877,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020451176847006115,
      "loss": 4.8522,
      "mean_token_accuracy": 0.23345444053411485,
      "num_tokens": 138412417.0,
      "step": 66695
    },
    {
      "entropy": 5.635405874252319,
      "epoch": 6.050435413642961,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020448104158637953,
      "loss": 4.7287,
      "mean_token_accuracy": 0.24407438486814498,
      "num_tokens": 138422996.0,
      "step": 66700
    },
    {
      "entropy": 5.687297201156616,
      "epoch": 6.050888969521045,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020445031616104798,
      "loss": 4.8449,
      "mean_token_accuracy": 0.23587649166584015,
      "num_tokens": 138432775.0,
      "step": 66705
    },
    {
      "entropy": 5.71309266090393,
      "epoch": 6.0513425253991295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020441959219470184,
      "loss": 4.837,
      "mean_token_accuracy": 0.23183569312095642,
      "num_tokens": 138443006.0,
      "step": 66710
    },
    {
      "entropy": 5.729897785186767,
      "epoch": 6.051796081277213,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020438886968797654,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22546723037958144,
      "num_tokens": 138453892.0,
      "step": 66715
    },
    {
      "entropy": 5.672634410858154,
      "epoch": 6.052249637155297,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002043581486415075,
      "loss": 4.7352,
      "mean_token_accuracy": 0.2455152153968811,
      "num_tokens": 138463355.0,
      "step": 66720
    },
    {
      "entropy": 5.6997980117797855,
      "epoch": 6.0527031930333814,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020432742905592983,
      "loss": 4.8075,
      "mean_token_accuracy": 0.23012545853853225,
      "num_tokens": 138473428.0,
      "step": 66725
    },
    {
      "entropy": 5.626162433624268,
      "epoch": 6.053156748911466,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002042967109318789,
      "loss": 4.7204,
      "mean_token_accuracy": 0.23358475267887116,
      "num_tokens": 138485083.0,
      "step": 66730
    },
    {
      "entropy": 5.5737669467926025,
      "epoch": 6.05361030478955,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020426599426999015,
      "loss": 4.6934,
      "mean_token_accuracy": 0.2501307278871536,
      "num_tokens": 138495645.0,
      "step": 66735
    },
    {
      "entropy": 5.646018552780151,
      "epoch": 6.054063860667634,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020423527907089856,
      "loss": 4.8405,
      "mean_token_accuracy": 0.23130932003259658,
      "num_tokens": 138505638.0,
      "step": 66740
    },
    {
      "entropy": 5.713876485824585,
      "epoch": 6.0545174165457185,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020420456533523952,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2402434080839157,
      "num_tokens": 138516072.0,
      "step": 66745
    },
    {
      "entropy": 5.695374155044556,
      "epoch": 6.054970972423803,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000204173853063648,
      "loss": 4.7688,
      "mean_token_accuracy": 0.23901621550321578,
      "num_tokens": 138525529.0,
      "step": 66750
    },
    {
      "entropy": 5.7215163230896,
      "epoch": 6.055424528301887,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002041431422567593,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23549802899360656,
      "num_tokens": 138535459.0,
      "step": 66755
    },
    {
      "entropy": 5.686954736709595,
      "epoch": 6.055878084179971,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020411243291520848,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22673770040273666,
      "num_tokens": 138546402.0,
      "step": 66760
    },
    {
      "entropy": 5.583535242080688,
      "epoch": 6.056331640058056,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020408172503963057,
      "loss": 4.7406,
      "mean_token_accuracy": 0.23470277339220047,
      "num_tokens": 138556510.0,
      "step": 66765
    },
    {
      "entropy": 5.634339427947998,
      "epoch": 6.056785195936139,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020405101863066072,
      "loss": 4.738,
      "mean_token_accuracy": 0.23888638764619827,
      "num_tokens": 138567029.0,
      "step": 66770
    },
    {
      "entropy": 5.660165691375733,
      "epoch": 6.057238751814223,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020402031368893387,
      "loss": 4.7846,
      "mean_token_accuracy": 0.23364911377429962,
      "num_tokens": 138577453.0,
      "step": 66775
    },
    {
      "entropy": 5.6400126934051515,
      "epoch": 6.0576923076923075,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000203989610215085,
      "loss": 4.7995,
      "mean_token_accuracy": 0.2394847109913826,
      "num_tokens": 138587607.0,
      "step": 66780
    },
    {
      "entropy": 5.70567364692688,
      "epoch": 6.058145863570392,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020395890820974904,
      "loss": 4.8145,
      "mean_token_accuracy": 0.22942672818899154,
      "num_tokens": 138597846.0,
      "step": 66785
    },
    {
      "entropy": 5.670610237121582,
      "epoch": 6.058599419448476,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020392820767356108,
      "loss": 4.781,
      "mean_token_accuracy": 0.2342354968190193,
      "num_tokens": 138608374.0,
      "step": 66790
    },
    {
      "entropy": 5.662143754959106,
      "epoch": 6.05905297532656,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020389750860715582,
      "loss": 4.7662,
      "mean_token_accuracy": 0.23141459971666337,
      "num_tokens": 138617857.0,
      "step": 66795
    },
    {
      "entropy": 5.659767961502075,
      "epoch": 6.0595065312046446,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020386681101116816,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23241615891456605,
      "num_tokens": 138627900.0,
      "step": 66800
    },
    {
      "entropy": 5.67933406829834,
      "epoch": 6.059960087082729,
      "grad_norm": 1.375,
      "learning_rate": 0.00020383611488623306,
      "loss": 4.7771,
      "mean_token_accuracy": 0.23885758072137833,
      "num_tokens": 138639520.0,
      "step": 66805
    },
    {
      "entropy": 5.681135511398315,
      "epoch": 6.060413642960813,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020380542023298526,
      "loss": 4.7328,
      "mean_token_accuracy": 0.24325101375579833,
      "num_tokens": 138648914.0,
      "step": 66810
    },
    {
      "entropy": 5.704254627227783,
      "epoch": 6.060867198838897,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020377472705205942,
      "loss": 4.781,
      "mean_token_accuracy": 0.24475665688514708,
      "num_tokens": 138659793.0,
      "step": 66815
    },
    {
      "entropy": 5.624878644943237,
      "epoch": 6.061320754716981,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020374403534409037,
      "loss": 4.7721,
      "mean_token_accuracy": 0.23691343814134597,
      "num_tokens": 138670108.0,
      "step": 66820
    },
    {
      "entropy": 5.675504350662232,
      "epoch": 6.061774310595065,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002037133451097129,
      "loss": 4.8925,
      "mean_token_accuracy": 0.221967913210392,
      "num_tokens": 138681112.0,
      "step": 66825
    },
    {
      "entropy": 5.645941638946534,
      "epoch": 6.062227866473149,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020368265634956162,
      "loss": 4.783,
      "mean_token_accuracy": 0.2325928047299385,
      "num_tokens": 138691075.0,
      "step": 66830
    },
    {
      "entropy": 5.715233182907104,
      "epoch": 6.0626814223512335,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020365196906427108,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23146281391382217,
      "num_tokens": 138700746.0,
      "step": 66835
    },
    {
      "entropy": 5.643641662597656,
      "epoch": 6.063134978229318,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002036212832544761,
      "loss": 4.8054,
      "mean_token_accuracy": 0.23688528686761856,
      "num_tokens": 138711319.0,
      "step": 66840
    },
    {
      "entropy": 5.646337795257568,
      "epoch": 6.063588534107402,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020359059892081116,
      "loss": 4.776,
      "mean_token_accuracy": 0.23035706877708434,
      "num_tokens": 138721298.0,
      "step": 66845
    },
    {
      "entropy": 5.743152284622193,
      "epoch": 6.064042089985486,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020355991606391078,
      "loss": 4.84,
      "mean_token_accuracy": 0.23721096217632293,
      "num_tokens": 138731173.0,
      "step": 66850
    },
    {
      "entropy": 5.647310638427735,
      "epoch": 6.064495645863571,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020352923468440958,
      "loss": 4.7633,
      "mean_token_accuracy": 0.24036917984485626,
      "num_tokens": 138742490.0,
      "step": 66855
    },
    {
      "entropy": 5.731290578842163,
      "epoch": 6.064949201741655,
      "grad_norm": 1.25,
      "learning_rate": 0.00020349855478294203,
      "loss": 4.9342,
      "mean_token_accuracy": 0.22720082700252534,
      "num_tokens": 138753885.0,
      "step": 66860
    },
    {
      "entropy": 5.682591295242309,
      "epoch": 6.065402757619739,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020346787636014264,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23967869877815245,
      "num_tokens": 138764330.0,
      "step": 66865
    },
    {
      "entropy": 5.684554386138916,
      "epoch": 6.0658563134978225,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020343719941664568,
      "loss": 4.833,
      "mean_token_accuracy": 0.23555974960327147,
      "num_tokens": 138774808.0,
      "step": 66870
    },
    {
      "entropy": 5.689710664749145,
      "epoch": 6.066309869375907,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020340652395308572,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23636410534381866,
      "num_tokens": 138784977.0,
      "step": 66875
    },
    {
      "entropy": 5.760881090164185,
      "epoch": 6.066763425253991,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020337584997009712,
      "loss": 4.8746,
      "mean_token_accuracy": 0.23506473004817963,
      "num_tokens": 138795730.0,
      "step": 66880
    },
    {
      "entropy": 5.652928638458252,
      "epoch": 6.067216981132075,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002033451774683141,
      "loss": 4.7431,
      "mean_token_accuracy": 0.23153735548257828,
      "num_tokens": 138806248.0,
      "step": 66885
    },
    {
      "entropy": 5.717310762405395,
      "epoch": 6.06767053701016,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020331450644837113,
      "loss": 4.904,
      "mean_token_accuracy": 0.2288309872150421,
      "num_tokens": 138816305.0,
      "step": 66890
    },
    {
      "entropy": 5.714952087402343,
      "epoch": 6.068124092888244,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020328383691090246,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22437614053487778,
      "num_tokens": 138826679.0,
      "step": 66895
    },
    {
      "entropy": 5.657682466506958,
      "epoch": 6.068577648766328,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020325316885654238,
      "loss": 4.7512,
      "mean_token_accuracy": 0.23582015931606293,
      "num_tokens": 138836848.0,
      "step": 66900
    },
    {
      "entropy": 5.750066900253296,
      "epoch": 6.069031204644412,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020322250228592493,
      "loss": 4.8791,
      "mean_token_accuracy": 0.232288958132267,
      "num_tokens": 138847898.0,
      "step": 66905
    },
    {
      "entropy": 5.693362712860107,
      "epoch": 6.069484760522497,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020319183719968442,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2238868772983551,
      "num_tokens": 138857893.0,
      "step": 66910
    },
    {
      "entropy": 5.72230167388916,
      "epoch": 6.069938316400581,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002031611735984551,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23881616741418837,
      "num_tokens": 138868289.0,
      "step": 66915
    },
    {
      "entropy": 5.755714750289917,
      "epoch": 6.070391872278665,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020313051148287094,
      "loss": 4.8481,
      "mean_token_accuracy": 0.2285938948392868,
      "num_tokens": 138878018.0,
      "step": 66920
    },
    {
      "entropy": 5.648716402053833,
      "epoch": 6.0708454281567485,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020309985085356614,
      "loss": 4.7578,
      "mean_token_accuracy": 0.23327567428350449,
      "num_tokens": 138888390.0,
      "step": 66925
    },
    {
      "entropy": 5.65207462310791,
      "epoch": 6.071298984034833,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020306919171117477,
      "loss": 4.8056,
      "mean_token_accuracy": 0.23171549141407013,
      "num_tokens": 138898512.0,
      "step": 66930
    },
    {
      "entropy": 5.7106578826904295,
      "epoch": 6.071752539912917,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020303853405633082,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2381778746843338,
      "num_tokens": 138908711.0,
      "step": 66935
    },
    {
      "entropy": 5.728688955307007,
      "epoch": 6.072206095791001,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020300787788966835,
      "loss": 4.858,
      "mean_token_accuracy": 0.2317335993051529,
      "num_tokens": 138919094.0,
      "step": 66940
    },
    {
      "entropy": 5.7555811405181885,
      "epoch": 6.072659651669086,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020297722321182133,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23751386404037475,
      "num_tokens": 138929064.0,
      "step": 66945
    },
    {
      "entropy": 5.725769567489624,
      "epoch": 6.07311320754717,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020294657002342364,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2347133368253708,
      "num_tokens": 138939705.0,
      "step": 66950
    },
    {
      "entropy": 5.6087390899658205,
      "epoch": 6.073566763425254,
      "grad_norm": 1.625,
      "learning_rate": 0.00020291591832510925,
      "loss": 4.7288,
      "mean_token_accuracy": 0.24156671166419982,
      "num_tokens": 138949642.0,
      "step": 66955
    },
    {
      "entropy": 5.749217700958252,
      "epoch": 6.074020319303338,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000202885268117512,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23235135078430175,
      "num_tokens": 138958670.0,
      "step": 66960
    },
    {
      "entropy": 5.723877573013306,
      "epoch": 6.074473875181423,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020285461940126588,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22928048968315123,
      "num_tokens": 138968731.0,
      "step": 66965
    },
    {
      "entropy": 5.7632993221282955,
      "epoch": 6.074927431059507,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020282397217700454,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22378408312797546,
      "num_tokens": 138979118.0,
      "step": 66970
    },
    {
      "entropy": 5.728323316574096,
      "epoch": 6.07538098693759,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020279332644536187,
      "loss": 4.8478,
      "mean_token_accuracy": 0.2436651647090912,
      "num_tokens": 138990369.0,
      "step": 66975
    },
    {
      "entropy": 5.769313907623291,
      "epoch": 6.075834542815675,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020276268220697163,
      "loss": 4.8609,
      "mean_token_accuracy": 0.2339174836874008,
      "num_tokens": 139001041.0,
      "step": 66980
    },
    {
      "entropy": 5.722331714630127,
      "epoch": 6.076288098693759,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020273203946246754,
      "loss": 4.8512,
      "mean_token_accuracy": 0.23298603743314744,
      "num_tokens": 139010921.0,
      "step": 66985
    },
    {
      "entropy": 5.725670862197876,
      "epoch": 6.076741654571843,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020270139821248323,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22754329144954683,
      "num_tokens": 139021423.0,
      "step": 66990
    },
    {
      "entropy": 5.760165691375732,
      "epoch": 6.077195210449927,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002026707584576525,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2274021327495575,
      "num_tokens": 139032218.0,
      "step": 66995
    },
    {
      "entropy": 5.700982189178466,
      "epoch": 6.077648766328012,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002026401201986089,
      "loss": 4.8147,
      "mean_token_accuracy": 0.23314134031534195,
      "num_tokens": 139041910.0,
      "step": 67000
    },
    {
      "entropy": 5.7646135807037355,
      "epoch": 6.078102322206096,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020260948343598596,
      "loss": 4.8248,
      "mean_token_accuracy": 0.2340657204389572,
      "num_tokens": 139051939.0,
      "step": 67005
    },
    {
      "entropy": 5.701890277862549,
      "epoch": 6.07855587808418,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020257884817041744,
      "loss": 4.855,
      "mean_token_accuracy": 0.23011746406555175,
      "num_tokens": 139063219.0,
      "step": 67010
    },
    {
      "entropy": 5.697453308105469,
      "epoch": 6.0790094339622645,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020254821440253678,
      "loss": 4.8145,
      "mean_token_accuracy": 0.2349799707531929,
      "num_tokens": 139074396.0,
      "step": 67015
    },
    {
      "entropy": 5.744212484359741,
      "epoch": 6.079462989840349,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002025175821329775,
      "loss": 4.89,
      "mean_token_accuracy": 0.22982007563114165,
      "num_tokens": 139084697.0,
      "step": 67020
    },
    {
      "entropy": 5.676651334762573,
      "epoch": 6.079916545718432,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020248695136237312,
      "loss": 4.7965,
      "mean_token_accuracy": 0.24186878949403762,
      "num_tokens": 139094957.0,
      "step": 67025
    },
    {
      "entropy": 5.761084127426147,
      "epoch": 6.080370101596516,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020245632209135706,
      "loss": 4.9371,
      "mean_token_accuracy": 0.22400783151388168,
      "num_tokens": 139105214.0,
      "step": 67030
    },
    {
      "entropy": 5.706796789169312,
      "epoch": 6.080823657474601,
      "grad_norm": 1.625,
      "learning_rate": 0.00020242569432056274,
      "loss": 4.8164,
      "mean_token_accuracy": 0.23728105574846267,
      "num_tokens": 139114574.0,
      "step": 67035
    },
    {
      "entropy": 5.7087358951568605,
      "epoch": 6.081277213352685,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020239506805062353,
      "loss": 4.754,
      "mean_token_accuracy": 0.239484940469265,
      "num_tokens": 139124823.0,
      "step": 67040
    },
    {
      "entropy": 5.644578075408935,
      "epoch": 6.081730769230769,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020236444328217278,
      "loss": 4.718,
      "mean_token_accuracy": 0.24372972249984742,
      "num_tokens": 139134870.0,
      "step": 67045
    },
    {
      "entropy": 5.63224720954895,
      "epoch": 6.082184325108853,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020233382001584393,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23766354471445084,
      "num_tokens": 139145195.0,
      "step": 67050
    },
    {
      "entropy": 5.686309814453125,
      "epoch": 6.082637880986938,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020230319825227018,
      "loss": 4.8246,
      "mean_token_accuracy": 0.23023488670587539,
      "num_tokens": 139156259.0,
      "step": 67055
    },
    {
      "entropy": 5.761583518981934,
      "epoch": 6.083091436865022,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002022725779920848,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22625633031129838,
      "num_tokens": 139166515.0,
      "step": 67060
    },
    {
      "entropy": 5.688245725631714,
      "epoch": 6.083544992743106,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020224195923592109,
      "loss": 4.856,
      "mean_token_accuracy": 0.22740104794502258,
      "num_tokens": 139176927.0,
      "step": 67065
    },
    {
      "entropy": 5.750839471817017,
      "epoch": 6.0839985486211905,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020221134198441216,
      "loss": 4.8553,
      "mean_token_accuracy": 0.22877784371376036,
      "num_tokens": 139187196.0,
      "step": 67070
    },
    {
      "entropy": 5.713904619216919,
      "epoch": 6.084452104499274,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020218072623819116,
      "loss": 4.9368,
      "mean_token_accuracy": 0.22796464264392852,
      "num_tokens": 139198503.0,
      "step": 67075
    },
    {
      "entropy": 5.704023170471191,
      "epoch": 6.084905660377358,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002021501119978914,
      "loss": 4.8174,
      "mean_token_accuracy": 0.23001689314842225,
      "num_tokens": 139208449.0,
      "step": 67080
    },
    {
      "entropy": 5.830355882644653,
      "epoch": 6.085359216255442,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020211949926414585,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2258099615573883,
      "num_tokens": 139217725.0,
      "step": 67085
    },
    {
      "entropy": 5.758458995819092,
      "epoch": 6.085812772133527,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020208888803758762,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22676844745874405,
      "num_tokens": 139227864.0,
      "step": 67090
    },
    {
      "entropy": 5.71276159286499,
      "epoch": 6.086266328011611,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020205827831884987,
      "loss": 4.8543,
      "mean_token_accuracy": 0.23006399273872374,
      "num_tokens": 139238679.0,
      "step": 67095
    },
    {
      "entropy": 5.661569356918335,
      "epoch": 6.086719883889695,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020202767010856542,
      "loss": 4.7406,
      "mean_token_accuracy": 0.2423327997326851,
      "num_tokens": 139248945.0,
      "step": 67100
    },
    {
      "entropy": 5.679501247406006,
      "epoch": 6.0871734397677795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020199706340736734,
      "loss": 4.8261,
      "mean_token_accuracy": 0.23367264717817307,
      "num_tokens": 139258180.0,
      "step": 67105
    },
    {
      "entropy": 5.706524324417114,
      "epoch": 6.087626995645864,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002019664582158886,
      "loss": 4.8209,
      "mean_token_accuracy": 0.23790159970521926,
      "num_tokens": 139267955.0,
      "step": 67110
    },
    {
      "entropy": 5.723002815246582,
      "epoch": 6.088080551523948,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020193585453476216,
      "loss": 4.8247,
      "mean_token_accuracy": 0.22473050355911256,
      "num_tokens": 139277787.0,
      "step": 67115
    },
    {
      "entropy": 5.689231586456299,
      "epoch": 6.088534107402032,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020190525236462076,
      "loss": 4.8096,
      "mean_token_accuracy": 0.23549890965223313,
      "num_tokens": 139288205.0,
      "step": 67120
    },
    {
      "entropy": 5.703114032745361,
      "epoch": 6.0889876632801165,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020187465170609748,
      "loss": 4.8009,
      "mean_token_accuracy": 0.24142030477523804,
      "num_tokens": 139298124.0,
      "step": 67125
    },
    {
      "entropy": 5.829217195510864,
      "epoch": 6.0894412191582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020184405255982497,
      "loss": 5.0106,
      "mean_token_accuracy": 0.21529214978218078,
      "num_tokens": 139308692.0,
      "step": 67130
    },
    {
      "entropy": 5.727618455886841,
      "epoch": 6.089894775036284,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002018134549264361,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23475711345672606,
      "num_tokens": 139317932.0,
      "step": 67135
    },
    {
      "entropy": 5.7304784774780275,
      "epoch": 6.0903483309143684,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002017828588065636,
      "loss": 4.8102,
      "mean_token_accuracy": 0.23720233142375946,
      "num_tokens": 139327419.0,
      "step": 67140
    },
    {
      "entropy": 5.665414142608642,
      "epoch": 6.090801886792453,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002017522642008403,
      "loss": 4.804,
      "mean_token_accuracy": 0.23510629534721375,
      "num_tokens": 139338258.0,
      "step": 67145
    },
    {
      "entropy": 5.7805098533630375,
      "epoch": 6.091255442670537,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002017216711098988,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23424971848726273,
      "num_tokens": 139347591.0,
      "step": 67150
    },
    {
      "entropy": 5.678393888473511,
      "epoch": 6.091708998548621,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002016910795343719,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23239237666130066,
      "num_tokens": 139357639.0,
      "step": 67155
    },
    {
      "entropy": 5.687078189849854,
      "epoch": 6.0921625544267055,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020166048947489206,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23868878930807114,
      "num_tokens": 139367988.0,
      "step": 67160
    },
    {
      "entropy": 5.668776512145996,
      "epoch": 6.09261611030479,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020162990093209205,
      "loss": 4.7792,
      "mean_token_accuracy": 0.23937090188264848,
      "num_tokens": 139377726.0,
      "step": 67165
    },
    {
      "entropy": 5.71937108039856,
      "epoch": 6.093069666182874,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002015993139066043,
      "loss": 4.8361,
      "mean_token_accuracy": 0.24004908800125122,
      "num_tokens": 139387361.0,
      "step": 67170
    },
    {
      "entropy": 5.714446067810059,
      "epoch": 6.093523222060958,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020156872839906153,
      "loss": 4.8586,
      "mean_token_accuracy": 0.2304424077272415,
      "num_tokens": 139397279.0,
      "step": 67175
    },
    {
      "entropy": 5.696230936050415,
      "epoch": 6.093976777939042,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020153814441009615,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23073593527078629,
      "num_tokens": 139406715.0,
      "step": 67180
    },
    {
      "entropy": 5.781562948226929,
      "epoch": 6.094430333817126,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020150756194034063,
      "loss": 4.9553,
      "mean_token_accuracy": 0.22730724066495894,
      "num_tokens": 139417485.0,
      "step": 67185
    },
    {
      "entropy": 5.7143477439880375,
      "epoch": 6.09488388969521,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002014769809904275,
      "loss": 4.8402,
      "mean_token_accuracy": 0.23060708492994308,
      "num_tokens": 139428174.0,
      "step": 67190
    },
    {
      "entropy": 5.6911626815795895,
      "epoch": 6.0953374455732945,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020144640156098914,
      "loss": 4.8045,
      "mean_token_accuracy": 0.23536928594112397,
      "num_tokens": 139438349.0,
      "step": 67195
    },
    {
      "entropy": 5.739307641983032,
      "epoch": 6.095791001451379,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002014158236526579,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22884681224822997,
      "num_tokens": 139448193.0,
      "step": 67200
    },
    {
      "entropy": 5.718907403945923,
      "epoch": 6.096244557329463,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002013852472660662,
      "loss": 4.8993,
      "mean_token_accuracy": 0.2235947757959366,
      "num_tokens": 139458418.0,
      "step": 67205
    },
    {
      "entropy": 5.705457925796509,
      "epoch": 6.096698113207547,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020135467240184636,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23300090730190276,
      "num_tokens": 139468085.0,
      "step": 67210
    },
    {
      "entropy": 5.754956531524658,
      "epoch": 6.0971516690856316,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020132409906063064,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2267980992794037,
      "num_tokens": 139479250.0,
      "step": 67215
    },
    {
      "entropy": 5.717527437210083,
      "epoch": 6.097605224963716,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020129352724305134,
      "loss": 4.8498,
      "mean_token_accuracy": 0.24045152366161346,
      "num_tokens": 139488602.0,
      "step": 67220
    },
    {
      "entropy": 5.703580904006958,
      "epoch": 6.0980587808418,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020126295694974068,
      "loss": 4.8617,
      "mean_token_accuracy": 0.22871570587158202,
      "num_tokens": 139499629.0,
      "step": 67225
    },
    {
      "entropy": 5.69329080581665,
      "epoch": 6.0985123367198835,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020123238818133083,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23475176393985747,
      "num_tokens": 139510567.0,
      "step": 67230
    },
    {
      "entropy": 5.701822996139526,
      "epoch": 6.098965892597968,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020120182093845402,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23594889640808106,
      "num_tokens": 139519893.0,
      "step": 67235
    },
    {
      "entropy": 5.768895149230957,
      "epoch": 6.099419448476052,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020117125522174236,
      "loss": 4.8704,
      "mean_token_accuracy": 0.22787535935640335,
      "num_tokens": 139530653.0,
      "step": 67240
    },
    {
      "entropy": 5.70025601387024,
      "epoch": 6.099873004354136,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002011406910318279,
      "loss": 4.7686,
      "mean_token_accuracy": 0.23806840628385545,
      "num_tokens": 139541032.0,
      "step": 67245
    },
    {
      "entropy": 5.716918420791626,
      "epoch": 6.1003265602322205,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020111012836934283,
      "loss": 4.827,
      "mean_token_accuracy": 0.23650583922863005,
      "num_tokens": 139551838.0,
      "step": 67250
    },
    {
      "entropy": 5.698369073867798,
      "epoch": 6.100780116110305,
      "grad_norm": 1.5,
      "learning_rate": 0.00020107956723491917,
      "loss": 4.7367,
      "mean_token_accuracy": 0.23461893051862717,
      "num_tokens": 139561672.0,
      "step": 67255
    },
    {
      "entropy": 5.642456436157227,
      "epoch": 6.101233671988389,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020104900762918883,
      "loss": 4.7677,
      "mean_token_accuracy": 0.23788135945796968,
      "num_tokens": 139572710.0,
      "step": 67260
    },
    {
      "entropy": 5.587803792953491,
      "epoch": 6.101687227866473,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020101844955278393,
      "loss": 4.6802,
      "mean_token_accuracy": 0.23941538333892823,
      "num_tokens": 139583020.0,
      "step": 67265
    },
    {
      "entropy": 5.758774757385254,
      "epoch": 6.102140783744558,
      "grad_norm": 1.65625,
      "learning_rate": 0.00020098789300633624,
      "loss": 4.8666,
      "mean_token_accuracy": 0.22844922691583633,
      "num_tokens": 139592944.0,
      "step": 67270
    },
    {
      "entropy": 5.709507036209106,
      "epoch": 6.102594339622642,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020095733799047782,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23568847477436067,
      "num_tokens": 139604415.0,
      "step": 67275
    },
    {
      "entropy": 5.750989007949829,
      "epoch": 6.103047895500726,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020092678450584056,
      "loss": 4.8505,
      "mean_token_accuracy": 0.235572250187397,
      "num_tokens": 139615456.0,
      "step": 67280
    },
    {
      "entropy": 5.710552501678467,
      "epoch": 6.1035014513788095,
      "grad_norm": 1.5,
      "learning_rate": 0.0002008962325530563,
      "loss": 4.8623,
      "mean_token_accuracy": 0.227780282497406,
      "num_tokens": 139625513.0,
      "step": 67285
    },
    {
      "entropy": 5.684914875030517,
      "epoch": 6.103955007256894,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002008656821327567,
      "loss": 4.8327,
      "mean_token_accuracy": 0.230044361948967,
      "num_tokens": 139635524.0,
      "step": 67290
    },
    {
      "entropy": 5.668376827239991,
      "epoch": 6.104408563134978,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002008351332455738,
      "loss": 4.6848,
      "mean_token_accuracy": 0.24131717532873154,
      "num_tokens": 139645958.0,
      "step": 67295
    },
    {
      "entropy": 5.740470695495605,
      "epoch": 6.104862119013062,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020080458589213918,
      "loss": 4.859,
      "mean_token_accuracy": 0.22829100489616394,
      "num_tokens": 139655022.0,
      "step": 67300
    },
    {
      "entropy": 5.7596625804901125,
      "epoch": 6.105315674891147,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020077404007308463,
      "loss": 4.9524,
      "mean_token_accuracy": 0.22282337099313737,
      "num_tokens": 139665913.0,
      "step": 67305
    },
    {
      "entropy": 5.719062089920044,
      "epoch": 6.105769230769231,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020074349578904195,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23387212604284285,
      "num_tokens": 139675260.0,
      "step": 67310
    },
    {
      "entropy": 5.711226081848144,
      "epoch": 6.106222786647315,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020071295304064258,
      "loss": 4.8005,
      "mean_token_accuracy": 0.24246039539575576,
      "num_tokens": 139685337.0,
      "step": 67315
    },
    {
      "entropy": 5.623147821426391,
      "epoch": 6.106676342525399,
      "grad_norm": 1.375,
      "learning_rate": 0.00020068241182851838,
      "loss": 4.7336,
      "mean_token_accuracy": 0.24799549877643584,
      "num_tokens": 139694977.0,
      "step": 67320
    },
    {
      "entropy": 5.719218349456787,
      "epoch": 6.107129898403484,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002006518721533007,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22384763658046722,
      "num_tokens": 139705708.0,
      "step": 67325
    },
    {
      "entropy": 5.750829362869263,
      "epoch": 6.107583454281568,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020062133401562137,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22265085130929946,
      "num_tokens": 139717268.0,
      "step": 67330
    },
    {
      "entropy": 5.686965656280518,
      "epoch": 6.108037010159651,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002005907974161118,
      "loss": 4.7688,
      "mean_token_accuracy": 0.24043578803539276,
      "num_tokens": 139728576.0,
      "step": 67335
    },
    {
      "entropy": 5.724533462524414,
      "epoch": 6.1084905660377355,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020056026235540342,
      "loss": 4.8974,
      "mean_token_accuracy": 0.231522636115551,
      "num_tokens": 139741136.0,
      "step": 67340
    },
    {
      "entropy": 5.637422752380371,
      "epoch": 6.10894412191582,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020052972883412783,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23739634156227113,
      "num_tokens": 139751244.0,
      "step": 67345
    },
    {
      "entropy": 5.723801565170288,
      "epoch": 6.109397677793904,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020049919685291646,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2306814670562744,
      "num_tokens": 139760584.0,
      "step": 67350
    },
    {
      "entropy": 5.735038137435913,
      "epoch": 6.109851233671988,
      "grad_norm": 1.5,
      "learning_rate": 0.00020046866641240064,
      "loss": 4.7628,
      "mean_token_accuracy": 0.24136192947626114,
      "num_tokens": 139770101.0,
      "step": 67355
    },
    {
      "entropy": 5.689612007141113,
      "epoch": 6.110304789550073,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020043813751321183,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2376016139984131,
      "num_tokens": 139780423.0,
      "step": 67360
    },
    {
      "entropy": 5.699858903884888,
      "epoch": 6.110758345428157,
      "grad_norm": 1.375,
      "learning_rate": 0.0002004076101559813,
      "loss": 4.8618,
      "mean_token_accuracy": 0.23196362406015397,
      "num_tokens": 139790999.0,
      "step": 67365
    },
    {
      "entropy": 5.652073335647583,
      "epoch": 6.111211901306241,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002003770843413404,
      "loss": 4.7533,
      "mean_token_accuracy": 0.23965562433004378,
      "num_tokens": 139800842.0,
      "step": 67370
    },
    {
      "entropy": 5.612607765197754,
      "epoch": 6.111665457184325,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002003465600699204,
      "loss": 4.7228,
      "mean_token_accuracy": 0.24814339876174926,
      "num_tokens": 139810831.0,
      "step": 67375
    },
    {
      "entropy": 5.670009565353394,
      "epoch": 6.11211901306241,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002003160373423526,
      "loss": 4.8291,
      "mean_token_accuracy": 0.2465319201350212,
      "num_tokens": 139822127.0,
      "step": 67380
    },
    {
      "entropy": 5.723679733276367,
      "epoch": 6.112572568940493,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002002855161592682,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23173681199550628,
      "num_tokens": 139832312.0,
      "step": 67385
    },
    {
      "entropy": 5.701530265808105,
      "epoch": 6.113026124818577,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020025499652129826,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22442688047885895,
      "num_tokens": 139842755.0,
      "step": 67390
    },
    {
      "entropy": 5.695860528945923,
      "epoch": 6.113479680696662,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00020022447842907415,
      "loss": 4.7679,
      "mean_token_accuracy": 0.24184100776910783,
      "num_tokens": 139854540.0,
      "step": 67395
    },
    {
      "entropy": 5.7001368522644045,
      "epoch": 6.113933236574746,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020019396188322686,
      "loss": 4.8266,
      "mean_token_accuracy": 0.23500894904136657,
      "num_tokens": 139864867.0,
      "step": 67400
    },
    {
      "entropy": 5.694049406051636,
      "epoch": 6.11438679245283,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002001634468843875,
      "loss": 4.8222,
      "mean_token_accuracy": 0.22965617924928666,
      "num_tokens": 139876230.0,
      "step": 67405
    },
    {
      "entropy": 5.671526384353638,
      "epoch": 6.114840348330914,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002001329334331871,
      "loss": 4.7764,
      "mean_token_accuracy": 0.23837778121232986,
      "num_tokens": 139887931.0,
      "step": 67410
    },
    {
      "entropy": 5.699833488464355,
      "epoch": 6.115293904208999,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002001024215302567,
      "loss": 4.8354,
      "mean_token_accuracy": 0.22805413752794265,
      "num_tokens": 139899003.0,
      "step": 67415
    },
    {
      "entropy": 5.78930778503418,
      "epoch": 6.115747460087083,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020007191117622737,
      "loss": 4.9043,
      "mean_token_accuracy": 0.23272642642259597,
      "num_tokens": 139909343.0,
      "step": 67420
    },
    {
      "entropy": 5.742742395401001,
      "epoch": 6.116201015965167,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002000414023717299,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2373853802680969,
      "num_tokens": 139920208.0,
      "step": 67425
    },
    {
      "entropy": 5.657450866699219,
      "epoch": 6.1166545718432515,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020001089511739547,
      "loss": 4.7492,
      "mean_token_accuracy": 0.24063433855772018,
      "num_tokens": 139931064.0,
      "step": 67430
    },
    {
      "entropy": 5.680905675888061,
      "epoch": 6.117108127721335,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019998038941385478,
      "loss": 4.741,
      "mean_token_accuracy": 0.2375318467617035,
      "num_tokens": 139940630.0,
      "step": 67435
    },
    {
      "entropy": 5.695903491973877,
      "epoch": 6.117561683599419,
      "grad_norm": 1.6875,
      "learning_rate": 0.0001999498852617387,
      "loss": 4.8743,
      "mean_token_accuracy": 0.22585879415273666,
      "num_tokens": 139950333.0,
      "step": 67440
    },
    {
      "entropy": 5.716772365570068,
      "epoch": 6.118015239477503,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001999193826616782,
      "loss": 4.9123,
      "mean_token_accuracy": 0.2290532648563385,
      "num_tokens": 139960298.0,
      "step": 67445
    },
    {
      "entropy": 5.766395139694214,
      "epoch": 6.118468795355588,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019988888161430397,
      "loss": 4.8171,
      "mean_token_accuracy": 0.2338327169418335,
      "num_tokens": 139970422.0,
      "step": 67450
    },
    {
      "entropy": 5.647141408920288,
      "epoch": 6.118922351233672,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019985838212024682,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2346251904964447,
      "num_tokens": 139980197.0,
      "step": 67455
    },
    {
      "entropy": 5.618058538436889,
      "epoch": 6.119375907111756,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019982788418013741,
      "loss": 4.7242,
      "mean_token_accuracy": 0.24343276172876357,
      "num_tokens": 139989474.0,
      "step": 67460
    },
    {
      "entropy": 5.770511722564697,
      "epoch": 6.11982946298984,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001997973877946066,
      "loss": 4.9308,
      "mean_token_accuracy": 0.23543926179409028,
      "num_tokens": 139999256.0,
      "step": 67465
    },
    {
      "entropy": 5.656657695770264,
      "epoch": 6.120283018867925,
      "grad_norm": 1.5,
      "learning_rate": 0.0001997668929642849,
      "loss": 4.7618,
      "mean_token_accuracy": 0.2349076345562935,
      "num_tokens": 140010256.0,
      "step": 67470
    },
    {
      "entropy": 5.672376441955566,
      "epoch": 6.120736574746009,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019973639968980306,
      "loss": 4.7758,
      "mean_token_accuracy": 0.24237747639417648,
      "num_tokens": 140021175.0,
      "step": 67475
    },
    {
      "entropy": 5.794150686264038,
      "epoch": 6.121190130624093,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019970590797179167,
      "loss": 4.9244,
      "mean_token_accuracy": 0.223647777736187,
      "num_tokens": 140030943.0,
      "step": 67480
    },
    {
      "entropy": 5.718950176239014,
      "epoch": 6.1216436865021775,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019967541781088129,
      "loss": 4.8072,
      "mean_token_accuracy": 0.22903668731451035,
      "num_tokens": 140040919.0,
      "step": 67485
    },
    {
      "entropy": 5.756234550476075,
      "epoch": 6.122097242380261,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001996449292077025,
      "loss": 4.8813,
      "mean_token_accuracy": 0.23368462026119233,
      "num_tokens": 140051656.0,
      "step": 67490
    },
    {
      "entropy": 5.698817729949951,
      "epoch": 6.122550798258345,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019961444216288565,
      "loss": 4.8032,
      "mean_token_accuracy": 0.23438399732112886,
      "num_tokens": 140061655.0,
      "step": 67495
    },
    {
      "entropy": 5.634574127197266,
      "epoch": 6.123004354136429,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019958395667706147,
      "loss": 4.7715,
      "mean_token_accuracy": 0.2391667515039444,
      "num_tokens": 140072985.0,
      "step": 67500
    },
    {
      "entropy": 5.646518850326538,
      "epoch": 6.123457910014514,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019955347275086023,
      "loss": 4.8067,
      "mean_token_accuracy": 0.23224214911460878,
      "num_tokens": 140083814.0,
      "step": 67505
    },
    {
      "entropy": 5.690139198303223,
      "epoch": 6.123911465892598,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001995229903849124,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23341808915138246,
      "num_tokens": 140093664.0,
      "step": 67510
    },
    {
      "entropy": 5.728716135025024,
      "epoch": 6.124365021770682,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019949250957984838,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2336490586400032,
      "num_tokens": 140103585.0,
      "step": 67515
    },
    {
      "entropy": 5.668175268173218,
      "epoch": 6.1248185776487665,
      "grad_norm": 1.625,
      "learning_rate": 0.00019946203033629844,
      "loss": 4.8317,
      "mean_token_accuracy": 0.22825641930103302,
      "num_tokens": 140112802.0,
      "step": 67520
    },
    {
      "entropy": 5.7249187469482425,
      "epoch": 6.125272133526851,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001994315526548931,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22767917662858964,
      "num_tokens": 140122980.0,
      "step": 67525
    },
    {
      "entropy": 5.736728572845459,
      "epoch": 6.125725689404935,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019940107653626245,
      "loss": 4.8499,
      "mean_token_accuracy": 0.22725210934877396,
      "num_tokens": 140134336.0,
      "step": 67530
    },
    {
      "entropy": 5.662375831604004,
      "epoch": 6.126179245283019,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019937060198103678,
      "loss": 4.8139,
      "mean_token_accuracy": 0.23353849202394486,
      "num_tokens": 140144683.0,
      "step": 67535
    },
    {
      "entropy": 5.718917751312256,
      "epoch": 6.126632801161103,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019934012898984634,
      "loss": 4.8479,
      "mean_token_accuracy": 0.23007434457540513,
      "num_tokens": 140155536.0,
      "step": 67540
    },
    {
      "entropy": 5.773644304275512,
      "epoch": 6.127086357039187,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001993096575633213,
      "loss": 4.8653,
      "mean_token_accuracy": 0.23061612844467164,
      "num_tokens": 140166391.0,
      "step": 67545
    },
    {
      "entropy": 5.743206024169922,
      "epoch": 6.127539912917271,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001992791877020918,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22216806262731553,
      "num_tokens": 140176136.0,
      "step": 67550
    },
    {
      "entropy": 5.720436334609985,
      "epoch": 6.1279934687953554,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000199248719406788,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2322528675198555,
      "num_tokens": 140187174.0,
      "step": 67555
    },
    {
      "entropy": 5.655912923812866,
      "epoch": 6.12844702467344,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019921825267804006,
      "loss": 4.8061,
      "mean_token_accuracy": 0.23724014461040496,
      "num_tokens": 140197512.0,
      "step": 67560
    },
    {
      "entropy": 5.696937799453735,
      "epoch": 6.128900580551524,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019918778751647792,
      "loss": 4.7557,
      "mean_token_accuracy": 0.24169720709323883,
      "num_tokens": 140208008.0,
      "step": 67565
    },
    {
      "entropy": 5.82158932685852,
      "epoch": 6.129354136429608,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019915732392273162,
      "loss": 4.9441,
      "mean_token_accuracy": 0.22452896237373351,
      "num_tokens": 140218278.0,
      "step": 67570
    },
    {
      "entropy": 5.692870235443115,
      "epoch": 6.1298076923076925,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019912686189743124,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23684967905282975,
      "num_tokens": 140228055.0,
      "step": 67575
    },
    {
      "entropy": 5.667782163619995,
      "epoch": 6.130261248185777,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019909640144120667,
      "loss": 4.7472,
      "mean_token_accuracy": 0.24310486018657684,
      "num_tokens": 140237339.0,
      "step": 67580
    },
    {
      "entropy": 5.720561361312866,
      "epoch": 6.130714804063861,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001990659425546879,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22792785316705705,
      "num_tokens": 140249220.0,
      "step": 67585
    },
    {
      "entropy": 5.750981330871582,
      "epoch": 6.131168359941944,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019903548523850473,
      "loss": 4.8869,
      "mean_token_accuracy": 0.23897096067667006,
      "num_tokens": 140259597.0,
      "step": 67590
    },
    {
      "entropy": 5.615643882751465,
      "epoch": 6.131621915820029,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019900502949328714,
      "loss": 4.7367,
      "mean_token_accuracy": 0.23981404751539231,
      "num_tokens": 140269574.0,
      "step": 67595
    },
    {
      "entropy": 5.648217248916626,
      "epoch": 6.132075471698113,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019897457531966483,
      "loss": 4.7271,
      "mean_token_accuracy": 0.24310789108276368,
      "num_tokens": 140279001.0,
      "step": 67600
    },
    {
      "entropy": 5.672075986862183,
      "epoch": 6.132529027576197,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019894412271826773,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2341894805431366,
      "num_tokens": 140289107.0,
      "step": 67605
    },
    {
      "entropy": 5.716253232955933,
      "epoch": 6.1329825834542815,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001989136716897255,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2233036518096924,
      "num_tokens": 140298796.0,
      "step": 67610
    },
    {
      "entropy": 5.670476913452148,
      "epoch": 6.133436139332366,
      "grad_norm": 1.5625,
      "learning_rate": 0.000198883222234668,
      "loss": 4.7862,
      "mean_token_accuracy": 0.24276862442493438,
      "num_tokens": 140308533.0,
      "step": 67615
    },
    {
      "entropy": 5.6875396251678465,
      "epoch": 6.13388969521045,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019885277435372485,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23369518667459488,
      "num_tokens": 140318454.0,
      "step": 67620
    },
    {
      "entropy": 5.745335674285888,
      "epoch": 6.134343251088534,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019882232804752564,
      "loss": 4.9288,
      "mean_token_accuracy": 0.22910068035125733,
      "num_tokens": 140329636.0,
      "step": 67625
    },
    {
      "entropy": 5.674749279022217,
      "epoch": 6.1347968069666186,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019879188331670022,
      "loss": 4.7594,
      "mean_token_accuracy": 0.23402052223682404,
      "num_tokens": 140339562.0,
      "step": 67630
    },
    {
      "entropy": 5.7796426773071286,
      "epoch": 6.135250362844703,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019876144016187802,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22686037868261338,
      "num_tokens": 140351276.0,
      "step": 67635
    },
    {
      "entropy": 5.638597011566162,
      "epoch": 6.135703918722786,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019873099858368866,
      "loss": 4.7626,
      "mean_token_accuracy": 0.23657905161380768,
      "num_tokens": 140361786.0,
      "step": 67640
    },
    {
      "entropy": 5.722605848312378,
      "epoch": 6.1361574746008705,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001987005585827617,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22514507174491882,
      "num_tokens": 140371982.0,
      "step": 67645
    },
    {
      "entropy": 5.7781556129455565,
      "epoch": 6.136611030478955,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019867012015972662,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22909651547670365,
      "num_tokens": 140382845.0,
      "step": 67650
    },
    {
      "entropy": 5.728400611877442,
      "epoch": 6.137064586357039,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019863968331521286,
      "loss": 4.9159,
      "mean_token_accuracy": 0.22065474092960358,
      "num_tokens": 140394161.0,
      "step": 67655
    },
    {
      "entropy": 5.758952474594116,
      "epoch": 6.137518142235123,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019860924804984998,
      "loss": 4.833,
      "mean_token_accuracy": 0.23548270612955094,
      "num_tokens": 140404938.0,
      "step": 67660
    },
    {
      "entropy": 5.702709197998047,
      "epoch": 6.1379716981132075,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019857881436426733,
      "loss": 4.7814,
      "mean_token_accuracy": 0.23564815223217012,
      "num_tokens": 140414569.0,
      "step": 67665
    },
    {
      "entropy": 5.646749210357666,
      "epoch": 6.138425253991292,
      "grad_norm": 1.375,
      "learning_rate": 0.00019854838225909422,
      "loss": 4.7341,
      "mean_token_accuracy": 0.2400654599070549,
      "num_tokens": 140424501.0,
      "step": 67670
    },
    {
      "entropy": 5.666451072692871,
      "epoch": 6.138878809869376,
      "grad_norm": 1.375,
      "learning_rate": 0.00019851795173496007,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23614473193883895,
      "num_tokens": 140435357.0,
      "step": 67675
    },
    {
      "entropy": 5.659175062179566,
      "epoch": 6.13933236574746,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019848752279249415,
      "loss": 4.7541,
      "mean_token_accuracy": 0.23763150423765184,
      "num_tokens": 140445125.0,
      "step": 67680
    },
    {
      "entropy": 5.817374229431152,
      "epoch": 6.139785921625545,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001984570954323258,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22815082520246505,
      "num_tokens": 140456024.0,
      "step": 67685
    },
    {
      "entropy": 5.70125675201416,
      "epoch": 6.140239477503629,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019842666965508415,
      "loss": 4.881,
      "mean_token_accuracy": 0.22888649851083756,
      "num_tokens": 140466356.0,
      "step": 67690
    },
    {
      "entropy": 5.664522647857666,
      "epoch": 6.140693033381712,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019839624546139856,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23466759473085402,
      "num_tokens": 140475560.0,
      "step": 67695
    },
    {
      "entropy": 5.62443962097168,
      "epoch": 6.1411465892597965,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019836582285189807,
      "loss": 4.7308,
      "mean_token_accuracy": 0.2426290199160576,
      "num_tokens": 140485521.0,
      "step": 67700
    },
    {
      "entropy": 5.751557683944702,
      "epoch": 6.141600145137881,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000198335401827212,
      "loss": 4.9155,
      "mean_token_accuracy": 0.2244908407330513,
      "num_tokens": 140495617.0,
      "step": 67705
    },
    {
      "entropy": 5.721890783309936,
      "epoch": 6.142053701015965,
      "grad_norm": 1.625,
      "learning_rate": 0.00019830498238796924,
      "loss": 4.7987,
      "mean_token_accuracy": 0.22949241548776628,
      "num_tokens": 140505760.0,
      "step": 67710
    },
    {
      "entropy": 5.766033601760864,
      "epoch": 6.142507256894049,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019827456453479903,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2295863449573517,
      "num_tokens": 140517135.0,
      "step": 67715
    },
    {
      "entropy": 5.7859130859375,
      "epoch": 6.142960812772134,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019824414826833037,
      "loss": 4.8536,
      "mean_token_accuracy": 0.23198671042919158,
      "num_tokens": 140527171.0,
      "step": 67720
    },
    {
      "entropy": 5.676467514038086,
      "epoch": 6.143414368650218,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019821373358919225,
      "loss": 4.7836,
      "mean_token_accuracy": 0.2328723669052124,
      "num_tokens": 140537871.0,
      "step": 67725
    },
    {
      "entropy": 5.679934787750244,
      "epoch": 6.143867924528302,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019818332049801373,
      "loss": 4.853,
      "mean_token_accuracy": 0.23492381721735,
      "num_tokens": 140547649.0,
      "step": 67730
    },
    {
      "entropy": 5.683159065246582,
      "epoch": 6.144321480406386,
      "grad_norm": 1.375,
      "learning_rate": 0.00019815290899542377,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23808281868696213,
      "num_tokens": 140557506.0,
      "step": 67735
    },
    {
      "entropy": 5.651753854751587,
      "epoch": 6.144775036284471,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001981224990820511,
      "loss": 4.7838,
      "mean_token_accuracy": 0.23828888088464736,
      "num_tokens": 140567378.0,
      "step": 67740
    },
    {
      "entropy": 5.730645561218262,
      "epoch": 6.145228592162554,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019809209075852485,
      "loss": 4.8042,
      "mean_token_accuracy": 0.2360678568482399,
      "num_tokens": 140578342.0,
      "step": 67745
    },
    {
      "entropy": 5.690159463882447,
      "epoch": 6.145682148040638,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019806168402547374,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22976778745651244,
      "num_tokens": 140588280.0,
      "step": 67750
    },
    {
      "entropy": 5.6934887886047365,
      "epoch": 6.1461357039187225,
      "grad_norm": 1.5,
      "learning_rate": 0.0001980312788835266,
      "loss": 4.7656,
      "mean_token_accuracy": 0.23820333182811737,
      "num_tokens": 140597789.0,
      "step": 67755
    },
    {
      "entropy": 5.663674926757812,
      "epoch": 6.146589259796807,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019800087533331222,
      "loss": 4.7556,
      "mean_token_accuracy": 0.23523871898651122,
      "num_tokens": 140607994.0,
      "step": 67760
    },
    {
      "entropy": 5.676571846008301,
      "epoch": 6.147042815674891,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019797047337545936,
      "loss": 4.8087,
      "mean_token_accuracy": 0.23321864753961563,
      "num_tokens": 140617989.0,
      "step": 67765
    },
    {
      "entropy": 5.766951227188111,
      "epoch": 6.147496371552975,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001979400730105968,
      "loss": 4.9859,
      "mean_token_accuracy": 0.21809102445840836,
      "num_tokens": 140629558.0,
      "step": 67770
    },
    {
      "entropy": 5.6967072010040285,
      "epoch": 6.14794992743106,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001979096742393531,
      "loss": 4.7649,
      "mean_token_accuracy": 0.24102618992328645,
      "num_tokens": 140639356.0,
      "step": 67775
    },
    {
      "entropy": 5.602591228485108,
      "epoch": 6.148403483309144,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019787927706235698,
      "loss": 4.6881,
      "mean_token_accuracy": 0.24634842574596405,
      "num_tokens": 140649821.0,
      "step": 67780
    },
    {
      "entropy": 5.673313617706299,
      "epoch": 6.148857039187228,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019784888148023706,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23493805825710296,
      "num_tokens": 140661341.0,
      "step": 67785
    },
    {
      "entropy": 5.697390413284301,
      "epoch": 6.149310595065312,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019781848749362196,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23323246836662292,
      "num_tokens": 140672011.0,
      "step": 67790
    },
    {
      "entropy": 5.77291841506958,
      "epoch": 6.149764150943396,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019778809510314016,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22648840993642808,
      "num_tokens": 140682155.0,
      "step": 67795
    },
    {
      "entropy": 5.667986345291138,
      "epoch": 6.15021770682148,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001977577043094203,
      "loss": 4.8308,
      "mean_token_accuracy": 0.2293156549334526,
      "num_tokens": 140692758.0,
      "step": 67800
    },
    {
      "entropy": 5.759743881225586,
      "epoch": 6.150671262699564,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019772731511309079,
      "loss": 4.8845,
      "mean_token_accuracy": 0.23234996646642686,
      "num_tokens": 140702385.0,
      "step": 67805
    },
    {
      "entropy": 5.783350610733033,
      "epoch": 6.151124818577649,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019769692751478007,
      "loss": 4.9241,
      "mean_token_accuracy": 0.22266885936260222,
      "num_tokens": 140713857.0,
      "step": 67810
    },
    {
      "entropy": 5.740213346481323,
      "epoch": 6.151578374455733,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001976665415151166,
      "loss": 4.8721,
      "mean_token_accuracy": 0.22646019607782364,
      "num_tokens": 140725406.0,
      "step": 67815
    },
    {
      "entropy": 5.709455919265747,
      "epoch": 6.152031930333817,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001976361571147287,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23136508464813232,
      "num_tokens": 140735365.0,
      "step": 67820
    },
    {
      "entropy": 5.707433700561523,
      "epoch": 6.152485486211901,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019760577431424486,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23559532761573793,
      "num_tokens": 140745581.0,
      "step": 67825
    },
    {
      "entropy": 5.720851421356201,
      "epoch": 6.152939042089986,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019757539311429328,
      "loss": 4.8691,
      "mean_token_accuracy": 0.23162650167942048,
      "num_tokens": 140755528.0,
      "step": 67830
    },
    {
      "entropy": 5.754171228408813,
      "epoch": 6.15339259796807,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019754501351550236,
      "loss": 4.8926,
      "mean_token_accuracy": 0.23095309287309645,
      "num_tokens": 140766892.0,
      "step": 67835
    },
    {
      "entropy": 5.772802734375,
      "epoch": 6.153846153846154,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019751463551850024,
      "loss": 4.9289,
      "mean_token_accuracy": 0.23159560114145278,
      "num_tokens": 140777327.0,
      "step": 67840
    },
    {
      "entropy": 5.748111200332642,
      "epoch": 6.1542997097242385,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019748425912391522,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22576113641262055,
      "num_tokens": 140787555.0,
      "step": 67845
    },
    {
      "entropy": 5.792949914932251,
      "epoch": 6.154753265602322,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001974538843323755,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22763627022504807,
      "num_tokens": 140797058.0,
      "step": 67850
    },
    {
      "entropy": 5.6615832328796385,
      "epoch": 6.155206821480406,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001974235111445092,
      "loss": 4.747,
      "mean_token_accuracy": 0.24063193202018737,
      "num_tokens": 140808420.0,
      "step": 67855
    },
    {
      "entropy": 5.638969373703003,
      "epoch": 6.15566037735849,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001973931395609444,
      "loss": 4.7345,
      "mean_token_accuracy": 0.2424511879682541,
      "num_tokens": 140818224.0,
      "step": 67860
    },
    {
      "entropy": 5.63839111328125,
      "epoch": 6.156113933236575,
      "grad_norm": 1.375,
      "learning_rate": 0.0001973627695823093,
      "loss": 4.796,
      "mean_token_accuracy": 0.2386150375008583,
      "num_tokens": 140827457.0,
      "step": 67865
    },
    {
      "entropy": 5.7508384704589846,
      "epoch": 6.156567489114659,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001973324012092319,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23667736649513244,
      "num_tokens": 140837675.0,
      "step": 67870
    },
    {
      "entropy": 5.717095184326172,
      "epoch": 6.157021044992743,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019730203444234029,
      "loss": 4.8006,
      "mean_token_accuracy": 0.24159766584634781,
      "num_tokens": 140847621.0,
      "step": 67875
    },
    {
      "entropy": 5.719616842269898,
      "epoch": 6.157474600870827,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019727166928226237,
      "loss": 4.8289,
      "mean_token_accuracy": 0.23638182878494263,
      "num_tokens": 140858193.0,
      "step": 67880
    },
    {
      "entropy": 5.655114698410034,
      "epoch": 6.157928156748912,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019724130572962612,
      "loss": 4.8173,
      "mean_token_accuracy": 0.2350546360015869,
      "num_tokens": 140868472.0,
      "step": 67885
    },
    {
      "entropy": 5.6798765659332275,
      "epoch": 6.158381712626996,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019721094378505948,
      "loss": 4.8025,
      "mean_token_accuracy": 0.23350827395915985,
      "num_tokens": 140878409.0,
      "step": 67890
    },
    {
      "entropy": 5.6629781246185305,
      "epoch": 6.15883526850508,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001971805834491904,
      "loss": 4.743,
      "mean_token_accuracy": 0.24071122854948043,
      "num_tokens": 140889724.0,
      "step": 67895
    },
    {
      "entropy": 5.671080875396728,
      "epoch": 6.159288824383164,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001971502247226466,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2391919180750847,
      "num_tokens": 140899603.0,
      "step": 67900
    },
    {
      "entropy": 5.699382162094116,
      "epoch": 6.159742380261248,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019711986760605606,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23631747663021088,
      "num_tokens": 140909628.0,
      "step": 67905
    },
    {
      "entropy": 5.684098196029663,
      "epoch": 6.160195936139332,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019708951210004643,
      "loss": 4.82,
      "mean_token_accuracy": 0.23619365841150283,
      "num_tokens": 140920023.0,
      "step": 67910
    },
    {
      "entropy": 5.73544020652771,
      "epoch": 6.160649492017416,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019705915820524561,
      "loss": 4.9343,
      "mean_token_accuracy": 0.22737802416086197,
      "num_tokens": 140930522.0,
      "step": 67915
    },
    {
      "entropy": 5.7185265064239506,
      "epoch": 6.161103047895501,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019702880592228122,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23584057837724687,
      "num_tokens": 140939660.0,
      "step": 67920
    },
    {
      "entropy": 5.7358465671539305,
      "epoch": 6.161556603773585,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019699845525178106,
      "loss": 4.8902,
      "mean_token_accuracy": 0.22590768784284593,
      "num_tokens": 140949807.0,
      "step": 67925
    },
    {
      "entropy": 5.7249712467193605,
      "epoch": 6.162010159651669,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019696810619437266,
      "loss": 4.8792,
      "mean_token_accuracy": 0.22988710999488832,
      "num_tokens": 140960155.0,
      "step": 67930
    },
    {
      "entropy": 5.787154197692871,
      "epoch": 6.1624637155297535,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001969377587506837,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22770071625709534,
      "num_tokens": 140970281.0,
      "step": 67935
    },
    {
      "entropy": 5.729632043838501,
      "epoch": 6.162917271407838,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019690741292134186,
      "loss": 4.82,
      "mean_token_accuracy": 0.23551400005817413,
      "num_tokens": 140980518.0,
      "step": 67940
    },
    {
      "entropy": 5.735444974899292,
      "epoch": 6.163370827285922,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019687706870697452,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22797603160142899,
      "num_tokens": 140991219.0,
      "step": 67945
    },
    {
      "entropy": 5.780756187438965,
      "epoch": 6.163824383164005,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019684672610820948,
      "loss": 5.0021,
      "mean_token_accuracy": 0.21900466978549957,
      "num_tokens": 141001308.0,
      "step": 67950
    },
    {
      "entropy": 5.700767230987549,
      "epoch": 6.16427793904209,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001968163851256739,
      "loss": 4.8515,
      "mean_token_accuracy": 0.2266002357006073,
      "num_tokens": 141012305.0,
      "step": 67955
    },
    {
      "entropy": 5.724942016601562,
      "epoch": 6.164731494920174,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001967860457599954,
      "loss": 4.7728,
      "mean_token_accuracy": 0.24012479186058044,
      "num_tokens": 141023374.0,
      "step": 67960
    },
    {
      "entropy": 5.681568670272827,
      "epoch": 6.165185050798258,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019675570801180149,
      "loss": 4.841,
      "mean_token_accuracy": 0.2334829568862915,
      "num_tokens": 141035056.0,
      "step": 67965
    },
    {
      "entropy": 5.645052766799926,
      "epoch": 6.1656386066763424,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001967253718817194,
      "loss": 4.7783,
      "mean_token_accuracy": 0.23755111694335937,
      "num_tokens": 141045721.0,
      "step": 67970
    },
    {
      "entropy": 5.705559253692627,
      "epoch": 6.166092162554427,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019669503737037658,
      "loss": 4.8657,
      "mean_token_accuracy": 0.22590356171131135,
      "num_tokens": 141057078.0,
      "step": 67975
    },
    {
      "entropy": 5.742174100875855,
      "epoch": 6.166545718432511,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019666470447840035,
      "loss": 4.8361,
      "mean_token_accuracy": 0.2346130132675171,
      "num_tokens": 141067059.0,
      "step": 67980
    },
    {
      "entropy": 5.835783576965332,
      "epoch": 6.166999274310595,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019663437320641803,
      "loss": 4.9756,
      "mean_token_accuracy": 0.22841860949993134,
      "num_tokens": 141078299.0,
      "step": 67985
    },
    {
      "entropy": 5.737079000473022,
      "epoch": 6.1674528301886795,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001966040435550568,
      "loss": 4.8113,
      "mean_token_accuracy": 0.22890915721654892,
      "num_tokens": 141088455.0,
      "step": 67990
    },
    {
      "entropy": 5.740402555465698,
      "epoch": 6.167906386066764,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019657371552494397,
      "loss": 4.8556,
      "mean_token_accuracy": 0.225116591155529,
      "num_tokens": 141098660.0,
      "step": 67995
    },
    {
      "entropy": 5.693703651428223,
      "epoch": 6.168359941944848,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019654338911670666,
      "loss": 4.7622,
      "mean_token_accuracy": 0.24063196927309036,
      "num_tokens": 141109226.0,
      "step": 68000
    },
    {
      "entropy": 5.781810474395752,
      "epoch": 6.168813497822931,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019651306433097205,
      "loss": 4.9471,
      "mean_token_accuracy": 0.2233913615345955,
      "num_tokens": 141120702.0,
      "step": 68005
    },
    {
      "entropy": 5.695836544036865,
      "epoch": 6.169267053701016,
      "grad_norm": 1.5,
      "learning_rate": 0.00019648274116836723,
      "loss": 4.864,
      "mean_token_accuracy": 0.22480394542217255,
      "num_tokens": 141131562.0,
      "step": 68010
    },
    {
      "entropy": 5.816996431350708,
      "epoch": 6.1697206095791,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019645241962951937,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22671701461076738,
      "num_tokens": 141141552.0,
      "step": 68015
    },
    {
      "entropy": 5.762571907043457,
      "epoch": 6.170174165457184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019642209971505555,
      "loss": 4.809,
      "mean_token_accuracy": 0.2347367912530899,
      "num_tokens": 141151731.0,
      "step": 68020
    },
    {
      "entropy": 5.637078475952149,
      "epoch": 6.1706277213352685,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001963917814256026,
      "loss": 4.7568,
      "mean_token_accuracy": 0.23689395040273667,
      "num_tokens": 141161654.0,
      "step": 68025
    },
    {
      "entropy": 5.6910400390625,
      "epoch": 6.171081277213353,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019636146476178774,
      "loss": 4.8391,
      "mean_token_accuracy": 0.23392308205366136,
      "num_tokens": 141172305.0,
      "step": 68030
    },
    {
      "entropy": 5.655898857116699,
      "epoch": 6.171534833091437,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001963311497242378,
      "loss": 4.8526,
      "mean_token_accuracy": 0.22694904804229737,
      "num_tokens": 141184422.0,
      "step": 68035
    },
    {
      "entropy": 5.747969770431519,
      "epoch": 6.171988388969521,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019630083631357975,
      "loss": 4.8405,
      "mean_token_accuracy": 0.2324291691184044,
      "num_tokens": 141194933.0,
      "step": 68040
    },
    {
      "entropy": 5.818314456939698,
      "epoch": 6.1724419448476056,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019627052453044047,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2357236459851265,
      "num_tokens": 141204868.0,
      "step": 68045
    },
    {
      "entropy": 5.67916350364685,
      "epoch": 6.17289550072569,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019624021437544676,
      "loss": 4.75,
      "mean_token_accuracy": 0.2342206969857216,
      "num_tokens": 141214953.0,
      "step": 68050
    },
    {
      "entropy": 5.686773777008057,
      "epoch": 6.173349056603773,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019620990584922555,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2295063003897667,
      "num_tokens": 141224727.0,
      "step": 68055
    },
    {
      "entropy": 5.7215108394622805,
      "epoch": 6.1738026124818575,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019617959895240346,
      "loss": 4.8544,
      "mean_token_accuracy": 0.2364455297589302,
      "num_tokens": 141234967.0,
      "step": 68060
    },
    {
      "entropy": 5.723119926452637,
      "epoch": 6.174256168359942,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019614929368560742,
      "loss": 4.9009,
      "mean_token_accuracy": 0.2247968003153801,
      "num_tokens": 141244993.0,
      "step": 68065
    },
    {
      "entropy": 5.644999599456787,
      "epoch": 6.174709724238026,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001961189900494641,
      "loss": 4.7738,
      "mean_token_accuracy": 0.24255088716745377,
      "num_tokens": 141254785.0,
      "step": 68070
    },
    {
      "entropy": 5.703164386749267,
      "epoch": 6.17516328011611,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001960886880446001,
      "loss": 4.8094,
      "mean_token_accuracy": 0.23417950421571732,
      "num_tokens": 141263951.0,
      "step": 68075
    },
    {
      "entropy": 5.722254657745362,
      "epoch": 6.1756168359941945,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001960583876716422,
      "loss": 4.7817,
      "mean_token_accuracy": 0.23759575933218002,
      "num_tokens": 141274528.0,
      "step": 68080
    },
    {
      "entropy": 5.687105131149292,
      "epoch": 6.176070391872279,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019602808893121694,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23144396543502807,
      "num_tokens": 141285188.0,
      "step": 68085
    },
    {
      "entropy": 5.673144721984864,
      "epoch": 6.176523947750363,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019599779182395094,
      "loss": 4.7335,
      "mean_token_accuracy": 0.24056257605552672,
      "num_tokens": 141295687.0,
      "step": 68090
    },
    {
      "entropy": 5.702981090545654,
      "epoch": 6.176977503628447,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019596749635047072,
      "loss": 4.8534,
      "mean_token_accuracy": 0.2315269261598587,
      "num_tokens": 141306458.0,
      "step": 68095
    },
    {
      "entropy": 5.7136742115020756,
      "epoch": 6.177431059506532,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019593720251140278,
      "loss": 4.861,
      "mean_token_accuracy": 0.22979894876480103,
      "num_tokens": 141317542.0,
      "step": 68100
    },
    {
      "entropy": 5.662292432785034,
      "epoch": 6.177884615384615,
      "grad_norm": 1.375,
      "learning_rate": 0.00019590691030737373,
      "loss": 4.8306,
      "mean_token_accuracy": 0.22992719411849977,
      "num_tokens": 141328358.0,
      "step": 68105
    },
    {
      "entropy": 5.772874927520752,
      "epoch": 6.178338171262699,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019587661973900983,
      "loss": 4.9188,
      "mean_token_accuracy": 0.2265625238418579,
      "num_tokens": 141338403.0,
      "step": 68110
    },
    {
      "entropy": 5.781896114349365,
      "epoch": 6.1787917271407835,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019584633080693765,
      "loss": 4.8893,
      "mean_token_accuracy": 0.2319344237446785,
      "num_tokens": 141349601.0,
      "step": 68115
    },
    {
      "entropy": 5.734843921661377,
      "epoch": 6.179245283018868,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001958160435117835,
      "loss": 4.8446,
      "mean_token_accuracy": 0.23563073724508285,
      "num_tokens": 141359505.0,
      "step": 68120
    },
    {
      "entropy": 5.754593324661255,
      "epoch": 6.179698838896952,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001957857578541738,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23031049072742463,
      "num_tokens": 141368785.0,
      "step": 68125
    },
    {
      "entropy": 5.729844665527343,
      "epoch": 6.180152394775036,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019575547383473475,
      "loss": 4.678,
      "mean_token_accuracy": 0.24696976840496063,
      "num_tokens": 141378079.0,
      "step": 68130
    },
    {
      "entropy": 5.682519483566284,
      "epoch": 6.180605950653121,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019572519145409273,
      "loss": 4.7794,
      "mean_token_accuracy": 0.23695330172777176,
      "num_tokens": 141389032.0,
      "step": 68135
    },
    {
      "entropy": 5.71339259147644,
      "epoch": 6.181059506531205,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019569491071287398,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22833989411592484,
      "num_tokens": 141398914.0,
      "step": 68140
    },
    {
      "entropy": 5.702907371520996,
      "epoch": 6.181513062409289,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001956646316117046,
      "loss": 4.7869,
      "mean_token_accuracy": 0.23295214027166367,
      "num_tokens": 141408363.0,
      "step": 68145
    },
    {
      "entropy": 5.746046447753907,
      "epoch": 6.181966618287373,
      "grad_norm": 1.5,
      "learning_rate": 0.00019563435415121088,
      "loss": 4.8473,
      "mean_token_accuracy": 0.2305851086974144,
      "num_tokens": 141418308.0,
      "step": 68150
    },
    {
      "entropy": 5.7702985286712645,
      "epoch": 6.182420174165458,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019560407833201903,
      "loss": 4.9301,
      "mean_token_accuracy": 0.228873810172081,
      "num_tokens": 141428599.0,
      "step": 68155
    },
    {
      "entropy": 5.732838201522827,
      "epoch": 6.182873730043541,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019557380415475496,
      "loss": 4.8505,
      "mean_token_accuracy": 0.2291821211576462,
      "num_tokens": 141438813.0,
      "step": 68160
    },
    {
      "entropy": 5.718004274368286,
      "epoch": 6.183327285921625,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019554353162004495,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22918113321065903,
      "num_tokens": 141449601.0,
      "step": 68165
    },
    {
      "entropy": 5.759047317504883,
      "epoch": 6.1837808417997095,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001955132607285149,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22376090586185454,
      "num_tokens": 141461275.0,
      "step": 68170
    },
    {
      "entropy": 5.697601175308227,
      "epoch": 6.184234397677794,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001954829914807909,
      "loss": 4.7895,
      "mean_token_accuracy": 0.23324380815029144,
      "num_tokens": 141471897.0,
      "step": 68175
    },
    {
      "entropy": 5.660521602630615,
      "epoch": 6.184687953555878,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019545272387749886,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2378561317920685,
      "num_tokens": 141481527.0,
      "step": 68180
    },
    {
      "entropy": 5.758808851242065,
      "epoch": 6.185141509433962,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019542245791926476,
      "loss": 4.9578,
      "mean_token_accuracy": 0.21714886724948884,
      "num_tokens": 141492060.0,
      "step": 68185
    },
    {
      "entropy": 5.655472040176392,
      "epoch": 6.185595065312047,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019539219360671455,
      "loss": 4.7868,
      "mean_token_accuracy": 0.2364722490310669,
      "num_tokens": 141502046.0,
      "step": 68190
    },
    {
      "entropy": 5.607492828369141,
      "epoch": 6.186048621190131,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019536193094047406,
      "loss": 4.7132,
      "mean_token_accuracy": 0.243633633852005,
      "num_tokens": 141511635.0,
      "step": 68195
    },
    {
      "entropy": 5.709384155273438,
      "epoch": 6.186502177068215,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019533166992116913,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22887132614850997,
      "num_tokens": 141520947.0,
      "step": 68200
    },
    {
      "entropy": 5.647162914276123,
      "epoch": 6.186955732946299,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019530141054942558,
      "loss": 4.7822,
      "mean_token_accuracy": 0.23507262766361237,
      "num_tokens": 141531098.0,
      "step": 68205
    },
    {
      "entropy": 5.763899612426758,
      "epoch": 6.187409288824383,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019527115282586916,
      "loss": 4.9811,
      "mean_token_accuracy": 0.22273797988891603,
      "num_tokens": 141542992.0,
      "step": 68210
    },
    {
      "entropy": 5.741075849533081,
      "epoch": 6.187862844702467,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019524089675112554,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22697626948356628,
      "num_tokens": 141553240.0,
      "step": 68215
    },
    {
      "entropy": 5.689839124679565,
      "epoch": 6.188316400580551,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001952106423258206,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23422041088342666,
      "num_tokens": 141564074.0,
      "step": 68220
    },
    {
      "entropy": 5.711842060089111,
      "epoch": 6.188769956458636,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019518038955057987,
      "loss": 4.789,
      "mean_token_accuracy": 0.23355743587017058,
      "num_tokens": 141574377.0,
      "step": 68225
    },
    {
      "entropy": 5.843049621582031,
      "epoch": 6.18922351233672,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019515013842602902,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21721508353948593,
      "num_tokens": 141584010.0,
      "step": 68230
    },
    {
      "entropy": 5.705858945846558,
      "epoch": 6.189677068214804,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001951198889527937,
      "loss": 4.7699,
      "mean_token_accuracy": 0.2355535387992859,
      "num_tokens": 141593820.0,
      "step": 68235
    },
    {
      "entropy": 5.683940029144287,
      "epoch": 6.190130624092888,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019508964113149936,
      "loss": 4.8265,
      "mean_token_accuracy": 0.23097802847623825,
      "num_tokens": 141603977.0,
      "step": 68240
    },
    {
      "entropy": 5.6724278926849365,
      "epoch": 6.190584179970973,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001950593949627717,
      "loss": 4.7773,
      "mean_token_accuracy": 0.23877708613872528,
      "num_tokens": 141614008.0,
      "step": 68245
    },
    {
      "entropy": 5.673081159591675,
      "epoch": 6.191037735849057,
      "grad_norm": 1.5,
      "learning_rate": 0.00019502915044723601,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2354784518480301,
      "num_tokens": 141625387.0,
      "step": 68250
    },
    {
      "entropy": 5.7263885021209715,
      "epoch": 6.191491291727141,
      "grad_norm": 1.375,
      "learning_rate": 0.0001949989075855179,
      "loss": 4.879,
      "mean_token_accuracy": 0.2210407793521881,
      "num_tokens": 141635904.0,
      "step": 68255
    },
    {
      "entropy": 5.721313333511352,
      "epoch": 6.191944847605225,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019496866637824285,
      "loss": 4.8329,
      "mean_token_accuracy": 0.23568331450223923,
      "num_tokens": 141647506.0,
      "step": 68260
    },
    {
      "entropy": 5.712257814407349,
      "epoch": 6.192398403483309,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019493842682603607,
      "loss": 4.915,
      "mean_token_accuracy": 0.2311409294605255,
      "num_tokens": 141660070.0,
      "step": 68265
    },
    {
      "entropy": 5.745078229904175,
      "epoch": 6.192851959361393,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001949081889295231,
      "loss": 4.8818,
      "mean_token_accuracy": 0.23209353238344194,
      "num_tokens": 141670833.0,
      "step": 68270
    },
    {
      "entropy": 5.727519845962524,
      "epoch": 6.193305515239477,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019487795268932918,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23469393849372863,
      "num_tokens": 141680426.0,
      "step": 68275
    },
    {
      "entropy": 5.691016960144043,
      "epoch": 6.193759071117562,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019484771810607954,
      "loss": 4.7828,
      "mean_token_accuracy": 0.2399241089820862,
      "num_tokens": 141691064.0,
      "step": 68280
    },
    {
      "entropy": 5.622486925125122,
      "epoch": 6.194212626995646,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001948174851803996,
      "loss": 4.7607,
      "mean_token_accuracy": 0.23397320806980132,
      "num_tokens": 141700852.0,
      "step": 68285
    },
    {
      "entropy": 5.696832752227783,
      "epoch": 6.19466618287373,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019478725391291452,
      "loss": 4.7925,
      "mean_token_accuracy": 0.24130522608757018,
      "num_tokens": 141710686.0,
      "step": 68290
    },
    {
      "entropy": 5.642184209823609,
      "epoch": 6.195119738751814,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019475702430424941,
      "loss": 4.772,
      "mean_token_accuracy": 0.24132533371448517,
      "num_tokens": 141721004.0,
      "step": 68295
    },
    {
      "entropy": 5.749057292938232,
      "epoch": 6.195573294629899,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001947267963550295,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22559824585914612,
      "num_tokens": 141732052.0,
      "step": 68300
    },
    {
      "entropy": 5.7883992195129395,
      "epoch": 6.196026850507983,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019469657006587987,
      "loss": 4.8754,
      "mean_token_accuracy": 0.23768904209136962,
      "num_tokens": 141743205.0,
      "step": 68305
    },
    {
      "entropy": 5.696283149719238,
      "epoch": 6.196480406386066,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019466634543742568,
      "loss": 4.7379,
      "mean_token_accuracy": 0.2412893518805504,
      "num_tokens": 141753270.0,
      "step": 68310
    },
    {
      "entropy": 5.694037532806396,
      "epoch": 6.196933962264151,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001946361224702919,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22904345840215684,
      "num_tokens": 141763421.0,
      "step": 68315
    },
    {
      "entropy": 5.685874176025391,
      "epoch": 6.197387518142235,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019460590116510357,
      "loss": 4.8005,
      "mean_token_accuracy": 0.2382381811738014,
      "num_tokens": 141772916.0,
      "step": 68320
    },
    {
      "entropy": 5.739017677307129,
      "epoch": 6.197841074020319,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019457568152248577,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2349826768040657,
      "num_tokens": 141783641.0,
      "step": 68325
    },
    {
      "entropy": 5.688175201416016,
      "epoch": 6.198294629898403,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019454546354306335,
      "loss": 4.7935,
      "mean_token_accuracy": 0.23900321125984192,
      "num_tokens": 141796413.0,
      "step": 68330
    },
    {
      "entropy": 5.641474676132202,
      "epoch": 6.198748185776488,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019451524722746116,
      "loss": 4.7362,
      "mean_token_accuracy": 0.24057557433843613,
      "num_tokens": 141805869.0,
      "step": 68335
    },
    {
      "entropy": 5.753029632568359,
      "epoch": 6.199201741654572,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019448503257630428,
      "loss": 4.8357,
      "mean_token_accuracy": 0.23056997805833818,
      "num_tokens": 141814766.0,
      "step": 68340
    },
    {
      "entropy": 5.631391954421997,
      "epoch": 6.199655297532656,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019445481959021744,
      "loss": 4.7929,
      "mean_token_accuracy": 0.2296081528067589,
      "num_tokens": 141825758.0,
      "step": 68345
    },
    {
      "entropy": 5.6948915958404545,
      "epoch": 6.2001088534107405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019442460826982533,
      "loss": 4.857,
      "mean_token_accuracy": 0.23413368314504623,
      "num_tokens": 141836825.0,
      "step": 68350
    },
    {
      "entropy": 5.664844036102295,
      "epoch": 6.200562409288825,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019439439861575292,
      "loss": 4.8199,
      "mean_token_accuracy": 0.22935360968112944,
      "num_tokens": 141847603.0,
      "step": 68355
    },
    {
      "entropy": 5.70045952796936,
      "epoch": 6.201015965166908,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019436419062862493,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22949670255184174,
      "num_tokens": 141857582.0,
      "step": 68360
    },
    {
      "entropy": 5.753337764739991,
      "epoch": 6.201469521044992,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019433398430906596,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22631898522377014,
      "num_tokens": 141868018.0,
      "step": 68365
    },
    {
      "entropy": 5.752689218521118,
      "epoch": 6.201923076923077,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019430377965770084,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2297607883810997,
      "num_tokens": 141879135.0,
      "step": 68370
    },
    {
      "entropy": 5.719089078903198,
      "epoch": 6.202376632801161,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001942735766751541,
      "loss": 4.817,
      "mean_token_accuracy": 0.2379404827952385,
      "num_tokens": 141889734.0,
      "step": 68375
    },
    {
      "entropy": 5.632276821136474,
      "epoch": 6.202830188679245,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001942433753620504,
      "loss": 4.7353,
      "mean_token_accuracy": 0.24260735213756562,
      "num_tokens": 141900665.0,
      "step": 68380
    },
    {
      "entropy": 5.71610951423645,
      "epoch": 6.2032837445573294,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001942131757190142,
      "loss": 4.8385,
      "mean_token_accuracy": 0.23892322778701783,
      "num_tokens": 141911620.0,
      "step": 68385
    },
    {
      "entropy": 5.762193107604981,
      "epoch": 6.203737300435414,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001941829777466701,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22952163070440293,
      "num_tokens": 141923289.0,
      "step": 68390
    },
    {
      "entropy": 5.711341905593872,
      "epoch": 6.204190856313498,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019415278144564263,
      "loss": 4.826,
      "mean_token_accuracy": 0.2349855825304985,
      "num_tokens": 141934034.0,
      "step": 68395
    },
    {
      "entropy": 5.676509141921997,
      "epoch": 6.204644412191582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019412258681655625,
      "loss": 4.7579,
      "mean_token_accuracy": 0.24480770975351335,
      "num_tokens": 141944581.0,
      "step": 68400
    },
    {
      "entropy": 5.636684989929199,
      "epoch": 6.2050979680696665,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019409239386003542,
      "loss": 4.744,
      "mean_token_accuracy": 0.24219780564308166,
      "num_tokens": 141955825.0,
      "step": 68405
    },
    {
      "entropy": 5.699303245544433,
      "epoch": 6.205551523947751,
      "grad_norm": 1.375,
      "learning_rate": 0.00019406220257670443,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23284305334091188,
      "num_tokens": 141966818.0,
      "step": 68410
    },
    {
      "entropy": 5.654456996917725,
      "epoch": 6.206005079825834,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019403201296718775,
      "loss": 4.8587,
      "mean_token_accuracy": 0.23098514080047608,
      "num_tokens": 141977211.0,
      "step": 68415
    },
    {
      "entropy": 5.716504383087158,
      "epoch": 6.206458635703918,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019400182503210973,
      "loss": 4.7953,
      "mean_token_accuracy": 0.23697117418050767,
      "num_tokens": 141988366.0,
      "step": 68420
    },
    {
      "entropy": 5.686081218719482,
      "epoch": 6.206912191582003,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019397163877209455,
      "loss": 4.8101,
      "mean_token_accuracy": 0.23375469893217088,
      "num_tokens": 141997915.0,
      "step": 68425
    },
    {
      "entropy": 5.67099084854126,
      "epoch": 6.207365747460087,
      "grad_norm": 1.375,
      "learning_rate": 0.0001939414541877665,
      "loss": 4.7729,
      "mean_token_accuracy": 0.2382734715938568,
      "num_tokens": 142009082.0,
      "step": 68430
    },
    {
      "entropy": 5.726237678527832,
      "epoch": 6.207819303338171,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001939112712797499,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23094429969787597,
      "num_tokens": 142020110.0,
      "step": 68435
    },
    {
      "entropy": 5.709772777557373,
      "epoch": 6.2082728592162555,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019388109004866882,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22792344987392427,
      "num_tokens": 142030926.0,
      "step": 68440
    },
    {
      "entropy": 5.714758634567261,
      "epoch": 6.20872641509434,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019385091049514748,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23364246040582656,
      "num_tokens": 142041207.0,
      "step": 68445
    },
    {
      "entropy": 5.720254278182983,
      "epoch": 6.209179970972424,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019382073261981004,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23172653168439866,
      "num_tokens": 142051044.0,
      "step": 68450
    },
    {
      "entropy": 5.744034957885742,
      "epoch": 6.209633526850508,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019379055642328052,
      "loss": 4.838,
      "mean_token_accuracy": 0.2325100153684616,
      "num_tokens": 142061303.0,
      "step": 68455
    },
    {
      "entropy": 5.710982036590576,
      "epoch": 6.2100870827285926,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019376038190618294,
      "loss": 4.836,
      "mean_token_accuracy": 0.22476520389318466,
      "num_tokens": 142071120.0,
      "step": 68460
    },
    {
      "entropy": 5.765153455734253,
      "epoch": 6.210540638606676,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001937302090691414,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22843344658613204,
      "num_tokens": 142082400.0,
      "step": 68465
    },
    {
      "entropy": 5.728313636779785,
      "epoch": 6.21099419448476,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019370003791277984,
      "loss": 4.8358,
      "mean_token_accuracy": 0.2302333191037178,
      "num_tokens": 142092720.0,
      "step": 68470
    },
    {
      "entropy": 5.6674905776977536,
      "epoch": 6.2114477503628445,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019366986843772223,
      "loss": 4.7624,
      "mean_token_accuracy": 0.24259959459304808,
      "num_tokens": 142103739.0,
      "step": 68475
    },
    {
      "entropy": 5.710055160522461,
      "epoch": 6.211901306240929,
      "grad_norm": 1.375,
      "learning_rate": 0.0001936397006445924,
      "loss": 4.8364,
      "mean_token_accuracy": 0.23496917635202408,
      "num_tokens": 142113781.0,
      "step": 68480
    },
    {
      "entropy": 5.726393318176269,
      "epoch": 6.212354862119013,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019360953453401438,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22787421345710754,
      "num_tokens": 142125551.0,
      "step": 68485
    },
    {
      "entropy": 5.722930955886841,
      "epoch": 6.212808417997097,
      "grad_norm": 1.734375,
      "learning_rate": 0.00019357937010661187,
      "loss": 4.806,
      "mean_token_accuracy": 0.23795555680990219,
      "num_tokens": 142135999.0,
      "step": 68490
    },
    {
      "entropy": 5.748543357849121,
      "epoch": 6.2132619738751815,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019354920736300874,
      "loss": 4.8496,
      "mean_token_accuracy": 0.23494156450033188,
      "num_tokens": 142146245.0,
      "step": 68495
    },
    {
      "entropy": 5.675063848495483,
      "epoch": 6.213715529753266,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019351904630382881,
      "loss": 4.845,
      "mean_token_accuracy": 0.22492608428001404,
      "num_tokens": 142156189.0,
      "step": 68500
    },
    {
      "entropy": 5.7343912601470945,
      "epoch": 6.21416908563135,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019348888692969574,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2313958704471588,
      "num_tokens": 142165997.0,
      "step": 68505
    },
    {
      "entropy": 5.720377254486084,
      "epoch": 6.214622641509434,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019345872924123327,
      "loss": 4.8481,
      "mean_token_accuracy": 0.22857255786657332,
      "num_tokens": 142176976.0,
      "step": 68510
    },
    {
      "entropy": 5.715360069274903,
      "epoch": 6.215076197387518,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019342857323906499,
      "loss": 4.8549,
      "mean_token_accuracy": 0.230117529630661,
      "num_tokens": 142188010.0,
      "step": 68515
    },
    {
      "entropy": 5.7572096824646,
      "epoch": 6.215529753265602,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019339841892381464,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2262549802660942,
      "num_tokens": 142198170.0,
      "step": 68520
    },
    {
      "entropy": 5.758052635192871,
      "epoch": 6.215983309143686,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019336826629610582,
      "loss": 4.8052,
      "mean_token_accuracy": 0.2328550010919571,
      "num_tokens": 142207534.0,
      "step": 68525
    },
    {
      "entropy": 5.737438154220581,
      "epoch": 6.2164368650217705,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000193338115356562,
      "loss": 4.8356,
      "mean_token_accuracy": 0.22800031304359436,
      "num_tokens": 142217752.0,
      "step": 68530
    },
    {
      "entropy": 5.724473762512207,
      "epoch": 6.216890420899855,
      "grad_norm": 1.5,
      "learning_rate": 0.0001933079661058068,
      "loss": 4.8201,
      "mean_token_accuracy": 0.2341575637459755,
      "num_tokens": 142227488.0,
      "step": 68535
    },
    {
      "entropy": 5.6862129211425785,
      "epoch": 6.217343976777939,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019327781854446368,
      "loss": 4.8341,
      "mean_token_accuracy": 0.23332665115594864,
      "num_tokens": 142237384.0,
      "step": 68540
    },
    {
      "entropy": 5.699624538421631,
      "epoch": 6.217797532656023,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019324767267315612,
      "loss": 4.8722,
      "mean_token_accuracy": 0.23043390214443207,
      "num_tokens": 142248588.0,
      "step": 68545
    },
    {
      "entropy": 5.773460817337036,
      "epoch": 6.218251088534108,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019321752849250746,
      "loss": 4.8712,
      "mean_token_accuracy": 0.23246156871318818,
      "num_tokens": 142259019.0,
      "step": 68550
    },
    {
      "entropy": 5.750768804550171,
      "epoch": 6.218704644412192,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019318738600314117,
      "loss": 4.8461,
      "mean_token_accuracy": 0.22985291182994844,
      "num_tokens": 142269154.0,
      "step": 68555
    },
    {
      "entropy": 5.725172662734986,
      "epoch": 6.219158200290276,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001931572452056806,
      "loss": 4.8466,
      "mean_token_accuracy": 0.23125855326652528,
      "num_tokens": 142279253.0,
      "step": 68560
    },
    {
      "entropy": 5.691882991790772,
      "epoch": 6.21961175616836,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000193127106100749,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23283152878284455,
      "num_tokens": 142289671.0,
      "step": 68565
    },
    {
      "entropy": 5.627656602859497,
      "epoch": 6.220065312046444,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001930969686889698,
      "loss": 4.7581,
      "mean_token_accuracy": 0.23926566541194916,
      "num_tokens": 142300716.0,
      "step": 68570
    },
    {
      "entropy": 5.689513301849365,
      "epoch": 6.220518867924528,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019306683297096617,
      "loss": 4.786,
      "mean_token_accuracy": 0.23062117397785187,
      "num_tokens": 142311144.0,
      "step": 68575
    },
    {
      "entropy": 5.736263942718506,
      "epoch": 6.220972423802612,
      "grad_norm": 1.375,
      "learning_rate": 0.0001930366989473612,
      "loss": 4.8583,
      "mean_token_accuracy": 0.2350495085120201,
      "num_tokens": 142320781.0,
      "step": 68580
    },
    {
      "entropy": 5.689935398101807,
      "epoch": 6.2214259796806965,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019300656661877825,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23524103164672852,
      "num_tokens": 142331197.0,
      "step": 68585
    },
    {
      "entropy": 5.69676194190979,
      "epoch": 6.221879535558781,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019297643598584042,
      "loss": 4.8095,
      "mean_token_accuracy": 0.22818985134363173,
      "num_tokens": 142341593.0,
      "step": 68590
    },
    {
      "entropy": 5.668300914764404,
      "epoch": 6.222333091436865,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019294630704917078,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23147343695163727,
      "num_tokens": 142351920.0,
      "step": 68595
    },
    {
      "entropy": 5.731163883209229,
      "epoch": 6.222786647314949,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019291617980939235,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22408971935510635,
      "num_tokens": 142361889.0,
      "step": 68600
    },
    {
      "entropy": 5.684784603118897,
      "epoch": 6.223240203193034,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019288605426712836,
      "loss": 4.773,
      "mean_token_accuracy": 0.23891256004571915,
      "num_tokens": 142372588.0,
      "step": 68605
    },
    {
      "entropy": 5.79763126373291,
      "epoch": 6.223693759071118,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019285593042300163,
      "loss": 4.937,
      "mean_token_accuracy": 0.22319513857364653,
      "num_tokens": 142382957.0,
      "step": 68610
    },
    {
      "entropy": 5.79462833404541,
      "epoch": 6.224147314949202,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019282580827763518,
      "loss": 4.8782,
      "mean_token_accuracy": 0.23282968997955322,
      "num_tokens": 142392928.0,
      "step": 68615
    },
    {
      "entropy": 5.727640438079834,
      "epoch": 6.2246008708272855,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001927956878316519,
      "loss": 4.8543,
      "mean_token_accuracy": 0.23660728484392166,
      "num_tokens": 142402788.0,
      "step": 68620
    },
    {
      "entropy": 5.738114690780639,
      "epoch": 6.22505442670537,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001927655690856749,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22795897126197814,
      "num_tokens": 142412791.0,
      "step": 68625
    },
    {
      "entropy": 5.660733652114868,
      "epoch": 6.225507982583454,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019273545204032678,
      "loss": 4.8337,
      "mean_token_accuracy": 0.22712952345609666,
      "num_tokens": 142423499.0,
      "step": 68630
    },
    {
      "entropy": 5.744814062118531,
      "epoch": 6.225961538461538,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019270533669623046,
      "loss": 4.8777,
      "mean_token_accuracy": 0.23122249096632003,
      "num_tokens": 142434055.0,
      "step": 68635
    },
    {
      "entropy": 5.696682071685791,
      "epoch": 6.226415094339623,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001926752230540088,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23273904919624328,
      "num_tokens": 142444501.0,
      "step": 68640
    },
    {
      "entropy": 5.763762903213501,
      "epoch": 6.226868650217707,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019264511111428446,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22832544893026352,
      "num_tokens": 142454702.0,
      "step": 68645
    },
    {
      "entropy": 5.6964517593383786,
      "epoch": 6.227322206095791,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019261500087768024,
      "loss": 4.8753,
      "mean_token_accuracy": 0.23045990020036697,
      "num_tokens": 142465415.0,
      "step": 68650
    },
    {
      "entropy": 5.63374342918396,
      "epoch": 6.227775761973875,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001925848923448188,
      "loss": 4.7198,
      "mean_token_accuracy": 0.23007305413484574,
      "num_tokens": 142475681.0,
      "step": 68655
    },
    {
      "entropy": 5.7305491924285885,
      "epoch": 6.22822931785196,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019255478551632278,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23098668158054353,
      "num_tokens": 142485120.0,
      "step": 68660
    },
    {
      "entropy": 5.692706203460693,
      "epoch": 6.228682873730044,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019252468039281478,
      "loss": 4.8452,
      "mean_token_accuracy": 0.23470177352428437,
      "num_tokens": 142494580.0,
      "step": 68665
    },
    {
      "entropy": 5.691557931900024,
      "epoch": 6.229136429608127,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019249457697491745,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22817928493022918,
      "num_tokens": 142504472.0,
      "step": 68670
    },
    {
      "entropy": 5.712571477890014,
      "epoch": 6.229589985486212,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019246447526325328,
      "loss": 4.8733,
      "mean_token_accuracy": 0.23179101198911667,
      "num_tokens": 142515355.0,
      "step": 68675
    },
    {
      "entropy": 5.681384563446045,
      "epoch": 6.230043541364296,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019243437525844482,
      "loss": 4.7691,
      "mean_token_accuracy": 0.23898616880178453,
      "num_tokens": 142524948.0,
      "step": 68680
    },
    {
      "entropy": 5.740105104446411,
      "epoch": 6.23049709724238,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019240427696111447,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23819949477910995,
      "num_tokens": 142536117.0,
      "step": 68685
    },
    {
      "entropy": 5.680611038208008,
      "epoch": 6.230950653120464,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019237418037188476,
      "loss": 4.769,
      "mean_token_accuracy": 0.23032658100128173,
      "num_tokens": 142546571.0,
      "step": 68690
    },
    {
      "entropy": 5.677080345153809,
      "epoch": 6.231404208998549,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000192344085491378,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2362999886274338,
      "num_tokens": 142556839.0,
      "step": 68695
    },
    {
      "entropy": 5.711142492294312,
      "epoch": 6.231857764876633,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001923139923202166,
      "loss": 4.772,
      "mean_token_accuracy": 0.23544572442770004,
      "num_tokens": 142566959.0,
      "step": 68700
    },
    {
      "entropy": 5.696617317199707,
      "epoch": 6.232311320754717,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019228390085902303,
      "loss": 4.8647,
      "mean_token_accuracy": 0.23201130777597428,
      "num_tokens": 142576894.0,
      "step": 68705
    },
    {
      "entropy": 5.755514478683471,
      "epoch": 6.232764876632801,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019225381110841945,
      "loss": 4.7847,
      "mean_token_accuracy": 0.22959758043289186,
      "num_tokens": 142587636.0,
      "step": 68710
    },
    {
      "entropy": 5.708876323699951,
      "epoch": 6.233218432510886,
      "grad_norm": 1.34375,
      "learning_rate": 0.000192223723069028,
      "loss": 4.824,
      "mean_token_accuracy": 0.22394466400146484,
      "num_tokens": 142598179.0,
      "step": 68715
    },
    {
      "entropy": 5.668489837646485,
      "epoch": 6.23367198838897,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019219363674147116,
      "loss": 4.7936,
      "mean_token_accuracy": 0.24211177825927735,
      "num_tokens": 142606919.0,
      "step": 68720
    },
    {
      "entropy": 5.634809780120849,
      "epoch": 6.234125544267053,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019216355212637104,
      "loss": 4.7463,
      "mean_token_accuracy": 0.24236707985401154,
      "num_tokens": 142617595.0,
      "step": 68725
    },
    {
      "entropy": 5.723748874664307,
      "epoch": 6.234579100145138,
      "grad_norm": 1.5,
      "learning_rate": 0.0001921334692243497,
      "loss": 4.8614,
      "mean_token_accuracy": 0.23390244394540788,
      "num_tokens": 142627948.0,
      "step": 68730
    },
    {
      "entropy": 5.694192790985108,
      "epoch": 6.235032656023222,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019210338803602933,
      "loss": 4.7807,
      "mean_token_accuracy": 0.2328990563750267,
      "num_tokens": 142638542.0,
      "step": 68735
    },
    {
      "entropy": 5.715362071990967,
      "epoch": 6.235486211901306,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001920733085620321,
      "loss": 4.7723,
      "mean_token_accuracy": 0.23611923158168793,
      "num_tokens": 142648454.0,
      "step": 68740
    },
    {
      "entropy": 5.64538950920105,
      "epoch": 6.23593976777939,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019204323080297992,
      "loss": 4.7426,
      "mean_token_accuracy": 0.24484184384346008,
      "num_tokens": 142658834.0,
      "step": 68745
    },
    {
      "entropy": 5.639278173446655,
      "epoch": 6.236393323657475,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019201315475949487,
      "loss": 4.7894,
      "mean_token_accuracy": 0.23034063577651978,
      "num_tokens": 142669667.0,
      "step": 68750
    },
    {
      "entropy": 5.705422496795654,
      "epoch": 6.236846879535559,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019198308043219891,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22835206240415573,
      "num_tokens": 142678994.0,
      "step": 68755
    },
    {
      "entropy": 5.671349573135376,
      "epoch": 6.237300435413643,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019195300782171403,
      "loss": 4.8324,
      "mean_token_accuracy": 0.23685871809720993,
      "num_tokens": 142690069.0,
      "step": 68760
    },
    {
      "entropy": 5.7276771545410154,
      "epoch": 6.2377539912917275,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019192293692866215,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23760911524295808,
      "num_tokens": 142700086.0,
      "step": 68765
    },
    {
      "entropy": 5.725411224365234,
      "epoch": 6.238207547169812,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019189286775366505,
      "loss": 4.8503,
      "mean_token_accuracy": 0.23625889420509338,
      "num_tokens": 142710035.0,
      "step": 68770
    },
    {
      "entropy": 5.691992378234863,
      "epoch": 6.238661103047895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001918628002973446,
      "loss": 4.821,
      "mean_token_accuracy": 0.23082524836063384,
      "num_tokens": 142720421.0,
      "step": 68775
    },
    {
      "entropy": 5.754815244674683,
      "epoch": 6.239114658925979,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019183273456032269,
      "loss": 4.903,
      "mean_token_accuracy": 0.22602873295545578,
      "num_tokens": 142731053.0,
      "step": 68780
    },
    {
      "entropy": 5.746997404098511,
      "epoch": 6.239568214804064,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019180267054322097,
      "loss": 4.8134,
      "mean_token_accuracy": 0.23465152084827423,
      "num_tokens": 142741510.0,
      "step": 68785
    },
    {
      "entropy": 5.732949733734131,
      "epoch": 6.240021770682148,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019177260824666126,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2376658648252487,
      "num_tokens": 142752817.0,
      "step": 68790
    },
    {
      "entropy": 5.666044282913208,
      "epoch": 6.240475326560232,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001917425476712652,
      "loss": 4.7878,
      "mean_token_accuracy": 0.23726363927125932,
      "num_tokens": 142762742.0,
      "step": 68795
    },
    {
      "entropy": 5.708791828155517,
      "epoch": 6.2409288824383164,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001917124888176545,
      "loss": 4.8717,
      "mean_token_accuracy": 0.22659501135349275,
      "num_tokens": 142773424.0,
      "step": 68800
    },
    {
      "entropy": 5.609237384796143,
      "epoch": 6.241382438316401,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019168243168645084,
      "loss": 4.7124,
      "mean_token_accuracy": 0.2471695825457573,
      "num_tokens": 142783530.0,
      "step": 68805
    },
    {
      "entropy": 5.695934438705445,
      "epoch": 6.241835994194485,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019165237627827557,
      "loss": 4.7953,
      "mean_token_accuracy": 0.229081629216671,
      "num_tokens": 142793074.0,
      "step": 68810
    },
    {
      "entropy": 5.730735874176025,
      "epoch": 6.242289550072569,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019162232259375045,
      "loss": 4.7932,
      "mean_token_accuracy": 0.23654786497354507,
      "num_tokens": 142803127.0,
      "step": 68815
    },
    {
      "entropy": 5.738964653015136,
      "epoch": 6.2427431059506535,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019159227063349698,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2316644996404648,
      "num_tokens": 142813473.0,
      "step": 68820
    },
    {
      "entropy": 5.725329637527466,
      "epoch": 6.243196661828737,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019156222039813658,
      "loss": 4.843,
      "mean_token_accuracy": 0.23808929920196534,
      "num_tokens": 142824117.0,
      "step": 68825
    },
    {
      "entropy": 5.753931474685669,
      "epoch": 6.243650217706821,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001915321718882907,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23129579722881316,
      "num_tokens": 142834801.0,
      "step": 68830
    },
    {
      "entropy": 5.735434532165527,
      "epoch": 6.244103773584905,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019150212510458077,
      "loss": 4.7915,
      "mean_token_accuracy": 0.2370927318930626,
      "num_tokens": 142845185.0,
      "step": 68835
    },
    {
      "entropy": 5.652476739883423,
      "epoch": 6.24455732946299,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019147208004762823,
      "loss": 4.7648,
      "mean_token_accuracy": 0.23788216859102249,
      "num_tokens": 142855341.0,
      "step": 68840
    },
    {
      "entropy": 5.719329404830932,
      "epoch": 6.245010885341074,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019144203671805433,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2358976975083351,
      "num_tokens": 142865022.0,
      "step": 68845
    },
    {
      "entropy": 5.679967164993286,
      "epoch": 6.245464441219158,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019141199511648032,
      "loss": 4.797,
      "mean_token_accuracy": 0.23829361647367478,
      "num_tokens": 142875409.0,
      "step": 68850
    },
    {
      "entropy": 5.637836933135986,
      "epoch": 6.2459179970972425,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001913819552435277,
      "loss": 4.7586,
      "mean_token_accuracy": 0.23522773534059524,
      "num_tokens": 142885852.0,
      "step": 68855
    },
    {
      "entropy": 5.73634524345398,
      "epoch": 6.246371552975327,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019135191709981748,
      "loss": 4.8562,
      "mean_token_accuracy": 0.23122540563344957,
      "num_tokens": 142894827.0,
      "step": 68860
    },
    {
      "entropy": 5.78660135269165,
      "epoch": 6.246825108853411,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001913218806859709,
      "loss": 4.9555,
      "mean_token_accuracy": 0.22173238098621367,
      "num_tokens": 142904495.0,
      "step": 68865
    },
    {
      "entropy": 5.682945585250854,
      "epoch": 6.247278664731495,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019129184600260918,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2423837125301361,
      "num_tokens": 142915105.0,
      "step": 68870
    },
    {
      "entropy": 5.6941296577453615,
      "epoch": 6.24773222060958,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019126181305035344,
      "loss": 4.8083,
      "mean_token_accuracy": 0.235399729013443,
      "num_tokens": 142925807.0,
      "step": 68875
    },
    {
      "entropy": 5.6821941375732425,
      "epoch": 6.248185776487663,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001912317818298247,
      "loss": 4.7631,
      "mean_token_accuracy": 0.23662749230861663,
      "num_tokens": 142936967.0,
      "step": 68880
    },
    {
      "entropy": 5.718036651611328,
      "epoch": 6.248639332365747,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019120175234164412,
      "loss": 4.8304,
      "mean_token_accuracy": 0.23652565330266953,
      "num_tokens": 142946332.0,
      "step": 68885
    },
    {
      "entropy": 5.681544303894043,
      "epoch": 6.2490928882438315,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019117172458643262,
      "loss": 4.7866,
      "mean_token_accuracy": 0.22918304651975632,
      "num_tokens": 142956472.0,
      "step": 68890
    },
    {
      "entropy": 5.757187128067017,
      "epoch": 6.249546444121916,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019114169856481123,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2365996301174164,
      "num_tokens": 142966281.0,
      "step": 68895
    },
    {
      "entropy": 5.661343574523926,
      "epoch": 6.25,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019111167427740083,
      "loss": 4.7823,
      "mean_token_accuracy": 0.23881106525659562,
      "num_tokens": 142976423.0,
      "step": 68900
    },
    {
      "entropy": 5.612199783325195,
      "epoch": 6.250453555878084,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00019108165172482245,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23705695420503617,
      "num_tokens": 142986711.0,
      "step": 68905
    },
    {
      "entropy": 5.721958827972412,
      "epoch": 6.2509071117561685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001910516309076969,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22941331267356874,
      "num_tokens": 142996624.0,
      "step": 68910
    },
    {
      "entropy": 5.7078148365020756,
      "epoch": 6.251360667634253,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019102161182664492,
      "loss": 4.8264,
      "mean_token_accuracy": 0.23214029967784883,
      "num_tokens": 143006627.0,
      "step": 68915
    },
    {
      "entropy": 5.723923873901367,
      "epoch": 6.251814223512337,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019099159448228754,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23811287581920623,
      "num_tokens": 143016920.0,
      "step": 68920
    },
    {
      "entropy": 5.7456827640533445,
      "epoch": 6.25226777939042,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019096157887524528,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23272311240434645,
      "num_tokens": 143026077.0,
      "step": 68925
    },
    {
      "entropy": 5.691325950622558,
      "epoch": 6.252721335268505,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019093156500613908,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23066130131483079,
      "num_tokens": 143036923.0,
      "step": 68930
    },
    {
      "entropy": 5.724421453475952,
      "epoch": 6.253174891146589,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019090155287558946,
      "loss": 4.8698,
      "mean_token_accuracy": 0.23131303042173385,
      "num_tokens": 143047717.0,
      "step": 68935
    },
    {
      "entropy": 5.694077873229981,
      "epoch": 6.253628447024673,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019087154248421717,
      "loss": 4.8469,
      "mean_token_accuracy": 0.23199039995670317,
      "num_tokens": 143057830.0,
      "step": 68940
    },
    {
      "entropy": 5.714629030227661,
      "epoch": 6.2540820029027575,
      "grad_norm": 1.5,
      "learning_rate": 0.00019084153383264286,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22899936139583588,
      "num_tokens": 143067353.0,
      "step": 68945
    },
    {
      "entropy": 5.7211791515350345,
      "epoch": 6.254535558780842,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000190811526921487,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23245684504508973,
      "num_tokens": 143077830.0,
      "step": 68950
    },
    {
      "entropy": 5.7619343280792235,
      "epoch": 6.254989114658926,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019078152175137038,
      "loss": 4.9133,
      "mean_token_accuracy": 0.22376570850610733,
      "num_tokens": 143088266.0,
      "step": 68955
    },
    {
      "entropy": 5.65132966041565,
      "epoch": 6.25544267053701,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001907515183229132,
      "loss": 4.7697,
      "mean_token_accuracy": 0.24002683013677598,
      "num_tokens": 143098814.0,
      "step": 68960
    },
    {
      "entropy": 5.624191236495972,
      "epoch": 6.255896226415095,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019072151663673608,
      "loss": 4.7624,
      "mean_token_accuracy": 0.23516296744346618,
      "num_tokens": 143109575.0,
      "step": 68965
    },
    {
      "entropy": 5.748601961135864,
      "epoch": 6.256349782293179,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019069151669345958,
      "loss": 4.8499,
      "mean_token_accuracy": 0.23904743045568466,
      "num_tokens": 143119865.0,
      "step": 68970
    },
    {
      "entropy": 5.682428407669067,
      "epoch": 6.256803338171263,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019066151849370395,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22677136212587357,
      "num_tokens": 143132059.0,
      "step": 68975
    },
    {
      "entropy": 5.6622387886047365,
      "epoch": 6.2572568940493465,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001906315220380896,
      "loss": 4.7641,
      "mean_token_accuracy": 0.2503108397126198,
      "num_tokens": 143142172.0,
      "step": 68980
    },
    {
      "entropy": 5.726080226898193,
      "epoch": 6.257710449927431,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019060152732723685,
      "loss": 4.8415,
      "mean_token_accuracy": 0.2326410546898842,
      "num_tokens": 143152017.0,
      "step": 68985
    },
    {
      "entropy": 5.718566131591797,
      "epoch": 6.258164005805515,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001905715343617661,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23967019766569136,
      "num_tokens": 143163482.0,
      "step": 68990
    },
    {
      "entropy": 5.654187870025635,
      "epoch": 6.258617561683599,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019054154314229749,
      "loss": 4.7771,
      "mean_token_accuracy": 0.2301197201013565,
      "num_tokens": 143173448.0,
      "step": 68995
    },
    {
      "entropy": 5.634386825561523,
      "epoch": 6.2590711175616836,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019051155366945128,
      "loss": 4.7499,
      "mean_token_accuracy": 0.23410994112491607,
      "num_tokens": 143183770.0,
      "step": 69000
    },
    {
      "epoch": 6.2590711175616836,
      "eval_entropy": 5.4108885896691366,
      "eval_loss": 4.930366039276123,
      "eval_mean_token_accuracy": 0.23564153250256276,
      "eval_num_tokens": 143183770.0,
      "eval_runtime": 20.7272,
      "eval_samples_per_second": 1705.968,
      "eval_steps_per_second": 213.246,
      "step": 69000
    },
    {
      "entropy": 5.632071876525879,
      "epoch": 6.259524673439768,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019048156594384763,
      "loss": 4.8153,
      "mean_token_accuracy": 0.22877019345760347,
      "num_tokens": 143194945.0,
      "step": 69005
    },
    {
      "entropy": 5.738817977905273,
      "epoch": 6.259978229317852,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019045157996610678,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2343387708067894,
      "num_tokens": 143204614.0,
      "step": 69010
    },
    {
      "entropy": 5.7940061569213865,
      "epoch": 6.260431785195936,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019042159573684887,
      "loss": 4.9573,
      "mean_token_accuracy": 0.22768549174070357,
      "num_tokens": 143214971.0,
      "step": 69015
    },
    {
      "entropy": 5.731793737411499,
      "epoch": 6.260885341074021,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019039161325669379,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22547492384910583,
      "num_tokens": 143225072.0,
      "step": 69020
    },
    {
      "entropy": 5.7665773868560795,
      "epoch": 6.261338896952105,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001903616325262618,
      "loss": 4.8153,
      "mean_token_accuracy": 0.23165659010410308,
      "num_tokens": 143234946.0,
      "step": 69025
    },
    {
      "entropy": 5.6846860408782955,
      "epoch": 6.261792452830189,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019033165354617276,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23257536441087723,
      "num_tokens": 143245236.0,
      "step": 69030
    },
    {
      "entropy": 5.667640495300293,
      "epoch": 6.2622460087082725,
      "grad_norm": 1.5,
      "learning_rate": 0.0001903016763170467,
      "loss": 4.7787,
      "mean_token_accuracy": 0.2364660918712616,
      "num_tokens": 143254977.0,
      "step": 69035
    },
    {
      "entropy": 5.667363786697388,
      "epoch": 6.262699564586357,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001902717008395036,
      "loss": 4.7899,
      "mean_token_accuracy": 0.23553307801485063,
      "num_tokens": 143265121.0,
      "step": 69040
    },
    {
      "entropy": 5.720791530609131,
      "epoch": 6.263153120464441,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019024172711416332,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2355584606528282,
      "num_tokens": 143276340.0,
      "step": 69045
    },
    {
      "entropy": 5.624916028976441,
      "epoch": 6.263606676342525,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019021175514164575,
      "loss": 4.7222,
      "mean_token_accuracy": 0.23930599093437194,
      "num_tokens": 143286740.0,
      "step": 69050
    },
    {
      "entropy": 5.7226848125457765,
      "epoch": 6.26406023222061,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019018178492257053,
      "loss": 4.8285,
      "mean_token_accuracy": 0.23492769449949263,
      "num_tokens": 143298144.0,
      "step": 69055
    },
    {
      "entropy": 5.782055473327636,
      "epoch": 6.264513788098694,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019015181645755776,
      "loss": 4.9207,
      "mean_token_accuracy": 0.22824618667364122,
      "num_tokens": 143308182.0,
      "step": 69060
    },
    {
      "entropy": 5.6969749450683596,
      "epoch": 6.264967343976778,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019012184974722703,
      "loss": 4.8128,
      "mean_token_accuracy": 0.23699456006288527,
      "num_tokens": 143318127.0,
      "step": 69065
    },
    {
      "entropy": 5.761108827590943,
      "epoch": 6.265420899854862,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000190091884792198,
      "loss": 4.9059,
      "mean_token_accuracy": 0.2241680160164833,
      "num_tokens": 143328564.0,
      "step": 69070
    },
    {
      "entropy": 5.792020797729492,
      "epoch": 6.265874455732947,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019006192159309055,
      "loss": 4.9067,
      "mean_token_accuracy": 0.2267729699611664,
      "num_tokens": 143338492.0,
      "step": 69075
    },
    {
      "entropy": 5.8083983898162845,
      "epoch": 6.26632801161103,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019003196015052415,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22843596935272217,
      "num_tokens": 143348866.0,
      "step": 69080
    },
    {
      "entropy": 5.655117845535278,
      "epoch": 6.266781567489114,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019000200046511838,
      "loss": 4.7604,
      "mean_token_accuracy": 0.236371947824955,
      "num_tokens": 143359409.0,
      "step": 69085
    },
    {
      "entropy": 5.647051525115967,
      "epoch": 6.267235123367199,
      "grad_norm": 1.484375,
      "learning_rate": 0.000189972042537493,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22626013457775115,
      "num_tokens": 143369922.0,
      "step": 69090
    },
    {
      "entropy": 5.600306415557862,
      "epoch": 6.267688679245283,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018994208636826748,
      "loss": 4.7561,
      "mean_token_accuracy": 0.23518714904785157,
      "num_tokens": 143379902.0,
      "step": 69095
    },
    {
      "entropy": 5.7175134181976315,
      "epoch": 6.268142235123367,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001899121319580613,
      "loss": 4.8911,
      "mean_token_accuracy": 0.22358963638544083,
      "num_tokens": 143390146.0,
      "step": 69100
    },
    {
      "entropy": 5.691286611557007,
      "epoch": 6.268595791001451,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018988217930749373,
      "loss": 4.7405,
      "mean_token_accuracy": 0.24176313877105712,
      "num_tokens": 143400927.0,
      "step": 69105
    },
    {
      "entropy": 5.695578336715698,
      "epoch": 6.269049346879536,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001898522284171846,
      "loss": 4.8647,
      "mean_token_accuracy": 0.22385180443525315,
      "num_tokens": 143411169.0,
      "step": 69110
    },
    {
      "entropy": 5.677926015853882,
      "epoch": 6.26950290275762,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018982227928775297,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2272980347275734,
      "num_tokens": 143422317.0,
      "step": 69115
    },
    {
      "entropy": 5.716152334213257,
      "epoch": 6.269956458635704,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018979233191981825,
      "loss": 4.8788,
      "mean_token_accuracy": 0.2323358803987503,
      "num_tokens": 143433657.0,
      "step": 69120
    },
    {
      "entropy": 5.791782808303833,
      "epoch": 6.270410014513788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018976238631399994,
      "loss": 4.9409,
      "mean_token_accuracy": 0.22382037490606307,
      "num_tokens": 143445669.0,
      "step": 69125
    },
    {
      "entropy": 5.712229871749878,
      "epoch": 6.270863570391873,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018973244247091716,
      "loss": 4.8256,
      "mean_token_accuracy": 0.23199427574872972,
      "num_tokens": 143455425.0,
      "step": 69130
    },
    {
      "entropy": 5.775795888900757,
      "epoch": 6.271317126269956,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018970250039118924,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2251964122056961,
      "num_tokens": 143465898.0,
      "step": 69135
    },
    {
      "entropy": 5.722259187698365,
      "epoch": 6.27177068214804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018967256007543527,
      "loss": 4.782,
      "mean_token_accuracy": 0.2381770893931389,
      "num_tokens": 143476739.0,
      "step": 69140
    },
    {
      "entropy": 5.737080097198486,
      "epoch": 6.272224238026125,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018964262152427458,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22588144689798356,
      "num_tokens": 143486360.0,
      "step": 69145
    },
    {
      "entropy": 5.705930280685425,
      "epoch": 6.272677793904209,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018961268473832615,
      "loss": 4.8152,
      "mean_token_accuracy": 0.24082891196012496,
      "num_tokens": 143496499.0,
      "step": 69150
    },
    {
      "entropy": 5.684082317352295,
      "epoch": 6.273131349782293,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018958274971820921,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22864419519901275,
      "num_tokens": 143506304.0,
      "step": 69155
    },
    {
      "entropy": 5.69429497718811,
      "epoch": 6.273584905660377,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018955281646454276,
      "loss": 4.8319,
      "mean_token_accuracy": 0.2317617192864418,
      "num_tokens": 143518182.0,
      "step": 69160
    },
    {
      "entropy": 5.742557668685913,
      "epoch": 6.274038461538462,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018952288497794584,
      "loss": 4.8386,
      "mean_token_accuracy": 0.23807520419359207,
      "num_tokens": 143527706.0,
      "step": 69165
    },
    {
      "entropy": 5.700177478790283,
      "epoch": 6.274492017416546,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018949295525903743,
      "loss": 4.7788,
      "mean_token_accuracy": 0.23327751457691193,
      "num_tokens": 143538219.0,
      "step": 69170
    },
    {
      "entropy": 5.651032638549805,
      "epoch": 6.27494557329463,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001894630273084365,
      "loss": 4.7425,
      "mean_token_accuracy": 0.23985906392335893,
      "num_tokens": 143548187.0,
      "step": 69175
    },
    {
      "entropy": 5.674950695037841,
      "epoch": 6.2753991291727145,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000189433101126762,
      "loss": 4.8134,
      "mean_token_accuracy": 0.24017827808856965,
      "num_tokens": 143558565.0,
      "step": 69180
    },
    {
      "entropy": 5.680096578598023,
      "epoch": 6.275852685050798,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018940317671463274,
      "loss": 4.843,
      "mean_token_accuracy": 0.22864576876163484,
      "num_tokens": 143568634.0,
      "step": 69185
    },
    {
      "entropy": 5.703913736343384,
      "epoch": 6.276306240928882,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018937325407266753,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2314653530716896,
      "num_tokens": 143578936.0,
      "step": 69190
    },
    {
      "entropy": 5.655046033859253,
      "epoch": 6.276759796806966,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018934333320148533,
      "loss": 4.703,
      "mean_token_accuracy": 0.24845698475837708,
      "num_tokens": 143588367.0,
      "step": 69195
    },
    {
      "entropy": 5.667413854598999,
      "epoch": 6.277213352685051,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001893134141017049,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2279138445854187,
      "num_tokens": 143598666.0,
      "step": 69200
    },
    {
      "entropy": 5.70428695678711,
      "epoch": 6.277666908563135,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018928349677394473,
      "loss": 4.8579,
      "mean_token_accuracy": 0.2334460660815239,
      "num_tokens": 143608463.0,
      "step": 69205
    },
    {
      "entropy": 5.718106985092163,
      "epoch": 6.278120464441219,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001892535812188238,
      "loss": 4.8106,
      "mean_token_accuracy": 0.2303337648510933,
      "num_tokens": 143618513.0,
      "step": 69210
    },
    {
      "entropy": 5.69996132850647,
      "epoch": 6.2785740203193035,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018922366743696063,
      "loss": 4.7713,
      "mean_token_accuracy": 0.2451622739434242,
      "num_tokens": 143628944.0,
      "step": 69215
    },
    {
      "entropy": 5.675320196151733,
      "epoch": 6.279027576197388,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018919375542897387,
      "loss": 4.775,
      "mean_token_accuracy": 0.24376852363348006,
      "num_tokens": 143638979.0,
      "step": 69220
    },
    {
      "entropy": 5.666946220397949,
      "epoch": 6.279481132075472,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018916384519548213,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23573147654533386,
      "num_tokens": 143648956.0,
      "step": 69225
    },
    {
      "entropy": 5.6754645824432375,
      "epoch": 6.279934687953556,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018913393673710406,
      "loss": 4.7771,
      "mean_token_accuracy": 0.22980378866195678,
      "num_tokens": 143658610.0,
      "step": 69230
    },
    {
      "entropy": 5.682068252563477,
      "epoch": 6.28038824383164,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018910403005445792,
      "loss": 4.7784,
      "mean_token_accuracy": 0.23923404663801193,
      "num_tokens": 143668548.0,
      "step": 69235
    },
    {
      "entropy": 5.655928754806519,
      "epoch": 6.280841799709724,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001890741251481623,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2450791358947754,
      "num_tokens": 143678462.0,
      "step": 69240
    },
    {
      "entropy": 5.710971641540527,
      "epoch": 6.281295355587808,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018904422201883574,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23035134375095367,
      "num_tokens": 143687767.0,
      "step": 69245
    },
    {
      "entropy": 5.6469591617584225,
      "epoch": 6.281748911465892,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018901432066709657,
      "loss": 4.7578,
      "mean_token_accuracy": 0.24119563400745392,
      "num_tokens": 143697866.0,
      "step": 69250
    },
    {
      "entropy": 5.681787204742432,
      "epoch": 6.282202467343977,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00018898442109356313,
      "loss": 4.8246,
      "mean_token_accuracy": 0.2289406180381775,
      "num_tokens": 143708099.0,
      "step": 69255
    },
    {
      "entropy": 5.6782622814178465,
      "epoch": 6.282656023222061,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018895452329885381,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23494193702936172,
      "num_tokens": 143718395.0,
      "step": 69260
    },
    {
      "entropy": 5.638819646835327,
      "epoch": 6.283109579100145,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018892462728358685,
      "loss": 4.7522,
      "mean_token_accuracy": 0.23539116084575654,
      "num_tokens": 143728950.0,
      "step": 69265
    },
    {
      "entropy": 5.74396448135376,
      "epoch": 6.2835631349782295,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001888947330483805,
      "loss": 4.9308,
      "mean_token_accuracy": 0.22449105829000474,
      "num_tokens": 143739621.0,
      "step": 69270
    },
    {
      "entropy": 5.731377792358399,
      "epoch": 6.284016690856314,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001888648405938531,
      "loss": 4.8363,
      "mean_token_accuracy": 0.2341867759823799,
      "num_tokens": 143749958.0,
      "step": 69275
    },
    {
      "entropy": 5.675999021530151,
      "epoch": 6.284470246734398,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018883494992062266,
      "loss": 4.7535,
      "mean_token_accuracy": 0.23866234868764877,
      "num_tokens": 143760454.0,
      "step": 69280
    },
    {
      "entropy": 5.800620365142822,
      "epoch": 6.284923802612482,
      "grad_norm": 1.640625,
      "learning_rate": 0.00018880506102930745,
      "loss": 4.86,
      "mean_token_accuracy": 0.23399696350097657,
      "num_tokens": 143770206.0,
      "step": 69285
    },
    {
      "entropy": 5.672335433959961,
      "epoch": 6.285377358490566,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018877517392052557,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22963858544826507,
      "num_tokens": 143780076.0,
      "step": 69290
    },
    {
      "entropy": 5.745244646072388,
      "epoch": 6.28583091436865,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018874528859489503,
      "loss": 4.9126,
      "mean_token_accuracy": 0.23393951058387757,
      "num_tokens": 143790024.0,
      "step": 69295
    },
    {
      "entropy": 5.693928480148315,
      "epoch": 6.286284470246734,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018871540505303392,
      "loss": 4.838,
      "mean_token_accuracy": 0.23332827687263488,
      "num_tokens": 143801755.0,
      "step": 69300
    },
    {
      "entropy": 5.7075989723205565,
      "epoch": 6.2867380261248185,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018868552329556016,
      "loss": 4.8426,
      "mean_token_accuracy": 0.23191869854927064,
      "num_tokens": 143812150.0,
      "step": 69305
    },
    {
      "entropy": 5.807513236999512,
      "epoch": 6.287191582002903,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018865564332309188,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22685440629720688,
      "num_tokens": 143823290.0,
      "step": 69310
    },
    {
      "entropy": 5.775135564804077,
      "epoch": 6.287645137880987,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00018862576513624686,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22382020354270935,
      "num_tokens": 143833626.0,
      "step": 69315
    },
    {
      "entropy": 5.720083570480346,
      "epoch": 6.288098693759071,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018859588873564297,
      "loss": 4.8756,
      "mean_token_accuracy": 0.23157150745391847,
      "num_tokens": 143844630.0,
      "step": 69320
    },
    {
      "entropy": 5.723723125457764,
      "epoch": 6.2885522496371555,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018856601412189816,
      "loss": 4.8189,
      "mean_token_accuracy": 0.23910756260156632,
      "num_tokens": 143854352.0,
      "step": 69325
    },
    {
      "entropy": 5.72510085105896,
      "epoch": 6.28900580551524,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018853614129563022,
      "loss": 4.8263,
      "mean_token_accuracy": 0.22934939712285995,
      "num_tokens": 143864147.0,
      "step": 69330
    },
    {
      "entropy": 5.691050243377686,
      "epoch": 6.289459361393324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018850627025745686,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23502744138240814,
      "num_tokens": 143874438.0,
      "step": 69335
    },
    {
      "entropy": 5.756023120880127,
      "epoch": 6.2899129172714074,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018847640100799596,
      "loss": 4.9019,
      "mean_token_accuracy": 0.23542007505893708,
      "num_tokens": 143885352.0,
      "step": 69340
    },
    {
      "entropy": 5.721328353881836,
      "epoch": 6.290366473149492,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018844653354786506,
      "loss": 4.7579,
      "mean_token_accuracy": 0.23706144392490386,
      "num_tokens": 143895353.0,
      "step": 69345
    },
    {
      "entropy": 5.682582378387451,
      "epoch": 6.290820029027576,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018841666787768196,
      "loss": 4.7346,
      "mean_token_accuracy": 0.23537892401218413,
      "num_tokens": 143905009.0,
      "step": 69350
    },
    {
      "entropy": 5.683664131164551,
      "epoch": 6.29127358490566,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001883868039980642,
      "loss": 4.8209,
      "mean_token_accuracy": 0.23534178584814072,
      "num_tokens": 143915788.0,
      "step": 69355
    },
    {
      "entropy": 5.636912059783936,
      "epoch": 6.2917271407837445,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018835694190962938,
      "loss": 4.8425,
      "mean_token_accuracy": 0.2305716872215271,
      "num_tokens": 143926388.0,
      "step": 69360
    },
    {
      "entropy": 5.751584005355835,
      "epoch": 6.292180696661829,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018832708161299516,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22621885687112808,
      "num_tokens": 143936284.0,
      "step": 69365
    },
    {
      "entropy": 5.7599709033966064,
      "epoch": 6.292634252539913,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001882972231087789,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21378742754459382,
      "num_tokens": 143946740.0,
      "step": 69370
    },
    {
      "entropy": 5.671859216690064,
      "epoch": 6.293087808417997,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018826736639759818,
      "loss": 4.7152,
      "mean_token_accuracy": 0.24388140439987183,
      "num_tokens": 143956818.0,
      "step": 69375
    },
    {
      "entropy": 5.712087202072143,
      "epoch": 6.293541364296082,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018823751148007046,
      "loss": 4.8384,
      "mean_token_accuracy": 0.2277571126818657,
      "num_tokens": 143967382.0,
      "step": 69380
    },
    {
      "entropy": 5.692970275878906,
      "epoch": 6.293994920174166,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018820765835681302,
      "loss": 4.7849,
      "mean_token_accuracy": 0.2419281303882599,
      "num_tokens": 143977561.0,
      "step": 69385
    },
    {
      "entropy": 5.754446172714234,
      "epoch": 6.294448476052249,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018817780702844346,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22732783257961273,
      "num_tokens": 143987506.0,
      "step": 69390
    },
    {
      "entropy": 5.6978710174560545,
      "epoch": 6.2949020319303335,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001881479574955789,
      "loss": 4.8576,
      "mean_token_accuracy": 0.22599196285009385,
      "num_tokens": 143997287.0,
      "step": 69395
    },
    {
      "entropy": 5.71471004486084,
      "epoch": 6.295355587808418,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001881181097588367,
      "loss": 4.8386,
      "mean_token_accuracy": 0.22903962582349777,
      "num_tokens": 144007858.0,
      "step": 69400
    },
    {
      "entropy": 5.772709941864013,
      "epoch": 6.295809143686502,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001880882638188341,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2269686982035637,
      "num_tokens": 144018061.0,
      "step": 69405
    },
    {
      "entropy": 5.726772785186768,
      "epoch": 6.296262699564586,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018805841967618842,
      "loss": 4.8114,
      "mean_token_accuracy": 0.23468356132507323,
      "num_tokens": 144028268.0,
      "step": 69410
    },
    {
      "entropy": 5.736799669265747,
      "epoch": 6.2967162554426706,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001880285773315168,
      "loss": 4.8237,
      "mean_token_accuracy": 0.22923197746276855,
      "num_tokens": 144039713.0,
      "step": 69415
    },
    {
      "entropy": 5.6618711948394775,
      "epoch": 6.297169811320755,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018799873678543628,
      "loss": 4.8435,
      "mean_token_accuracy": 0.23639959394931792,
      "num_tokens": 144050978.0,
      "step": 69420
    },
    {
      "entropy": 5.70859432220459,
      "epoch": 6.297623367198839,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018796889803856409,
      "loss": 4.9217,
      "mean_token_accuracy": 0.2199353262782097,
      "num_tokens": 144061671.0,
      "step": 69425
    },
    {
      "entropy": 5.772086763381958,
      "epoch": 6.298076923076923,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018793906109151727,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22807921171188356,
      "num_tokens": 144071958.0,
      "step": 69430
    },
    {
      "entropy": 5.725177812576294,
      "epoch": 6.298530478955008,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018790922594491284,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23218655437231064,
      "num_tokens": 144082337.0,
      "step": 69435
    },
    {
      "entropy": 5.723983526229858,
      "epoch": 6.298984034833092,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018787939259936782,
      "loss": 4.8478,
      "mean_token_accuracy": 0.23887892961502075,
      "num_tokens": 144091711.0,
      "step": 69440
    },
    {
      "entropy": 5.6777068138122555,
      "epoch": 6.299437590711175,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018784956105549918,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23517279773950578,
      "num_tokens": 144102886.0,
      "step": 69445
    },
    {
      "entropy": 5.691696357727051,
      "epoch": 6.2998911465892595,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018781973131392382,
      "loss": 4.7632,
      "mean_token_accuracy": 0.23511848449707032,
      "num_tokens": 144113157.0,
      "step": 69450
    },
    {
      "entropy": 5.748024225234985,
      "epoch": 6.300344702467344,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018778990337525858,
      "loss": 4.9604,
      "mean_token_accuracy": 0.23304659873247147,
      "num_tokens": 144123940.0,
      "step": 69455
    },
    {
      "entropy": 5.715226745605468,
      "epoch": 6.300798258345428,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018776007724012045,
      "loss": 4.8262,
      "mean_token_accuracy": 0.237128409743309,
      "num_tokens": 144133977.0,
      "step": 69460
    },
    {
      "entropy": 5.765193939208984,
      "epoch": 6.301251814223512,
      "grad_norm": 1.5,
      "learning_rate": 0.0001877302529091261,
      "loss": 4.8841,
      "mean_token_accuracy": 0.23328723311424254,
      "num_tokens": 144143575.0,
      "step": 69465
    },
    {
      "entropy": 5.632579612731933,
      "epoch": 6.301705370101597,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018770043038289237,
      "loss": 4.7114,
      "mean_token_accuracy": 0.24672336429357528,
      "num_tokens": 144153466.0,
      "step": 69470
    },
    {
      "entropy": 5.714065074920654,
      "epoch": 6.302158925979681,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001876706096620359,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2258319675922394,
      "num_tokens": 144164463.0,
      "step": 69475
    },
    {
      "entropy": 5.711749839782715,
      "epoch": 6.302612481857765,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018764079074717362,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2270869106054306,
      "num_tokens": 144174184.0,
      "step": 69480
    },
    {
      "entropy": 5.804592657089233,
      "epoch": 6.303066037735849,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018761097363892198,
      "loss": 4.947,
      "mean_token_accuracy": 0.2182214766740799,
      "num_tokens": 144185299.0,
      "step": 69485
    },
    {
      "entropy": 5.718217277526856,
      "epoch": 6.303519593613933,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018758115833789768,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22870820313692092,
      "num_tokens": 144196137.0,
      "step": 69490
    },
    {
      "entropy": 5.697477340698242,
      "epoch": 6.303973149492017,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001875513448447173,
      "loss": 4.789,
      "mean_token_accuracy": 0.23953619450330735,
      "num_tokens": 144205910.0,
      "step": 69495
    },
    {
      "entropy": 5.724014854431152,
      "epoch": 6.304426705370101,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018752153315999742,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2325324684381485,
      "num_tokens": 144216159.0,
      "step": 69500
    },
    {
      "entropy": 5.713852453231811,
      "epoch": 6.304880261248186,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001874917232843545,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2287030413746834,
      "num_tokens": 144227611.0,
      "step": 69505
    },
    {
      "entropy": 5.69041256904602,
      "epoch": 6.30533381712627,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018746191521840506,
      "loss": 4.7747,
      "mean_token_accuracy": 0.23710108995437623,
      "num_tokens": 144237696.0,
      "step": 69510
    },
    {
      "entropy": 5.653715896606445,
      "epoch": 6.305787373004354,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018743210896276548,
      "loss": 4.7488,
      "mean_token_accuracy": 0.2372844010591507,
      "num_tokens": 144248083.0,
      "step": 69515
    },
    {
      "entropy": 5.668255567550659,
      "epoch": 6.306240928882438,
      "grad_norm": 1.640625,
      "learning_rate": 0.00018740230451805225,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22944216579198837,
      "num_tokens": 144258470.0,
      "step": 69520
    },
    {
      "entropy": 5.704874801635742,
      "epoch": 6.306694484760523,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018737250188488163,
      "loss": 4.8343,
      "mean_token_accuracy": 0.23090581148862838,
      "num_tokens": 144270151.0,
      "step": 69525
    },
    {
      "entropy": 5.761707782745361,
      "epoch": 6.307148040638607,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018734270106387003,
      "loss": 4.8788,
      "mean_token_accuracy": 0.23037322014570236,
      "num_tokens": 144280096.0,
      "step": 69530
    },
    {
      "entropy": 5.692492198944092,
      "epoch": 6.307601596516691,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018731290205563372,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23712970614433287,
      "num_tokens": 144290663.0,
      "step": 69535
    },
    {
      "entropy": 5.670037698745728,
      "epoch": 6.308055152394775,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018728310486078897,
      "loss": 4.7699,
      "mean_token_accuracy": 0.23512330502271653,
      "num_tokens": 144301083.0,
      "step": 69540
    },
    {
      "entropy": 5.727655649185181,
      "epoch": 6.308508708272859,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018725330947995188,
      "loss": 4.9039,
      "mean_token_accuracy": 0.2309576615691185,
      "num_tokens": 144311698.0,
      "step": 69545
    },
    {
      "entropy": 5.77444372177124,
      "epoch": 6.308962264150943,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018722351591373877,
      "loss": 4.9804,
      "mean_token_accuracy": 0.22496961951255798,
      "num_tokens": 144322678.0,
      "step": 69550
    },
    {
      "entropy": 5.840793466567993,
      "epoch": 6.309415820029027,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018719372416276574,
      "loss": 4.9039,
      "mean_token_accuracy": 0.2203359603881836,
      "num_tokens": 144332200.0,
      "step": 69555
    },
    {
      "entropy": 5.693044281005859,
      "epoch": 6.309869375907112,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018716393422764884,
      "loss": 4.8161,
      "mean_token_accuracy": 0.22826451808214188,
      "num_tokens": 144343125.0,
      "step": 69560
    },
    {
      "entropy": 5.759302806854248,
      "epoch": 6.310322931785196,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018713414610900414,
      "loss": 4.858,
      "mean_token_accuracy": 0.2322767421603203,
      "num_tokens": 144353874.0,
      "step": 69565
    },
    {
      "entropy": 5.726332139968872,
      "epoch": 6.31077648766328,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018710435980744778,
      "loss": 4.816,
      "mean_token_accuracy": 0.23571571111679077,
      "num_tokens": 144363813.0,
      "step": 69570
    },
    {
      "entropy": 5.722552490234375,
      "epoch": 6.311230043541364,
      "grad_norm": 1.375,
      "learning_rate": 0.0001870745753235955,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23919051438570021,
      "num_tokens": 144374612.0,
      "step": 69575
    },
    {
      "entropy": 5.753951597213745,
      "epoch": 6.311683599419449,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00018704479265806352,
      "loss": 4.8639,
      "mean_token_accuracy": 0.23006850033998488,
      "num_tokens": 144385545.0,
      "step": 69580
    },
    {
      "entropy": 5.670970392227173,
      "epoch": 6.312137155297533,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018701501181146764,
      "loss": 4.7554,
      "mean_token_accuracy": 0.236422498524189,
      "num_tokens": 144395519.0,
      "step": 69585
    },
    {
      "entropy": 5.658001232147217,
      "epoch": 6.312590711175617,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001869852327844237,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23137336820364,
      "num_tokens": 144405921.0,
      "step": 69590
    },
    {
      "entropy": 5.623881244659424,
      "epoch": 6.3130442670537015,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018695545557754757,
      "loss": 4.7076,
      "mean_token_accuracy": 0.24715080112218857,
      "num_tokens": 144415080.0,
      "step": 69595
    },
    {
      "entropy": 5.698985385894775,
      "epoch": 6.313497822931785,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001869256801914551,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23718321919441224,
      "num_tokens": 144426085.0,
      "step": 69600
    },
    {
      "entropy": 5.693345832824707,
      "epoch": 6.313951378809869,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018689590662676205,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23896073400974274,
      "num_tokens": 144435503.0,
      "step": 69605
    },
    {
      "entropy": 5.718071317672729,
      "epoch": 6.314404934687953,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000186866134884084,
      "loss": 4.7646,
      "mean_token_accuracy": 0.24268862158060073,
      "num_tokens": 144445654.0,
      "step": 69610
    },
    {
      "entropy": 5.752972602844238,
      "epoch": 6.314858490566038,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001868363649640368,
      "loss": 4.855,
      "mean_token_accuracy": 0.22967728674411775,
      "num_tokens": 144456155.0,
      "step": 69615
    },
    {
      "entropy": 5.682955741882324,
      "epoch": 6.315312046444122,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000186806596867236,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2283874198794365,
      "num_tokens": 144466386.0,
      "step": 69620
    },
    {
      "entropy": 5.694012880325317,
      "epoch": 6.315765602322206,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018677683059429724,
      "loss": 4.806,
      "mean_token_accuracy": 0.23590486496686935,
      "num_tokens": 144476674.0,
      "step": 69625
    },
    {
      "entropy": 5.6400364398956295,
      "epoch": 6.3162191582002905,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018674706614583614,
      "loss": 4.7181,
      "mean_token_accuracy": 0.24505894780158996,
      "num_tokens": 144486484.0,
      "step": 69630
    },
    {
      "entropy": 5.661855936050415,
      "epoch": 6.316672714078375,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018671730352246818,
      "loss": 4.8427,
      "mean_token_accuracy": 0.23507307171821595,
      "num_tokens": 144497146.0,
      "step": 69635
    },
    {
      "entropy": 5.73178768157959,
      "epoch": 6.317126269956459,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001866875427248089,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23188378810882568,
      "num_tokens": 144506736.0,
      "step": 69640
    },
    {
      "entropy": 5.707584095001221,
      "epoch": 6.317579825834542,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001866577837534737,
      "loss": 4.799,
      "mean_token_accuracy": 0.2325536608695984,
      "num_tokens": 144517004.0,
      "step": 69645
    },
    {
      "entropy": 5.684377908706665,
      "epoch": 6.318033381712627,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018662802660907813,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23091079592704772,
      "num_tokens": 144526589.0,
      "step": 69650
    },
    {
      "entropy": 5.751738119125366,
      "epoch": 6.318486937590711,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018659827129223749,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2383214846253395,
      "num_tokens": 144536890.0,
      "step": 69655
    },
    {
      "entropy": 5.699126148223877,
      "epoch": 6.318940493468795,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001865685178035671,
      "loss": 4.8309,
      "mean_token_accuracy": 0.22882273495197297,
      "num_tokens": 144547226.0,
      "step": 69660
    },
    {
      "entropy": 5.66297173500061,
      "epoch": 6.319394049346879,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018653876614368225,
      "loss": 4.7518,
      "mean_token_accuracy": 0.23775212317705155,
      "num_tokens": 144558076.0,
      "step": 69665
    },
    {
      "entropy": 5.717262935638428,
      "epoch": 6.319847605224964,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00018650901631319828,
      "loss": 4.8407,
      "mean_token_accuracy": 0.2306547224521637,
      "num_tokens": 144569208.0,
      "step": 69670
    },
    {
      "entropy": 5.6870513439178465,
      "epoch": 6.320301161103048,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018647926831273043,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23723114132881165,
      "num_tokens": 144579259.0,
      "step": 69675
    },
    {
      "entropy": 5.688019275665283,
      "epoch": 6.320754716981132,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018644952214289385,
      "loss": 4.8232,
      "mean_token_accuracy": 0.22986948788166045,
      "num_tokens": 144589330.0,
      "step": 69680
    },
    {
      "entropy": 5.733973121643066,
      "epoch": 6.3212082728592165,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001864197778043037,
      "loss": 4.7962,
      "mean_token_accuracy": 0.23283744603395462,
      "num_tokens": 144599420.0,
      "step": 69685
    },
    {
      "entropy": 5.784787845611572,
      "epoch": 6.321661828737301,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001863900352975751,
      "loss": 4.933,
      "mean_token_accuracy": 0.221104995906353,
      "num_tokens": 144610162.0,
      "step": 69690
    },
    {
      "entropy": 5.72698860168457,
      "epoch": 6.322115384615385,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001863602946233232,
      "loss": 4.8795,
      "mean_token_accuracy": 0.23130964487791061,
      "num_tokens": 144620962.0,
      "step": 69695
    },
    {
      "entropy": 5.680918550491333,
      "epoch": 6.322568940493468,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000186330555782163,
      "loss": 4.7495,
      "mean_token_accuracy": 0.2421787202358246,
      "num_tokens": 144631691.0,
      "step": 69700
    },
    {
      "entropy": 5.707267570495605,
      "epoch": 6.323022496371553,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018630081877470944,
      "loss": 4.8189,
      "mean_token_accuracy": 0.23263752609491348,
      "num_tokens": 144642426.0,
      "step": 69705
    },
    {
      "entropy": 5.699931335449219,
      "epoch": 6.323476052249637,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018627108360157757,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22680552899837494,
      "num_tokens": 144652359.0,
      "step": 69710
    },
    {
      "entropy": 5.752998018264771,
      "epoch": 6.323929608127721,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018624135026338234,
      "loss": 4.8891,
      "mean_token_accuracy": 0.23525680601596832,
      "num_tokens": 144662868.0,
      "step": 69715
    },
    {
      "entropy": 5.703775691986084,
      "epoch": 6.3243831640058055,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018621161876073857,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23498543053865434,
      "num_tokens": 144672705.0,
      "step": 69720
    },
    {
      "entropy": 5.713349485397339,
      "epoch": 6.32483671988389,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001861818890942612,
      "loss": 4.927,
      "mean_token_accuracy": 0.22044449299573898,
      "num_tokens": 144684859.0,
      "step": 69725
    },
    {
      "entropy": 5.7044844150543215,
      "epoch": 6.325290275761974,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018615216126456496,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22860324531793594,
      "num_tokens": 144695227.0,
      "step": 69730
    },
    {
      "entropy": 5.674829053878784,
      "epoch": 6.325743831640058,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001861224352722647,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23824437707662582,
      "num_tokens": 144705852.0,
      "step": 69735
    },
    {
      "entropy": 5.682053422927856,
      "epoch": 6.3261973875181425,
      "grad_norm": 1.5,
      "learning_rate": 0.00018609271111797503,
      "loss": 4.8025,
      "mean_token_accuracy": 0.23277491629123687,
      "num_tokens": 144715733.0,
      "step": 69740
    },
    {
      "entropy": 5.714254331588745,
      "epoch": 6.326650943396227,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018606298880231089,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2245865434408188,
      "num_tokens": 144727359.0,
      "step": 69745
    },
    {
      "entropy": 5.637652969360351,
      "epoch": 6.327104499274311,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018603326832588674,
      "loss": 4.6774,
      "mean_token_accuracy": 0.2463843360543251,
      "num_tokens": 144737764.0,
      "step": 69750
    },
    {
      "entropy": 5.664326524734497,
      "epoch": 6.3275580551523944,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018600354968931722,
      "loss": 4.7647,
      "mean_token_accuracy": 0.24266693592071534,
      "num_tokens": 144747784.0,
      "step": 69755
    },
    {
      "entropy": 5.665355014801025,
      "epoch": 6.328011611030479,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018597383289321702,
      "loss": 4.7792,
      "mean_token_accuracy": 0.23735352158546447,
      "num_tokens": 144757806.0,
      "step": 69760
    },
    {
      "entropy": 5.67782883644104,
      "epoch": 6.328465166908563,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018594411793820066,
      "loss": 4.8641,
      "mean_token_accuracy": 0.23290115147829055,
      "num_tokens": 144768245.0,
      "step": 69765
    },
    {
      "entropy": 5.688329982757568,
      "epoch": 6.328918722786647,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001859144048248826,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2356724038720131,
      "num_tokens": 144779218.0,
      "step": 69770
    },
    {
      "entropy": 5.713956451416015,
      "epoch": 6.3293722786647315,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018588469355387732,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22954051345586776,
      "num_tokens": 144789930.0,
      "step": 69775
    },
    {
      "entropy": 5.679361629486084,
      "epoch": 6.329825834542816,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018585498412579938,
      "loss": 4.7468,
      "mean_token_accuracy": 0.23276254683732986,
      "num_tokens": 144799847.0,
      "step": 69780
    },
    {
      "entropy": 5.755019235610962,
      "epoch": 6.3302793904209,
      "grad_norm": 1.453125,
      "learning_rate": 0.000185825276541263,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22491479963064193,
      "num_tokens": 144810727.0,
      "step": 69785
    },
    {
      "entropy": 5.682917261123658,
      "epoch": 6.330732946298984,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018579557080088265,
      "loss": 4.7499,
      "mean_token_accuracy": 0.24511424005031585,
      "num_tokens": 144820828.0,
      "step": 69790
    },
    {
      "entropy": 5.642300748825074,
      "epoch": 6.331186502177069,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018576586690527265,
      "loss": 4.7321,
      "mean_token_accuracy": 0.2422398954629898,
      "num_tokens": 144830972.0,
      "step": 69795
    },
    {
      "entropy": 5.664015531539917,
      "epoch": 6.331640058055152,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001857361648550473,
      "loss": 4.7722,
      "mean_token_accuracy": 0.23514128774404525,
      "num_tokens": 144841540.0,
      "step": 69800
    },
    {
      "entropy": 5.710032796859741,
      "epoch": 6.332093613933236,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001857064646508207,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23626655638217925,
      "num_tokens": 144851075.0,
      "step": 69805
    },
    {
      "entropy": 5.745505046844483,
      "epoch": 6.3325471698113205,
      "grad_norm": 1.765625,
      "learning_rate": 0.00018567676629320724,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23787504732608794,
      "num_tokens": 144860549.0,
      "step": 69810
    },
    {
      "entropy": 5.733305740356445,
      "epoch": 6.333000725689405,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018564706978282098,
      "loss": 4.8135,
      "mean_token_accuracy": 0.23380369395017625,
      "num_tokens": 144870853.0,
      "step": 69815
    },
    {
      "entropy": 5.659478998184204,
      "epoch": 6.333454281567489,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018561737512027616,
      "loss": 4.7717,
      "mean_token_accuracy": 0.2355077549815178,
      "num_tokens": 144881239.0,
      "step": 69820
    },
    {
      "entropy": 5.651845836639405,
      "epoch": 6.333907837445573,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018558768230618667,
      "loss": 4.7246,
      "mean_token_accuracy": 0.24375042021274568,
      "num_tokens": 144891136.0,
      "step": 69825
    },
    {
      "entropy": 5.73990650177002,
      "epoch": 6.3343613933236576,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001855579913411668,
      "loss": 4.8736,
      "mean_token_accuracy": 0.23520870953798295,
      "num_tokens": 144900811.0,
      "step": 69830
    },
    {
      "entropy": 5.650855779647827,
      "epoch": 6.334814949201742,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018552830222583045,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2353281706571579,
      "num_tokens": 144911376.0,
      "step": 69835
    },
    {
      "entropy": 5.700569915771484,
      "epoch": 6.335268505079826,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018549861496079158,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23223581314086914,
      "num_tokens": 144921764.0,
      "step": 69840
    },
    {
      "entropy": 5.645887994766236,
      "epoch": 6.33572206095791,
      "grad_norm": 1.5,
      "learning_rate": 0.00018546892954666416,
      "loss": 4.7495,
      "mean_token_accuracy": 0.24323057532310485,
      "num_tokens": 144931364.0,
      "step": 69845
    },
    {
      "entropy": 5.682774019241333,
      "epoch": 6.336175616835995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018543924598406214,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2307319834828377,
      "num_tokens": 144941403.0,
      "step": 69850
    },
    {
      "entropy": 5.709167575836181,
      "epoch": 6.336629172714078,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001854095642735993,
      "loss": 4.8186,
      "mean_token_accuracy": 0.2315560981631279,
      "num_tokens": 144952115.0,
      "step": 69855
    },
    {
      "entropy": 5.664198589324951,
      "epoch": 6.337082728592162,
      "grad_norm": 1.375,
      "learning_rate": 0.00018537988441588948,
      "loss": 4.8404,
      "mean_token_accuracy": 0.23117802739143373,
      "num_tokens": 144962837.0,
      "step": 69860
    },
    {
      "entropy": 5.643884706497192,
      "epoch": 6.3375362844702465,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001853502064115465,
      "loss": 4.804,
      "mean_token_accuracy": 0.2362341672182083,
      "num_tokens": 144974300.0,
      "step": 69865
    },
    {
      "entropy": 5.705239200592041,
      "epoch": 6.337989840348331,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001853205302611841,
      "loss": 4.8055,
      "mean_token_accuracy": 0.2402034029364586,
      "num_tokens": 144984911.0,
      "step": 69870
    },
    {
      "entropy": 5.685149097442627,
      "epoch": 6.338443396226415,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018529085596541602,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2326686203479767,
      "num_tokens": 144995145.0,
      "step": 69875
    },
    {
      "entropy": 5.681478881835938,
      "epoch": 6.338896952104499,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001852611835248559,
      "loss": 4.7972,
      "mean_token_accuracy": 0.24265554249286653,
      "num_tokens": 145005117.0,
      "step": 69880
    },
    {
      "entropy": 5.696357154846192,
      "epoch": 6.339350507982584,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001852315129401173,
      "loss": 4.8202,
      "mean_token_accuracy": 0.2341443970799446,
      "num_tokens": 145014747.0,
      "step": 69885
    },
    {
      "entropy": 5.725050687789917,
      "epoch": 6.339804063860668,
      "grad_norm": 1.5,
      "learning_rate": 0.00018520184421181392,
      "loss": 4.778,
      "mean_token_accuracy": 0.237289097905159,
      "num_tokens": 145025136.0,
      "step": 69890
    },
    {
      "entropy": 5.628947448730469,
      "epoch": 6.340257619738752,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001851721773405593,
      "loss": 4.7573,
      "mean_token_accuracy": 0.24069731086492538,
      "num_tokens": 145035146.0,
      "step": 69895
    },
    {
      "entropy": 5.730430936813354,
      "epoch": 6.340711175616836,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018514251232696694,
      "loss": 4.8082,
      "mean_token_accuracy": 0.2389848053455353,
      "num_tokens": 145046740.0,
      "step": 69900
    },
    {
      "entropy": 5.682955551147461,
      "epoch": 6.34116473149492,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001851128491716504,
      "loss": 4.8716,
      "mean_token_accuracy": 0.22357165515422822,
      "num_tokens": 145057959.0,
      "step": 69905
    },
    {
      "entropy": 5.660880994796753,
      "epoch": 6.341618287373004,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018508318787522287,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2409690499305725,
      "num_tokens": 145068046.0,
      "step": 69910
    },
    {
      "entropy": 5.760783290863037,
      "epoch": 6.342071843251088,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001850535284382981,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22955686897039412,
      "num_tokens": 145078862.0,
      "step": 69915
    },
    {
      "entropy": 5.673556995391846,
      "epoch": 6.342525399129173,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018502387086148918,
      "loss": 4.8134,
      "mean_token_accuracy": 0.2328316941857338,
      "num_tokens": 145088824.0,
      "step": 69920
    },
    {
      "entropy": 5.716585683822632,
      "epoch": 6.342978955007257,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018499421514540958,
      "loss": 4.8189,
      "mean_token_accuracy": 0.23244086802005767,
      "num_tokens": 145098450.0,
      "step": 69925
    },
    {
      "entropy": 5.670581340789795,
      "epoch": 6.343432510885341,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018496456129067252,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2310157150030136,
      "num_tokens": 145108363.0,
      "step": 69930
    },
    {
      "entropy": 5.690232610702514,
      "epoch": 6.343886066763425,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001849349092978913,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2306363731622696,
      "num_tokens": 145119536.0,
      "step": 69935
    },
    {
      "entropy": 5.6666905879974365,
      "epoch": 6.34433962264151,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018490525916767914,
      "loss": 4.7945,
      "mean_token_accuracy": 0.23097265213727952,
      "num_tokens": 145130535.0,
      "step": 69940
    },
    {
      "entropy": 5.713274383544922,
      "epoch": 6.344793178519594,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00018487561090064914,
      "loss": 4.7768,
      "mean_token_accuracy": 0.24091741293668748,
      "num_tokens": 145140177.0,
      "step": 69945
    },
    {
      "entropy": 5.682727718353272,
      "epoch": 6.345246734397678,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018484596449741447,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23397144675254822,
      "num_tokens": 145151074.0,
      "step": 69950
    },
    {
      "entropy": 5.638821077346802,
      "epoch": 6.3457002902757615,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018481631995858827,
      "loss": 4.7519,
      "mean_token_accuracy": 0.2452814295887947,
      "num_tokens": 145162137.0,
      "step": 69955
    },
    {
      "entropy": 5.738331699371338,
      "epoch": 6.346153846153846,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018478667728478342,
      "loss": 4.9132,
      "mean_token_accuracy": 0.23594805598258972,
      "num_tokens": 145172871.0,
      "step": 69960
    },
    {
      "entropy": 5.628500270843506,
      "epoch": 6.34660740203193,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018475703647661323,
      "loss": 4.757,
      "mean_token_accuracy": 0.2335907205939293,
      "num_tokens": 145183423.0,
      "step": 69965
    },
    {
      "entropy": 5.666410636901856,
      "epoch": 6.347060957910014,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018472739753469048,
      "loss": 4.8785,
      "mean_token_accuracy": 0.24010483026504517,
      "num_tokens": 145193841.0,
      "step": 69970
    },
    {
      "entropy": 5.677509546279907,
      "epoch": 6.347514513788099,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001846977604596281,
      "loss": 4.7711,
      "mean_token_accuracy": 0.24247970581054687,
      "num_tokens": 145204283.0,
      "step": 69975
    },
    {
      "entropy": 5.694681024551391,
      "epoch": 6.347968069666183,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018466812525203907,
      "loss": 4.7889,
      "mean_token_accuracy": 0.24324388504028321,
      "num_tokens": 145214875.0,
      "step": 69980
    },
    {
      "entropy": 5.774785709381104,
      "epoch": 6.348421625544267,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001846384919125363,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22363875955343246,
      "num_tokens": 145225306.0,
      "step": 69985
    },
    {
      "entropy": 5.7022168159484865,
      "epoch": 6.348875181422351,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018460886044173246,
      "loss": 4.8662,
      "mean_token_accuracy": 0.2230447754263878,
      "num_tokens": 145235646.0,
      "step": 69990
    },
    {
      "entropy": 5.724163675308228,
      "epoch": 6.349328737300436,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018457923084024043,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22804831713438034,
      "num_tokens": 145246022.0,
      "step": 69995
    },
    {
      "entropy": 5.749662828445435,
      "epoch": 6.34978229317852,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018454960310867297,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2243976593017578,
      "num_tokens": 145258469.0,
      "step": 70000
    },
    {
      "entropy": 5.745873498916626,
      "epoch": 6.350235849056604,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018451997724764274,
      "loss": 4.8116,
      "mean_token_accuracy": 0.2340235933661461,
      "num_tokens": 145267974.0,
      "step": 70005
    },
    {
      "entropy": 5.6483947277069095,
      "epoch": 6.350689404934688,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018449035325776242,
      "loss": 4.7834,
      "mean_token_accuracy": 0.24117229431867598,
      "num_tokens": 145278360.0,
      "step": 70010
    },
    {
      "entropy": 5.709462022781372,
      "epoch": 6.351142960812772,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018446073113964474,
      "loss": 4.8779,
      "mean_token_accuracy": 0.2293476641178131,
      "num_tokens": 145289889.0,
      "step": 70015
    },
    {
      "entropy": 5.73007550239563,
      "epoch": 6.351596516690856,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001844311108939022,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2386433497071266,
      "num_tokens": 145300569.0,
      "step": 70020
    },
    {
      "entropy": 5.700740003585816,
      "epoch": 6.35205007256894,
      "grad_norm": 1.5,
      "learning_rate": 0.0001844014925211473,
      "loss": 4.776,
      "mean_token_accuracy": 0.24206384867429734,
      "num_tokens": 145310083.0,
      "step": 70025
    },
    {
      "entropy": 5.616424083709717,
      "epoch": 6.352503628447025,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018437187602199267,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2366520494222641,
      "num_tokens": 145320063.0,
      "step": 70030
    },
    {
      "entropy": 5.70401177406311,
      "epoch": 6.352957184325109,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018434226139705072,
      "loss": 4.8128,
      "mean_token_accuracy": 0.2331136330962181,
      "num_tokens": 145330533.0,
      "step": 70035
    },
    {
      "entropy": 5.631469488143921,
      "epoch": 6.353410740203193,
      "grad_norm": 1.375,
      "learning_rate": 0.00018431264864693391,
      "loss": 4.7988,
      "mean_token_accuracy": 0.22856221050024034,
      "num_tokens": 145341559.0,
      "step": 70040
    },
    {
      "entropy": 5.654885721206665,
      "epoch": 6.3538642960812775,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00018428303777225463,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23668671250343323,
      "num_tokens": 145352240.0,
      "step": 70045
    },
    {
      "entropy": 5.658485269546508,
      "epoch": 6.354317851959362,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018425342877362527,
      "loss": 4.7443,
      "mean_token_accuracy": 0.2411179170012474,
      "num_tokens": 145362154.0,
      "step": 70050
    },
    {
      "entropy": 5.75869607925415,
      "epoch": 6.354771407837446,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018422382165165813,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23233546912670136,
      "num_tokens": 145373329.0,
      "step": 70055
    },
    {
      "entropy": 5.733530807495117,
      "epoch": 6.355224963715529,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001841942164069655,
      "loss": 4.8531,
      "mean_token_accuracy": 0.23045427352190018,
      "num_tokens": 145382909.0,
      "step": 70060
    },
    {
      "entropy": 5.747710227966309,
      "epoch": 6.355678519593614,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018416461304015952,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22837016135454177,
      "num_tokens": 145393149.0,
      "step": 70065
    },
    {
      "entropy": 5.689214563369751,
      "epoch": 6.356132075471698,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018413501155185259,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2319369599223137,
      "num_tokens": 145404726.0,
      "step": 70070
    },
    {
      "entropy": 5.650358772277832,
      "epoch": 6.356585631349782,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018410541194265674,
      "loss": 4.7605,
      "mean_token_accuracy": 0.2399410367012024,
      "num_tokens": 145415466.0,
      "step": 70075
    },
    {
      "entropy": 5.669849967956543,
      "epoch": 6.357039187227866,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001840758142131841,
      "loss": 4.8544,
      "mean_token_accuracy": 0.23281147480010986,
      "num_tokens": 145426960.0,
      "step": 70080
    },
    {
      "entropy": 5.755230951309204,
      "epoch": 6.357492743105951,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018404621836404694,
      "loss": 4.932,
      "mean_token_accuracy": 0.2228967696428299,
      "num_tokens": 145437175.0,
      "step": 70085
    },
    {
      "entropy": 5.738829755783081,
      "epoch": 6.357946298984035,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018401662439585705,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2226560041308403,
      "num_tokens": 145447951.0,
      "step": 70090
    },
    {
      "entropy": 5.7623748779296875,
      "epoch": 6.358399854862119,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018398703230922654,
      "loss": 4.8252,
      "mean_token_accuracy": 0.2372948482632637,
      "num_tokens": 145457935.0,
      "step": 70095
    },
    {
      "entropy": 5.730972385406494,
      "epoch": 6.3588534107402035,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018395744210476746,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22914122939109802,
      "num_tokens": 145469456.0,
      "step": 70100
    },
    {
      "entropy": 5.709758186340332,
      "epoch": 6.359306966618288,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018392785378309158,
      "loss": 4.904,
      "mean_token_accuracy": 0.2324694037437439,
      "num_tokens": 145479400.0,
      "step": 70105
    },
    {
      "entropy": 5.65526385307312,
      "epoch": 6.359760522496371,
      "grad_norm": 1.625,
      "learning_rate": 0.00018389826734481097,
      "loss": 4.7626,
      "mean_token_accuracy": 0.23643006682395934,
      "num_tokens": 145489999.0,
      "step": 70110
    },
    {
      "entropy": 5.773889255523682,
      "epoch": 6.360214078374455,
      "grad_norm": 1.375,
      "learning_rate": 0.00018386868279053736,
      "loss": 4.8738,
      "mean_token_accuracy": 0.23092931210994722,
      "num_tokens": 145500884.0,
      "step": 70115
    },
    {
      "entropy": 5.774820041656494,
      "epoch": 6.36066763425254,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018383910012088268,
      "loss": 4.8906,
      "mean_token_accuracy": 0.21600917875766754,
      "num_tokens": 145512271.0,
      "step": 70120
    },
    {
      "entropy": 5.671259021759033,
      "epoch": 6.361121190130624,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001838095193364585,
      "loss": 4.7532,
      "mean_token_accuracy": 0.23880353569984436,
      "num_tokens": 145522974.0,
      "step": 70125
    },
    {
      "entropy": 5.722359275817871,
      "epoch": 6.361574746008708,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018377994043787681,
      "loss": 4.8505,
      "mean_token_accuracy": 0.23607783764600754,
      "num_tokens": 145534169.0,
      "step": 70130
    },
    {
      "entropy": 5.704586410522461,
      "epoch": 6.3620283018867925,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018375036342574918,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23244306147098542,
      "num_tokens": 145544720.0,
      "step": 70135
    },
    {
      "entropy": 5.7351487159729,
      "epoch": 6.362481857764877,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00018372078830068727,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2316556066274643,
      "num_tokens": 145555040.0,
      "step": 70140
    },
    {
      "entropy": 5.78014726638794,
      "epoch": 6.362935413642961,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018369121506330277,
      "loss": 4.8683,
      "mean_token_accuracy": 0.23834515362977982,
      "num_tokens": 145566247.0,
      "step": 70145
    },
    {
      "entropy": 5.630336761474609,
      "epoch": 6.363388969521045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018366164371420718,
      "loss": 4.7619,
      "mean_token_accuracy": 0.23726095557212828,
      "num_tokens": 145576970.0,
      "step": 70150
    },
    {
      "entropy": 5.687159729003906,
      "epoch": 6.3638425253991295,
      "grad_norm": 1.25,
      "learning_rate": 0.00018363207425401207,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2334800750017166,
      "num_tokens": 145587816.0,
      "step": 70155
    },
    {
      "entropy": 5.783616971969605,
      "epoch": 6.364296081277214,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001836025066833289,
      "loss": 4.9463,
      "mean_token_accuracy": 0.22308497428894042,
      "num_tokens": 145598572.0,
      "step": 70160
    },
    {
      "entropy": 5.7513867855072025,
      "epoch": 6.364749637155297,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018357294100276923,
      "loss": 4.814,
      "mean_token_accuracy": 0.23182342946529388,
      "num_tokens": 145609596.0,
      "step": 70165
    },
    {
      "entropy": 5.79822006225586,
      "epoch": 6.3652031930333814,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018354337721294446,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23155371248722076,
      "num_tokens": 145619516.0,
      "step": 70170
    },
    {
      "entropy": 5.679126214981079,
      "epoch": 6.365656748911466,
      "grad_norm": 1.625,
      "learning_rate": 0.00018351381531446587,
      "loss": 4.8087,
      "mean_token_accuracy": 0.2366683676838875,
      "num_tokens": 145629052.0,
      "step": 70175
    },
    {
      "entropy": 5.647406196594238,
      "epoch": 6.36611030478955,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018348425530794499,
      "loss": 4.7832,
      "mean_token_accuracy": 0.23986810445785522,
      "num_tokens": 145638821.0,
      "step": 70180
    },
    {
      "entropy": 5.690026712417603,
      "epoch": 6.366563860667634,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000183454697193993,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23511693626642227,
      "num_tokens": 145649295.0,
      "step": 70185
    },
    {
      "entropy": 5.686995792388916,
      "epoch": 6.3670174165457185,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018342514097322121,
      "loss": 4.8133,
      "mean_token_accuracy": 0.23725005090236664,
      "num_tokens": 145659556.0,
      "step": 70190
    },
    {
      "entropy": 5.767138385772705,
      "epoch": 6.367470972423803,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001833955866462408,
      "loss": 4.8833,
      "mean_token_accuracy": 0.2294931009411812,
      "num_tokens": 145669381.0,
      "step": 70195
    },
    {
      "entropy": 5.720695686340332,
      "epoch": 6.367924528301887,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00018336603421366304,
      "loss": 4.7799,
      "mean_token_accuracy": 0.23347796201705934,
      "num_tokens": 145678837.0,
      "step": 70200
    },
    {
      "entropy": 5.731848192214966,
      "epoch": 6.368378084179971,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018333648367609907,
      "loss": 4.8841,
      "mean_token_accuracy": 0.23087610006332399,
      "num_tokens": 145688751.0,
      "step": 70205
    },
    {
      "entropy": 5.686236810684204,
      "epoch": 6.368831640058055,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018330693503415985,
      "loss": 4.7597,
      "mean_token_accuracy": 0.24403734058141707,
      "num_tokens": 145698635.0,
      "step": 70210
    },
    {
      "entropy": 5.62174243927002,
      "epoch": 6.369285195936139,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018327738828845675,
      "loss": 4.7559,
      "mean_token_accuracy": 0.2315164178609848,
      "num_tokens": 145708309.0,
      "step": 70215
    },
    {
      "entropy": 5.687505865097046,
      "epoch": 6.369738751814223,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018324784343960054,
      "loss": 4.8036,
      "mean_token_accuracy": 0.23935557305812835,
      "num_tokens": 145718373.0,
      "step": 70220
    },
    {
      "entropy": 5.728559780120849,
      "epoch": 6.3701923076923075,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018321830048820227,
      "loss": 4.8375,
      "mean_token_accuracy": 0.2350381150841713,
      "num_tokens": 145728830.0,
      "step": 70225
    },
    {
      "entropy": 5.708759927749634,
      "epoch": 6.370645863570392,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018318875943487302,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23406125158071517,
      "num_tokens": 145739272.0,
      "step": 70230
    },
    {
      "entropy": 5.713709449768066,
      "epoch": 6.371099419448476,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018315922028022357,
      "loss": 4.8627,
      "mean_token_accuracy": 0.2351989343762398,
      "num_tokens": 145750834.0,
      "step": 70235
    },
    {
      "entropy": 5.63247766494751,
      "epoch": 6.37155297532656,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018312968302486489,
      "loss": 4.7812,
      "mean_token_accuracy": 0.2420559525489807,
      "num_tokens": 145762536.0,
      "step": 70240
    },
    {
      "entropy": 5.738820362091064,
      "epoch": 6.3720065312046446,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018310014766940774,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2310422569513321,
      "num_tokens": 145772889.0,
      "step": 70245
    },
    {
      "entropy": 5.7297117710113525,
      "epoch": 6.372460087082729,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018307061421446292,
      "loss": 4.849,
      "mean_token_accuracy": 0.2350467711687088,
      "num_tokens": 145783553.0,
      "step": 70250
    },
    {
      "entropy": 5.769515466690064,
      "epoch": 6.372913642960813,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018304108266064135,
      "loss": 4.9869,
      "mean_token_accuracy": 0.22518827617168427,
      "num_tokens": 145793056.0,
      "step": 70255
    },
    {
      "entropy": 5.797899055480957,
      "epoch": 6.373367198838897,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001830115530085535,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2319054365158081,
      "num_tokens": 145803245.0,
      "step": 70260
    },
    {
      "entropy": 5.741237163543701,
      "epoch": 6.373820754716981,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001829820252588102,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23998288363218306,
      "num_tokens": 145813603.0,
      "step": 70265
    },
    {
      "entropy": 5.724849271774292,
      "epoch": 6.374274310595065,
      "grad_norm": 1.65625,
      "learning_rate": 0.00018295249941202208,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2321416988968849,
      "num_tokens": 145823792.0,
      "step": 70270
    },
    {
      "entropy": 5.723431253433228,
      "epoch": 6.374727866473149,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018292297546879978,
      "loss": 4.8617,
      "mean_token_accuracy": 0.23108412325382233,
      "num_tokens": 145835329.0,
      "step": 70275
    },
    {
      "entropy": 5.723123550415039,
      "epoch": 6.3751814223512335,
      "grad_norm": 1.65625,
      "learning_rate": 0.00018289345342975373,
      "loss": 4.8147,
      "mean_token_accuracy": 0.23419201523065566,
      "num_tokens": 145846214.0,
      "step": 70280
    },
    {
      "entropy": 5.620950222015381,
      "epoch": 6.375634978229318,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018286393329549456,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23731584399938582,
      "num_tokens": 145857406.0,
      "step": 70285
    },
    {
      "entropy": 5.690269517898559,
      "epoch": 6.376088534107402,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018283441506663278,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23044320046901703,
      "num_tokens": 145867823.0,
      "step": 70290
    },
    {
      "entropy": 5.696664142608642,
      "epoch": 6.376542089985486,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001828048987437787,
      "loss": 4.861,
      "mean_token_accuracy": 0.2296757236123085,
      "num_tokens": 145878267.0,
      "step": 70295
    },
    {
      "entropy": 5.709789752960205,
      "epoch": 6.376995645863571,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018277538432754286,
      "loss": 4.837,
      "mean_token_accuracy": 0.23420707732439042,
      "num_tokens": 145888776.0,
      "step": 70300
    },
    {
      "entropy": 5.765915822982788,
      "epoch": 6.377449201741655,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018274587181853558,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2229839012026787,
      "num_tokens": 145899084.0,
      "step": 70305
    },
    {
      "entropy": 5.6998738765716555,
      "epoch": 6.377902757619739,
      "grad_norm": 1.5,
      "learning_rate": 0.0001827163612173671,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23339347839355468,
      "num_tokens": 145910105.0,
      "step": 70310
    },
    {
      "entropy": 5.687453413009644,
      "epoch": 6.378356313497823,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001826868525246479,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23609803020954132,
      "num_tokens": 145920877.0,
      "step": 70315
    },
    {
      "entropy": 5.671487760543823,
      "epoch": 6.378809869375907,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018265734574098808,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23677227199077605,
      "num_tokens": 145931567.0,
      "step": 70320
    },
    {
      "entropy": 5.667222595214843,
      "epoch": 6.379263425253991,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018262784086699784,
      "loss": 4.799,
      "mean_token_accuracy": 0.23211256712675093,
      "num_tokens": 145941450.0,
      "step": 70325
    },
    {
      "entropy": 5.754470682144165,
      "epoch": 6.379716981132075,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001825983379032874,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23132820874452592,
      "num_tokens": 145952229.0,
      "step": 70330
    },
    {
      "entropy": 5.606310224533081,
      "epoch": 6.38017053701016,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018256883685046694,
      "loss": 4.6566,
      "mean_token_accuracy": 0.25308469235897063,
      "num_tokens": 145962835.0,
      "step": 70335
    },
    {
      "entropy": 5.783209037780762,
      "epoch": 6.380624092888244,
      "grad_norm": 1.375,
      "learning_rate": 0.00018253933770914644,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2284923568367958,
      "num_tokens": 145974543.0,
      "step": 70340
    },
    {
      "entropy": 5.734971284866333,
      "epoch": 6.381077648766328,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018250984047993597,
      "loss": 4.8598,
      "mean_token_accuracy": 0.23454180210828782,
      "num_tokens": 145985076.0,
      "step": 70345
    },
    {
      "entropy": 5.750672006607056,
      "epoch": 6.381531204644412,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018248034516344557,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2258136823773384,
      "num_tokens": 145996338.0,
      "step": 70350
    },
    {
      "entropy": 5.694254541397095,
      "epoch": 6.381984760522497,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001824508517602853,
      "loss": 4.7547,
      "mean_token_accuracy": 0.2411262333393097,
      "num_tokens": 146007088.0,
      "step": 70355
    },
    {
      "entropy": 5.732867622375489,
      "epoch": 6.382438316400581,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018242136027106493,
      "loss": 4.8936,
      "mean_token_accuracy": 0.22749026119709015,
      "num_tokens": 146017907.0,
      "step": 70360
    },
    {
      "entropy": 5.72150182723999,
      "epoch": 6.382891872278664,
      "grad_norm": 1.625,
      "learning_rate": 0.00018239187069639441,
      "loss": 4.8433,
      "mean_token_accuracy": 0.23070623129606246,
      "num_tokens": 146027801.0,
      "step": 70365
    },
    {
      "entropy": 5.6849065780639645,
      "epoch": 6.3833454281567485,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018236238303688364,
      "loss": 4.7714,
      "mean_token_accuracy": 0.23400918245315552,
      "num_tokens": 146037532.0,
      "step": 70370
    },
    {
      "entropy": 5.669039154052735,
      "epoch": 6.383798984034833,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001823328972931424,
      "loss": 4.7726,
      "mean_token_accuracy": 0.23460208475589753,
      "num_tokens": 146047420.0,
      "step": 70375
    },
    {
      "entropy": 5.759241056442261,
      "epoch": 6.384252539912917,
      "grad_norm": 1.5,
      "learning_rate": 0.0001823034134657804,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23705968558788298,
      "num_tokens": 146057926.0,
      "step": 70380
    },
    {
      "entropy": 5.719886255264282,
      "epoch": 6.384706095791001,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018227393155540743,
      "loss": 4.8914,
      "mean_token_accuracy": 0.23170557469129563,
      "num_tokens": 146068734.0,
      "step": 70385
    },
    {
      "entropy": 5.6225341796875,
      "epoch": 6.385159651669086,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001822444515626333,
      "loss": 4.6983,
      "mean_token_accuracy": 0.2413717433810234,
      "num_tokens": 146079252.0,
      "step": 70390
    },
    {
      "entropy": 5.686128187179565,
      "epoch": 6.38561320754717,
      "grad_norm": 1.375,
      "learning_rate": 0.00018221497348806743,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23645755499601365,
      "num_tokens": 146089724.0,
      "step": 70395
    },
    {
      "entropy": 5.7774162769317625,
      "epoch": 6.386066763425254,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001821854973323196,
      "loss": 4.9177,
      "mean_token_accuracy": 0.2216261386871338,
      "num_tokens": 146100092.0,
      "step": 70400
    },
    {
      "entropy": 5.725089025497437,
      "epoch": 6.386520319303338,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001821560230959993,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23929842114448546,
      "num_tokens": 146110958.0,
      "step": 70405
    },
    {
      "entropy": 5.729790115356446,
      "epoch": 6.386973875181423,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018212655077971617,
      "loss": 4.7732,
      "mean_token_accuracy": 0.23526514023542405,
      "num_tokens": 146120880.0,
      "step": 70410
    },
    {
      "entropy": 5.6538347721099855,
      "epoch": 6.387427431059507,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018209708038407956,
      "loss": 4.8037,
      "mean_token_accuracy": 0.23867130279541016,
      "num_tokens": 146130625.0,
      "step": 70415
    },
    {
      "entropy": 5.664519214630127,
      "epoch": 6.38788098693759,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018206761190969905,
      "loss": 4.8103,
      "mean_token_accuracy": 0.24080341905355454,
      "num_tokens": 146141474.0,
      "step": 70420
    },
    {
      "entropy": 5.712248945236206,
      "epoch": 6.388334542815675,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018203814535718406,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2300987347960472,
      "num_tokens": 146152064.0,
      "step": 70425
    },
    {
      "entropy": 5.718474006652832,
      "epoch": 6.388788098693759,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018200868072714384,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22394658625125885,
      "num_tokens": 146162427.0,
      "step": 70430
    },
    {
      "entropy": 5.787578105926514,
      "epoch": 6.389241654571843,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001819792180201878,
      "loss": 4.9933,
      "mean_token_accuracy": 0.22297780215740204,
      "num_tokens": 146173569.0,
      "step": 70435
    },
    {
      "entropy": 5.701943349838257,
      "epoch": 6.389695210449927,
      "grad_norm": 1.375,
      "learning_rate": 0.00018194975723692525,
      "loss": 4.7771,
      "mean_token_accuracy": 0.23423608094453813,
      "num_tokens": 146183197.0,
      "step": 70440
    },
    {
      "entropy": 5.651017379760742,
      "epoch": 6.390148766328012,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018192029837796542,
      "loss": 4.7871,
      "mean_token_accuracy": 0.23125408887863158,
      "num_tokens": 146193332.0,
      "step": 70445
    },
    {
      "entropy": 5.737072467803955,
      "epoch": 6.390602322206096,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018189084144391754,
      "loss": 4.905,
      "mean_token_accuracy": 0.22622291147708892,
      "num_tokens": 146203345.0,
      "step": 70450
    },
    {
      "entropy": 5.801781940460205,
      "epoch": 6.39105587808418,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001818613864353908,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22411325126886367,
      "num_tokens": 146212988.0,
      "step": 70455
    },
    {
      "entropy": 5.855278587341308,
      "epoch": 6.3915094339622645,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018183193335299437,
      "loss": 4.9803,
      "mean_token_accuracy": 0.22321960628032683,
      "num_tokens": 146224216.0,
      "step": 70460
    },
    {
      "entropy": 5.717117547988892,
      "epoch": 6.391962989840349,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018180248219733719,
      "loss": 4.81,
      "mean_token_accuracy": 0.23677492439746856,
      "num_tokens": 146234955.0,
      "step": 70465
    },
    {
      "entropy": 5.678185939788818,
      "epoch": 6.392416545718432,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018177303296902853,
      "loss": 4.7914,
      "mean_token_accuracy": 0.24044689238071443,
      "num_tokens": 146245394.0,
      "step": 70470
    },
    {
      "entropy": 5.682695913314819,
      "epoch": 6.392870101596516,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018174358566867726,
      "loss": 4.7574,
      "mean_token_accuracy": 0.240975221991539,
      "num_tokens": 146254842.0,
      "step": 70475
    },
    {
      "entropy": 5.68006739616394,
      "epoch": 6.393323657474601,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018171414029689243,
      "loss": 4.7988,
      "mean_token_accuracy": 0.22505468875169754,
      "num_tokens": 146265627.0,
      "step": 70480
    },
    {
      "entropy": 5.6642004489898685,
      "epoch": 6.393777213352685,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018168469685428285,
      "loss": 4.7649,
      "mean_token_accuracy": 0.2369389608502388,
      "num_tokens": 146275263.0,
      "step": 70485
    },
    {
      "entropy": 5.705706071853638,
      "epoch": 6.394230769230769,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001816552553414576,
      "loss": 4.8278,
      "mean_token_accuracy": 0.2390711098909378,
      "num_tokens": 146285553.0,
      "step": 70490
    },
    {
      "entropy": 5.750981998443604,
      "epoch": 6.394684325108853,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018162581575902547,
      "loss": 4.8546,
      "mean_token_accuracy": 0.23646675944328308,
      "num_tokens": 146295256.0,
      "step": 70495
    },
    {
      "entropy": 5.6785746097564695,
      "epoch": 6.395137880986938,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018159637810759527,
      "loss": 4.8509,
      "mean_token_accuracy": 0.22823170721530914,
      "num_tokens": 146304818.0,
      "step": 70500
    },
    {
      "entropy": 5.777908658981323,
      "epoch": 6.395591436865022,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001815669423877757,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21933305263519287,
      "num_tokens": 146315349.0,
      "step": 70505
    },
    {
      "entropy": 5.698064947128296,
      "epoch": 6.396044992743106,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001815375086001757,
      "loss": 4.7802,
      "mean_token_accuracy": 0.2320660337805748,
      "num_tokens": 146325426.0,
      "step": 70510
    },
    {
      "entropy": 5.788116550445556,
      "epoch": 6.3964985486211905,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018150807674540383,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2347804054617882,
      "num_tokens": 146334807.0,
      "step": 70515
    },
    {
      "entropy": 5.719933414459229,
      "epoch": 6.396952104499274,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018147864682406867,
      "loss": 4.8058,
      "mean_token_accuracy": 0.2354842558503151,
      "num_tokens": 146344519.0,
      "step": 70520
    },
    {
      "entropy": 5.685208654403686,
      "epoch": 6.397405660377358,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001814492188367789,
      "loss": 4.7794,
      "mean_token_accuracy": 0.24078905284404756,
      "num_tokens": 146355283.0,
      "step": 70525
    },
    {
      "entropy": 5.681673145294189,
      "epoch": 6.397859216255442,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018141979278414318,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23479366153478623,
      "num_tokens": 146366050.0,
      "step": 70530
    },
    {
      "entropy": 5.65433349609375,
      "epoch": 6.398312772133527,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018139036866677,
      "loss": 4.8006,
      "mean_token_accuracy": 0.24062508195638657,
      "num_tokens": 146376131.0,
      "step": 70535
    },
    {
      "entropy": 5.680834102630615,
      "epoch": 6.398766328011611,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018136094648526786,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23496696650981902,
      "num_tokens": 146386328.0,
      "step": 70540
    },
    {
      "entropy": 5.752797794342041,
      "epoch": 6.399219883889695,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001813315262402452,
      "loss": 4.852,
      "mean_token_accuracy": 0.2393464282155037,
      "num_tokens": 146396498.0,
      "step": 70545
    },
    {
      "entropy": 5.711153316497803,
      "epoch": 6.3996734397677795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018130210793231033,
      "loss": 4.8451,
      "mean_token_accuracy": 0.23013011664152144,
      "num_tokens": 146406765.0,
      "step": 70550
    },
    {
      "entropy": 5.716722822189331,
      "epoch": 6.400126995645864,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018127269156207183,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2313359022140503,
      "num_tokens": 146417079.0,
      "step": 70555
    },
    {
      "entropy": 5.7537590026855465,
      "epoch": 6.400580551523948,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000181243277130138,
      "loss": 4.8461,
      "mean_token_accuracy": 0.23249889314174652,
      "num_tokens": 146426797.0,
      "step": 70560
    },
    {
      "entropy": 5.761621952056885,
      "epoch": 6.401034107402032,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018121386463711703,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2247883215546608,
      "num_tokens": 146437605.0,
      "step": 70565
    },
    {
      "entropy": 5.6909568309783936,
      "epoch": 6.4014876632801165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018118445408361728,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23960227370262147,
      "num_tokens": 146448280.0,
      "step": 70570
    },
    {
      "entropy": 5.689118051528931,
      "epoch": 6.4019412191582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001811550454702469,
      "loss": 4.836,
      "mean_token_accuracy": 0.23107632994651794,
      "num_tokens": 146458887.0,
      "step": 70575
    },
    {
      "entropy": 5.727928829193115,
      "epoch": 6.402394775036284,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018112563879761417,
      "loss": 4.7945,
      "mean_token_accuracy": 0.2340507388114929,
      "num_tokens": 146468715.0,
      "step": 70580
    },
    {
      "entropy": 5.732639932632447,
      "epoch": 6.4028483309143684,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018109623406632703,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22715349942445756,
      "num_tokens": 146479243.0,
      "step": 70585
    },
    {
      "entropy": 5.72926812171936,
      "epoch": 6.403301886792453,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018106683127699374,
      "loss": 4.8477,
      "mean_token_accuracy": 0.23651461452245712,
      "num_tokens": 146488824.0,
      "step": 70590
    },
    {
      "entropy": 5.66126070022583,
      "epoch": 6.403755442670537,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018103743043022236,
      "loss": 4.8218,
      "mean_token_accuracy": 0.23492249846458435,
      "num_tokens": 146498920.0,
      "step": 70595
    },
    {
      "entropy": 5.728295564651489,
      "epoch": 6.404208998548621,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018100803152662083,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2279101073741913,
      "num_tokens": 146509510.0,
      "step": 70600
    },
    {
      "entropy": 5.6998475074768065,
      "epoch": 6.4046625544267055,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001809786345667972,
      "loss": 4.8234,
      "mean_token_accuracy": 0.2360050082206726,
      "num_tokens": 146520426.0,
      "step": 70605
    },
    {
      "entropy": 5.730823993682861,
      "epoch": 6.40511611030479,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018094923955135934,
      "loss": 4.845,
      "mean_token_accuracy": 0.23722218871116638,
      "num_tokens": 146530012.0,
      "step": 70610
    },
    {
      "entropy": 5.756723690032959,
      "epoch": 6.405569666182874,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018091984648091513,
      "loss": 4.861,
      "mean_token_accuracy": 0.2347150519490242,
      "num_tokens": 146540594.0,
      "step": 70615
    },
    {
      "entropy": 5.681533861160278,
      "epoch": 6.406023222060958,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018089045535607247,
      "loss": 4.8283,
      "mean_token_accuracy": 0.2315276101231575,
      "num_tokens": 146551527.0,
      "step": 70620
    },
    {
      "entropy": 5.703603506088257,
      "epoch": 6.406476777939042,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001808610661774392,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23706597536802293,
      "num_tokens": 146562428.0,
      "step": 70625
    },
    {
      "entropy": 5.629204225540161,
      "epoch": 6.406930333817126,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001808316789456231,
      "loss": 4.7575,
      "mean_token_accuracy": 0.23209803253412248,
      "num_tokens": 146572923.0,
      "step": 70630
    },
    {
      "entropy": 5.685846185684204,
      "epoch": 6.40738388969521,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018080229366123184,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22924392074346542,
      "num_tokens": 146583949.0,
      "step": 70635
    },
    {
      "entropy": 5.670234632492066,
      "epoch": 6.4078374455732945,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001807729103248732,
      "loss": 4.8052,
      "mean_token_accuracy": 0.2346929430961609,
      "num_tokens": 146593586.0,
      "step": 70640
    },
    {
      "entropy": 5.686408567428589,
      "epoch": 6.408291001451379,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001807435289371548,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22823875099420549,
      "num_tokens": 146604133.0,
      "step": 70645
    },
    {
      "entropy": 5.67485203742981,
      "epoch": 6.408744557329463,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018071414949868413,
      "loss": 4.8146,
      "mean_token_accuracy": 0.2342243954539299,
      "num_tokens": 146614469.0,
      "step": 70650
    },
    {
      "entropy": 5.683402729034424,
      "epoch": 6.409198113207547,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018068477201006888,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23497851490974425,
      "num_tokens": 146624926.0,
      "step": 70655
    },
    {
      "entropy": 5.700759553909302,
      "epoch": 6.4096516690856316,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001806553964719167,
      "loss": 4.7512,
      "mean_token_accuracy": 0.24302223175764084,
      "num_tokens": 146634967.0,
      "step": 70660
    },
    {
      "entropy": 5.6676935195922855,
      "epoch": 6.410105224963716,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018062602288483492,
      "loss": 4.7813,
      "mean_token_accuracy": 0.2361939013004303,
      "num_tokens": 146645242.0,
      "step": 70665
    },
    {
      "entropy": 5.73809323310852,
      "epoch": 6.4105587808418,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018059665124943103,
      "loss": 4.8147,
      "mean_token_accuracy": 0.23518799096345902,
      "num_tokens": 146654480.0,
      "step": 70670
    },
    {
      "entropy": 5.729962539672852,
      "epoch": 6.4110123367198835,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018056728156631241,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22688669264316558,
      "num_tokens": 146664663.0,
      "step": 70675
    },
    {
      "entropy": 5.679760932922363,
      "epoch": 6.411465892597968,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018053791383608654,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23200079351663588,
      "num_tokens": 146674734.0,
      "step": 70680
    },
    {
      "entropy": 5.692693281173706,
      "epoch": 6.411919448476052,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018050854805936063,
      "loss": 4.8606,
      "mean_token_accuracy": 0.23218037635087968,
      "num_tokens": 146683682.0,
      "step": 70685
    },
    {
      "entropy": 5.7420401096344,
      "epoch": 6.412373004354136,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018047918423674198,
      "loss": 4.8543,
      "mean_token_accuracy": 0.22431911379098893,
      "num_tokens": 146693991.0,
      "step": 70690
    },
    {
      "entropy": 5.750612497329712,
      "epoch": 6.4128265602322205,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018044982236883794,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22932361215353012,
      "num_tokens": 146704711.0,
      "step": 70695
    },
    {
      "entropy": 5.652996015548706,
      "epoch": 6.413280116110305,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001804204624562557,
      "loss": 4.7366,
      "mean_token_accuracy": 0.23681730777025223,
      "num_tokens": 146714177.0,
      "step": 70700
    },
    {
      "entropy": 5.629668378829956,
      "epoch": 6.413733671988389,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001803911044996024,
      "loss": 4.7045,
      "mean_token_accuracy": 0.2390146717429161,
      "num_tokens": 146723563.0,
      "step": 70705
    },
    {
      "entropy": 5.761636018753052,
      "epoch": 6.414187227866473,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001803617484994851,
      "loss": 4.8929,
      "mean_token_accuracy": 0.22444778233766555,
      "num_tokens": 146732997.0,
      "step": 70710
    },
    {
      "entropy": 5.727337884902954,
      "epoch": 6.414640783744558,
      "grad_norm": 1.53125,
      "learning_rate": 0.000180332394456511,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22260825335979462,
      "num_tokens": 146742949.0,
      "step": 70715
    },
    {
      "entropy": 5.695150661468506,
      "epoch": 6.415094339622642,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018030304237128713,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23951640129089355,
      "num_tokens": 146753167.0,
      "step": 70720
    },
    {
      "entropy": 5.677291059494019,
      "epoch": 6.415547895500726,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018027369224442042,
      "loss": 4.8218,
      "mean_token_accuracy": 0.2362108737230301,
      "num_tokens": 146763028.0,
      "step": 70725
    },
    {
      "entropy": 5.666226053237915,
      "epoch": 6.4160014513788095,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001802443440765179,
      "loss": 4.8527,
      "mean_token_accuracy": 0.22786120623350142,
      "num_tokens": 146774620.0,
      "step": 70730
    },
    {
      "entropy": 5.71726975440979,
      "epoch": 6.416455007256894,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001802149978681865,
      "loss": 4.8536,
      "mean_token_accuracy": 0.23019926697015763,
      "num_tokens": 146785126.0,
      "step": 70735
    },
    {
      "entropy": 5.76117730140686,
      "epoch": 6.416908563134978,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018018565362003315,
      "loss": 4.7864,
      "mean_token_accuracy": 0.23862154632806779,
      "num_tokens": 146794297.0,
      "step": 70740
    },
    {
      "entropy": 5.706498432159424,
      "epoch": 6.417362119013062,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018015631133266465,
      "loss": 4.74,
      "mean_token_accuracy": 0.24243807047605515,
      "num_tokens": 146805513.0,
      "step": 70745
    },
    {
      "entropy": 5.685661315917969,
      "epoch": 6.417815674891147,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018012697100668772,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22953783124685287,
      "num_tokens": 146815106.0,
      "step": 70750
    },
    {
      "entropy": 5.683173799514771,
      "epoch": 6.418269230769231,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018009763264270924,
      "loss": 4.8132,
      "mean_token_accuracy": 0.2360678046941757,
      "num_tokens": 146825302.0,
      "step": 70755
    },
    {
      "entropy": 5.671889591217041,
      "epoch": 6.418722786647315,
      "grad_norm": 1.5,
      "learning_rate": 0.00018006829624133585,
      "loss": 4.7945,
      "mean_token_accuracy": 0.23800333887338637,
      "num_tokens": 146836389.0,
      "step": 70760
    },
    {
      "entropy": 5.752557706832886,
      "epoch": 6.419176342525399,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018003896180317436,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21976514160633087,
      "num_tokens": 146846778.0,
      "step": 70765
    },
    {
      "entropy": 5.712267589569092,
      "epoch": 6.419629898403484,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018000962932883136,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23278138637542725,
      "num_tokens": 146857042.0,
      "step": 70770
    },
    {
      "entropy": 5.705101728439331,
      "epoch": 6.420083454281568,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017998029881891337,
      "loss": 4.7895,
      "mean_token_accuracy": 0.2350945770740509,
      "num_tokens": 146866795.0,
      "step": 70775
    },
    {
      "entropy": 5.6891473770141605,
      "epoch": 6.420537010159651,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017995097027402707,
      "loss": 4.7818,
      "mean_token_accuracy": 0.23699991405010223,
      "num_tokens": 146876889.0,
      "step": 70780
    },
    {
      "entropy": 5.725992679595947,
      "epoch": 6.4209905660377355,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017992164369477892,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22722319960594178,
      "num_tokens": 146887250.0,
      "step": 70785
    },
    {
      "entropy": 5.626702451705933,
      "epoch": 6.42144412191582,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001798923190817753,
      "loss": 4.718,
      "mean_token_accuracy": 0.24258360862731934,
      "num_tokens": 146897237.0,
      "step": 70790
    },
    {
      "entropy": 5.766167449951172,
      "epoch": 6.421897677793904,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017986299643562282,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2260460928082466,
      "num_tokens": 146910547.0,
      "step": 70795
    },
    {
      "entropy": 5.7891950607299805,
      "epoch": 6.422351233671988,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017983367575692788,
      "loss": 4.9208,
      "mean_token_accuracy": 0.22901972085237504,
      "num_tokens": 146922809.0,
      "step": 70800
    },
    {
      "entropy": 5.715320825576782,
      "epoch": 6.422804789550073,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001798043570462967,
      "loss": 4.8891,
      "mean_token_accuracy": 0.22505521327257155,
      "num_tokens": 146933804.0,
      "step": 70805
    },
    {
      "entropy": 5.817822408676148,
      "epoch": 6.423258345428157,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017977504030433573,
      "loss": 4.8971,
      "mean_token_accuracy": 0.2263585552573204,
      "num_tokens": 146944071.0,
      "step": 70810
    },
    {
      "entropy": 5.724524974822998,
      "epoch": 6.423711901306241,
      "grad_norm": 1.375,
      "learning_rate": 0.0001797457255316512,
      "loss": 4.8515,
      "mean_token_accuracy": 0.2314885899424553,
      "num_tokens": 146954918.0,
      "step": 70815
    },
    {
      "entropy": 5.660332489013672,
      "epoch": 6.424165457184325,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017971641272884932,
      "loss": 4.7387,
      "mean_token_accuracy": 0.23378390073776245,
      "num_tokens": 146964984.0,
      "step": 70820
    },
    {
      "entropy": 5.680329370498657,
      "epoch": 6.42461901306241,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001796871018965362,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2350657179951668,
      "num_tokens": 146974407.0,
      "step": 70825
    },
    {
      "entropy": 5.700923728942871,
      "epoch": 6.425072568940493,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017965779303531822,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23462052941322326,
      "num_tokens": 146985037.0,
      "step": 70830
    },
    {
      "entropy": 5.627496719360352,
      "epoch": 6.425526124818577,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017962848614580136,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23220048248767852,
      "num_tokens": 146995539.0,
      "step": 70835
    },
    {
      "entropy": 5.691107130050659,
      "epoch": 6.425979680696662,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001795991812285917,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23321322798728944,
      "num_tokens": 147005418.0,
      "step": 70840
    },
    {
      "entropy": 5.68921217918396,
      "epoch": 6.426433236574746,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017956987828429534,
      "loss": 4.7527,
      "mean_token_accuracy": 0.2349328577518463,
      "num_tokens": 147016209.0,
      "step": 70845
    },
    {
      "entropy": 5.698661661148071,
      "epoch": 6.42688679245283,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017954057731351815,
      "loss": 4.7746,
      "mean_token_accuracy": 0.23956458270549774,
      "num_tokens": 147027745.0,
      "step": 70850
    },
    {
      "entropy": 5.718867588043213,
      "epoch": 6.427340348330914,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017951127831686615,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23543084263801575,
      "num_tokens": 147038061.0,
      "step": 70855
    },
    {
      "entropy": 5.677894830703735,
      "epoch": 6.427793904208999,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001794819812949452,
      "loss": 4.816,
      "mean_token_accuracy": 0.23717377930879593,
      "num_tokens": 147048791.0,
      "step": 70860
    },
    {
      "entropy": 5.739093589782715,
      "epoch": 6.428247460087083,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017945268624836125,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2242904394865036,
      "num_tokens": 147059561.0,
      "step": 70865
    },
    {
      "entropy": 5.662107133865357,
      "epoch": 6.428701015965167,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017942339317772015,
      "loss": 4.7602,
      "mean_token_accuracy": 0.23822778165340425,
      "num_tokens": 147069868.0,
      "step": 70870
    },
    {
      "entropy": 5.677222204208374,
      "epoch": 6.4291545718432515,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001793941020836276,
      "loss": 4.8091,
      "mean_token_accuracy": 0.23179276436567306,
      "num_tokens": 147079153.0,
      "step": 70875
    },
    {
      "entropy": 5.781295919418335,
      "epoch": 6.429608127721336,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017936481296668938,
      "loss": 4.888,
      "mean_token_accuracy": 0.2261670097708702,
      "num_tokens": 147090191.0,
      "step": 70880
    },
    {
      "entropy": 5.624687719345093,
      "epoch": 6.430061683599419,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017933552582751123,
      "loss": 4.7307,
      "mean_token_accuracy": 0.2370669052004814,
      "num_tokens": 147100076.0,
      "step": 70885
    },
    {
      "entropy": 5.663266372680664,
      "epoch": 6.430515239477503,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017930624066669882,
      "loss": 4.753,
      "mean_token_accuracy": 0.24308146238327027,
      "num_tokens": 147109639.0,
      "step": 70890
    },
    {
      "entropy": 5.672377204895019,
      "epoch": 6.430968795355588,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017927695748485756,
      "loss": 4.8352,
      "mean_token_accuracy": 0.22931897938251494,
      "num_tokens": 147120325.0,
      "step": 70895
    },
    {
      "entropy": 5.695535898208618,
      "epoch": 6.431422351233672,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017924767628259338,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2346455782651901,
      "num_tokens": 147130638.0,
      "step": 70900
    },
    {
      "entropy": 5.697269773483276,
      "epoch": 6.431875907111756,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017921839706051156,
      "loss": 4.8028,
      "mean_token_accuracy": 0.2319232627749443,
      "num_tokens": 147141128.0,
      "step": 70905
    },
    {
      "entropy": 5.678957080841064,
      "epoch": 6.43232946298984,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017918911981921776,
      "loss": 4.7456,
      "mean_token_accuracy": 0.24070467203855514,
      "num_tokens": 147151490.0,
      "step": 70910
    },
    {
      "entropy": 5.650431156158447,
      "epoch": 6.432783018867925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001791598445593174,
      "loss": 4.7569,
      "mean_token_accuracy": 0.23825024664402009,
      "num_tokens": 147162741.0,
      "step": 70915
    },
    {
      "entropy": 5.651977634429931,
      "epoch": 6.433236574746009,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017913057128141586,
      "loss": 4.7918,
      "mean_token_accuracy": 0.23894250839948655,
      "num_tokens": 147172622.0,
      "step": 70920
    },
    {
      "entropy": 5.678831100463867,
      "epoch": 6.433690130624093,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017910129998611852,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2387595757842064,
      "num_tokens": 147183133.0,
      "step": 70925
    },
    {
      "entropy": 5.68528380393982,
      "epoch": 6.434143686502177,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001790720306740306,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23511189669370652,
      "num_tokens": 147192766.0,
      "step": 70930
    },
    {
      "entropy": 5.753165292739868,
      "epoch": 6.434597242380261,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001790427633457578,
      "loss": 4.8514,
      "mean_token_accuracy": 0.23301711082458496,
      "num_tokens": 147204765.0,
      "step": 70935
    },
    {
      "entropy": 5.692088413238525,
      "epoch": 6.435050798258345,
      "grad_norm": 1.625,
      "learning_rate": 0.00017901349800190496,
      "loss": 4.8476,
      "mean_token_accuracy": 0.23799291849136353,
      "num_tokens": 147215092.0,
      "step": 70940
    },
    {
      "entropy": 5.696939897537232,
      "epoch": 6.435504354136429,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017898423464307752,
      "loss": 4.848,
      "mean_token_accuracy": 0.232370787858963,
      "num_tokens": 147226678.0,
      "step": 70945
    },
    {
      "entropy": 5.696112489700317,
      "epoch": 6.435957910014514,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017895497326988048,
      "loss": 4.845,
      "mean_token_accuracy": 0.22500832974910737,
      "num_tokens": 147236575.0,
      "step": 70950
    },
    {
      "entropy": 5.652151870727539,
      "epoch": 6.436411465892598,
      "grad_norm": 1.5,
      "learning_rate": 0.000178925713882919,
      "loss": 4.832,
      "mean_token_accuracy": 0.23669513165950776,
      "num_tokens": 147246968.0,
      "step": 70955
    },
    {
      "entropy": 5.6197608470916744,
      "epoch": 6.436865021770682,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017889645648279834,
      "loss": 4.7294,
      "mean_token_accuracy": 0.24403427839279174,
      "num_tokens": 147257370.0,
      "step": 70960
    },
    {
      "entropy": 5.730651092529297,
      "epoch": 6.4373185776487665,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001788672010701235,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2321549892425537,
      "num_tokens": 147268422.0,
      "step": 70965
    },
    {
      "entropy": 5.700937938690186,
      "epoch": 6.437772133526851,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001788379476454993,
      "loss": 4.7992,
      "mean_token_accuracy": 0.23563717752695085,
      "num_tokens": 147279057.0,
      "step": 70970
    },
    {
      "entropy": 5.641712951660156,
      "epoch": 6.438225689404935,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000178808696209531,
      "loss": 4.7716,
      "mean_token_accuracy": 0.2401732623577118,
      "num_tokens": 147290734.0,
      "step": 70975
    },
    {
      "entropy": 5.684827613830566,
      "epoch": 6.438679245283019,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017877944676282333,
      "loss": 4.8109,
      "mean_token_accuracy": 0.2303181156516075,
      "num_tokens": 147301361.0,
      "step": 70980
    },
    {
      "entropy": 5.744139719009399,
      "epoch": 6.439132801161103,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017875019930598118,
      "loss": 4.8444,
      "mean_token_accuracy": 0.2371288776397705,
      "num_tokens": 147311232.0,
      "step": 70985
    },
    {
      "entropy": 5.753862619400024,
      "epoch": 6.439586357039187,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017872095383960934,
      "loss": 4.9048,
      "mean_token_accuracy": 0.22838907390832902,
      "num_tokens": 147320810.0,
      "step": 70990
    },
    {
      "entropy": 5.669847536087036,
      "epoch": 6.440039912917271,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017869171036431286,
      "loss": 4.7532,
      "mean_token_accuracy": 0.2422761633992195,
      "num_tokens": 147331290.0,
      "step": 70995
    },
    {
      "entropy": 5.668242692947388,
      "epoch": 6.4404934687953554,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017866246888069632,
      "loss": 4.742,
      "mean_token_accuracy": 0.23321637213230134,
      "num_tokens": 147341239.0,
      "step": 71000
    },
    {
      "entropy": 5.640512752532959,
      "epoch": 6.44094702467344,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001786332293893645,
      "loss": 4.793,
      "mean_token_accuracy": 0.23955928534269333,
      "num_tokens": 147351877.0,
      "step": 71005
    },
    {
      "entropy": 5.639070987701416,
      "epoch": 6.441400580551524,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017860399189092196,
      "loss": 4.8156,
      "mean_token_accuracy": 0.23349910527467727,
      "num_tokens": 147362382.0,
      "step": 71010
    },
    {
      "entropy": 5.6619163990020756,
      "epoch": 6.441854136429608,
      "grad_norm": 1.609375,
      "learning_rate": 0.00017857475638597352,
      "loss": 4.7244,
      "mean_token_accuracy": 0.23820935636758805,
      "num_tokens": 147372441.0,
      "step": 71015
    },
    {
      "entropy": 5.718052911758423,
      "epoch": 6.4423076923076925,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001785455228751237,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2376662462949753,
      "num_tokens": 147383544.0,
      "step": 71020
    },
    {
      "entropy": 5.733886003494263,
      "epoch": 6.442761248185777,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001785162913589769,
      "loss": 4.7656,
      "mean_token_accuracy": 0.24309698790311812,
      "num_tokens": 147394047.0,
      "step": 71025
    },
    {
      "entropy": 5.7066246509552006,
      "epoch": 6.443214804063861,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001784870618381379,
      "loss": 4.889,
      "mean_token_accuracy": 0.23156434595584868,
      "num_tokens": 147405881.0,
      "step": 71030
    },
    {
      "entropy": 5.695529508590698,
      "epoch": 6.443668359941945,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00017845783431321105,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22693922221660615,
      "num_tokens": 147415807.0,
      "step": 71035
    },
    {
      "entropy": 5.758642768859863,
      "epoch": 6.444121915820029,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017842860878480082,
      "loss": 4.9509,
      "mean_token_accuracy": 0.22923022806644439,
      "num_tokens": 147425878.0,
      "step": 71040
    },
    {
      "entropy": 5.7666082859039305,
      "epoch": 6.444575471698113,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017839938525351156,
      "loss": 4.8059,
      "mean_token_accuracy": 0.2346212834119797,
      "num_tokens": 147435518.0,
      "step": 71045
    },
    {
      "entropy": 5.765372276306152,
      "epoch": 6.445029027576197,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017837016371994763,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2278841346502304,
      "num_tokens": 147447475.0,
      "step": 71050
    },
    {
      "entropy": 5.757219648361206,
      "epoch": 6.4454825834542815,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001783409441847133,
      "loss": 4.9018,
      "mean_token_accuracy": 0.23007603436708451,
      "num_tokens": 147458301.0,
      "step": 71055
    },
    {
      "entropy": 5.694183444976806,
      "epoch": 6.445936139332366,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017831172664841278,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23456026017665862,
      "num_tokens": 147468092.0,
      "step": 71060
    },
    {
      "entropy": 5.766922330856323,
      "epoch": 6.44638969521045,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017828251111165055,
      "loss": 4.8657,
      "mean_token_accuracy": 0.23162149935960769,
      "num_tokens": 147478602.0,
      "step": 71065
    },
    {
      "entropy": 5.681288576126098,
      "epoch": 6.446843251088534,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017825329757503055,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24471033960580826,
      "num_tokens": 147488333.0,
      "step": 71070
    },
    {
      "entropy": 5.723830604553223,
      "epoch": 6.4472968069666186,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017822408603915707,
      "loss": 4.7618,
      "mean_token_accuracy": 0.24328127205371858,
      "num_tokens": 147498306.0,
      "step": 71075
    },
    {
      "entropy": 5.698511266708374,
      "epoch": 6.447750362844703,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017819487650463413,
      "loss": 4.834,
      "mean_token_accuracy": 0.2290411040186882,
      "num_tokens": 147508339.0,
      "step": 71080
    },
    {
      "entropy": 5.673756265640259,
      "epoch": 6.448203918722786,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017816566897206583,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2363726630806923,
      "num_tokens": 147519631.0,
      "step": 71085
    },
    {
      "entropy": 5.692460250854492,
      "epoch": 6.4486574746008705,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001781364634420561,
      "loss": 4.8651,
      "mean_token_accuracy": 0.23007919937372207,
      "num_tokens": 147529824.0,
      "step": 71090
    },
    {
      "entropy": 5.682571077346802,
      "epoch": 6.449111030478955,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017810725991520903,
      "loss": 4.724,
      "mean_token_accuracy": 0.24917980134487153,
      "num_tokens": 147539902.0,
      "step": 71095
    },
    {
      "entropy": 5.70580677986145,
      "epoch": 6.449564586357039,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017807805839212853,
      "loss": 4.889,
      "mean_token_accuracy": 0.22996657490730285,
      "num_tokens": 147551083.0,
      "step": 71100
    },
    {
      "entropy": 5.725620174407959,
      "epoch": 6.450018142235123,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017804885887341847,
      "loss": 4.9207,
      "mean_token_accuracy": 0.22658213078975678,
      "num_tokens": 147561751.0,
      "step": 71105
    },
    {
      "entropy": 5.7225110054016115,
      "epoch": 6.4504716981132075,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017801966135968272,
      "loss": 4.873,
      "mean_token_accuracy": 0.2230396658182144,
      "num_tokens": 147572563.0,
      "step": 71110
    },
    {
      "entropy": 5.77375226020813,
      "epoch": 6.450925253991292,
      "grad_norm": 1.578125,
      "learning_rate": 0.00017799046585152513,
      "loss": 4.8731,
      "mean_token_accuracy": 0.24240701347589494,
      "num_tokens": 147582537.0,
      "step": 71115
    },
    {
      "entropy": 5.793610858917236,
      "epoch": 6.451378809869376,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001779612723495494,
      "loss": 4.8953,
      "mean_token_accuracy": 0.22771514803171158,
      "num_tokens": 147594516.0,
      "step": 71120
    },
    {
      "entropy": 5.741343832015991,
      "epoch": 6.45183236574746,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017793208085435918,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22752640694379805,
      "num_tokens": 147606151.0,
      "step": 71125
    },
    {
      "entropy": 5.702996873855591,
      "epoch": 6.452285921625545,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001779028913665584,
      "loss": 4.84,
      "mean_token_accuracy": 0.2274933785200119,
      "num_tokens": 147616138.0,
      "step": 71130
    },
    {
      "entropy": 5.735477924346924,
      "epoch": 6.452739477503629,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017787370388675056,
      "loss": 4.8297,
      "mean_token_accuracy": 0.2333693102002144,
      "num_tokens": 147626597.0,
      "step": 71135
    },
    {
      "entropy": 5.662895345687867,
      "epoch": 6.453193033381712,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017784451841553923,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2324958249926567,
      "num_tokens": 147637234.0,
      "step": 71140
    },
    {
      "entropy": 5.752129650115966,
      "epoch": 6.4536465892597965,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017781533495352803,
      "loss": 4.9317,
      "mean_token_accuracy": 0.22682391107082367,
      "num_tokens": 147648112.0,
      "step": 71145
    },
    {
      "entropy": 5.687780475616455,
      "epoch": 6.454100145137881,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001777861535013205,
      "loss": 4.8289,
      "mean_token_accuracy": 0.23696191012859344,
      "num_tokens": 147658605.0,
      "step": 71150
    },
    {
      "entropy": 5.70307674407959,
      "epoch": 6.454553701015965,
      "grad_norm": 1.65625,
      "learning_rate": 0.00017775697405952002,
      "loss": 4.7666,
      "mean_token_accuracy": 0.24435267746448516,
      "num_tokens": 147669192.0,
      "step": 71155
    },
    {
      "entropy": 5.729100561141967,
      "epoch": 6.455007256894049,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017772779662873008,
      "loss": 4.7604,
      "mean_token_accuracy": 0.23712778091430664,
      "num_tokens": 147678532.0,
      "step": 71160
    },
    {
      "entropy": 5.728821849822998,
      "epoch": 6.455460812772134,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017769862120955415,
      "loss": 4.846,
      "mean_token_accuracy": 0.22800759673118592,
      "num_tokens": 147688792.0,
      "step": 71165
    },
    {
      "entropy": 5.623590564727783,
      "epoch": 6.455914368650218,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017766944780259552,
      "loss": 4.7544,
      "mean_token_accuracy": 0.23349268585443497,
      "num_tokens": 147699111.0,
      "step": 71170
    },
    {
      "entropy": 5.730262041091919,
      "epoch": 6.456367924528302,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017764027640845752,
      "loss": 4.8994,
      "mean_token_accuracy": 0.2251575246453285,
      "num_tokens": 147709404.0,
      "step": 71175
    },
    {
      "entropy": 5.717969703674316,
      "epoch": 6.456821480406386,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017761110702774346,
      "loss": 4.8125,
      "mean_token_accuracy": 0.23508322685956956,
      "num_tokens": 147720865.0,
      "step": 71180
    },
    {
      "entropy": 5.716017436981201,
      "epoch": 6.457275036284471,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017758193966105646,
      "loss": 4.7926,
      "mean_token_accuracy": 0.23812198042869567,
      "num_tokens": 147730317.0,
      "step": 71185
    },
    {
      "entropy": 5.706196737289429,
      "epoch": 6.457728592162554,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001775527743089998,
      "loss": 4.8159,
      "mean_token_accuracy": 0.2409118890762329,
      "num_tokens": 147740536.0,
      "step": 71190
    },
    {
      "entropy": 5.705328035354614,
      "epoch": 6.458182148040638,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001775236109721765,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23682954758405686,
      "num_tokens": 147749918.0,
      "step": 71195
    },
    {
      "entropy": 5.623924493789673,
      "epoch": 6.4586357039187225,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001774944496511899,
      "loss": 4.6993,
      "mean_token_accuracy": 0.24888624250888824,
      "num_tokens": 147761258.0,
      "step": 71200
    },
    {
      "entropy": 5.770474243164062,
      "epoch": 6.459089259796807,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017746529034664288,
      "loss": 4.9391,
      "mean_token_accuracy": 0.22964007705450057,
      "num_tokens": 147772808.0,
      "step": 71205
    },
    {
      "entropy": 5.7523890972137455,
      "epoch": 6.459542815674891,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017743613305913852,
      "loss": 4.8689,
      "mean_token_accuracy": 0.2237837702035904,
      "num_tokens": 147782906.0,
      "step": 71210
    },
    {
      "entropy": 5.628430032730103,
      "epoch": 6.459996371552975,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001774069777892798,
      "loss": 4.7389,
      "mean_token_accuracy": 0.23935502767562866,
      "num_tokens": 147792638.0,
      "step": 71215
    },
    {
      "entropy": 5.6045904636383055,
      "epoch": 6.46044992743106,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017737782453766966,
      "loss": 4.7936,
      "mean_token_accuracy": 0.23057248592376708,
      "num_tokens": 147803649.0,
      "step": 71220
    },
    {
      "entropy": 5.635188961029053,
      "epoch": 6.460903483309144,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017734867330491095,
      "loss": 4.8048,
      "mean_token_accuracy": 0.23102509677410127,
      "num_tokens": 147813872.0,
      "step": 71225
    },
    {
      "entropy": 5.714914989471436,
      "epoch": 6.461357039187228,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001773195240916065,
      "loss": 4.902,
      "mean_token_accuracy": 0.22919033467769623,
      "num_tokens": 147825260.0,
      "step": 71230
    },
    {
      "entropy": 5.6626218318939205,
      "epoch": 6.461810595065312,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017729037689835926,
      "loss": 4.7539,
      "mean_token_accuracy": 0.23905389308929442,
      "num_tokens": 147835607.0,
      "step": 71235
    },
    {
      "entropy": 5.7382666110992435,
      "epoch": 6.462264150943396,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001772612317257719,
      "loss": 4.874,
      "mean_token_accuracy": 0.2305854618549347,
      "num_tokens": 147845435.0,
      "step": 71240
    },
    {
      "entropy": 5.715415716171265,
      "epoch": 6.46271770682148,
      "grad_norm": 1.5,
      "learning_rate": 0.00017723208857444722,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2375014066696167,
      "num_tokens": 147855974.0,
      "step": 71245
    },
    {
      "entropy": 5.708133125305176,
      "epoch": 6.463171262699564,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001772029474449879,
      "loss": 4.8004,
      "mean_token_accuracy": 0.2425512731075287,
      "num_tokens": 147865658.0,
      "step": 71250
    },
    {
      "entropy": 5.697375297546387,
      "epoch": 6.463624818577649,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017717380833799645,
      "loss": 4.8448,
      "mean_token_accuracy": 0.23308936804533004,
      "num_tokens": 147876159.0,
      "step": 71255
    },
    {
      "entropy": 5.677821922302246,
      "epoch": 6.464078374455733,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017714467125407558,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22692082226276397,
      "num_tokens": 147885608.0,
      "step": 71260
    },
    {
      "entropy": 5.694507789611817,
      "epoch": 6.464531930333817,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001771155361938278,
      "loss": 4.8421,
      "mean_token_accuracy": 0.23626298308372498,
      "num_tokens": 147896664.0,
      "step": 71265
    },
    {
      "entropy": 5.738249731063843,
      "epoch": 6.464985486211901,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017708640315785578,
      "loss": 4.8798,
      "mean_token_accuracy": 0.2247924968600273,
      "num_tokens": 147908639.0,
      "step": 71270
    },
    {
      "entropy": 5.68536524772644,
      "epoch": 6.465439042089986,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017705727214676188,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22730724811553954,
      "num_tokens": 147918721.0,
      "step": 71275
    },
    {
      "entropy": 5.660280752182007,
      "epoch": 6.46589259796807,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017702814316114857,
      "loss": 4.811,
      "mean_token_accuracy": 0.24015709310770034,
      "num_tokens": 147929965.0,
      "step": 71280
    },
    {
      "entropy": 5.669826936721802,
      "epoch": 6.466346153846154,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017699901620161818,
      "loss": 4.7807,
      "mean_token_accuracy": 0.23646599650382996,
      "num_tokens": 147940102.0,
      "step": 71285
    },
    {
      "entropy": 5.75422101020813,
      "epoch": 6.4667997097242385,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001769698912687731,
      "loss": 4.8668,
      "mean_token_accuracy": 0.23182375133037567,
      "num_tokens": 147950411.0,
      "step": 71290
    },
    {
      "entropy": 5.69544415473938,
      "epoch": 6.467253265602322,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001769407683632156,
      "loss": 4.8173,
      "mean_token_accuracy": 0.2314540207386017,
      "num_tokens": 147962618.0,
      "step": 71295
    },
    {
      "entropy": 5.678116798400879,
      "epoch": 6.467706821480406,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017691164748554806,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23496493846178054,
      "num_tokens": 147973305.0,
      "step": 71300
    },
    {
      "entropy": 5.674277973175049,
      "epoch": 6.46816037735849,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017688252863637266,
      "loss": 4.7776,
      "mean_token_accuracy": 0.23689865022897721,
      "num_tokens": 147983442.0,
      "step": 71305
    },
    {
      "entropy": 5.6681255340576175,
      "epoch": 6.468613933236575,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001768534118162916,
      "loss": 4.7607,
      "mean_token_accuracy": 0.2350291758775711,
      "num_tokens": 147993606.0,
      "step": 71310
    },
    {
      "entropy": 5.749480533599853,
      "epoch": 6.469067489114659,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017682429702590698,
      "loss": 4.8417,
      "mean_token_accuracy": 0.2319296509027481,
      "num_tokens": 148005021.0,
      "step": 71315
    },
    {
      "entropy": 5.686738538742065,
      "epoch": 6.469521044992743,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017679518426582085,
      "loss": 4.8011,
      "mean_token_accuracy": 0.23734811544418336,
      "num_tokens": 148015432.0,
      "step": 71320
    },
    {
      "entropy": 5.75280499458313,
      "epoch": 6.469974600870827,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017676607353663538,
      "loss": 4.873,
      "mean_token_accuracy": 0.23264372497797012,
      "num_tokens": 148025521.0,
      "step": 71325
    },
    {
      "entropy": 5.708115148544311,
      "epoch": 6.470428156748912,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017673696483895245,
      "loss": 4.8823,
      "mean_token_accuracy": 0.23280919939279557,
      "num_tokens": 148037436.0,
      "step": 71330
    },
    {
      "entropy": 5.694458246231079,
      "epoch": 6.470881712626996,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017670785817337426,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2342131569981575,
      "num_tokens": 148048090.0,
      "step": 71335
    },
    {
      "entropy": 5.690529489517212,
      "epoch": 6.47133526850508,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017667875354050256,
      "loss": 4.7991,
      "mean_token_accuracy": 0.23426274508237838,
      "num_tokens": 148057783.0,
      "step": 71340
    },
    {
      "entropy": 5.693074798583984,
      "epoch": 6.471788824383164,
      "grad_norm": 1.375,
      "learning_rate": 0.00017664965094093933,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22912612706422805,
      "num_tokens": 148068259.0,
      "step": 71345
    },
    {
      "entropy": 5.7840193748474125,
      "epoch": 6.472242380261248,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017662055037528636,
      "loss": 4.927,
      "mean_token_accuracy": 0.2212515264749527,
      "num_tokens": 148079127.0,
      "step": 71350
    },
    {
      "entropy": 5.680845594406128,
      "epoch": 6.472695936139332,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001765914518441455,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2405536115169525,
      "num_tokens": 148089588.0,
      "step": 71355
    },
    {
      "entropy": 5.72136607170105,
      "epoch": 6.473149492017416,
      "grad_norm": 1.5,
      "learning_rate": 0.00017656235534811842,
      "loss": 4.8385,
      "mean_token_accuracy": 0.23578199595212937,
      "num_tokens": 148099880.0,
      "step": 71360
    },
    {
      "entropy": 5.696648836135864,
      "epoch": 6.473603047895501,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017653326088780698,
      "loss": 4.8335,
      "mean_token_accuracy": 0.22698826640844344,
      "num_tokens": 148111484.0,
      "step": 71365
    },
    {
      "entropy": 5.6695550918579105,
      "epoch": 6.474056603773585,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001765041684638128,
      "loss": 4.8426,
      "mean_token_accuracy": 0.23409832119941712,
      "num_tokens": 148123676.0,
      "step": 71370
    },
    {
      "entropy": 5.690120553970337,
      "epoch": 6.474510159651669,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001764750780767375,
      "loss": 4.7967,
      "mean_token_accuracy": 0.23349715173244476,
      "num_tokens": 148133257.0,
      "step": 71375
    },
    {
      "entropy": 5.717583227157593,
      "epoch": 6.4749637155297535,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017644598972718268,
      "loss": 4.7743,
      "mean_token_accuracy": 0.23637299090623856,
      "num_tokens": 148143548.0,
      "step": 71380
    },
    {
      "entropy": 5.602769565582276,
      "epoch": 6.475417271407838,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017641690341574996,
      "loss": 4.7331,
      "mean_token_accuracy": 0.23588496595621108,
      "num_tokens": 148153837.0,
      "step": 71385
    },
    {
      "entropy": 5.695160007476806,
      "epoch": 6.475870827285922,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017638781914304063,
      "loss": 4.8255,
      "mean_token_accuracy": 0.22810632586479188,
      "num_tokens": 148164672.0,
      "step": 71390
    },
    {
      "entropy": 5.7161726474761965,
      "epoch": 6.476324383164005,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017635873690965647,
      "loss": 4.8104,
      "mean_token_accuracy": 0.2359452098608017,
      "num_tokens": 148174330.0,
      "step": 71395
    },
    {
      "entropy": 5.766201114654541,
      "epoch": 6.47677793904209,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017632965671619878,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22616257071495055,
      "num_tokens": 148185227.0,
      "step": 71400
    },
    {
      "entropy": 5.845149517059326,
      "epoch": 6.477231494920174,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017630057856326886,
      "loss": 4.9832,
      "mean_token_accuracy": 0.2251088634133339,
      "num_tokens": 148196487.0,
      "step": 71405
    },
    {
      "entropy": 5.690674686431885,
      "epoch": 6.477685050798258,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017627150245146822,
      "loss": 4.8549,
      "mean_token_accuracy": 0.2281600669026375,
      "num_tokens": 148206170.0,
      "step": 71410
    },
    {
      "entropy": 5.701501131057739,
      "epoch": 6.4781386066763424,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017624242838139805,
      "loss": 4.8538,
      "mean_token_accuracy": 0.2357405200600624,
      "num_tokens": 148216930.0,
      "step": 71415
    },
    {
      "entropy": 5.754270839691162,
      "epoch": 6.478592162554427,
      "grad_norm": 1.5,
      "learning_rate": 0.00017621335635365954,
      "loss": 4.9035,
      "mean_token_accuracy": 0.22653175741434098,
      "num_tokens": 148227401.0,
      "step": 71420
    },
    {
      "entropy": 5.654256772994995,
      "epoch": 6.479045718432511,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017618428636885402,
      "loss": 4.7857,
      "mean_token_accuracy": 0.23970262557268143,
      "num_tokens": 148237655.0,
      "step": 71425
    },
    {
      "entropy": 5.711587619781494,
      "epoch": 6.479499274310595,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017615521842758263,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22802733033895492,
      "num_tokens": 148248417.0,
      "step": 71430
    },
    {
      "entropy": 5.710081624984741,
      "epoch": 6.4799528301886795,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017612615253044655,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22785090506076813,
      "num_tokens": 148258070.0,
      "step": 71435
    },
    {
      "entropy": 5.7217100143432615,
      "epoch": 6.480406386066764,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017609708867804684,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23472338169813156,
      "num_tokens": 148267877.0,
      "step": 71440
    },
    {
      "entropy": 5.681832027435303,
      "epoch": 6.480859941944848,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017606802687098456,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23800572603940964,
      "num_tokens": 148277946.0,
      "step": 71445
    },
    {
      "entropy": 5.740044212341308,
      "epoch": 6.481313497822931,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017603896710986067,
      "loss": 4.9012,
      "mean_token_accuracy": 0.23549850583076476,
      "num_tokens": 148289158.0,
      "step": 71450
    },
    {
      "entropy": 5.731240510940552,
      "epoch": 6.481767053701016,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017600990939527618,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22551062405109407,
      "num_tokens": 148299398.0,
      "step": 71455
    },
    {
      "entropy": 5.720298910140992,
      "epoch": 6.4822206095791,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017598085372783184,
      "loss": 4.8657,
      "mean_token_accuracy": 0.23138342946767806,
      "num_tokens": 148309935.0,
      "step": 71460
    },
    {
      "entropy": 5.673817300796509,
      "epoch": 6.482674165457184,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017595180010812882,
      "loss": 4.7771,
      "mean_token_accuracy": 0.2333161100745201,
      "num_tokens": 148320851.0,
      "step": 71465
    },
    {
      "entropy": 5.686814069747925,
      "epoch": 6.4831277213352685,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017592274853676782,
      "loss": 4.7989,
      "mean_token_accuracy": 0.233808171749115,
      "num_tokens": 148330807.0,
      "step": 71470
    },
    {
      "entropy": 5.6356700420379635,
      "epoch": 6.483581277213353,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017589369901434965,
      "loss": 4.787,
      "mean_token_accuracy": 0.2382635533809662,
      "num_tokens": 148342193.0,
      "step": 71475
    },
    {
      "entropy": 5.661986541748047,
      "epoch": 6.484034833091437,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017586465154147496,
      "loss": 4.7941,
      "mean_token_accuracy": 0.23306164294481277,
      "num_tokens": 148351867.0,
      "step": 71480
    },
    {
      "entropy": 5.6757566928863525,
      "epoch": 6.484488388969521,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001758356061187446,
      "loss": 4.7621,
      "mean_token_accuracy": 0.2371470510959625,
      "num_tokens": 148362213.0,
      "step": 71485
    },
    {
      "entropy": 5.715110158920288,
      "epoch": 6.4849419448476056,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017580656274675916,
      "loss": 4.8659,
      "mean_token_accuracy": 0.23361371755599974,
      "num_tokens": 148373172.0,
      "step": 71490
    },
    {
      "entropy": 5.676829624176025,
      "epoch": 6.485395500725689,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001757775214261192,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23643743246793747,
      "num_tokens": 148382780.0,
      "step": 71495
    },
    {
      "entropy": 5.6830919742584225,
      "epoch": 6.485849056603773,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017574848215742546,
      "loss": 4.8326,
      "mean_token_accuracy": 0.23547711074352265,
      "num_tokens": 148392652.0,
      "step": 71500
    },
    {
      "entropy": 5.676751136779785,
      "epoch": 6.4863026124818575,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001757194449412784,
      "loss": 4.7828,
      "mean_token_accuracy": 0.2371935561299324,
      "num_tokens": 148402536.0,
      "step": 71505
    },
    {
      "entropy": 5.6179427146911625,
      "epoch": 6.486756168359942,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001756904097782785,
      "loss": 4.7688,
      "mean_token_accuracy": 0.23743234425783158,
      "num_tokens": 148413202.0,
      "step": 71510
    },
    {
      "entropy": 5.744229698181153,
      "epoch": 6.487209724238026,
      "grad_norm": 1.5,
      "learning_rate": 0.00017566137666902626,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22671953290700914,
      "num_tokens": 148424741.0,
      "step": 71515
    },
    {
      "entropy": 5.681464529037475,
      "epoch": 6.48766328011611,
      "grad_norm": 1.375,
      "learning_rate": 0.00017563234561412207,
      "loss": 4.8407,
      "mean_token_accuracy": 0.22905945181846618,
      "num_tokens": 148435235.0,
      "step": 71520
    },
    {
      "entropy": 5.699384593963623,
      "epoch": 6.4881168359941945,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017560331661416623,
      "loss": 4.8279,
      "mean_token_accuracy": 0.22873536497354507,
      "num_tokens": 148445940.0,
      "step": 71525
    },
    {
      "entropy": 5.700925493240357,
      "epoch": 6.488570391872279,
      "grad_norm": 1.375,
      "learning_rate": 0.00017557428966975914,
      "loss": 4.8116,
      "mean_token_accuracy": 0.23354265838861465,
      "num_tokens": 148456383.0,
      "step": 71530
    },
    {
      "entropy": 5.6808828830719,
      "epoch": 6.489023947750363,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017554526478150113,
      "loss": 4.8825,
      "mean_token_accuracy": 0.22884398102760314,
      "num_tokens": 148467465.0,
      "step": 71535
    },
    {
      "entropy": 5.698465490341187,
      "epoch": 6.489477503628447,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017551624194999234,
      "loss": 4.7916,
      "mean_token_accuracy": 0.23835870772600173,
      "num_tokens": 148477127.0,
      "step": 71540
    },
    {
      "entropy": 5.639472198486328,
      "epoch": 6.489931059506532,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017548722117583304,
      "loss": 4.765,
      "mean_token_accuracy": 0.22583670020103455,
      "num_tokens": 148487390.0,
      "step": 71545
    },
    {
      "entropy": 5.732473993301392,
      "epoch": 6.490384615384615,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001754582024596234,
      "loss": 4.9499,
      "mean_token_accuracy": 0.2271898567676544,
      "num_tokens": 148497215.0,
      "step": 71550
    },
    {
      "entropy": 5.7014086723327635,
      "epoch": 6.490838171262699,
      "grad_norm": 1.5,
      "learning_rate": 0.0001754291858019635,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23997109979391099,
      "num_tokens": 148506922.0,
      "step": 71555
    },
    {
      "entropy": 5.640878534317016,
      "epoch": 6.4912917271407835,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017540017120345336,
      "loss": 4.812,
      "mean_token_accuracy": 0.22586410641670226,
      "num_tokens": 148517403.0,
      "step": 71560
    },
    {
      "entropy": 5.665939664840698,
      "epoch": 6.491745283018868,
      "grad_norm": 1.453125,
      "learning_rate": 0.000175371158664693,
      "loss": 4.7561,
      "mean_token_accuracy": 0.23748359829187393,
      "num_tokens": 148527676.0,
      "step": 71565
    },
    {
      "entropy": 5.745523357391358,
      "epoch": 6.492198838896952,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017534214818628257,
      "loss": 4.8441,
      "mean_token_accuracy": 0.2381623610854149,
      "num_tokens": 148537990.0,
      "step": 71570
    },
    {
      "entropy": 5.7289917945861815,
      "epoch": 6.492652394775036,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017531313976882192,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23886471688747407,
      "num_tokens": 148547467.0,
      "step": 71575
    },
    {
      "entropy": 5.736178731918335,
      "epoch": 6.493105950653121,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001752841334129109,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22977407276630402,
      "num_tokens": 148557206.0,
      "step": 71580
    },
    {
      "entropy": 5.72425103187561,
      "epoch": 6.493559506531205,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017525512911914943,
      "loss": 4.9674,
      "mean_token_accuracy": 0.21785872876644136,
      "num_tokens": 148568357.0,
      "step": 71585
    },
    {
      "entropy": 5.748760128021241,
      "epoch": 6.494013062409289,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017522612688813732,
      "loss": 4.84,
      "mean_token_accuracy": 0.22950478494167328,
      "num_tokens": 148577644.0,
      "step": 71590
    },
    {
      "entropy": 5.768467617034912,
      "epoch": 6.494466618287373,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001751971267204742,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2209431856870651,
      "num_tokens": 148588871.0,
      "step": 71595
    },
    {
      "entropy": 5.76017918586731,
      "epoch": 6.494920174165458,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017516812861676008,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2239644080400467,
      "num_tokens": 148598862.0,
      "step": 71600
    },
    {
      "entropy": 5.742852210998535,
      "epoch": 6.495373730043541,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001751391325775945,
      "loss": 4.8893,
      "mean_token_accuracy": 0.2301328480243683,
      "num_tokens": 148609912.0,
      "step": 71605
    },
    {
      "entropy": 5.7367900848388675,
      "epoch": 6.495827285921625,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017511013860357704,
      "loss": 4.8479,
      "mean_token_accuracy": 0.23246921151876448,
      "num_tokens": 148620390.0,
      "step": 71610
    },
    {
      "entropy": 5.741549825668335,
      "epoch": 6.4962808417997095,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017508114669530746,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22674290239810943,
      "num_tokens": 148630597.0,
      "step": 71615
    },
    {
      "entropy": 5.701026201248169,
      "epoch": 6.496734397677794,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017505215685338516,
      "loss": 4.7955,
      "mean_token_accuracy": 0.22995498776435852,
      "num_tokens": 148640550.0,
      "step": 71620
    },
    {
      "entropy": 5.7106099128723145,
      "epoch": 6.497187953555878,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017502316907840976,
      "loss": 4.7404,
      "mean_token_accuracy": 0.24402310699224472,
      "num_tokens": 148650038.0,
      "step": 71625
    },
    {
      "entropy": 5.760454225540161,
      "epoch": 6.497641509433962,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017499418337098058,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22184017151594163,
      "num_tokens": 148661013.0,
      "step": 71630
    },
    {
      "entropy": 5.671788835525513,
      "epoch": 6.498095065312047,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001749651997316973,
      "loss": 4.8487,
      "mean_token_accuracy": 0.22615660279989241,
      "num_tokens": 148671611.0,
      "step": 71635
    },
    {
      "entropy": 5.774231863021851,
      "epoch": 6.498548621190131,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017493621816115913,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2349378749728203,
      "num_tokens": 148681321.0,
      "step": 71640
    },
    {
      "entropy": 5.730592250823975,
      "epoch": 6.499002177068215,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017490723865996549,
      "loss": 4.826,
      "mean_token_accuracy": 0.23714344203472137,
      "num_tokens": 148691650.0,
      "step": 71645
    },
    {
      "entropy": 5.680110788345337,
      "epoch": 6.4994557329462985,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017487826122871566,
      "loss": 4.8188,
      "mean_token_accuracy": 0.23308492302894593,
      "num_tokens": 148702497.0,
      "step": 71650
    },
    {
      "entropy": 5.716888284683227,
      "epoch": 6.499909288824383,
      "grad_norm": 1.5,
      "learning_rate": 0.00017484928586800893,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2280501738190651,
      "num_tokens": 148712634.0,
      "step": 71655
    },
    {
      "entropy": 5.7548205852508545,
      "epoch": 6.500362844702467,
      "grad_norm": 1.5,
      "learning_rate": 0.00017482031257844447,
      "loss": 4.8937,
      "mean_token_accuracy": 0.23547202199697495,
      "num_tokens": 148722349.0,
      "step": 71660
    },
    {
      "entropy": 5.666384363174439,
      "epoch": 6.500816400580551,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001747913413606214,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23693014681339264,
      "num_tokens": 148733085.0,
      "step": 71665
    },
    {
      "entropy": 5.747865152359009,
      "epoch": 6.501269956458636,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000174762372215139,
      "loss": 4.8189,
      "mean_token_accuracy": 0.24107296168804168,
      "num_tokens": 148743695.0,
      "step": 71670
    },
    {
      "entropy": 5.7396022319793705,
      "epoch": 6.50172351233672,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017473340514259628,
      "loss": 4.8453,
      "mean_token_accuracy": 0.23208147585391997,
      "num_tokens": 148753980.0,
      "step": 71675
    },
    {
      "entropy": 5.673021984100342,
      "epoch": 6.502177068214804,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001747044401435923,
      "loss": 4.7817,
      "mean_token_accuracy": 0.23947502076625823,
      "num_tokens": 148765236.0,
      "step": 71680
    },
    {
      "entropy": 5.696552181243897,
      "epoch": 6.502630624092888,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017467547721872607,
      "loss": 4.8313,
      "mean_token_accuracy": 0.2268936350941658,
      "num_tokens": 148775316.0,
      "step": 71685
    },
    {
      "entropy": 5.706180572509766,
      "epoch": 6.503084179970973,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017464651636859652,
      "loss": 4.8517,
      "mean_token_accuracy": 0.22243440747261048,
      "num_tokens": 148784800.0,
      "step": 71690
    },
    {
      "entropy": 5.72543420791626,
      "epoch": 6.503537735849057,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001746175575938026,
      "loss": 4.819,
      "mean_token_accuracy": 0.2310737833380699,
      "num_tokens": 148794052.0,
      "step": 71695
    },
    {
      "entropy": 5.72935209274292,
      "epoch": 6.503991291727141,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017458860089494306,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22767609357833862,
      "num_tokens": 148805809.0,
      "step": 71700
    },
    {
      "entropy": 5.7116889476776125,
      "epoch": 6.504444847605225,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001745596462726169,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2370714008808136,
      "num_tokens": 148817003.0,
      "step": 71705
    },
    {
      "entropy": 5.674324893951416,
      "epoch": 6.504898403483309,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017453069372742292,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23802342265844345,
      "num_tokens": 148827301.0,
      "step": 71710
    },
    {
      "entropy": 5.689039802551269,
      "epoch": 6.505351959361393,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017450174325995977,
      "loss": 4.7964,
      "mean_token_accuracy": 0.23081085383892058,
      "num_tokens": 148837607.0,
      "step": 71715
    },
    {
      "entropy": 5.728352785110474,
      "epoch": 6.505805515239477,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017447279487082618,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2327496588230133,
      "num_tokens": 148847826.0,
      "step": 71720
    },
    {
      "entropy": 5.7060133934021,
      "epoch": 6.506259071117562,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001744438485606208,
      "loss": 4.8183,
      "mean_token_accuracy": 0.23345744907855986,
      "num_tokens": 148858271.0,
      "step": 71725
    },
    {
      "entropy": 5.680156278610229,
      "epoch": 6.506712626995646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017441490432994224,
      "loss": 4.8153,
      "mean_token_accuracy": 0.23459224104881288,
      "num_tokens": 148867840.0,
      "step": 71730
    },
    {
      "entropy": 5.7039495468139645,
      "epoch": 6.50716618287373,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017438596217938905,
      "loss": 4.7956,
      "mean_token_accuracy": 0.23299494087696077,
      "num_tokens": 148878820.0,
      "step": 71735
    },
    {
      "entropy": 5.6704448699951175,
      "epoch": 6.507619738751814,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017435702210955983,
      "loss": 4.7368,
      "mean_token_accuracy": 0.23959792852401735,
      "num_tokens": 148889076.0,
      "step": 71740
    },
    {
      "entropy": 5.768798398971557,
      "epoch": 6.508073294629899,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017432808412105308,
      "loss": 4.9209,
      "mean_token_accuracy": 0.23606379330158234,
      "num_tokens": 148898783.0,
      "step": 71745
    },
    {
      "entropy": 5.744433164596558,
      "epoch": 6.508526850507983,
      "grad_norm": 1.65625,
      "learning_rate": 0.00017429914821446722,
      "loss": 4.8806,
      "mean_token_accuracy": 0.23216688632965088,
      "num_tokens": 148908133.0,
      "step": 71750
    },
    {
      "entropy": 5.701012849807739,
      "epoch": 6.508980406386067,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001742702143904006,
      "loss": 4.8396,
      "mean_token_accuracy": 0.23826718032360078,
      "num_tokens": 148919104.0,
      "step": 71755
    },
    {
      "entropy": 5.722887468338013,
      "epoch": 6.509433962264151,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017424128264945164,
      "loss": 4.8547,
      "mean_token_accuracy": 0.2332828998565674,
      "num_tokens": 148930088.0,
      "step": 71760
    },
    {
      "entropy": 5.759407234191895,
      "epoch": 6.509887518142235,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001742123529922186,
      "loss": 4.859,
      "mean_token_accuracy": 0.22410778552293778,
      "num_tokens": 148939885.0,
      "step": 71765
    },
    {
      "entropy": 5.708763647079468,
      "epoch": 6.510341074020319,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001741834254192997,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22811733037233353,
      "num_tokens": 148950877.0,
      "step": 71770
    },
    {
      "entropy": 5.673473501205445,
      "epoch": 6.510794629898403,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001741544999312934,
      "loss": 4.7921,
      "mean_token_accuracy": 0.2353540077805519,
      "num_tokens": 148960022.0,
      "step": 71775
    },
    {
      "entropy": 5.675828981399536,
      "epoch": 6.511248185776488,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017412557652879767,
      "loss": 4.8248,
      "mean_token_accuracy": 0.23123179823160173,
      "num_tokens": 148970935.0,
      "step": 71780
    },
    {
      "entropy": 5.663238286972046,
      "epoch": 6.511701741654572,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017409665521241074,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23549281805753708,
      "num_tokens": 148982191.0,
      "step": 71785
    },
    {
      "entropy": 5.647547483444214,
      "epoch": 6.512155297532656,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017406773598273079,
      "loss": 4.7665,
      "mean_token_accuracy": 0.24579842239618302,
      "num_tokens": 148994093.0,
      "step": 71790
    },
    {
      "entropy": 5.728768539428711,
      "epoch": 6.5126088534107405,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017403881884035563,
      "loss": 4.8056,
      "mean_token_accuracy": 0.2350648045539856,
      "num_tokens": 149004595.0,
      "step": 71795
    },
    {
      "entropy": 5.7542150020599365,
      "epoch": 6.513062409288825,
      "grad_norm": 1.625,
      "learning_rate": 0.0001740099037858835,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2369346559047699,
      "num_tokens": 149014111.0,
      "step": 71800
    },
    {
      "entropy": 5.6470178127288815,
      "epoch": 6.513515965166908,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017398099081991232,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22745247930288315,
      "num_tokens": 149025597.0,
      "step": 71805
    },
    {
      "entropy": 5.7896435260772705,
      "epoch": 6.513969521044992,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017395207994303993,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2358950272202492,
      "num_tokens": 149036244.0,
      "step": 71810
    },
    {
      "entropy": 5.708968734741211,
      "epoch": 6.514423076923077,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017392317115586428,
      "loss": 4.7937,
      "mean_token_accuracy": 0.23333110362291337,
      "num_tokens": 149046196.0,
      "step": 71815
    },
    {
      "entropy": 5.703542232513428,
      "epoch": 6.514876632801161,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001738942644589832,
      "loss": 4.7882,
      "mean_token_accuracy": 0.23545939177274705,
      "num_tokens": 149057117.0,
      "step": 71820
    },
    {
      "entropy": 5.727682399749756,
      "epoch": 6.515330188679245,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017386535985299456,
      "loss": 4.8718,
      "mean_token_accuracy": 0.23421724885702133,
      "num_tokens": 149067360.0,
      "step": 71825
    },
    {
      "entropy": 5.733049583435059,
      "epoch": 6.5157837445573294,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017383645733849593,
      "loss": 4.8381,
      "mean_token_accuracy": 0.23847337812185287,
      "num_tokens": 149078226.0,
      "step": 71830
    },
    {
      "entropy": 5.614093160629272,
      "epoch": 6.516237300435414,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017380755691608524,
      "loss": 4.6976,
      "mean_token_accuracy": 0.24096086621284485,
      "num_tokens": 149088403.0,
      "step": 71835
    },
    {
      "entropy": 5.775616788864136,
      "epoch": 6.516690856313498,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017377865858636004,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2356860488653183,
      "num_tokens": 149099327.0,
      "step": 71840
    },
    {
      "entropy": 5.747296476364136,
      "epoch": 6.517144412191582,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000173749762349918,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22242130488157272,
      "num_tokens": 149109557.0,
      "step": 71845
    },
    {
      "entropy": 5.668387746810913,
      "epoch": 6.5175979680696665,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017372086820735666,
      "loss": 4.7631,
      "mean_token_accuracy": 0.23517660349607467,
      "num_tokens": 149120105.0,
      "step": 71850
    },
    {
      "entropy": 5.674514293670654,
      "epoch": 6.518051523947751,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00017369197615927357,
      "loss": 4.7804,
      "mean_token_accuracy": 0.24016808420419694,
      "num_tokens": 149130574.0,
      "step": 71855
    },
    {
      "entropy": 5.7054376125335695,
      "epoch": 6.518505079825834,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017366308620626623,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23728396445512773,
      "num_tokens": 149140970.0,
      "step": 71860
    },
    {
      "entropy": 5.713532018661499,
      "epoch": 6.518958635703918,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017363419834893202,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22932238727808,
      "num_tokens": 149151456.0,
      "step": 71865
    },
    {
      "entropy": 5.6602388381958,
      "epoch": 6.519412191582003,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001736053125878685,
      "loss": 4.7632,
      "mean_token_accuracy": 0.234443861246109,
      "num_tokens": 149162017.0,
      "step": 71870
    },
    {
      "entropy": 5.723539590835571,
      "epoch": 6.519865747460087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001735764289236729,
      "loss": 4.8177,
      "mean_token_accuracy": 0.23723961859941484,
      "num_tokens": 149173724.0,
      "step": 71875
    },
    {
      "entropy": 5.713969326019287,
      "epoch": 6.520319303338171,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017354754735694267,
      "loss": 4.8774,
      "mean_token_accuracy": 0.23261469900608062,
      "num_tokens": 149185288.0,
      "step": 71880
    },
    {
      "entropy": 5.638553333282471,
      "epoch": 6.5207728592162555,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017351866788827496,
      "loss": 4.7116,
      "mean_token_accuracy": 0.23796905875205993,
      "num_tokens": 149194994.0,
      "step": 71885
    },
    {
      "entropy": 5.738799619674682,
      "epoch": 6.52122641509434,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017348979051826712,
      "loss": 4.8724,
      "mean_token_accuracy": 0.2288164898753166,
      "num_tokens": 149205492.0,
      "step": 71890
    },
    {
      "entropy": 5.71259503364563,
      "epoch": 6.521679970972424,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017346091524751613,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23858111798763276,
      "num_tokens": 149214990.0,
      "step": 71895
    },
    {
      "entropy": 5.744389581680298,
      "epoch": 6.522133526850508,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017343204207661943,
      "loss": 4.8737,
      "mean_token_accuracy": 0.23567089140415193,
      "num_tokens": 149224218.0,
      "step": 71900
    },
    {
      "entropy": 5.718636131286621,
      "epoch": 6.522587082728592,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017340317100617396,
      "loss": 4.8527,
      "mean_token_accuracy": 0.23279243111610412,
      "num_tokens": 149235121.0,
      "step": 71905
    },
    {
      "entropy": 5.722566080093384,
      "epoch": 6.523040638606677,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017337430203677678,
      "loss": 4.8267,
      "mean_token_accuracy": 0.22899113595485687,
      "num_tokens": 149245743.0,
      "step": 71910
    },
    {
      "entropy": 5.747587490081787,
      "epoch": 6.52349419448476,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000173345435169025,
      "loss": 4.8077,
      "mean_token_accuracy": 0.2379833474755287,
      "num_tokens": 149256064.0,
      "step": 71915
    },
    {
      "entropy": 5.720508718490601,
      "epoch": 6.5239477503628445,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017331657040351546,
      "loss": 4.8403,
      "mean_token_accuracy": 0.2339043691754341,
      "num_tokens": 149265770.0,
      "step": 71920
    },
    {
      "entropy": 5.719367027282715,
      "epoch": 6.524401306240929,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017328770774084517,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22853144854307175,
      "num_tokens": 149275972.0,
      "step": 71925
    },
    {
      "entropy": 5.712953042984009,
      "epoch": 6.524854862119013,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001732588471816109,
      "loss": 4.827,
      "mean_token_accuracy": 0.23620035946369172,
      "num_tokens": 149286339.0,
      "step": 71930
    },
    {
      "entropy": 5.651980781555176,
      "epoch": 6.525308417997097,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017322998872640972,
      "loss": 4.7314,
      "mean_token_accuracy": 0.23913915902376176,
      "num_tokens": 149296802.0,
      "step": 71935
    },
    {
      "entropy": 5.709076309204102,
      "epoch": 6.5257619738751815,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001732011323758383,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2281265988945961,
      "num_tokens": 149308051.0,
      "step": 71940
    },
    {
      "entropy": 5.66795015335083,
      "epoch": 6.526215529753266,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017317227813049336,
      "loss": 4.7477,
      "mean_token_accuracy": 0.2361699789762497,
      "num_tokens": 149318254.0,
      "step": 71945
    },
    {
      "entropy": 5.687368631362915,
      "epoch": 6.52666908563135,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017314342599097172,
      "loss": 4.8556,
      "mean_token_accuracy": 0.23050604164600372,
      "num_tokens": 149329032.0,
      "step": 71950
    },
    {
      "entropy": 5.7371625900268555,
      "epoch": 6.527122641509434,
      "grad_norm": 1.375,
      "learning_rate": 0.00017311457595786995,
      "loss": 4.8687,
      "mean_token_accuracy": 0.23851078748703003,
      "num_tokens": 149339209.0,
      "step": 71955
    },
    {
      "entropy": 5.7339273452758786,
      "epoch": 6.527576197387518,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017308572803178472,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23483708947896959,
      "num_tokens": 149349067.0,
      "step": 71960
    },
    {
      "entropy": 5.670236253738404,
      "epoch": 6.528029753265602,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001730568822133125,
      "loss": 4.7394,
      "mean_token_accuracy": 0.24005953818559647,
      "num_tokens": 149359551.0,
      "step": 71965
    },
    {
      "entropy": 5.694903087615967,
      "epoch": 6.528483309143686,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017302803850305004,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22498073130846025,
      "num_tokens": 149369783.0,
      "step": 71970
    },
    {
      "entropy": 5.7166955947875975,
      "epoch": 6.5289368650217705,
      "grad_norm": 1.5,
      "learning_rate": 0.00017299919690159376,
      "loss": 4.8271,
      "mean_token_accuracy": 0.22958080768585204,
      "num_tokens": 149379414.0,
      "step": 71975
    },
    {
      "entropy": 5.785369777679444,
      "epoch": 6.529390420899855,
      "grad_norm": 1.625,
      "learning_rate": 0.00017297035740954003,
      "loss": 4.983,
      "mean_token_accuracy": 0.22607865780591965,
      "num_tokens": 149390024.0,
      "step": 71980
    },
    {
      "entropy": 5.696232175827026,
      "epoch": 6.529843976777939,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017294152002748534,
      "loss": 4.7944,
      "mean_token_accuracy": 0.24024925529956817,
      "num_tokens": 149401154.0,
      "step": 71985
    },
    {
      "entropy": 5.70639328956604,
      "epoch": 6.530297532656023,
      "grad_norm": 1.375,
      "learning_rate": 0.000172912684756026,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23578567653894425,
      "num_tokens": 149411067.0,
      "step": 71990
    },
    {
      "entropy": 5.748448657989502,
      "epoch": 6.530751088534108,
      "grad_norm": 1.5,
      "learning_rate": 0.0001728838515957584,
      "loss": 4.9353,
      "mean_token_accuracy": 0.22488908022642135,
      "num_tokens": 149421743.0,
      "step": 71995
    },
    {
      "entropy": 5.6530375480651855,
      "epoch": 6.531204644412192,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017285502054727864,
      "loss": 4.7511,
      "mean_token_accuracy": 0.24426826387643813,
      "num_tokens": 149432422.0,
      "step": 72000
    },
    {
      "epoch": 6.531204644412192,
      "eval_entropy": 5.3929178650023175,
      "eval_loss": 4.927300453186035,
      "eval_mean_token_accuracy": 0.23606229218827113,
      "eval_num_tokens": 149432422.0,
      "eval_runtime": 20.4946,
      "eval_samples_per_second": 1725.33,
      "eval_steps_per_second": 215.666,
      "step": 72000
    },
    {
      "entropy": 5.713057518005371,
      "epoch": 6.531658200290276,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001728261916111832,
      "loss": 4.83,
      "mean_token_accuracy": 0.23261955976486207,
      "num_tokens": 149443677.0,
      "step": 72005
    },
    {
      "entropy": 5.699325895309448,
      "epoch": 6.53211175616836,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001727973647880681,
      "loss": 4.7539,
      "mean_token_accuracy": 0.24245361983776093,
      "num_tokens": 149453574.0,
      "step": 72010
    },
    {
      "entropy": 5.649574661254883,
      "epoch": 6.532565312046444,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017276854007852957,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23111411184072495,
      "num_tokens": 149463714.0,
      "step": 72015
    },
    {
      "entropy": 5.6480310440063475,
      "epoch": 6.533018867924528,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017273971748316372,
      "loss": 4.792,
      "mean_token_accuracy": 0.23935135304927826,
      "num_tokens": 149474241.0,
      "step": 72020
    },
    {
      "entropy": 5.679630470275879,
      "epoch": 6.533472423802612,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017271089700256654,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22839495241642,
      "num_tokens": 149486032.0,
      "step": 72025
    },
    {
      "entropy": 5.6701171875,
      "epoch": 6.5339259796806965,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017268207863733407,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2369838535785675,
      "num_tokens": 149496829.0,
      "step": 72030
    },
    {
      "entropy": 5.687762212753296,
      "epoch": 6.534379535558781,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001726532623880622,
      "loss": 4.842,
      "mean_token_accuracy": 0.23934153318405152,
      "num_tokens": 149506930.0,
      "step": 72035
    },
    {
      "entropy": 5.694679069519043,
      "epoch": 6.534833091436865,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017262444825534706,
      "loss": 4.787,
      "mean_token_accuracy": 0.23195414245128632,
      "num_tokens": 149517259.0,
      "step": 72040
    },
    {
      "entropy": 5.686942100524902,
      "epoch": 6.535286647314949,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001725956362397844,
      "loss": 4.7772,
      "mean_token_accuracy": 0.2349265083670616,
      "num_tokens": 149528011.0,
      "step": 72045
    },
    {
      "entropy": 5.696550178527832,
      "epoch": 6.535740203193034,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017256682634197008,
      "loss": 4.8567,
      "mean_token_accuracy": 0.23382608294487,
      "num_tokens": 149538034.0,
      "step": 72050
    },
    {
      "entropy": 5.645427083969116,
      "epoch": 6.536193759071118,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001725380185624999,
      "loss": 4.8178,
      "mean_token_accuracy": 0.2294309377670288,
      "num_tokens": 149548059.0,
      "step": 72055
    },
    {
      "entropy": 5.6989155292510985,
      "epoch": 6.536647314949201,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00017250921290196957,
      "loss": 4.8784,
      "mean_token_accuracy": 0.23364082127809524,
      "num_tokens": 149557837.0,
      "step": 72060
    },
    {
      "entropy": 5.695302104949951,
      "epoch": 6.537100870827286,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017248040936097486,
      "loss": 4.7869,
      "mean_token_accuracy": 0.24384714365005494,
      "num_tokens": 149567478.0,
      "step": 72065
    },
    {
      "entropy": 5.706568670272827,
      "epoch": 6.53755442670537,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017245160794011137,
      "loss": 4.7648,
      "mean_token_accuracy": 0.23373015373945236,
      "num_tokens": 149578468.0,
      "step": 72070
    },
    {
      "entropy": 5.704473972320557,
      "epoch": 6.538007982583454,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001724228086399748,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23287260681390762,
      "num_tokens": 149588178.0,
      "step": 72075
    },
    {
      "entropy": 5.756757736206055,
      "epoch": 6.538461538461538,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001723940114611607,
      "loss": 4.9048,
      "mean_token_accuracy": 0.23328547924757004,
      "num_tokens": 149598281.0,
      "step": 72080
    },
    {
      "entropy": 5.689513254165649,
      "epoch": 6.538915094339623,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00017236521640426455,
      "loss": 4.7753,
      "mean_token_accuracy": 0.23824233710765838,
      "num_tokens": 149609312.0,
      "step": 72085
    },
    {
      "entropy": 5.701389503479004,
      "epoch": 6.539368650217707,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001723364234698819,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23515657484531402,
      "num_tokens": 149619736.0,
      "step": 72090
    },
    {
      "entropy": 5.6892369270324705,
      "epoch": 6.539822206095791,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017230763265860822,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23281458020210266,
      "num_tokens": 149629782.0,
      "step": 72095
    },
    {
      "entropy": 5.728103685379028,
      "epoch": 6.540275761973875,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017227884397103872,
      "loss": 4.832,
      "mean_token_accuracy": 0.23696476519107817,
      "num_tokens": 149638888.0,
      "step": 72100
    },
    {
      "entropy": 5.726718759536743,
      "epoch": 6.54072931785196,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017225005740776904,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22360638082027434,
      "num_tokens": 149649840.0,
      "step": 72105
    },
    {
      "entropy": 5.679835987091065,
      "epoch": 6.541182873730044,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017222127296939428,
      "loss": 4.8133,
      "mean_token_accuracy": 0.23435441106557847,
      "num_tokens": 149660004.0,
      "step": 72110
    },
    {
      "entropy": 5.712972211837768,
      "epoch": 6.541636429608127,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017219249065650985,
      "loss": 4.8476,
      "mean_token_accuracy": 0.23389933109283448,
      "num_tokens": 149669640.0,
      "step": 72115
    },
    {
      "entropy": 5.6898932456970215,
      "epoch": 6.542089985486212,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017216371046971092,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23560573011636735,
      "num_tokens": 149680985.0,
      "step": 72120
    },
    {
      "entropy": 5.731432199478149,
      "epoch": 6.542543541364296,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001721349324095926,
      "loss": 4.937,
      "mean_token_accuracy": 0.2309147909283638,
      "num_tokens": 149692239.0,
      "step": 72125
    },
    {
      "entropy": 5.746819353103637,
      "epoch": 6.54299709724238,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017210615647675014,
      "loss": 4.8032,
      "mean_token_accuracy": 0.2389887109398842,
      "num_tokens": 149703086.0,
      "step": 72130
    },
    {
      "entropy": 5.66050386428833,
      "epoch": 6.543450653120464,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017207738267177851,
      "loss": 4.7844,
      "mean_token_accuracy": 0.2312371924519539,
      "num_tokens": 149714855.0,
      "step": 72135
    },
    {
      "entropy": 5.711235094070434,
      "epoch": 6.543904208998549,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017204861099527288,
      "loss": 4.8415,
      "mean_token_accuracy": 0.22550837248563765,
      "num_tokens": 149724648.0,
      "step": 72140
    },
    {
      "entropy": 5.76874041557312,
      "epoch": 6.544357764876633,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017201984144782822,
      "loss": 4.9257,
      "mean_token_accuracy": 0.22842127531766893,
      "num_tokens": 149735239.0,
      "step": 72145
    },
    {
      "entropy": 5.657539319992066,
      "epoch": 6.544811320754717,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017199107403003956,
      "loss": 4.7276,
      "mean_token_accuracy": 0.24535485804080964,
      "num_tokens": 149745523.0,
      "step": 72150
    },
    {
      "entropy": 5.62187123298645,
      "epoch": 6.545264876632801,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017196230874250162,
      "loss": 4.7714,
      "mean_token_accuracy": 0.25115336030721663,
      "num_tokens": 149755582.0,
      "step": 72155
    },
    {
      "entropy": 5.696204900741577,
      "epoch": 6.545718432510886,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017193354558580942,
      "loss": 4.8056,
      "mean_token_accuracy": 0.23230873495340348,
      "num_tokens": 149766578.0,
      "step": 72160
    },
    {
      "entropy": 5.635182952880859,
      "epoch": 6.54617198838897,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017190478456055784,
      "loss": 4.7564,
      "mean_token_accuracy": 0.23360179364681244,
      "num_tokens": 149777198.0,
      "step": 72165
    },
    {
      "entropy": 5.68300609588623,
      "epoch": 6.546625544267053,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017187602566734144,
      "loss": 4.7616,
      "mean_token_accuracy": 0.23937374651432036,
      "num_tokens": 149787791.0,
      "step": 72170
    },
    {
      "entropy": 5.654554939270019,
      "epoch": 6.547079100145138,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017184726890675516,
      "loss": 4.7844,
      "mean_token_accuracy": 0.2373127043247223,
      "num_tokens": 149797862.0,
      "step": 72175
    },
    {
      "entropy": 5.611088371276855,
      "epoch": 6.547532656023222,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017181851427939372,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2407965123653412,
      "num_tokens": 149807943.0,
      "step": 72180
    },
    {
      "entropy": 5.724078273773193,
      "epoch": 6.547986211901306,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017178976178585165,
      "loss": 4.9265,
      "mean_token_accuracy": 0.22881275415420532,
      "num_tokens": 149817619.0,
      "step": 72185
    },
    {
      "entropy": 5.687651681900024,
      "epoch": 6.54843976777939,
      "grad_norm": 1.5,
      "learning_rate": 0.00017176101142672362,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23028352856636047,
      "num_tokens": 149827007.0,
      "step": 72190
    },
    {
      "entropy": 5.661147546768189,
      "epoch": 6.548893323657475,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001717322632026042,
      "loss": 4.7823,
      "mean_token_accuracy": 0.23763234913349152,
      "num_tokens": 149836626.0,
      "step": 72195
    },
    {
      "entropy": 5.667944669723511,
      "epoch": 6.549346879535559,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017170351711408784,
      "loss": 4.7451,
      "mean_token_accuracy": 0.24140601307153703,
      "num_tokens": 149847268.0,
      "step": 72200
    },
    {
      "entropy": 5.722063827514648,
      "epoch": 6.549800435413643,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017167477316176906,
      "loss": 4.8427,
      "mean_token_accuracy": 0.23512313216924668,
      "num_tokens": 149857716.0,
      "step": 72205
    },
    {
      "entropy": 5.727155160903931,
      "epoch": 6.5502539912917275,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017164603134624232,
      "loss": 4.8154,
      "mean_token_accuracy": 0.22867701798677445,
      "num_tokens": 149867394.0,
      "step": 72210
    },
    {
      "entropy": 5.7161589622497555,
      "epoch": 6.550707547169811,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001716172916681021,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2346205785870552,
      "num_tokens": 149877794.0,
      "step": 72215
    },
    {
      "entropy": 5.733375167846679,
      "epoch": 6.551161103047895,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00017158855412794259,
      "loss": 4.8056,
      "mean_token_accuracy": 0.2420044794678688,
      "num_tokens": 149887198.0,
      "step": 72220
    },
    {
      "entropy": 5.708791494369507,
      "epoch": 6.551614658925979,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017155981872635806,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22845568805932998,
      "num_tokens": 149896601.0,
      "step": 72225
    },
    {
      "entropy": 5.631677198410034,
      "epoch": 6.552068214804064,
      "grad_norm": 1.5,
      "learning_rate": 0.0001715310854639428,
      "loss": 4.7574,
      "mean_token_accuracy": 0.23620755225419998,
      "num_tokens": 149907699.0,
      "step": 72230
    },
    {
      "entropy": 5.640610218048096,
      "epoch": 6.552521770682148,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017150235434129114,
      "loss": 4.8646,
      "mean_token_accuracy": 0.23854122161865235,
      "num_tokens": 149918583.0,
      "step": 72235
    },
    {
      "entropy": 5.6793825149536135,
      "epoch": 6.552975326560232,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017147362535899712,
      "loss": 4.7948,
      "mean_token_accuracy": 0.23387733101844788,
      "num_tokens": 149928743.0,
      "step": 72240
    },
    {
      "entropy": 5.710197257995605,
      "epoch": 6.5534288824383164,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017144489851765498,
      "loss": 4.8454,
      "mean_token_accuracy": 0.233577224612236,
      "num_tokens": 149939532.0,
      "step": 72245
    },
    {
      "entropy": 5.715352153778076,
      "epoch": 6.553882438316401,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017141617381785868,
      "loss": 4.855,
      "mean_token_accuracy": 0.23455416411161423,
      "num_tokens": 149950644.0,
      "step": 72250
    },
    {
      "entropy": 5.7224555015563965,
      "epoch": 6.554335994194485,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017138745126020233,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23713795989751815,
      "num_tokens": 149960783.0,
      "step": 72255
    },
    {
      "entropy": 5.726727104187011,
      "epoch": 6.554789550072569,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017135873084527987,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23351465165615082,
      "num_tokens": 149972203.0,
      "step": 72260
    },
    {
      "entropy": 5.688212966918945,
      "epoch": 6.5552431059506535,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017133001257368518,
      "loss": 4.8035,
      "mean_token_accuracy": 0.24024507850408555,
      "num_tokens": 149981921.0,
      "step": 72265
    },
    {
      "entropy": 5.6780554294586185,
      "epoch": 6.555696661828737,
      "grad_norm": 1.5,
      "learning_rate": 0.00017130129644601233,
      "loss": 4.7672,
      "mean_token_accuracy": 0.2438584253191948,
      "num_tokens": 149991528.0,
      "step": 72270
    },
    {
      "entropy": 5.723074007034302,
      "epoch": 6.556150217706821,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017127258246285514,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22562436908483505,
      "num_tokens": 150002300.0,
      "step": 72275
    },
    {
      "entropy": 5.689615249633789,
      "epoch": 6.556603773584905,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017124387062480734,
      "loss": 4.8371,
      "mean_token_accuracy": 0.233815498650074,
      "num_tokens": 150013227.0,
      "step": 72280
    },
    {
      "entropy": 5.758282089233399,
      "epoch": 6.55705732946299,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017121516093246273,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22979819476604463,
      "num_tokens": 150023448.0,
      "step": 72285
    },
    {
      "entropy": 5.653907394409179,
      "epoch": 6.557510885341074,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017118645338641504,
      "loss": 4.7489,
      "mean_token_accuracy": 0.2377116322517395,
      "num_tokens": 150033095.0,
      "step": 72290
    },
    {
      "entropy": 5.801639175415039,
      "epoch": 6.557964441219158,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017115774798725795,
      "loss": 4.9554,
      "mean_token_accuracy": 0.227422097325325,
      "num_tokens": 150042808.0,
      "step": 72295
    },
    {
      "entropy": 5.76988034248352,
      "epoch": 6.5584179970972425,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000171129044735585,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23270216286182405,
      "num_tokens": 150052063.0,
      "step": 72300
    },
    {
      "entropy": 5.69601993560791,
      "epoch": 6.558871552975327,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017110034363199,
      "loss": 4.8167,
      "mean_token_accuracy": 0.23033954799175263,
      "num_tokens": 150062348.0,
      "step": 72305
    },
    {
      "entropy": 5.685674667358398,
      "epoch": 6.559325108853411,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017107164467706639,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23293882608413696,
      "num_tokens": 150072877.0,
      "step": 72310
    },
    {
      "entropy": 5.650580930709839,
      "epoch": 6.559778664731495,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017104294787140755,
      "loss": 4.771,
      "mean_token_accuracy": 0.23186508417129517,
      "num_tokens": 150083977.0,
      "step": 72315
    },
    {
      "entropy": 5.689244079589844,
      "epoch": 6.56023222060958,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001710142532156071,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23424462080001832,
      "num_tokens": 150094821.0,
      "step": 72320
    },
    {
      "entropy": 5.747851705551147,
      "epoch": 6.560685776487663,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017098556071025843,
      "loss": 4.8868,
      "mean_token_accuracy": 0.23288564383983612,
      "num_tokens": 150105391.0,
      "step": 72325
    },
    {
      "entropy": 5.6275430679321286,
      "epoch": 6.561139332365747,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001709568703559548,
      "loss": 4.7546,
      "mean_token_accuracy": 0.23948789685964583,
      "num_tokens": 150116045.0,
      "step": 72330
    },
    {
      "entropy": 5.655219078063965,
      "epoch": 6.5615928882438315,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017092818215328956,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23123055845499038,
      "num_tokens": 150125983.0,
      "step": 72335
    },
    {
      "entropy": 5.648747158050537,
      "epoch": 6.562046444121916,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017089949610285613,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23301033824682235,
      "num_tokens": 150136575.0,
      "step": 72340
    },
    {
      "entropy": 5.701452350616455,
      "epoch": 6.5625,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017087081220524763,
      "loss": 4.7845,
      "mean_token_accuracy": 0.2373380705714226,
      "num_tokens": 150146463.0,
      "step": 72345
    },
    {
      "entropy": 5.721449899673462,
      "epoch": 6.562953555878084,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017084213046105726,
      "loss": 4.9263,
      "mean_token_accuracy": 0.22716274857521057,
      "num_tokens": 150156437.0,
      "step": 72350
    },
    {
      "entropy": 5.732723140716553,
      "epoch": 6.5634071117561685,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001708134508708782,
      "loss": 4.8544,
      "mean_token_accuracy": 0.2269733026623726,
      "num_tokens": 150166235.0,
      "step": 72355
    },
    {
      "entropy": 5.711224365234375,
      "epoch": 6.563860667634253,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017078477343530346,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22791800796985626,
      "num_tokens": 150176131.0,
      "step": 72360
    },
    {
      "entropy": 5.709299945831299,
      "epoch": 6.564314223512337,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001707560981549262,
      "loss": 4.7538,
      "mean_token_accuracy": 0.236695297062397,
      "num_tokens": 150185942.0,
      "step": 72365
    },
    {
      "entropy": 5.684591007232666,
      "epoch": 6.56476777939042,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001707274250303394,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2404743254184723,
      "num_tokens": 150195378.0,
      "step": 72370
    },
    {
      "entropy": 5.727300786972046,
      "epoch": 6.565221335268505,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017069875406213597,
      "loss": 4.8083,
      "mean_token_accuracy": 0.23668829053640367,
      "num_tokens": 150205424.0,
      "step": 72375
    },
    {
      "entropy": 5.692717885971069,
      "epoch": 6.565674891146589,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017067008525090897,
      "loss": 4.8326,
      "mean_token_accuracy": 0.239689140021801,
      "num_tokens": 150214635.0,
      "step": 72380
    },
    {
      "entropy": 5.702355813980103,
      "epoch": 6.566128447024673,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017064141859725112,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2308818593621254,
      "num_tokens": 150224517.0,
      "step": 72385
    },
    {
      "entropy": 5.72013144493103,
      "epoch": 6.5665820029027575,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017061275410175543,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2276448428630829,
      "num_tokens": 150235593.0,
      "step": 72390
    },
    {
      "entropy": 5.768556451797485,
      "epoch": 6.567035558780842,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001705840917650145,
      "loss": 4.8716,
      "mean_token_accuracy": 0.23704079836606978,
      "num_tokens": 150247067.0,
      "step": 72395
    },
    {
      "entropy": 5.7424718856811525,
      "epoch": 6.567489114658926,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001705554315876211,
      "loss": 4.8338,
      "mean_token_accuracy": 0.2337277501821518,
      "num_tokens": 150257089.0,
      "step": 72400
    },
    {
      "entropy": 5.732801246643066,
      "epoch": 6.56794267053701,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001705267735701681,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23820631206035614,
      "num_tokens": 150267247.0,
      "step": 72405
    },
    {
      "entropy": 5.687901210784912,
      "epoch": 6.568396226415095,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017049811771324802,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23753687292337416,
      "num_tokens": 150278901.0,
      "step": 72410
    },
    {
      "entropy": 5.7209725856781,
      "epoch": 6.568849782293179,
      "grad_norm": 1.5,
      "learning_rate": 0.0001704694640174535,
      "loss": 4.8473,
      "mean_token_accuracy": 0.23240963816642762,
      "num_tokens": 150288992.0,
      "step": 72415
    },
    {
      "entropy": 5.716869497299195,
      "epoch": 6.569303338171263,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001704408124833771,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2326132118701935,
      "num_tokens": 150300014.0,
      "step": 72420
    },
    {
      "entropy": 5.746783638000489,
      "epoch": 6.5697568940493465,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00017041216311161134,
      "loss": 4.836,
      "mean_token_accuracy": 0.22990678548812865,
      "num_tokens": 150309156.0,
      "step": 72425
    },
    {
      "entropy": 5.677720594406128,
      "epoch": 6.570210449927431,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001703835159027487,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2382649526000023,
      "num_tokens": 150319022.0,
      "step": 72430
    },
    {
      "entropy": 5.69286036491394,
      "epoch": 6.570664005805515,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017035487085738154,
      "loss": 4.8537,
      "mean_token_accuracy": 0.23267170041799545,
      "num_tokens": 150329283.0,
      "step": 72435
    },
    {
      "entropy": 5.724183797836304,
      "epoch": 6.571117561683599,
      "grad_norm": 1.5,
      "learning_rate": 0.00017032622797610243,
      "loss": 4.9059,
      "mean_token_accuracy": 0.23056194335222244,
      "num_tokens": 150339024.0,
      "step": 72440
    },
    {
      "entropy": 5.641886568069458,
      "epoch": 6.5715711175616836,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00017029758725950353,
      "loss": 4.8057,
      "mean_token_accuracy": 0.22870104759931564,
      "num_tokens": 150349161.0,
      "step": 72445
    },
    {
      "entropy": 5.6505934715271,
      "epoch": 6.572024673439768,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017026894870817726,
      "loss": 4.7863,
      "mean_token_accuracy": 0.2332190379500389,
      "num_tokens": 150359817.0,
      "step": 72450
    },
    {
      "entropy": 5.664076566696167,
      "epoch": 6.572478229317852,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001702403123227158,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2336039125919342,
      "num_tokens": 150370961.0,
      "step": 72455
    },
    {
      "entropy": 5.6847672939300535,
      "epoch": 6.572931785195936,
      "grad_norm": 1.5,
      "learning_rate": 0.00017021167810371137,
      "loss": 4.813,
      "mean_token_accuracy": 0.23522995859384538,
      "num_tokens": 150381110.0,
      "step": 72460
    },
    {
      "entropy": 5.771377992630005,
      "epoch": 6.573385341074021,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017018304605175612,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23839448392391205,
      "num_tokens": 150391159.0,
      "step": 72465
    },
    {
      "entropy": 5.732053279876709,
      "epoch": 6.573838896952105,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017015441616744214,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2312560573220253,
      "num_tokens": 150401378.0,
      "step": 72470
    },
    {
      "entropy": 5.675827121734619,
      "epoch": 6.574292452830189,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017012578845136163,
      "loss": 4.7875,
      "mean_token_accuracy": 0.2391842409968376,
      "num_tokens": 150411321.0,
      "step": 72475
    },
    {
      "entropy": 5.793768405914307,
      "epoch": 6.5747460087082725,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017009716290410654,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22479472756385804,
      "num_tokens": 150421969.0,
      "step": 72480
    },
    {
      "entropy": 5.733856153488159,
      "epoch": 6.575199564586357,
      "grad_norm": 1.375,
      "learning_rate": 0.00017006853952626882,
      "loss": 4.8562,
      "mean_token_accuracy": 0.23399157971143722,
      "num_tokens": 150432423.0,
      "step": 72485
    },
    {
      "entropy": 5.698465824127197,
      "epoch": 6.575653120464441,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017003991831844047,
      "loss": 4.8296,
      "mean_token_accuracy": 0.2348013699054718,
      "num_tokens": 150443092.0,
      "step": 72490
    },
    {
      "entropy": 5.743837833404541,
      "epoch": 6.576106676342525,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001700112992812134,
      "loss": 4.9039,
      "mean_token_accuracy": 0.23586080670356752,
      "num_tokens": 150454701.0,
      "step": 72495
    },
    {
      "entropy": 5.692415189743042,
      "epoch": 6.57656023222061,
      "grad_norm": 1.546875,
      "learning_rate": 0.00016998268241517932,
      "loss": 4.8013,
      "mean_token_accuracy": 0.23200731724500656,
      "num_tokens": 150463593.0,
      "step": 72500
    },
    {
      "entropy": 5.699705696105957,
      "epoch": 6.577013788098694,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016995406772093015,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23092663288116455,
      "num_tokens": 150473257.0,
      "step": 72505
    },
    {
      "entropy": 5.6355553150177,
      "epoch": 6.577467343976778,
      "grad_norm": 1.78125,
      "learning_rate": 0.00016992545519905764,
      "loss": 4.6512,
      "mean_token_accuracy": 0.24721065610647203,
      "num_tokens": 150483354.0,
      "step": 72510
    },
    {
      "entropy": 5.671248245239258,
      "epoch": 6.577920899854862,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016989684485015354,
      "loss": 4.8342,
      "mean_token_accuracy": 0.2293059378862381,
      "num_tokens": 150493747.0,
      "step": 72515
    },
    {
      "entropy": 5.684698390960693,
      "epoch": 6.578374455732947,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001698682366748094,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2359684154391289,
      "num_tokens": 150503717.0,
      "step": 72520
    },
    {
      "entropy": 5.657890653610229,
      "epoch": 6.57882801161103,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016983963067361695,
      "loss": 4.7775,
      "mean_token_accuracy": 0.24318470805883408,
      "num_tokens": 150514267.0,
      "step": 72525
    },
    {
      "entropy": 5.63765811920166,
      "epoch": 6.579281567489114,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001698110268471678,
      "loss": 4.7444,
      "mean_token_accuracy": 0.24225241392850877,
      "num_tokens": 150524303.0,
      "step": 72530
    },
    {
      "entropy": 5.7071154594421385,
      "epoch": 6.579735123367199,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016978242519605336,
      "loss": 4.8891,
      "mean_token_accuracy": 0.23461734503507614,
      "num_tokens": 150534207.0,
      "step": 72535
    },
    {
      "entropy": 5.738918304443359,
      "epoch": 6.580188679245283,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016975382572086508,
      "loss": 4.8525,
      "mean_token_accuracy": 0.2361239418387413,
      "num_tokens": 150544536.0,
      "step": 72540
    },
    {
      "entropy": 5.770876407623291,
      "epoch": 6.580642235123367,
      "grad_norm": 1.5,
      "learning_rate": 0.00016972522842219462,
      "loss": 4.9105,
      "mean_token_accuracy": 0.22296653240919112,
      "num_tokens": 150555230.0,
      "step": 72545
    },
    {
      "entropy": 5.756689453125,
      "epoch": 6.581095791001451,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016969663330063329,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2274173766374588,
      "num_tokens": 150564839.0,
      "step": 72550
    },
    {
      "entropy": 5.6487569332122805,
      "epoch": 6.581549346879536,
      "grad_norm": 1.59375,
      "learning_rate": 0.00016966804035677236,
      "loss": 4.7642,
      "mean_token_accuracy": 0.23822353929281234,
      "num_tokens": 150575106.0,
      "step": 72555
    },
    {
      "entropy": 5.699617385864258,
      "epoch": 6.58200290275762,
      "grad_norm": 1.625,
      "learning_rate": 0.0001696394495912032,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22716952711343766,
      "num_tokens": 150585427.0,
      "step": 72560
    },
    {
      "entropy": 5.656291913986206,
      "epoch": 6.582456458635704,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001696108610045171,
      "loss": 4.7515,
      "mean_token_accuracy": 0.242903171479702,
      "num_tokens": 150595795.0,
      "step": 72565
    },
    {
      "entropy": 5.727199745178223,
      "epoch": 6.582910014513788,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016958227459730525,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23552878499031066,
      "num_tokens": 150606357.0,
      "step": 72570
    },
    {
      "entropy": 5.745443677902221,
      "epoch": 6.583363570391873,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016955369037015874,
      "loss": 4.9142,
      "mean_token_accuracy": 0.22276604026556016,
      "num_tokens": 150617445.0,
      "step": 72575
    },
    {
      "entropy": 5.705176877975464,
      "epoch": 6.583817126269956,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016952510832366884,
      "loss": 4.7969,
      "mean_token_accuracy": 0.2363715022802353,
      "num_tokens": 150627470.0,
      "step": 72580
    },
    {
      "entropy": 5.692406892776489,
      "epoch": 6.58427068214804,
      "grad_norm": 1.375,
      "learning_rate": 0.00016949652845842661,
      "loss": 4.7619,
      "mean_token_accuracy": 0.23924749940633774,
      "num_tokens": 150638829.0,
      "step": 72585
    },
    {
      "entropy": 5.732613611221313,
      "epoch": 6.584724238026125,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00016946795077502305,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23568745106458663,
      "num_tokens": 150648115.0,
      "step": 72590
    },
    {
      "entropy": 5.736706733703613,
      "epoch": 6.585177793904209,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016943937527404913,
      "loss": 4.8179,
      "mean_token_accuracy": 0.23312382847070695,
      "num_tokens": 150659253.0,
      "step": 72595
    },
    {
      "entropy": 5.747865676879883,
      "epoch": 6.585631349782293,
      "grad_norm": 1.375,
      "learning_rate": 0.00016941080195609587,
      "loss": 4.8699,
      "mean_token_accuracy": 0.23229345232248305,
      "num_tokens": 150670755.0,
      "step": 72600
    },
    {
      "entropy": 5.682966470718384,
      "epoch": 6.586084905660377,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016938223082175403,
      "loss": 4.8378,
      "mean_token_accuracy": 0.23644421845674515,
      "num_tokens": 150681068.0,
      "step": 72605
    },
    {
      "entropy": 5.7068016052246096,
      "epoch": 6.586538461538462,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016935366187161466,
      "loss": 4.8033,
      "mean_token_accuracy": 0.23011485934257508,
      "num_tokens": 150691502.0,
      "step": 72610
    },
    {
      "entropy": 5.667045783996582,
      "epoch": 6.586992017416546,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001693250951062685,
      "loss": 4.793,
      "mean_token_accuracy": 0.236271171271801,
      "num_tokens": 150701573.0,
      "step": 72615
    },
    {
      "entropy": 5.6348069190979,
      "epoch": 6.58744557329463,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016929653052630628,
      "loss": 4.7267,
      "mean_token_accuracy": 0.2398097813129425,
      "num_tokens": 150711251.0,
      "step": 72620
    },
    {
      "entropy": 5.764076185226441,
      "epoch": 6.587899129172714,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016926796813231874,
      "loss": 4.9177,
      "mean_token_accuracy": 0.22438569217920304,
      "num_tokens": 150721947.0,
      "step": 72625
    },
    {
      "entropy": 5.737332963943482,
      "epoch": 6.588352685050799,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016923940792489662,
      "loss": 4.901,
      "mean_token_accuracy": 0.23085505962371827,
      "num_tokens": 150731986.0,
      "step": 72630
    },
    {
      "entropy": 5.771316862106323,
      "epoch": 6.588806240928882,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016921084990463044,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22929960936307908,
      "num_tokens": 150742518.0,
      "step": 72635
    },
    {
      "entropy": 5.7193773746490475,
      "epoch": 6.589259796806966,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016918229407211072,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21619307696819307,
      "num_tokens": 150753432.0,
      "step": 72640
    },
    {
      "entropy": 5.754888343811035,
      "epoch": 6.589713352685051,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016915374042792837,
      "loss": 4.9391,
      "mean_token_accuracy": 0.22644102275371553,
      "num_tokens": 150765613.0,
      "step": 72645
    },
    {
      "entropy": 5.674430656433105,
      "epoch": 6.590166908563135,
      "grad_norm": 1.59375,
      "learning_rate": 0.00016912518897267353,
      "loss": 4.8189,
      "mean_token_accuracy": 0.23540622740983963,
      "num_tokens": 150775861.0,
      "step": 72650
    },
    {
      "entropy": 5.6883350849151615,
      "epoch": 6.590620464441219,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016909663970693676,
      "loss": 4.8075,
      "mean_token_accuracy": 0.23717712312936784,
      "num_tokens": 150785313.0,
      "step": 72655
    },
    {
      "entropy": 5.715034341812133,
      "epoch": 6.5910740203193035,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016906809263130847,
      "loss": 4.7945,
      "mean_token_accuracy": 0.2329096958041191,
      "num_tokens": 150795926.0,
      "step": 72660
    },
    {
      "entropy": 5.692601442337036,
      "epoch": 6.591527576197388,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016903954774637902,
      "loss": 4.8186,
      "mean_token_accuracy": 0.2375083938241005,
      "num_tokens": 150806149.0,
      "step": 72665
    },
    {
      "entropy": 5.744310188293457,
      "epoch": 6.591981132075472,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016901100505273864,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2293948709964752,
      "num_tokens": 150816087.0,
      "step": 72670
    },
    {
      "entropy": 5.689078283309937,
      "epoch": 6.592434687953556,
      "grad_norm": 1.5,
      "learning_rate": 0.00016898246455097774,
      "loss": 4.8008,
      "mean_token_accuracy": 0.2324522241950035,
      "num_tokens": 150827490.0,
      "step": 72675
    },
    {
      "entropy": 5.723266983032227,
      "epoch": 6.59288824383164,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016895392624168652,
      "loss": 4.8768,
      "mean_token_accuracy": 0.23661767989397048,
      "num_tokens": 150838188.0,
      "step": 72680
    },
    {
      "entropy": 5.7681725025177,
      "epoch": 6.593341799709724,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016892539012545513,
      "loss": 4.8824,
      "mean_token_accuracy": 0.23255650997161864,
      "num_tokens": 150849068.0,
      "step": 72685
    },
    {
      "entropy": 5.706495666503907,
      "epoch": 6.593795355587808,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016889685620287371,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2399112954735756,
      "num_tokens": 150859567.0,
      "step": 72690
    },
    {
      "entropy": 5.688956928253174,
      "epoch": 6.594248911465892,
      "grad_norm": 1.5,
      "learning_rate": 0.0001688683244745323,
      "loss": 4.7676,
      "mean_token_accuracy": 0.2400890916585922,
      "num_tokens": 150870186.0,
      "step": 72695
    },
    {
      "entropy": 5.802611398696899,
      "epoch": 6.594702467343977,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016883979494102092,
      "loss": 4.8993,
      "mean_token_accuracy": 0.23404129445552826,
      "num_tokens": 150880758.0,
      "step": 72700
    },
    {
      "entropy": 5.7256628513336185,
      "epoch": 6.595156023222061,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016881126760292975,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22723172307014466,
      "num_tokens": 150891720.0,
      "step": 72705
    },
    {
      "entropy": 5.677957963943482,
      "epoch": 6.595609579100145,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001687827424608486,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23861677944660187,
      "num_tokens": 150902712.0,
      "step": 72710
    },
    {
      "entropy": 5.679570531845092,
      "epoch": 6.5960631349782295,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001687542195153674,
      "loss": 4.8001,
      "mean_token_accuracy": 0.2317885622382164,
      "num_tokens": 150911782.0,
      "step": 72715
    },
    {
      "entropy": 5.715060997009277,
      "epoch": 6.596516690856314,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016872569876707604,
      "loss": 4.7456,
      "mean_token_accuracy": 0.2437054842710495,
      "num_tokens": 150924381.0,
      "step": 72720
    },
    {
      "entropy": 5.664087772369385,
      "epoch": 6.596970246734398,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016869718021656432,
      "loss": 4.7534,
      "mean_token_accuracy": 0.23678730726242064,
      "num_tokens": 150935146.0,
      "step": 72725
    },
    {
      "entropy": 5.737681484222412,
      "epoch": 6.597423802612482,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016866866386442198,
      "loss": 4.8546,
      "mean_token_accuracy": 0.23214062601327895,
      "num_tokens": 150945799.0,
      "step": 72730
    },
    {
      "entropy": 5.6971179962158205,
      "epoch": 6.597877358490566,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016864014971123867,
      "loss": 4.803,
      "mean_token_accuracy": 0.23782337456941605,
      "num_tokens": 150956450.0,
      "step": 72735
    },
    {
      "entropy": 5.674326801300049,
      "epoch": 6.59833091436865,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016861163775760423,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2293666660785675,
      "num_tokens": 150967563.0,
      "step": 72740
    },
    {
      "entropy": 5.732677268981933,
      "epoch": 6.598784470246734,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001685831280041083,
      "loss": 4.8358,
      "mean_token_accuracy": 0.23661968857049942,
      "num_tokens": 150977381.0,
      "step": 72745
    },
    {
      "entropy": 5.715435743331909,
      "epoch": 6.5992380261248185,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016855462045134036,
      "loss": 4.8632,
      "mean_token_accuracy": 0.23152562081813813,
      "num_tokens": 150986986.0,
      "step": 72750
    },
    {
      "entropy": 5.783113718032837,
      "epoch": 6.599691582002903,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016852611509989,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2276251122355461,
      "num_tokens": 150997695.0,
      "step": 72755
    },
    {
      "entropy": 5.746082162857055,
      "epoch": 6.600145137880987,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001684976119503467,
      "loss": 4.8893,
      "mean_token_accuracy": 0.22510594576597215,
      "num_tokens": 151008122.0,
      "step": 72760
    },
    {
      "entropy": 5.73475079536438,
      "epoch": 6.600598693759071,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016846911100329995,
      "loss": 4.831,
      "mean_token_accuracy": 0.23188257664442063,
      "num_tokens": 151017095.0,
      "step": 72765
    },
    {
      "entropy": 5.716335344314575,
      "epoch": 6.6010522496371555,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016844061225933905,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2330981507897377,
      "num_tokens": 151027660.0,
      "step": 72770
    },
    {
      "entropy": 5.7065873622894285,
      "epoch": 6.60150580551524,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016841211571905348,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23566617965698242,
      "num_tokens": 151036970.0,
      "step": 72775
    },
    {
      "entropy": 5.637796258926391,
      "epoch": 6.601959361393323,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016838362138303257,
      "loss": 4.7931,
      "mean_token_accuracy": 0.23364713191986083,
      "num_tokens": 151048946.0,
      "step": 72780
    },
    {
      "entropy": 5.613930606842041,
      "epoch": 6.602412917271408,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016835512925186553,
      "loss": 4.7368,
      "mean_token_accuracy": 0.23533292412757872,
      "num_tokens": 151059775.0,
      "step": 72785
    },
    {
      "entropy": 5.6965656757354735,
      "epoch": 6.602866473149492,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016832663932614156,
      "loss": 4.7802,
      "mean_token_accuracy": 0.24107449352741242,
      "num_tokens": 151069481.0,
      "step": 72790
    },
    {
      "entropy": 5.752678537368775,
      "epoch": 6.603320029027576,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001682981516064499,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2356856569647789,
      "num_tokens": 151078987.0,
      "step": 72795
    },
    {
      "entropy": 5.7315428256988525,
      "epoch": 6.60377358490566,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016826966609337962,
      "loss": 4.948,
      "mean_token_accuracy": 0.22674308270215987,
      "num_tokens": 151089629.0,
      "step": 72800
    },
    {
      "entropy": 5.695126056671143,
      "epoch": 6.6042271407837445,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001682411827875198,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22527575939893724,
      "num_tokens": 151100181.0,
      "step": 72805
    },
    {
      "entropy": 5.729460048675537,
      "epoch": 6.604680696661829,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016821270168945962,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2326389044523239,
      "num_tokens": 151111483.0,
      "step": 72810
    },
    {
      "entropy": 5.679951238632202,
      "epoch": 6.605134252539913,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016818422279978797,
      "loss": 4.731,
      "mean_token_accuracy": 0.23490749001502992,
      "num_tokens": 151121456.0,
      "step": 72815
    },
    {
      "entropy": 5.706127882003784,
      "epoch": 6.605587808417997,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001681557461190938,
      "loss": 4.8019,
      "mean_token_accuracy": 0.22968794256448746,
      "num_tokens": 151132662.0,
      "step": 72820
    },
    {
      "entropy": 5.72541651725769,
      "epoch": 6.606041364296082,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016812727164796604,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2272854521870613,
      "num_tokens": 151143118.0,
      "step": 72825
    },
    {
      "entropy": 5.706947278976441,
      "epoch": 6.606494920174166,
      "grad_norm": 1.5,
      "learning_rate": 0.00016809879938699352,
      "loss": 4.8277,
      "mean_token_accuracy": 0.2381410762667656,
      "num_tokens": 151152884.0,
      "step": 72830
    },
    {
      "entropy": 5.674201440811157,
      "epoch": 6.606948476052249,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001680703293367651,
      "loss": 4.7625,
      "mean_token_accuracy": 0.23327672779560088,
      "num_tokens": 151162918.0,
      "step": 72835
    },
    {
      "entropy": 5.657916402816772,
      "epoch": 6.6074020319303335,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016804186149786937,
      "loss": 4.8238,
      "mean_token_accuracy": 0.2390560269355774,
      "num_tokens": 151174093.0,
      "step": 72840
    },
    {
      "entropy": 5.66225380897522,
      "epoch": 6.607855587808418,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016801339587089534,
      "loss": 4.7207,
      "mean_token_accuracy": 0.24197387099266052,
      "num_tokens": 151184085.0,
      "step": 72845
    },
    {
      "entropy": 5.747978973388672,
      "epoch": 6.608309143686502,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016798493245643155,
      "loss": 4.885,
      "mean_token_accuracy": 0.21889235973358154,
      "num_tokens": 151194394.0,
      "step": 72850
    },
    {
      "entropy": 5.685960531234741,
      "epoch": 6.608762699564586,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001679564712550666,
      "loss": 4.8079,
      "mean_token_accuracy": 0.2386814072728157,
      "num_tokens": 151205114.0,
      "step": 72855
    },
    {
      "entropy": 5.671371746063232,
      "epoch": 6.6092162554426706,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001679280122673891,
      "loss": 4.7522,
      "mean_token_accuracy": 0.23378522247076033,
      "num_tokens": 151215332.0,
      "step": 72860
    },
    {
      "entropy": 5.618901491165161,
      "epoch": 6.609669811320755,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016789955549398761,
      "loss": 4.8381,
      "mean_token_accuracy": 0.24018425345420838,
      "num_tokens": 151226240.0,
      "step": 72865
    },
    {
      "entropy": 5.647942733764649,
      "epoch": 6.610123367198839,
      "grad_norm": 1.625,
      "learning_rate": 0.0001678711009354506,
      "loss": 4.7705,
      "mean_token_accuracy": 0.22847193330526352,
      "num_tokens": 151236017.0,
      "step": 72870
    },
    {
      "entropy": 5.68246169090271,
      "epoch": 6.610576923076923,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016784264859236644,
      "loss": 4.7696,
      "mean_token_accuracy": 0.23409924656152725,
      "num_tokens": 151247275.0,
      "step": 72875
    },
    {
      "entropy": 5.714103031158447,
      "epoch": 6.611030478955008,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001678141984653237,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22664543688297273,
      "num_tokens": 151258142.0,
      "step": 72880
    },
    {
      "entropy": 5.748696041107178,
      "epoch": 6.611484034833092,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016778575055491064,
      "loss": 4.8425,
      "mean_token_accuracy": 0.22969766110181808,
      "num_tokens": 151267561.0,
      "step": 72885
    },
    {
      "entropy": 5.627288818359375,
      "epoch": 6.611937590711175,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016775730486171558,
      "loss": 4.7533,
      "mean_token_accuracy": 0.23724275082349777,
      "num_tokens": 151278351.0,
      "step": 72890
    },
    {
      "entropy": 5.708839416503906,
      "epoch": 6.6123911465892595,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016772886138632681,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22736536264419555,
      "num_tokens": 151288542.0,
      "step": 72895
    },
    {
      "entropy": 5.6313159465789795,
      "epoch": 6.612844702467344,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001677004201293325,
      "loss": 4.7811,
      "mean_token_accuracy": 0.2343734860420227,
      "num_tokens": 151298550.0,
      "step": 72900
    },
    {
      "entropy": 5.66116156578064,
      "epoch": 6.613298258345428,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016767198109132075,
      "loss": 4.7562,
      "mean_token_accuracy": 0.23931602090597154,
      "num_tokens": 151308388.0,
      "step": 72905
    },
    {
      "entropy": 5.689555263519287,
      "epoch": 6.613751814223512,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001676435442728799,
      "loss": 4.7897,
      "mean_token_accuracy": 0.24546639919281005,
      "num_tokens": 151317846.0,
      "step": 72910
    },
    {
      "entropy": 5.70825309753418,
      "epoch": 6.614205370101597,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016761510967459788,
      "loss": 4.8404,
      "mean_token_accuracy": 0.22712114453315735,
      "num_tokens": 151330003.0,
      "step": 72915
    },
    {
      "entropy": 5.591966009140014,
      "epoch": 6.614658925979681,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016758667729706279,
      "loss": 4.647,
      "mean_token_accuracy": 0.24022201895713807,
      "num_tokens": 151340237.0,
      "step": 72920
    },
    {
      "entropy": 5.673927736282349,
      "epoch": 6.615112481857765,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016755824714086254,
      "loss": 4.7591,
      "mean_token_accuracy": 0.23307252079248428,
      "num_tokens": 151350141.0,
      "step": 72925
    },
    {
      "entropy": 5.740627479553223,
      "epoch": 6.615566037735849,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016752981920658516,
      "loss": 4.8433,
      "mean_token_accuracy": 0.23340240120887756,
      "num_tokens": 151360812.0,
      "step": 72930
    },
    {
      "entropy": 5.693246173858642,
      "epoch": 6.616019593613933,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016750139349481846,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22776975929737092,
      "num_tokens": 151370712.0,
      "step": 72935
    },
    {
      "entropy": 5.750940942764283,
      "epoch": 6.616473149492017,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001674729700061503,
      "loss": 4.82,
      "mean_token_accuracy": 0.23403716087341309,
      "num_tokens": 151380675.0,
      "step": 72940
    },
    {
      "entropy": 5.628984022140503,
      "epoch": 6.616926705370101,
      "grad_norm": 1.5,
      "learning_rate": 0.00016744454874116857,
      "loss": 4.7381,
      "mean_token_accuracy": 0.24520465731620789,
      "num_tokens": 151390738.0,
      "step": 72945
    },
    {
      "entropy": 5.689851379394531,
      "epoch": 6.617380261248186,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000167416129700461,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2332070901989937,
      "num_tokens": 151402616.0,
      "step": 72950
    },
    {
      "entropy": 5.659261751174927,
      "epoch": 6.61783381712627,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001673877128846153,
      "loss": 4.7663,
      "mean_token_accuracy": 0.23417799025774003,
      "num_tokens": 151412770.0,
      "step": 72955
    },
    {
      "entropy": 5.760413408279419,
      "epoch": 6.618287373004354,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016735929829421909,
      "loss": 4.8605,
      "mean_token_accuracy": 0.23104440569877624,
      "num_tokens": 151423023.0,
      "step": 72960
    },
    {
      "entropy": 5.657692289352417,
      "epoch": 6.618740928882438,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016733088592986006,
      "loss": 4.7604,
      "mean_token_accuracy": 0.24318794310092925,
      "num_tokens": 151433574.0,
      "step": 72965
    },
    {
      "entropy": 5.726697683334351,
      "epoch": 6.619194484760523,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001673024757921257,
      "loss": 4.8682,
      "mean_token_accuracy": 0.23186723440885543,
      "num_tokens": 151443817.0,
      "step": 72970
    },
    {
      "entropy": 5.740680789947509,
      "epoch": 6.619648040638607,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001672740678816035,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2312163755297661,
      "num_tokens": 151454164.0,
      "step": 72975
    },
    {
      "entropy": 5.704058027267456,
      "epoch": 6.620101596516691,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016724566219888115,
      "loss": 4.7693,
      "mean_token_accuracy": 0.24160690307617189,
      "num_tokens": 151464157.0,
      "step": 72980
    },
    {
      "entropy": 5.748938751220703,
      "epoch": 6.620555152394775,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001672172587445459,
      "loss": 4.9024,
      "mean_token_accuracy": 0.22836268395185472,
      "num_tokens": 151474531.0,
      "step": 72985
    },
    {
      "entropy": 5.722777843475342,
      "epoch": 6.621008708272859,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001671888575191853,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2359430283308029,
      "num_tokens": 151485147.0,
      "step": 72990
    },
    {
      "entropy": 5.714015483856201,
      "epoch": 6.621462264150943,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016716045852338654,
      "loss": 4.7809,
      "mean_token_accuracy": 0.23523824661970139,
      "num_tokens": 151495074.0,
      "step": 72995
    },
    {
      "entropy": 5.737928438186645,
      "epoch": 6.621915820029027,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016713206175773698,
      "loss": 4.8684,
      "mean_token_accuracy": 0.23074829429388047,
      "num_tokens": 151505551.0,
      "step": 73000
    },
    {
      "entropy": 5.698632860183716,
      "epoch": 6.622369375907112,
      "grad_norm": 1.59375,
      "learning_rate": 0.00016710366722282388,
      "loss": 4.7658,
      "mean_token_accuracy": 0.24564643502235411,
      "num_tokens": 151514601.0,
      "step": 73005
    },
    {
      "entropy": 5.726531314849853,
      "epoch": 6.622822931785196,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016707527491923439,
      "loss": 4.9011,
      "mean_token_accuracy": 0.2231115520000458,
      "num_tokens": 151525125.0,
      "step": 73010
    },
    {
      "entropy": 5.719416284561158,
      "epoch": 6.62327648766328,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016704688484755575,
      "loss": 4.8741,
      "mean_token_accuracy": 0.2313395380973816,
      "num_tokens": 151535364.0,
      "step": 73015
    },
    {
      "entropy": 5.7148504734039305,
      "epoch": 6.623730043541364,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016701849700837505,
      "loss": 4.7493,
      "mean_token_accuracy": 0.24137728214263915,
      "num_tokens": 151545634.0,
      "step": 73020
    },
    {
      "entropy": 5.712407827377319,
      "epoch": 6.624183599419449,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016699011140227935,
      "loss": 4.8264,
      "mean_token_accuracy": 0.22631563395261764,
      "num_tokens": 151555564.0,
      "step": 73025
    },
    {
      "entropy": 5.677314281463623,
      "epoch": 6.624637155297533,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016696172802985567,
      "loss": 4.782,
      "mean_token_accuracy": 0.23468479365110398,
      "num_tokens": 151566387.0,
      "step": 73030
    },
    {
      "entropy": 5.6671384334564205,
      "epoch": 6.625090711175617,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016693334689169099,
      "loss": 4.7967,
      "mean_token_accuracy": 0.2295251354575157,
      "num_tokens": 151576331.0,
      "step": 73035
    },
    {
      "entropy": 5.6403450012207035,
      "epoch": 6.6255442670537015,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016690496798837224,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23379406630992888,
      "num_tokens": 151586771.0,
      "step": 73040
    },
    {
      "entropy": 5.610440874099732,
      "epoch": 6.625997822931785,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016687659132048622,
      "loss": 4.7439,
      "mean_token_accuracy": 0.2388744056224823,
      "num_tokens": 151597233.0,
      "step": 73045
    },
    {
      "entropy": 5.649420976638794,
      "epoch": 6.626451378809869,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016684821688861992,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23959779739379883,
      "num_tokens": 151606712.0,
      "step": 73050
    },
    {
      "entropy": 5.697282314300537,
      "epoch": 6.626904934687953,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016681984469336003,
      "loss": 4.7947,
      "mean_token_accuracy": 0.2387610122561455,
      "num_tokens": 151616954.0,
      "step": 73055
    },
    {
      "entropy": 5.697718715667724,
      "epoch": 6.627358490566038,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016679147473529338,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23555756360292435,
      "num_tokens": 151627143.0,
      "step": 73060
    },
    {
      "entropy": 5.685942602157593,
      "epoch": 6.627812046444122,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016676310701500659,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23681259006261826,
      "num_tokens": 151636451.0,
      "step": 73065
    },
    {
      "entropy": 5.665895175933838,
      "epoch": 6.628265602322206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016673474153308627,
      "loss": 4.7693,
      "mean_token_accuracy": 0.23006830364465714,
      "num_tokens": 151645981.0,
      "step": 73070
    },
    {
      "entropy": 5.738363409042359,
      "epoch": 6.6287191582002905,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016670637829011912,
      "loss": 4.7932,
      "mean_token_accuracy": 0.23531256169080733,
      "num_tokens": 151656777.0,
      "step": 73075
    },
    {
      "entropy": 5.665886449813843,
      "epoch": 6.629172714078375,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016667801728669168,
      "loss": 4.827,
      "mean_token_accuracy": 0.2362750306725502,
      "num_tokens": 151668426.0,
      "step": 73080
    },
    {
      "entropy": 5.710175085067749,
      "epoch": 6.629626269956459,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016664965852339044,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23000371903181077,
      "num_tokens": 151678765.0,
      "step": 73085
    },
    {
      "entropy": 5.710592412948609,
      "epoch": 6.630079825834542,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016662130200080182,
      "loss": 4.8457,
      "mean_token_accuracy": 0.23453081250190735,
      "num_tokens": 151689495.0,
      "step": 73090
    },
    {
      "entropy": 5.627464151382446,
      "epoch": 6.630533381712627,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001665929477195124,
      "loss": 4.7632,
      "mean_token_accuracy": 0.23640727549791335,
      "num_tokens": 151699067.0,
      "step": 73095
    },
    {
      "entropy": 5.726162242889404,
      "epoch": 6.630986937590711,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00016656459568010836,
      "loss": 4.8679,
      "mean_token_accuracy": 0.23529718518257142,
      "num_tokens": 151708930.0,
      "step": 73100
    },
    {
      "entropy": 5.673913526535034,
      "epoch": 6.631440493468795,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016653624588317608,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2322577267885208,
      "num_tokens": 151717965.0,
      "step": 73105
    },
    {
      "entropy": 5.657195091247559,
      "epoch": 6.631894049346879,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001665078983293019,
      "loss": 4.7957,
      "mean_token_accuracy": 0.23204618245363234,
      "num_tokens": 151727615.0,
      "step": 73110
    },
    {
      "entropy": 5.685066747665405,
      "epoch": 6.632347605224964,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016647955301907208,
      "loss": 4.7998,
      "mean_token_accuracy": 0.2333192691206932,
      "num_tokens": 151736911.0,
      "step": 73115
    },
    {
      "entropy": 5.697988367080688,
      "epoch": 6.632801161103048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016645120995307272,
      "loss": 4.8528,
      "mean_token_accuracy": 0.2250020682811737,
      "num_tokens": 151747696.0,
      "step": 73120
    },
    {
      "entropy": 5.7285804271698,
      "epoch": 6.633254716981132,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016642286913189002,
      "loss": 4.8216,
      "mean_token_accuracy": 0.2376688688993454,
      "num_tokens": 151757094.0,
      "step": 73125
    },
    {
      "entropy": 5.714438819885254,
      "epoch": 6.6337082728592165,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016639453055611008,
      "loss": 4.7842,
      "mean_token_accuracy": 0.23219016194343567,
      "num_tokens": 151767955.0,
      "step": 73130
    },
    {
      "entropy": 5.784292793273925,
      "epoch": 6.634161828737301,
      "grad_norm": 1.5,
      "learning_rate": 0.0001663661942263189,
      "loss": 4.9473,
      "mean_token_accuracy": 0.22560760825872422,
      "num_tokens": 151779393.0,
      "step": 73135
    },
    {
      "entropy": 5.612339782714844,
      "epoch": 6.634615384615385,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016633786014310243,
      "loss": 4.7142,
      "mean_token_accuracy": 0.252791465818882,
      "num_tokens": 151790261.0,
      "step": 73140
    },
    {
      "entropy": 5.759529066085816,
      "epoch": 6.635068940493468,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001663095283070468,
      "loss": 4.8615,
      "mean_token_accuracy": 0.23343775123357774,
      "num_tokens": 151800273.0,
      "step": 73145
    },
    {
      "entropy": 5.764356708526611,
      "epoch": 6.635522496371553,
      "grad_norm": 1.609375,
      "learning_rate": 0.00016628119871873776,
      "loss": 4.8838,
      "mean_token_accuracy": 0.23149701505899428,
      "num_tokens": 151811665.0,
      "step": 73150
    },
    {
      "entropy": 5.727956008911133,
      "epoch": 6.635976052249637,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001662528713787613,
      "loss": 4.8444,
      "mean_token_accuracy": 0.22855918407440184,
      "num_tokens": 151822844.0,
      "step": 73155
    },
    {
      "entropy": 5.665267467498779,
      "epoch": 6.636429608127721,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016622454628770314,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23031493425369262,
      "num_tokens": 151833256.0,
      "step": 73160
    },
    {
      "entropy": 5.658594226837158,
      "epoch": 6.6368831640058055,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001661962234461491,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23302369862794875,
      "num_tokens": 151843825.0,
      "step": 73165
    },
    {
      "entropy": 5.673654699325562,
      "epoch": 6.63733671988389,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016616790285468486,
      "loss": 4.7617,
      "mean_token_accuracy": 0.23518707901239394,
      "num_tokens": 151854716.0,
      "step": 73170
    },
    {
      "entropy": 5.753694009780884,
      "epoch": 6.637790275761974,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016613958451389604,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22944281846284867,
      "num_tokens": 151865419.0,
      "step": 73175
    },
    {
      "entropy": 5.7522351264953615,
      "epoch": 6.638243831640058,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016611126842436846,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2252444639801979,
      "num_tokens": 151876944.0,
      "step": 73180
    },
    {
      "entropy": 5.8071441650390625,
      "epoch": 6.6386973875181425,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001660829545866876,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2339979514479637,
      "num_tokens": 151886862.0,
      "step": 73185
    },
    {
      "entropy": 5.754555654525757,
      "epoch": 6.639150943396227,
      "grad_norm": 1.5,
      "learning_rate": 0.00016605464300143902,
      "loss": 4.9474,
      "mean_token_accuracy": 0.22943779528141023,
      "num_tokens": 151897979.0,
      "step": 73190
    },
    {
      "entropy": 5.637397718429566,
      "epoch": 6.639604499274311,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016602633366920815,
      "loss": 4.7687,
      "mean_token_accuracy": 0.2395181193947792,
      "num_tokens": 151907347.0,
      "step": 73195
    },
    {
      "entropy": 5.740890645980835,
      "epoch": 6.6400580551523944,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016599802659058045,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2297603189945221,
      "num_tokens": 151916811.0,
      "step": 73200
    },
    {
      "entropy": 5.67964129447937,
      "epoch": 6.640511611030479,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016596972176614137,
      "loss": 4.7603,
      "mean_token_accuracy": 0.24042351692914962,
      "num_tokens": 151927080.0,
      "step": 73205
    },
    {
      "entropy": 5.65992283821106,
      "epoch": 6.640965166908563,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001659414191964761,
      "loss": 4.7554,
      "mean_token_accuracy": 0.2348073825240135,
      "num_tokens": 151936838.0,
      "step": 73210
    },
    {
      "entropy": 5.716505289077759,
      "epoch": 6.641418722786647,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001659131188821702,
      "loss": 4.8851,
      "mean_token_accuracy": 0.23257524520158768,
      "num_tokens": 151947488.0,
      "step": 73215
    },
    {
      "entropy": 5.691204166412353,
      "epoch": 6.6418722786647315,
      "grad_norm": 1.5,
      "learning_rate": 0.00016588482082380878,
      "loss": 4.8681,
      "mean_token_accuracy": 0.2262652650475502,
      "num_tokens": 151957901.0,
      "step": 73220
    },
    {
      "entropy": 5.647391414642334,
      "epoch": 6.642325834542816,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016585652502197713,
      "loss": 4.7994,
      "mean_token_accuracy": 0.23937687277793884,
      "num_tokens": 151968443.0,
      "step": 73225
    },
    {
      "entropy": 5.722518539428711,
      "epoch": 6.6427793904209,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001658282314772603,
      "loss": 4.8405,
      "mean_token_accuracy": 0.23150643706321716,
      "num_tokens": 151978744.0,
      "step": 73230
    },
    {
      "entropy": 5.671969985961914,
      "epoch": 6.643232946298984,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016579994019024345,
      "loss": 4.798,
      "mean_token_accuracy": 0.23730442225933074,
      "num_tokens": 151988640.0,
      "step": 73235
    },
    {
      "entropy": 5.634556007385254,
      "epoch": 6.643686502177069,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016577165116151156,
      "loss": 4.7663,
      "mean_token_accuracy": 0.235809126496315,
      "num_tokens": 151998802.0,
      "step": 73240
    },
    {
      "entropy": 5.718212127685547,
      "epoch": 6.644140058055152,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001657433643916499,
      "loss": 4.8421,
      "mean_token_accuracy": 0.23306339830160142,
      "num_tokens": 152009085.0,
      "step": 73245
    },
    {
      "entropy": 5.681940317153931,
      "epoch": 6.644593613933236,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016571507988124328,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2341841608285904,
      "num_tokens": 152019239.0,
      "step": 73250
    },
    {
      "entropy": 5.682776927947998,
      "epoch": 6.6450471698113205,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001656867976308767,
      "loss": 4.8431,
      "mean_token_accuracy": 0.23669089376926422,
      "num_tokens": 152029202.0,
      "step": 73255
    },
    {
      "entropy": 5.639969253540039,
      "epoch": 6.645500725689405,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016565851764113493,
      "loss": 4.7154,
      "mean_token_accuracy": 0.2409015953540802,
      "num_tokens": 152039578.0,
      "step": 73260
    },
    {
      "entropy": 5.772208309173584,
      "epoch": 6.645954281567489,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001656302399126029,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22917351722717286,
      "num_tokens": 152049943.0,
      "step": 73265
    },
    {
      "entropy": 5.76683406829834,
      "epoch": 6.646407837445573,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001656019644458654,
      "loss": 4.9298,
      "mean_token_accuracy": 0.22459129393100738,
      "num_tokens": 152061401.0,
      "step": 73270
    },
    {
      "entropy": 5.785471773147583,
      "epoch": 6.6468613933236576,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001655736912415071,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22584091871976852,
      "num_tokens": 152071706.0,
      "step": 73275
    },
    {
      "entropy": 5.640091133117676,
      "epoch": 6.647314949201742,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016554542030011276,
      "loss": 4.7312,
      "mean_token_accuracy": 0.23901208043098449,
      "num_tokens": 152081812.0,
      "step": 73280
    },
    {
      "entropy": 5.685503578186035,
      "epoch": 6.647768505079826,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016551715162226708,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2322638913989067,
      "num_tokens": 152091875.0,
      "step": 73285
    },
    {
      "entropy": 5.6859029769897464,
      "epoch": 6.64822206095791,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001654888852085546,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23076315075159073,
      "num_tokens": 152102939.0,
      "step": 73290
    },
    {
      "entropy": 5.763209438323974,
      "epoch": 6.648675616835995,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016546062105955982,
      "loss": 4.9372,
      "mean_token_accuracy": 0.22495071738958358,
      "num_tokens": 152115277.0,
      "step": 73295
    },
    {
      "entropy": 5.742188405990601,
      "epoch": 6.649129172714078,
      "grad_norm": 1.375,
      "learning_rate": 0.00016543235917586738,
      "loss": 4.8435,
      "mean_token_accuracy": 0.2427151769399643,
      "num_tokens": 152125552.0,
      "step": 73300
    },
    {
      "entropy": 5.746438694000244,
      "epoch": 6.649582728592162,
      "grad_norm": 1.546875,
      "learning_rate": 0.00016540409955806163,
      "loss": 4.8594,
      "mean_token_accuracy": 0.23065048158168794,
      "num_tokens": 152134834.0,
      "step": 73305
    },
    {
      "entropy": 5.69749755859375,
      "epoch": 6.6500362844702465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016537584220672696,
      "loss": 4.8708,
      "mean_token_accuracy": 0.22877785414457322,
      "num_tokens": 152146405.0,
      "step": 73310
    },
    {
      "entropy": 5.643028402328492,
      "epoch": 6.650489840348331,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001653475871224479,
      "loss": 4.7632,
      "mean_token_accuracy": 0.2407899871468544,
      "num_tokens": 152156450.0,
      "step": 73315
    },
    {
      "entropy": 5.68131799697876,
      "epoch": 6.650943396226415,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016531933430580868,
      "loss": 4.7972,
      "mean_token_accuracy": 0.23994619846343995,
      "num_tokens": 152166586.0,
      "step": 73320
    },
    {
      "entropy": 5.712125062942505,
      "epoch": 6.651396952104499,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016529108375739357,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23764376044273378,
      "num_tokens": 152176880.0,
      "step": 73325
    },
    {
      "entropy": 5.701754856109619,
      "epoch": 6.651850507982584,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001652628354777868,
      "loss": 4.8165,
      "mean_token_accuracy": 0.2298336684703827,
      "num_tokens": 152186467.0,
      "step": 73330
    },
    {
      "entropy": 5.745971012115478,
      "epoch": 6.652304063860668,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016523458946757257,
      "loss": 4.8983,
      "mean_token_accuracy": 0.23041210770606996,
      "num_tokens": 152196367.0,
      "step": 73335
    },
    {
      "entropy": 5.717006683349609,
      "epoch": 6.652757619738752,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016520634572733502,
      "loss": 4.7574,
      "mean_token_accuracy": 0.2447450190782547,
      "num_tokens": 152206098.0,
      "step": 73340
    },
    {
      "entropy": 5.692621040344238,
      "epoch": 6.6532111756168355,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016517810425765805,
      "loss": 4.7835,
      "mean_token_accuracy": 0.2312995284795761,
      "num_tokens": 152215707.0,
      "step": 73345
    },
    {
      "entropy": 5.662269163131714,
      "epoch": 6.653664731494921,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016514986505912604,
      "loss": 4.7863,
      "mean_token_accuracy": 0.2390891745686531,
      "num_tokens": 152226420.0,
      "step": 73350
    },
    {
      "entropy": 5.722865009307862,
      "epoch": 6.654118287373004,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001651216281323228,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22880950272083284,
      "num_tokens": 152235607.0,
      "step": 73355
    },
    {
      "entropy": 5.696837663650513,
      "epoch": 6.654571843251088,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001650933934778323,
      "loss": 4.8576,
      "mean_token_accuracy": 0.23381619304418563,
      "num_tokens": 152244867.0,
      "step": 73360
    },
    {
      "entropy": 5.69116702079773,
      "epoch": 6.655025399129173,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016506516109623837,
      "loss": 4.8163,
      "mean_token_accuracy": 0.2373882055282593,
      "num_tokens": 152254646.0,
      "step": 73365
    },
    {
      "entropy": 5.667304039001465,
      "epoch": 6.655478955007257,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016503693098812497,
      "loss": 4.7184,
      "mean_token_accuracy": 0.24080202728509903,
      "num_tokens": 152264054.0,
      "step": 73370
    },
    {
      "entropy": 5.727405548095703,
      "epoch": 6.655932510885341,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016500870315407586,
      "loss": 4.8039,
      "mean_token_accuracy": 0.2317151665687561,
      "num_tokens": 152274231.0,
      "step": 73375
    },
    {
      "entropy": 5.765132761001587,
      "epoch": 6.656386066763425,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016498047759467473,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2300128310918808,
      "num_tokens": 152285145.0,
      "step": 73380
    },
    {
      "entropy": 5.6967747688293455,
      "epoch": 6.65683962264151,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00016495225431050538,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2319079667329788,
      "num_tokens": 152296140.0,
      "step": 73385
    },
    {
      "entropy": 5.671389722824097,
      "epoch": 6.657293178519594,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001649240333021515,
      "loss": 4.7609,
      "mean_token_accuracy": 0.24044249206781387,
      "num_tokens": 152305672.0,
      "step": 73390
    },
    {
      "entropy": 5.76292314529419,
      "epoch": 6.657746734397678,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016489581457019664,
      "loss": 4.9468,
      "mean_token_accuracy": 0.22278534919023513,
      "num_tokens": 152317000.0,
      "step": 73395
    },
    {
      "entropy": 5.709265899658203,
      "epoch": 6.6582002902757615,
      "grad_norm": 1.375,
      "learning_rate": 0.00016486759811522441,
      "loss": 4.7607,
      "mean_token_accuracy": 0.23822851926088334,
      "num_tokens": 152326446.0,
      "step": 73400
    },
    {
      "entropy": 5.694754266738892,
      "epoch": 6.658653846153846,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016483938393781834,
      "loss": 4.7991,
      "mean_token_accuracy": 0.23398952931165695,
      "num_tokens": 152337277.0,
      "step": 73405
    },
    {
      "entropy": 5.717377948760986,
      "epoch": 6.65910740203193,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016481117203856187,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22976155579090118,
      "num_tokens": 152347943.0,
      "step": 73410
    },
    {
      "entropy": 5.693431186676025,
      "epoch": 6.659560957910014,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016478296241803834,
      "loss": 4.836,
      "mean_token_accuracy": 0.2384708985686302,
      "num_tokens": 152358805.0,
      "step": 73415
    },
    {
      "entropy": 5.723751163482666,
      "epoch": 6.660014513788099,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001647547550768313,
      "loss": 4.8845,
      "mean_token_accuracy": 0.23042507320642472,
      "num_tokens": 152369249.0,
      "step": 73420
    },
    {
      "entropy": 5.750255346298218,
      "epoch": 6.660468069666183,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016472655001552406,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23044152110815047,
      "num_tokens": 152379722.0,
      "step": 73425
    },
    {
      "entropy": 5.705027723312378,
      "epoch": 6.660921625544267,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016469834723469983,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23592676371335983,
      "num_tokens": 152390506.0,
      "step": 73430
    },
    {
      "entropy": 5.66292552947998,
      "epoch": 6.661375181422351,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001646701467349419,
      "loss": 4.8169,
      "mean_token_accuracy": 0.244455748796463,
      "num_tokens": 152400425.0,
      "step": 73435
    },
    {
      "entropy": 5.707886266708374,
      "epoch": 6.661828737300436,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016464194851683347,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2298708140850067,
      "num_tokens": 152409843.0,
      "step": 73440
    },
    {
      "entropy": 5.678226041793823,
      "epoch": 6.66228229317852,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016461375258095757,
      "loss": 4.8594,
      "mean_token_accuracy": 0.23256205320358275,
      "num_tokens": 152420950.0,
      "step": 73445
    },
    {
      "entropy": 5.716143465042114,
      "epoch": 6.662735849056604,
      "grad_norm": 1.375,
      "learning_rate": 0.0001645855589278975,
      "loss": 4.8019,
      "mean_token_accuracy": 0.23972900807857514,
      "num_tokens": 152431383.0,
      "step": 73450
    },
    {
      "entropy": 5.668188190460205,
      "epoch": 6.663189404934688,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016455736755823617,
      "loss": 4.84,
      "mean_token_accuracy": 0.22809351235628128,
      "num_tokens": 152441119.0,
      "step": 73455
    },
    {
      "entropy": 5.7152698040008545,
      "epoch": 6.663642960812772,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016452917847255665,
      "loss": 4.8905,
      "mean_token_accuracy": 0.22718822658061982,
      "num_tokens": 152451272.0,
      "step": 73460
    },
    {
      "entropy": 5.690067386627197,
      "epoch": 6.664096516690856,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016450099167144184,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2352613925933838,
      "num_tokens": 152461475.0,
      "step": 73465
    },
    {
      "entropy": 5.725361394882202,
      "epoch": 6.66455007256894,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016447280715547468,
      "loss": 4.8252,
      "mean_token_accuracy": 0.24046750813722612,
      "num_tokens": 152471665.0,
      "step": 73470
    },
    {
      "entropy": 5.6278221130371096,
      "epoch": 6.665003628447025,
      "grad_norm": 1.5,
      "learning_rate": 0.0001644446249252381,
      "loss": 4.7284,
      "mean_token_accuracy": 0.25113736242055895,
      "num_tokens": 152481937.0,
      "step": 73475
    },
    {
      "entropy": 5.6778641700744625,
      "epoch": 6.665457184325109,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00016441644498131473,
      "loss": 4.7782,
      "mean_token_accuracy": 0.23942315578460693,
      "num_tokens": 152491607.0,
      "step": 73480
    },
    {
      "entropy": 5.67618203163147,
      "epoch": 6.665910740203193,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001643882673242875,
      "loss": 4.7835,
      "mean_token_accuracy": 0.2393004849553108,
      "num_tokens": 152501898.0,
      "step": 73485
    },
    {
      "entropy": 5.730347728729248,
      "epoch": 6.6663642960812775,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001643600919547391,
      "loss": 4.8538,
      "mean_token_accuracy": 0.23185974657535552,
      "num_tokens": 152512354.0,
      "step": 73490
    },
    {
      "entropy": 5.692304515838623,
      "epoch": 6.666817851959362,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016433191887325232,
      "loss": 4.8273,
      "mean_token_accuracy": 0.22835113108158112,
      "num_tokens": 152522695.0,
      "step": 73495
    },
    {
      "entropy": 5.726573848724366,
      "epoch": 6.667271407837445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016430374808040955,
      "loss": 4.8484,
      "mean_token_accuracy": 0.23091787099838257,
      "num_tokens": 152533354.0,
      "step": 73500
    },
    {
      "entropy": 5.7522745609283445,
      "epoch": 6.66772496371553,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001642755795767935,
      "loss": 4.8968,
      "mean_token_accuracy": 0.2328228771686554,
      "num_tokens": 152543695.0,
      "step": 73505
    },
    {
      "entropy": 5.685683965682983,
      "epoch": 6.668178519593614,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001642474133629866,
      "loss": 4.7368,
      "mean_token_accuracy": 0.24643722027540207,
      "num_tokens": 152553736.0,
      "step": 73510
    },
    {
      "entropy": 5.677708959579467,
      "epoch": 6.668632075471698,
      "grad_norm": 1.375,
      "learning_rate": 0.00016421924943957144,
      "loss": 4.8126,
      "mean_token_accuracy": 0.2369406118988991,
      "num_tokens": 152564028.0,
      "step": 73515
    },
    {
      "entropy": 5.6588170528411865,
      "epoch": 6.669085631349782,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016419108780713046,
      "loss": 4.771,
      "mean_token_accuracy": 0.2375771149992943,
      "num_tokens": 152573614.0,
      "step": 73520
    },
    {
      "entropy": 5.6961668014526365,
      "epoch": 6.669539187227866,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016416292846624603,
      "loss": 4.897,
      "mean_token_accuracy": 0.22718661278486252,
      "num_tokens": 152584077.0,
      "step": 73525
    },
    {
      "entropy": 5.757908296585083,
      "epoch": 6.669992743105951,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001641347714175005,
      "loss": 4.9696,
      "mean_token_accuracy": 0.22258879095315934,
      "num_tokens": 152594130.0,
      "step": 73530
    },
    {
      "entropy": 5.729808568954468,
      "epoch": 6.670446298984035,
      "grad_norm": 1.546875,
      "learning_rate": 0.00016410661666147613,
      "loss": 4.8362,
      "mean_token_accuracy": 0.23839647322893143,
      "num_tokens": 152604887.0,
      "step": 73535
    },
    {
      "entropy": 5.6790975570678714,
      "epoch": 6.670899854862119,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001640784641987551,
      "loss": 4.7658,
      "mean_token_accuracy": 0.23906605541706086,
      "num_tokens": 152614242.0,
      "step": 73540
    },
    {
      "entropy": 5.655331707000732,
      "epoch": 6.6713534107402035,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001640503140299198,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2430289715528488,
      "num_tokens": 152624105.0,
      "step": 73545
    },
    {
      "entropy": 5.636852550506592,
      "epoch": 6.671806966618288,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016402216615555226,
      "loss": 4.7254,
      "mean_token_accuracy": 0.24410470128059386,
      "num_tokens": 152633869.0,
      "step": 73550
    },
    {
      "entropy": 5.703981161117554,
      "epoch": 6.672260522496371,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016399402057623463,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2270257443189621,
      "num_tokens": 152643400.0,
      "step": 73555
    },
    {
      "entropy": 5.727384090423584,
      "epoch": 6.672714078374455,
      "grad_norm": 1.5,
      "learning_rate": 0.0001639658772925489,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22675511687994004,
      "num_tokens": 152653304.0,
      "step": 73560
    },
    {
      "entropy": 5.701131868362427,
      "epoch": 6.67316763425254,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016393773630507714,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23362574726343155,
      "num_tokens": 152662899.0,
      "step": 73565
    },
    {
      "entropy": 5.729264545440674,
      "epoch": 6.673621190130624,
      "grad_norm": 1.5,
      "learning_rate": 0.00016390959761440133,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22412341088056564,
      "num_tokens": 152672554.0,
      "step": 73570
    },
    {
      "entropy": 5.742924928665161,
      "epoch": 6.674074746008708,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016388146122110322,
      "loss": 4.902,
      "mean_token_accuracy": 0.23004448264837266,
      "num_tokens": 152683219.0,
      "step": 73575
    },
    {
      "entropy": 5.639815139770508,
      "epoch": 6.6745283018867925,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001638533271257649,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23677269667387008,
      "num_tokens": 152693550.0,
      "step": 73580
    },
    {
      "entropy": 5.740131092071533,
      "epoch": 6.674981857764877,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016382519532896813,
      "loss": 4.8978,
      "mean_token_accuracy": 0.22338780760765076,
      "num_tokens": 152704241.0,
      "step": 73585
    },
    {
      "entropy": 5.729410314559937,
      "epoch": 6.675435413642961,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001637970658312946,
      "loss": 4.7792,
      "mean_token_accuracy": 0.23524338454008104,
      "num_tokens": 152714648.0,
      "step": 73590
    },
    {
      "entropy": 5.652108192443848,
      "epoch": 6.675888969521045,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016376893863332616,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2360846996307373,
      "num_tokens": 152724276.0,
      "step": 73595
    },
    {
      "entropy": 5.701826381683349,
      "epoch": 6.6763425253991295,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016374081373564438,
      "loss": 4.8324,
      "mean_token_accuracy": 0.23607291877269745,
      "num_tokens": 152734055.0,
      "step": 73600
    },
    {
      "entropy": 5.724689149856568,
      "epoch": 6.676796081277214,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016371269113883087,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22709676921367644,
      "num_tokens": 152745435.0,
      "step": 73605
    },
    {
      "entropy": 5.694452476501465,
      "epoch": 6.677249637155297,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001636845708434672,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2307298004627228,
      "num_tokens": 152754909.0,
      "step": 73610
    },
    {
      "entropy": 5.713170909881592,
      "epoch": 6.6777031930333814,
      "grad_norm": 1.609375,
      "learning_rate": 0.00016365645285013514,
      "loss": 4.8704,
      "mean_token_accuracy": 0.23380965143442153,
      "num_tokens": 152764668.0,
      "step": 73615
    },
    {
      "entropy": 5.717723703384399,
      "epoch": 6.678156748911466,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016362833715941595,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23501888066530227,
      "num_tokens": 152775174.0,
      "step": 73620
    },
    {
      "entropy": 5.702842092514038,
      "epoch": 6.67861030478955,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001636002237718911,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2271185114979744,
      "num_tokens": 152785618.0,
      "step": 73625
    },
    {
      "entropy": 5.738274431228637,
      "epoch": 6.679063860667634,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016357211268814203,
      "loss": 4.8468,
      "mean_token_accuracy": 0.23348528742790223,
      "num_tokens": 152795930.0,
      "step": 73630
    },
    {
      "entropy": 5.795600891113281,
      "epoch": 6.6795174165457185,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016354400390875006,
      "loss": 4.953,
      "mean_token_accuracy": 0.22124177515506743,
      "num_tokens": 152807169.0,
      "step": 73635
    },
    {
      "entropy": 5.699259328842163,
      "epoch": 6.679970972423803,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016351589743429654,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2286711573600769,
      "num_tokens": 152817397.0,
      "step": 73640
    },
    {
      "entropy": 5.725169944763183,
      "epoch": 6.680424528301887,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016348779326536255,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22804758399724961,
      "num_tokens": 152827788.0,
      "step": 73645
    },
    {
      "entropy": 5.742950916290283,
      "epoch": 6.680878084179971,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00016345969140252952,
      "loss": 4.8301,
      "mean_token_accuracy": 0.23473363518714904,
      "num_tokens": 152837526.0,
      "step": 73650
    },
    {
      "entropy": 5.636748838424682,
      "epoch": 6.681331640058055,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016343159184637845,
      "loss": 4.7115,
      "mean_token_accuracy": 0.24589669853448867,
      "num_tokens": 152848607.0,
      "step": 73655
    },
    {
      "entropy": 5.683617401123047,
      "epoch": 6.681785195936139,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016340349459749052,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23803334981203078,
      "num_tokens": 152859873.0,
      "step": 73660
    },
    {
      "entropy": 5.715646743774414,
      "epoch": 6.682238751814223,
      "grad_norm": 1.625,
      "learning_rate": 0.00016337539965644677,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22966769188642502,
      "num_tokens": 152870281.0,
      "step": 73665
    },
    {
      "entropy": 5.722357130050659,
      "epoch": 6.6826923076923075,
      "grad_norm": 1.59375,
      "learning_rate": 0.00016334730702382817,
      "loss": 4.8248,
      "mean_token_accuracy": 0.22922955453395844,
      "num_tokens": 152880558.0,
      "step": 73670
    },
    {
      "entropy": 5.72389702796936,
      "epoch": 6.683145863570392,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001633192167002157,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22331023961305618,
      "num_tokens": 152891764.0,
      "step": 73675
    },
    {
      "entropy": 5.752257680892944,
      "epoch": 6.683599419448476,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016329112868619026,
      "loss": 4.8402,
      "mean_token_accuracy": 0.24507542401552201,
      "num_tokens": 152901862.0,
      "step": 73680
    },
    {
      "entropy": 5.743086290359497,
      "epoch": 6.68405297532656,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001632630429823328,
      "loss": 4.8919,
      "mean_token_accuracy": 0.23362889140844345,
      "num_tokens": 152912636.0,
      "step": 73685
    },
    {
      "entropy": 5.608486843109131,
      "epoch": 6.6845065312046446,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001632349595892241,
      "loss": 4.6361,
      "mean_token_accuracy": 0.2448476582765579,
      "num_tokens": 152922863.0,
      "step": 73690
    },
    {
      "entropy": 5.694629573822022,
      "epoch": 6.684960087082729,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016320687850744493,
      "loss": 4.8343,
      "mean_token_accuracy": 0.23667318671941756,
      "num_tokens": 152933258.0,
      "step": 73695
    },
    {
      "entropy": 5.745881032943726,
      "epoch": 6.685413642960813,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016317879973757593,
      "loss": 4.8544,
      "mean_token_accuracy": 0.23365900963544844,
      "num_tokens": 152944347.0,
      "step": 73700
    },
    {
      "entropy": 5.679413223266602,
      "epoch": 6.685867198838897,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016315072328019792,
      "loss": 4.8444,
      "mean_token_accuracy": 0.23134596645832062,
      "num_tokens": 152954591.0,
      "step": 73705
    },
    {
      "entropy": 5.708281660079956,
      "epoch": 6.686320754716981,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016312264913589142,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22376088052988052,
      "num_tokens": 152965888.0,
      "step": 73710
    },
    {
      "entropy": 5.664129114151001,
      "epoch": 6.686774310595065,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016309457730523697,
      "loss": 4.8164,
      "mean_token_accuracy": 0.22699255496263504,
      "num_tokens": 152976068.0,
      "step": 73715
    },
    {
      "entropy": 5.762937688827515,
      "epoch": 6.687227866473149,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016306650778881527,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2406255841255188,
      "num_tokens": 152985583.0,
      "step": 73720
    },
    {
      "entropy": 5.681034851074219,
      "epoch": 6.6876814223512335,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001630384405872067,
      "loss": 4.7833,
      "mean_token_accuracy": 0.2380860298871994,
      "num_tokens": 152995603.0,
      "step": 73725
    },
    {
      "entropy": 5.806798505783081,
      "epoch": 6.688134978229318,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016301037570099172,
      "loss": 4.9477,
      "mean_token_accuracy": 0.22554632872343064,
      "num_tokens": 153006276.0,
      "step": 73730
    },
    {
      "entropy": 5.706961584091187,
      "epoch": 6.688588534107402,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016298231313075068,
      "loss": 4.8291,
      "mean_token_accuracy": 0.24016245305538178,
      "num_tokens": 153015578.0,
      "step": 73735
    },
    {
      "entropy": 5.717144918441773,
      "epoch": 6.689042089985486,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016295425287706397,
      "loss": 4.7775,
      "mean_token_accuracy": 0.23188602030277253,
      "num_tokens": 153026033.0,
      "step": 73740
    },
    {
      "entropy": 5.683669805526733,
      "epoch": 6.689495645863571,
      "grad_norm": 1.375,
      "learning_rate": 0.0001629261949405118,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23356156796216965,
      "num_tokens": 153036360.0,
      "step": 73745
    },
    {
      "entropy": 5.720836448669433,
      "epoch": 6.689949201741655,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016289813932167453,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23395867347717286,
      "num_tokens": 153046360.0,
      "step": 73750
    },
    {
      "entropy": 5.631626415252685,
      "epoch": 6.690402757619739,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016287008602113235,
      "loss": 4.7673,
      "mean_token_accuracy": 0.23245521187782286,
      "num_tokens": 153056663.0,
      "step": 73755
    },
    {
      "entropy": 5.719799757003784,
      "epoch": 6.690856313497823,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016284203503946532,
      "loss": 4.846,
      "mean_token_accuracy": 0.22868212759494783,
      "num_tokens": 153066641.0,
      "step": 73760
    },
    {
      "entropy": 5.7053145408630375,
      "epoch": 6.691309869375907,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001628139863772536,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2225598856806755,
      "num_tokens": 153077478.0,
      "step": 73765
    },
    {
      "entropy": 5.624771022796631,
      "epoch": 6.691763425253991,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016278594003507724,
      "loss": 4.7245,
      "mean_token_accuracy": 0.24247135519981383,
      "num_tokens": 153087616.0,
      "step": 73770
    },
    {
      "entropy": 5.634509611129761,
      "epoch": 6.692216981132075,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016275789601351623,
      "loss": 4.7638,
      "mean_token_accuracy": 0.23477419912815095,
      "num_tokens": 153098359.0,
      "step": 73775
    },
    {
      "entropy": 5.749018955230713,
      "epoch": 6.69267053701016,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016272985431315043,
      "loss": 4.8558,
      "mean_token_accuracy": 0.23239511549472808,
      "num_tokens": 153107996.0,
      "step": 73780
    },
    {
      "entropy": 5.699700450897216,
      "epoch": 6.693124092888244,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016270181493455995,
      "loss": 4.8427,
      "mean_token_accuracy": 0.23222757428884505,
      "num_tokens": 153118674.0,
      "step": 73785
    },
    {
      "entropy": 5.6746338367462155,
      "epoch": 6.693577648766328,
      "grad_norm": 1.375,
      "learning_rate": 0.00016267377787832459,
      "loss": 4.7576,
      "mean_token_accuracy": 0.23733555227518083,
      "num_tokens": 153128609.0,
      "step": 73790
    },
    {
      "entropy": 5.603982877731323,
      "epoch": 6.694031204644412,
      "grad_norm": 1.625,
      "learning_rate": 0.00016264574314502406,
      "loss": 4.772,
      "mean_token_accuracy": 0.2450007513165474,
      "num_tokens": 153137937.0,
      "step": 73795
    },
    {
      "entropy": 5.730864000320435,
      "epoch": 6.694484760522497,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016261771073523823,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22650693655014037,
      "num_tokens": 153148443.0,
      "step": 73800
    },
    {
      "entropy": 5.678388547897339,
      "epoch": 6.694938316400581,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001625896806495468,
      "loss": 4.7839,
      "mean_token_accuracy": 0.24179973751306533,
      "num_tokens": 153158947.0,
      "step": 73805
    },
    {
      "entropy": 5.688528299331665,
      "epoch": 6.695391872278664,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001625616528885294,
      "loss": 4.8352,
      "mean_token_accuracy": 0.2331357255578041,
      "num_tokens": 153169483.0,
      "step": 73810
    },
    {
      "entropy": 5.695988702774048,
      "epoch": 6.6958454281567485,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016253362745276558,
      "loss": 4.7668,
      "mean_token_accuracy": 0.24022380262613297,
      "num_tokens": 153181039.0,
      "step": 73815
    },
    {
      "entropy": 5.690416240692139,
      "epoch": 6.696298984034833,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001625056043428351,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2371407136321068,
      "num_tokens": 153191021.0,
      "step": 73820
    },
    {
      "entropy": 5.68513126373291,
      "epoch": 6.696752539912917,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001624775835593174,
      "loss": 4.7807,
      "mean_token_accuracy": 0.24341676235198975,
      "num_tokens": 153201644.0,
      "step": 73825
    },
    {
      "entropy": 5.794989776611328,
      "epoch": 6.697206095791001,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001624495651027919,
      "loss": 4.9451,
      "mean_token_accuracy": 0.22742465138435364,
      "num_tokens": 153212185.0,
      "step": 73830
    },
    {
      "entropy": 5.709947204589843,
      "epoch": 6.697659651669086,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001624215489738381,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2343137741088867,
      "num_tokens": 153223197.0,
      "step": 73835
    },
    {
      "entropy": 5.725047922134399,
      "epoch": 6.69811320754717,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016239353517303538,
      "loss": 4.8362,
      "mean_token_accuracy": 0.2321936547756195,
      "num_tokens": 153233582.0,
      "step": 73840
    },
    {
      "entropy": 5.750746631622315,
      "epoch": 6.698566763425254,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016236552370096303,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2317337363958359,
      "num_tokens": 153245043.0,
      "step": 73845
    },
    {
      "entropy": 5.66613450050354,
      "epoch": 6.699020319303338,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016233751455820028,
      "loss": 4.7641,
      "mean_token_accuracy": 0.23563439548015594,
      "num_tokens": 153254848.0,
      "step": 73850
    },
    {
      "entropy": 5.754138517379761,
      "epoch": 6.699473875181423,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001623095077453265,
      "loss": 4.8505,
      "mean_token_accuracy": 0.23378751426935196,
      "num_tokens": 153264748.0,
      "step": 73855
    },
    {
      "entropy": 5.675392961502075,
      "epoch": 6.699927431059507,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001622815032629208,
      "loss": 4.8078,
      "mean_token_accuracy": 0.24015552103519439,
      "num_tokens": 153273920.0,
      "step": 73860
    },
    {
      "entropy": 5.69617657661438,
      "epoch": 6.70038098693759,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001622535011115624,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23248081803321838,
      "num_tokens": 153283505.0,
      "step": 73865
    },
    {
      "entropy": 5.677442598342895,
      "epoch": 6.700834542815675,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016222550129183028,
      "loss": 4.7827,
      "mean_token_accuracy": 0.24360829889774321,
      "num_tokens": 153294003.0,
      "step": 73870
    },
    {
      "entropy": 5.72383222579956,
      "epoch": 6.701288098693759,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016219750380430352,
      "loss": 4.8677,
      "mean_token_accuracy": 0.23362002223730088,
      "num_tokens": 153304040.0,
      "step": 73875
    },
    {
      "entropy": 5.663859701156616,
      "epoch": 6.701741654571843,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016216950864956115,
      "loss": 4.7248,
      "mean_token_accuracy": 0.24980342984199524,
      "num_tokens": 153314897.0,
      "step": 73880
    },
    {
      "entropy": 5.743944692611694,
      "epoch": 6.702195210449927,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016214151582818204,
      "loss": 4.8922,
      "mean_token_accuracy": 0.23884375244379044,
      "num_tokens": 153325300.0,
      "step": 73885
    },
    {
      "entropy": 5.703127145767212,
      "epoch": 6.702648766328012,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001621135253407452,
      "loss": 4.7906,
      "mean_token_accuracy": 0.24398779422044753,
      "num_tokens": 153335201.0,
      "step": 73890
    },
    {
      "entropy": 5.682218360900879,
      "epoch": 6.703102322206096,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016208553718782937,
      "loss": 4.7884,
      "mean_token_accuracy": 0.24040312319993973,
      "num_tokens": 153344979.0,
      "step": 73895
    },
    {
      "entropy": 5.640918588638305,
      "epoch": 6.70355587808418,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016205755137001349,
      "loss": 4.7614,
      "mean_token_accuracy": 0.24925082176923752,
      "num_tokens": 153354796.0,
      "step": 73900
    },
    {
      "entropy": 5.668312501907349,
      "epoch": 6.7040094339622645,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016202956788787616,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22783312946558,
      "num_tokens": 153364774.0,
      "step": 73905
    },
    {
      "entropy": 5.61656608581543,
      "epoch": 6.704462989840349,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001620015867419962,
      "loss": 4.7354,
      "mean_token_accuracy": 0.23887523710727693,
      "num_tokens": 153375196.0,
      "step": 73910
    },
    {
      "entropy": 5.792078495025635,
      "epoch": 6.704916545718433,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001619736079329522,
      "loss": 4.9304,
      "mean_token_accuracy": 0.2265590861439705,
      "num_tokens": 153385665.0,
      "step": 73915
    },
    {
      "entropy": 5.735355424880981,
      "epoch": 6.705370101596516,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016194563146132275,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22574643343687056,
      "num_tokens": 153396075.0,
      "step": 73920
    },
    {
      "entropy": 5.694345331192016,
      "epoch": 6.705823657474601,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016191765732768656,
      "loss": 4.7542,
      "mean_token_accuracy": 0.24326709061861038,
      "num_tokens": 153406025.0,
      "step": 73925
    },
    {
      "entropy": 5.656508064270019,
      "epoch": 6.706277213352685,
      "grad_norm": 1.5,
      "learning_rate": 0.00016188968553262196,
      "loss": 4.7785,
      "mean_token_accuracy": 0.2414724662899971,
      "num_tokens": 153415940.0,
      "step": 73930
    },
    {
      "entropy": 5.583708810806274,
      "epoch": 6.706730769230769,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016186171607670752,
      "loss": 4.6551,
      "mean_token_accuracy": 0.25334600210189817,
      "num_tokens": 153426562.0,
      "step": 73935
    },
    {
      "entropy": 5.751545143127442,
      "epoch": 6.707184325108853,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016183374896052162,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2274552121758461,
      "num_tokens": 153436866.0,
      "step": 73940
    },
    {
      "entropy": 5.6723246574401855,
      "epoch": 6.707637880986938,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001618057841846425,
      "loss": 4.7799,
      "mean_token_accuracy": 0.24345881640911102,
      "num_tokens": 153446547.0,
      "step": 73945
    },
    {
      "entropy": 5.734947919845581,
      "epoch": 6.708091436865022,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016177782174964873,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23488697707653045,
      "num_tokens": 153457491.0,
      "step": 73950
    },
    {
      "entropy": 5.643275260925293,
      "epoch": 6.708544992743106,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016174986165611843,
      "loss": 4.7735,
      "mean_token_accuracy": 0.2338003396987915,
      "num_tokens": 153467241.0,
      "step": 73955
    },
    {
      "entropy": 5.758356285095215,
      "epoch": 6.7089985486211905,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001617219039046299,
      "loss": 4.8627,
      "mean_token_accuracy": 0.23504533469676972,
      "num_tokens": 153478233.0,
      "step": 73960
    },
    {
      "entropy": 5.740468072891235,
      "epoch": 6.709452104499274,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016169394849576123,
      "loss": 4.9203,
      "mean_token_accuracy": 0.227863147854805,
      "num_tokens": 153488559.0,
      "step": 73965
    },
    {
      "entropy": 5.770613193511963,
      "epoch": 6.709905660377358,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001616659954300906,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22354158461093904,
      "num_tokens": 153500176.0,
      "step": 73970
    },
    {
      "entropy": 5.6549163341522215,
      "epoch": 6.710359216255442,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016163804470819605,
      "loss": 4.7379,
      "mean_token_accuracy": 0.2340419441461563,
      "num_tokens": 153510773.0,
      "step": 73975
    },
    {
      "entropy": 5.67246789932251,
      "epoch": 6.710812772133527,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001616100963306555,
      "loss": 4.7429,
      "mean_token_accuracy": 0.23394516855478287,
      "num_tokens": 153522320.0,
      "step": 73980
    },
    {
      "entropy": 5.718921756744384,
      "epoch": 6.711266328011611,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001615821502980472,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23458809703588485,
      "num_tokens": 153531806.0,
      "step": 73985
    },
    {
      "entropy": 5.674756765365601,
      "epoch": 6.711719883889695,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016155420661094892,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22665900141000747,
      "num_tokens": 153543116.0,
      "step": 73990
    },
    {
      "entropy": 5.7043044090271,
      "epoch": 6.7121734397677795,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00016152626526993853,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23249350786209105,
      "num_tokens": 153552225.0,
      "step": 73995
    },
    {
      "entropy": 5.654328870773315,
      "epoch": 6.712626995645864,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001614983262755939,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23429814577102662,
      "num_tokens": 153562190.0,
      "step": 74000
    },
    {
      "entropy": 5.627152967453003,
      "epoch": 6.713080551523948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001614703896284928,
      "loss": 4.7716,
      "mean_token_accuracy": 0.24127721935510635,
      "num_tokens": 153573134.0,
      "step": 74005
    },
    {
      "entropy": 5.772825574874878,
      "epoch": 6.713534107402032,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000161442455329213,
      "loss": 4.8134,
      "mean_token_accuracy": 0.23548368811607362,
      "num_tokens": 153583886.0,
      "step": 74010
    },
    {
      "entropy": 5.733354663848877,
      "epoch": 6.7139876632801165,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016141452337833207,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22541408389806747,
      "num_tokens": 153595673.0,
      "step": 74015
    },
    {
      "entropy": 5.7267557144165036,
      "epoch": 6.7144412191582,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001613865937764278,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22786572873592376,
      "num_tokens": 153606154.0,
      "step": 74020
    },
    {
      "entropy": 5.673788404464721,
      "epoch": 6.714894775036284,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016135866652407777,
      "loss": 4.7861,
      "mean_token_accuracy": 0.2352248787879944,
      "num_tokens": 153616253.0,
      "step": 74025
    },
    {
      "entropy": 5.6769567966461185,
      "epoch": 6.7153483309143684,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016133074162185947,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23527016192674638,
      "num_tokens": 153626595.0,
      "step": 74030
    },
    {
      "entropy": 5.714638614654541,
      "epoch": 6.715801886792453,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016130281907035036,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23486142456531525,
      "num_tokens": 153637164.0,
      "step": 74035
    },
    {
      "entropy": 5.59273509979248,
      "epoch": 6.716255442670537,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016127489887012792,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2533374309539795,
      "num_tokens": 153648341.0,
      "step": 74040
    },
    {
      "entropy": 5.7912262916564945,
      "epoch": 6.716708998548621,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001612469810217695,
      "loss": 4.9306,
      "mean_token_accuracy": 0.22623936235904693,
      "num_tokens": 153658374.0,
      "step": 74045
    },
    {
      "entropy": 5.718336153030395,
      "epoch": 6.7171625544267055,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016121906552585258,
      "loss": 4.8232,
      "mean_token_accuracy": 0.23727944493293762,
      "num_tokens": 153667644.0,
      "step": 74050
    },
    {
      "entropy": 5.661896753311157,
      "epoch": 6.71761611030479,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016119115238295435,
      "loss": 4.721,
      "mean_token_accuracy": 0.24145569354295732,
      "num_tokens": 153677537.0,
      "step": 74055
    },
    {
      "entropy": 5.670519876480102,
      "epoch": 6.718069666182874,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016116324159365215,
      "loss": 4.8403,
      "mean_token_accuracy": 0.23824632614851,
      "num_tokens": 153688146.0,
      "step": 74060
    },
    {
      "entropy": 5.737116050720215,
      "epoch": 6.718523222060957,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001611353331585231,
      "loss": 4.8746,
      "mean_token_accuracy": 0.23369110524654388,
      "num_tokens": 153698312.0,
      "step": 74065
    },
    {
      "entropy": 5.755515766143799,
      "epoch": 6.718976777939043,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016110742707814432,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22447153478860854,
      "num_tokens": 153709969.0,
      "step": 74070
    },
    {
      "entropy": 5.630433893203735,
      "epoch": 6.719430333817126,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016107952335309302,
      "loss": 4.6714,
      "mean_token_accuracy": 0.244423945248127,
      "num_tokens": 153719855.0,
      "step": 74075
    },
    {
      "entropy": 5.694650793075562,
      "epoch": 6.71988388969521,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016105162198394612,
      "loss": 4.8141,
      "mean_token_accuracy": 0.22404515147209167,
      "num_tokens": 153730084.0,
      "step": 74080
    },
    {
      "entropy": 5.6851582527160645,
      "epoch": 6.7203374455732945,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016102372297128082,
      "loss": 4.7881,
      "mean_token_accuracy": 0.22773291617631913,
      "num_tokens": 153740597.0,
      "step": 74085
    },
    {
      "entropy": 5.745826864242554,
      "epoch": 6.720791001451379,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016099582631567394,
      "loss": 4.9078,
      "mean_token_accuracy": 0.22471851706504822,
      "num_tokens": 153751583.0,
      "step": 74090
    },
    {
      "entropy": 5.691237258911133,
      "epoch": 6.721244557329463,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016096793201770244,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2327282026410103,
      "num_tokens": 153761933.0,
      "step": 74095
    },
    {
      "entropy": 5.70614333152771,
      "epoch": 6.721698113207547,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016094004007794315,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2247857615351677,
      "num_tokens": 153773233.0,
      "step": 74100
    },
    {
      "entropy": 5.629740047454834,
      "epoch": 6.7221516690856316,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016091215049697288,
      "loss": 4.7643,
      "mean_token_accuracy": 0.23169605880975724,
      "num_tokens": 153784176.0,
      "step": 74105
    },
    {
      "entropy": 5.700917100906372,
      "epoch": 6.722605224963716,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001608842632753685,
      "loss": 4.7696,
      "mean_token_accuracy": 0.2342546746134758,
      "num_tokens": 153794193.0,
      "step": 74110
    },
    {
      "entropy": 5.6748542308807375,
      "epoch": 6.7230587808418,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00016085637841370645,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23208565413951873,
      "num_tokens": 153803539.0,
      "step": 74115
    },
    {
      "entropy": 5.663325262069702,
      "epoch": 6.7235123367198835,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001608284959125637,
      "loss": 4.7302,
      "mean_token_accuracy": 0.24093592315912246,
      "num_tokens": 153813846.0,
      "step": 74120
    },
    {
      "entropy": 5.631997776031494,
      "epoch": 6.723965892597968,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016080061577251675,
      "loss": 4.7611,
      "mean_token_accuracy": 0.23995478600263595,
      "num_tokens": 153824273.0,
      "step": 74125
    },
    {
      "entropy": 5.679180335998535,
      "epoch": 6.724419448476052,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016077273799414222,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23661873936653138,
      "num_tokens": 153834501.0,
      "step": 74130
    },
    {
      "entropy": 5.7508772850036625,
      "epoch": 6.724873004354136,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00016074486257801652,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22057586461305617,
      "num_tokens": 153842939.0,
      "step": 74135
    },
    {
      "entropy": 5.6858278751373295,
      "epoch": 6.7253265602322205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016071698952471614,
      "loss": 4.8747,
      "mean_token_accuracy": 0.23222548365592957,
      "num_tokens": 153852913.0,
      "step": 74140
    },
    {
      "entropy": 5.704034900665283,
      "epoch": 6.725780116110305,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016068911883481763,
      "loss": 4.7991,
      "mean_token_accuracy": 0.24004730135202407,
      "num_tokens": 153862924.0,
      "step": 74145
    },
    {
      "entropy": 5.713697099685669,
      "epoch": 6.726233671988389,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016066125050889714,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2298784449696541,
      "num_tokens": 153874358.0,
      "step": 74150
    },
    {
      "entropy": 5.7009193897247314,
      "epoch": 6.726687227866473,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001606333845475312,
      "loss": 4.8076,
      "mean_token_accuracy": 0.2319143146276474,
      "num_tokens": 153884076.0,
      "step": 74155
    },
    {
      "entropy": 5.731790971755982,
      "epoch": 6.727140783744558,
      "grad_norm": 1.40625,
      "learning_rate": 0.000160605520951296,
      "loss": 4.8204,
      "mean_token_accuracy": 0.23890906870365142,
      "num_tokens": 153894536.0,
      "step": 74160
    },
    {
      "entropy": 5.6331054210662845,
      "epoch": 6.727594339622642,
      "grad_norm": 1.5,
      "learning_rate": 0.00016057765972076783,
      "loss": 4.7434,
      "mean_token_accuracy": 0.23769078105688096,
      "num_tokens": 153905007.0,
      "step": 74165
    },
    {
      "entropy": 5.694492053985596,
      "epoch": 6.728047895500726,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001605498008565228,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2279347822070122,
      "num_tokens": 153915595.0,
      "step": 74170
    },
    {
      "entropy": 5.710798740386963,
      "epoch": 6.7285014513788095,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016052194435913704,
      "loss": 4.8416,
      "mean_token_accuracy": 0.23159810155630112,
      "num_tokens": 153925892.0,
      "step": 74175
    },
    {
      "entropy": 5.713548994064331,
      "epoch": 6.728955007256894,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016049409022918663,
      "loss": 4.7773,
      "mean_token_accuracy": 0.2349509760737419,
      "num_tokens": 153935483.0,
      "step": 74180
    },
    {
      "entropy": 5.641081619262695,
      "epoch": 6.729408563134978,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016046623846724755,
      "loss": 4.7638,
      "mean_token_accuracy": 0.2402131289243698,
      "num_tokens": 153945999.0,
      "step": 74185
    },
    {
      "entropy": 5.682225704193115,
      "epoch": 6.729862119013062,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001604383890738959,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23980641961097718,
      "num_tokens": 153956243.0,
      "step": 74190
    },
    {
      "entropy": 5.696243524551392,
      "epoch": 6.730315674891147,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016041054204970763,
      "loss": 4.8418,
      "mean_token_accuracy": 0.2350976884365082,
      "num_tokens": 153967209.0,
      "step": 74195
    },
    {
      "entropy": 5.692945289611816,
      "epoch": 6.730769230769231,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001603826973952585,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22467875182628633,
      "num_tokens": 153977350.0,
      "step": 74200
    },
    {
      "entropy": 5.702161884307861,
      "epoch": 6.731222786647315,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016035485511112436,
      "loss": 4.8251,
      "mean_token_accuracy": 0.22803494483232498,
      "num_tokens": 153987764.0,
      "step": 74205
    },
    {
      "entropy": 5.733908176422119,
      "epoch": 6.731676342525399,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001603270151978811,
      "loss": 4.8421,
      "mean_token_accuracy": 0.23288582414388656,
      "num_tokens": 153997872.0,
      "step": 74210
    },
    {
      "entropy": 5.7154052734375,
      "epoch": 6.732129898403484,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016029917765610435,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2298692509531975,
      "num_tokens": 154008894.0,
      "step": 74215
    },
    {
      "entropy": 5.675086450576782,
      "epoch": 6.732583454281567,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016027134248636977,
      "loss": 4.754,
      "mean_token_accuracy": 0.23547986000776291,
      "num_tokens": 154019347.0,
      "step": 74220
    },
    {
      "entropy": 5.6887585639953615,
      "epoch": 6.733037010159651,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016024350968925311,
      "loss": 4.8616,
      "mean_token_accuracy": 0.232624451816082,
      "num_tokens": 154030793.0,
      "step": 74225
    },
    {
      "entropy": 5.685548782348633,
      "epoch": 6.7334905660377355,
      "grad_norm": 1.5,
      "learning_rate": 0.00016021567926532995,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22596227526664733,
      "num_tokens": 154041976.0,
      "step": 74230
    },
    {
      "entropy": 5.746297025680542,
      "epoch": 6.73394412191582,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001601878512151758,
      "loss": 4.829,
      "mean_token_accuracy": 0.2296900525689125,
      "num_tokens": 154051890.0,
      "step": 74235
    },
    {
      "entropy": 5.764937877655029,
      "epoch": 6.734397677793904,
      "grad_norm": 1.5,
      "learning_rate": 0.0001601600255393661,
      "loss": 4.8815,
      "mean_token_accuracy": 0.23480176776647568,
      "num_tokens": 154062462.0,
      "step": 74240
    },
    {
      "entropy": 5.626000118255615,
      "epoch": 6.734851233671988,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016013220223847635,
      "loss": 4.7332,
      "mean_token_accuracy": 0.242634317278862,
      "num_tokens": 154072113.0,
      "step": 74245
    },
    {
      "entropy": 5.6602214813232425,
      "epoch": 6.735304789550073,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016010438131308188,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23157997876405717,
      "num_tokens": 154081956.0,
      "step": 74250
    },
    {
      "entropy": 5.714801263809204,
      "epoch": 6.735758345428157,
      "grad_norm": 1.734375,
      "learning_rate": 0.00016007656276375814,
      "loss": 4.8563,
      "mean_token_accuracy": 0.23599500805139542,
      "num_tokens": 154091952.0,
      "step": 74255
    },
    {
      "entropy": 5.731163072586059,
      "epoch": 6.736211901306241,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016004874659108036,
      "loss": 4.8337,
      "mean_token_accuracy": 0.2326156660914421,
      "num_tokens": 154102031.0,
      "step": 74260
    },
    {
      "entropy": 5.711922454833984,
      "epoch": 6.736665457184325,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016002093279562384,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22640259861946105,
      "num_tokens": 154112706.0,
      "step": 74265
    },
    {
      "entropy": 5.660459280014038,
      "epoch": 6.73711901306241,
      "grad_norm": 1.5,
      "learning_rate": 0.0001599931213779637,
      "loss": 4.712,
      "mean_token_accuracy": 0.23862918466329575,
      "num_tokens": 154122547.0,
      "step": 74270
    },
    {
      "entropy": 5.693757724761963,
      "epoch": 6.737572568940493,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001599653123386751,
      "loss": 4.8358,
      "mean_token_accuracy": 0.22904132902622223,
      "num_tokens": 154133248.0,
      "step": 74275
    },
    {
      "entropy": 5.706321430206299,
      "epoch": 6.738026124818577,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015993750567833322,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22574436217546462,
      "num_tokens": 154144506.0,
      "step": 74280
    },
    {
      "entropy": 5.732943725585938,
      "epoch": 6.738479680696662,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015990970139751287,
      "loss": 4.8694,
      "mean_token_accuracy": 0.23111312538385392,
      "num_tokens": 154155098.0,
      "step": 74285
    },
    {
      "entropy": 5.735673189163208,
      "epoch": 6.738933236574746,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015988189949678937,
      "loss": 4.7961,
      "mean_token_accuracy": 0.2446872264146805,
      "num_tokens": 154164494.0,
      "step": 74290
    },
    {
      "entropy": 5.6740752220153805,
      "epoch": 6.73938679245283,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001598540999767375,
      "loss": 4.8053,
      "mean_token_accuracy": 0.23597876429557801,
      "num_tokens": 154175432.0,
      "step": 74295
    },
    {
      "entropy": 5.807412195205688,
      "epoch": 6.739840348330914,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001598263028379322,
      "loss": 5.0146,
      "mean_token_accuracy": 0.220061095058918,
      "num_tokens": 154186759.0,
      "step": 74300
    },
    {
      "entropy": 5.670079851150513,
      "epoch": 6.740293904208999,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015979850808094828,
      "loss": 4.737,
      "mean_token_accuracy": 0.2416263535618782,
      "num_tokens": 154197034.0,
      "step": 74305
    },
    {
      "entropy": 5.728830480575562,
      "epoch": 6.740747460087083,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001597707157063606,
      "loss": 4.8059,
      "mean_token_accuracy": 0.23450047969818116,
      "num_tokens": 154207317.0,
      "step": 74310
    },
    {
      "entropy": 5.725682163238526,
      "epoch": 6.741201015965167,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015974292571474386,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22997768521308898,
      "num_tokens": 154217272.0,
      "step": 74315
    },
    {
      "entropy": 5.704510259628296,
      "epoch": 6.7416545718432515,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015971513810667272,
      "loss": 4.7976,
      "mean_token_accuracy": 0.23965161740779878,
      "num_tokens": 154228118.0,
      "step": 74320
    },
    {
      "entropy": 5.698211002349853,
      "epoch": 6.742108127721336,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00015968735288272195,
      "loss": 4.8014,
      "mean_token_accuracy": 0.2403249502182007,
      "num_tokens": 154237466.0,
      "step": 74325
    },
    {
      "entropy": 5.667181587219238,
      "epoch": 6.742561683599419,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015965957004346616,
      "loss": 4.7771,
      "mean_token_accuracy": 0.228840234875679,
      "num_tokens": 154248347.0,
      "step": 74330
    },
    {
      "entropy": 5.664678525924683,
      "epoch": 6.743015239477503,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015963178958947982,
      "loss": 4.7503,
      "mean_token_accuracy": 0.25221625715494156,
      "num_tokens": 154258278.0,
      "step": 74335
    },
    {
      "entropy": 5.712762355804443,
      "epoch": 6.743468795355588,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015960401152133747,
      "loss": 4.8356,
      "mean_token_accuracy": 0.23488451093435286,
      "num_tokens": 154268671.0,
      "step": 74340
    },
    {
      "entropy": 5.65318717956543,
      "epoch": 6.743922351233672,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001595762358396136,
      "loss": 4.8459,
      "mean_token_accuracy": 0.23421700298786163,
      "num_tokens": 154279951.0,
      "step": 74345
    },
    {
      "entropy": 5.780076408386231,
      "epoch": 6.744375907111756,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015954846254488252,
      "loss": 4.9688,
      "mean_token_accuracy": 0.22566401809453965,
      "num_tokens": 154291782.0,
      "step": 74350
    },
    {
      "entropy": 5.6398597240448,
      "epoch": 6.74482946298984,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015952069163771877,
      "loss": 4.7239,
      "mean_token_accuracy": 0.23939345628023148,
      "num_tokens": 154301524.0,
      "step": 74355
    },
    {
      "entropy": 5.669930267333984,
      "epoch": 6.745283018867925,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015949292311869646,
      "loss": 4.7665,
      "mean_token_accuracy": 0.24724904596805572,
      "num_tokens": 154312076.0,
      "step": 74360
    },
    {
      "entropy": 5.683302211761474,
      "epoch": 6.745736574746009,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015946515698838992,
      "loss": 4.8386,
      "mean_token_accuracy": 0.2326451748609543,
      "num_tokens": 154323473.0,
      "step": 74365
    },
    {
      "entropy": 5.672780179977417,
      "epoch": 6.746190130624093,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015943739324737342,
      "loss": 4.8258,
      "mean_token_accuracy": 0.24006792157888412,
      "num_tokens": 154333235.0,
      "step": 74370
    },
    {
      "entropy": 5.664455699920654,
      "epoch": 6.746643686502177,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015940963189622104,
      "loss": 4.7849,
      "mean_token_accuracy": 0.2331676885485649,
      "num_tokens": 154343837.0,
      "step": 74375
    },
    {
      "entropy": 5.662339353561402,
      "epoch": 6.747097242380261,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001593818729355069,
      "loss": 4.7798,
      "mean_token_accuracy": 0.2372131749987602,
      "num_tokens": 154353955.0,
      "step": 74380
    },
    {
      "entropy": 5.753809022903442,
      "epoch": 6.747550798258345,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015935411636580513,
      "loss": 4.8798,
      "mean_token_accuracy": 0.23000428378582,
      "num_tokens": 154364876.0,
      "step": 74385
    },
    {
      "entropy": 5.685974550247193,
      "epoch": 6.748004354136429,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001593263621876897,
      "loss": 4.8081,
      "mean_token_accuracy": 0.23133932948112487,
      "num_tokens": 154374178.0,
      "step": 74390
    },
    {
      "entropy": 5.732316303253174,
      "epoch": 6.748457910014514,
      "grad_norm": 1.5,
      "learning_rate": 0.0001592986104017346,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23252707421779634,
      "num_tokens": 154383767.0,
      "step": 74395
    },
    {
      "entropy": 5.7058189868927,
      "epoch": 6.748911465892598,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015927086100851368,
      "loss": 4.8495,
      "mean_token_accuracy": 0.23099205046892166,
      "num_tokens": 154393460.0,
      "step": 74400
    },
    {
      "entropy": 5.7070986270904545,
      "epoch": 6.749365021770682,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001592431140086009,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23216235935688018,
      "num_tokens": 154403337.0,
      "step": 74405
    },
    {
      "entropy": 5.748113536834717,
      "epoch": 6.7498185776487665,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015921536940257002,
      "loss": 4.8501,
      "mean_token_accuracy": 0.23128001242876053,
      "num_tokens": 154414383.0,
      "step": 74410
    },
    {
      "entropy": 5.735575199127197,
      "epoch": 6.750272133526851,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015918762719099468,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22651145309209825,
      "num_tokens": 154424729.0,
      "step": 74415
    },
    {
      "entropy": 5.814986753463745,
      "epoch": 6.750725689404935,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001591598873744488,
      "loss": 5.0242,
      "mean_token_accuracy": 0.21515374779701232,
      "num_tokens": 154436341.0,
      "step": 74420
    },
    {
      "entropy": 5.7127738952636715,
      "epoch": 6.751179245283019,
      "grad_norm": 1.34375,
      "learning_rate": 0.000159132149953506,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22951442152261733,
      "num_tokens": 154447569.0,
      "step": 74425
    },
    {
      "entropy": 5.735399007797241,
      "epoch": 6.751632801161103,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015910441492873986,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2301063671708107,
      "num_tokens": 154457393.0,
      "step": 74430
    },
    {
      "entropy": 5.716166162490845,
      "epoch": 6.752086357039187,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00015907668230072397,
      "loss": 4.8028,
      "mean_token_accuracy": 0.2394646480679512,
      "num_tokens": 154467708.0,
      "step": 74435
    },
    {
      "entropy": 5.673102188110351,
      "epoch": 6.752539912917271,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015904895207003182,
      "loss": 4.7381,
      "mean_token_accuracy": 0.24087364077568055,
      "num_tokens": 154477048.0,
      "step": 74440
    },
    {
      "entropy": 5.669525480270385,
      "epoch": 6.7529934687953554,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015902122423723687,
      "loss": 4.7992,
      "mean_token_accuracy": 0.24582487493753433,
      "num_tokens": 154487118.0,
      "step": 74445
    },
    {
      "entropy": 5.723471784591675,
      "epoch": 6.75344702467344,
      "grad_norm": 1.59375,
      "learning_rate": 0.00015899349880291248,
      "loss": 4.8194,
      "mean_token_accuracy": 0.23753952533006667,
      "num_tokens": 154496235.0,
      "step": 74450
    },
    {
      "entropy": 5.681961870193481,
      "epoch": 6.753900580551524,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001589657757676322,
      "loss": 4.7578,
      "mean_token_accuracy": 0.2376547709107399,
      "num_tokens": 154506701.0,
      "step": 74455
    },
    {
      "entropy": 5.637529516220093,
      "epoch": 6.754354136429608,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015893805513196924,
      "loss": 4.7363,
      "mean_token_accuracy": 0.23961506634950638,
      "num_tokens": 154516565.0,
      "step": 74460
    },
    {
      "entropy": 5.693013095855713,
      "epoch": 6.7548076923076925,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015891033689649687,
      "loss": 4.8129,
      "mean_token_accuracy": 0.23376789540052414,
      "num_tokens": 154526435.0,
      "step": 74465
    },
    {
      "entropy": 5.73045539855957,
      "epoch": 6.755261248185777,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001588826210617883,
      "loss": 4.8852,
      "mean_token_accuracy": 0.23154502213001252,
      "num_tokens": 154537376.0,
      "step": 74470
    },
    {
      "entropy": 5.727739048004151,
      "epoch": 6.755714804063861,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015885490762841673,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22353834360837938,
      "num_tokens": 154548193.0,
      "step": 74475
    },
    {
      "entropy": 5.703621482849121,
      "epoch": 6.756168359941945,
      "grad_norm": 1.609375,
      "learning_rate": 0.00015882719659695527,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23347126096487045,
      "num_tokens": 154557601.0,
      "step": 74480
    },
    {
      "entropy": 5.738251733779907,
      "epoch": 6.756621915820029,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015879948796797692,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2353489652276039,
      "num_tokens": 154567240.0,
      "step": 74485
    },
    {
      "entropy": 5.631132698059082,
      "epoch": 6.757075471698113,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001587717817420548,
      "loss": 4.7533,
      "mean_token_accuracy": 0.23809243887662887,
      "num_tokens": 154577610.0,
      "step": 74490
    },
    {
      "entropy": 5.6984892845153805,
      "epoch": 6.757529027576197,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015874407791976192,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23712996542453765,
      "num_tokens": 154587396.0,
      "step": 74495
    },
    {
      "entropy": 5.723380613327026,
      "epoch": 6.7579825834542815,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001587163765016711,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2233245551586151,
      "num_tokens": 154597891.0,
      "step": 74500
    },
    {
      "entropy": 5.688806962966919,
      "epoch": 6.758436139332366,
      "grad_norm": 1.671875,
      "learning_rate": 0.00015868867748835525,
      "loss": 4.7552,
      "mean_token_accuracy": 0.2439547896385193,
      "num_tokens": 154607702.0,
      "step": 74505
    },
    {
      "entropy": 5.625900554656982,
      "epoch": 6.75888969521045,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001586609808803872,
      "loss": 4.7517,
      "mean_token_accuracy": 0.24588073939085006,
      "num_tokens": 154618806.0,
      "step": 74510
    },
    {
      "entropy": 5.695935535430908,
      "epoch": 6.759343251088534,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015863328667833971,
      "loss": 4.8318,
      "mean_token_accuracy": 0.23697419464588165,
      "num_tokens": 154630139.0,
      "step": 74515
    },
    {
      "entropy": 5.710346508026123,
      "epoch": 6.7597968069666186,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015860559488278542,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23714868873357772,
      "num_tokens": 154639973.0,
      "step": 74520
    },
    {
      "entropy": 5.743674850463867,
      "epoch": 6.760250362844703,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015857790549429716,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22715015560388566,
      "num_tokens": 154650666.0,
      "step": 74525
    },
    {
      "entropy": 5.634935903549194,
      "epoch": 6.760703918722786,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015855021851344747,
      "loss": 4.7981,
      "mean_token_accuracy": 0.23857494294643403,
      "num_tokens": 154661410.0,
      "step": 74530
    },
    {
      "entropy": 5.6549054145812985,
      "epoch": 6.7611574746008705,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015852253394080896,
      "loss": 4.7786,
      "mean_token_accuracy": 0.24481942504644394,
      "num_tokens": 154670961.0,
      "step": 74535
    },
    {
      "entropy": 5.679853773117065,
      "epoch": 6.761611030478955,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015849485177695414,
      "loss": 4.8237,
      "mean_token_accuracy": 0.22967022210359572,
      "num_tokens": 154681184.0,
      "step": 74540
    },
    {
      "entropy": 5.706052827835083,
      "epoch": 6.762064586357039,
      "grad_norm": 1.609375,
      "learning_rate": 0.00015846717202245544,
      "loss": 4.8065,
      "mean_token_accuracy": 0.24308725595474243,
      "num_tokens": 154692572.0,
      "step": 74545
    },
    {
      "entropy": 5.722026300430298,
      "epoch": 6.762518142235123,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015843949467788526,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2309614509344101,
      "num_tokens": 154702629.0,
      "step": 74550
    },
    {
      "entropy": 5.680447912216186,
      "epoch": 6.7629716981132075,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001584118197438161,
      "loss": 4.7178,
      "mean_token_accuracy": 0.23277585059404374,
      "num_tokens": 154713708.0,
      "step": 74555
    },
    {
      "entropy": 5.706203985214233,
      "epoch": 6.763425253991292,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015838414722082022,
      "loss": 4.7717,
      "mean_token_accuracy": 0.23765295445919038,
      "num_tokens": 154723259.0,
      "step": 74560
    },
    {
      "entropy": 5.783371114730835,
      "epoch": 6.763878809869376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015835647710946992,
      "loss": 4.8868,
      "mean_token_accuracy": 0.22270139753818513,
      "num_tokens": 154734384.0,
      "step": 74565
    },
    {
      "entropy": 5.681018257141114,
      "epoch": 6.76433236574746,
      "grad_norm": 1.375,
      "learning_rate": 0.0001583288094103374,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23469282686710358,
      "num_tokens": 154744317.0,
      "step": 74570
    },
    {
      "entropy": 5.6193763256073,
      "epoch": 6.764785921625545,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001583011441239948,
      "loss": 4.76,
      "mean_token_accuracy": 0.24116779118776321,
      "num_tokens": 154754819.0,
      "step": 74575
    },
    {
      "entropy": 5.662204313278198,
      "epoch": 6.765239477503629,
      "grad_norm": 1.625,
      "learning_rate": 0.00015827348125101428,
      "loss": 4.729,
      "mean_token_accuracy": 0.23721338361501693,
      "num_tokens": 154764837.0,
      "step": 74580
    },
    {
      "entropy": 5.682091188430786,
      "epoch": 6.765693033381712,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015824582079196787,
      "loss": 4.7775,
      "mean_token_accuracy": 0.24324487298727035,
      "num_tokens": 154774278.0,
      "step": 74585
    },
    {
      "entropy": 5.727855491638183,
      "epoch": 6.7661465892597965,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015821816274742769,
      "loss": 4.8843,
      "mean_token_accuracy": 0.2273190438747406,
      "num_tokens": 154785647.0,
      "step": 74590
    },
    {
      "entropy": 5.69824252128601,
      "epoch": 6.766600145137881,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015819050711796565,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24293381422758104,
      "num_tokens": 154795614.0,
      "step": 74595
    },
    {
      "entropy": 5.694895839691162,
      "epoch": 6.767053701015965,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015816285390415368,
      "loss": 4.7989,
      "mean_token_accuracy": 0.24209104180336,
      "num_tokens": 154805972.0,
      "step": 74600
    },
    {
      "entropy": 5.7037862777709964,
      "epoch": 6.767507256894049,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015813520310656372,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2350797936320305,
      "num_tokens": 154817185.0,
      "step": 74605
    },
    {
      "entropy": 5.691404104232788,
      "epoch": 6.767960812772134,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001581075547257675,
      "loss": 4.8674,
      "mean_token_accuracy": 0.23240315020084382,
      "num_tokens": 154827530.0,
      "step": 74610
    },
    {
      "entropy": 5.635840606689453,
      "epoch": 6.768414368650218,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015807990876233685,
      "loss": 4.7945,
      "mean_token_accuracy": 0.23538612127304076,
      "num_tokens": 154838166.0,
      "step": 74615
    },
    {
      "entropy": 5.664965963363647,
      "epoch": 6.768867924528302,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015805226521684338,
      "loss": 4.7734,
      "mean_token_accuracy": 0.2397976413369179,
      "num_tokens": 154848317.0,
      "step": 74620
    },
    {
      "entropy": 5.6873180866241455,
      "epoch": 6.769321480406386,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015802462408985896,
      "loss": 4.8112,
      "mean_token_accuracy": 0.2394749253988266,
      "num_tokens": 154859070.0,
      "step": 74625
    },
    {
      "entropy": 5.7260603427886965,
      "epoch": 6.76977503628447,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001579969853819551,
      "loss": 4.8569,
      "mean_token_accuracy": 0.22214989364147186,
      "num_tokens": 154869412.0,
      "step": 74630
    },
    {
      "entropy": 5.710358667373657,
      "epoch": 6.770228592162555,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00015796934909370348,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22619009912014007,
      "num_tokens": 154879555.0,
      "step": 74635
    },
    {
      "entropy": 5.722429466247559,
      "epoch": 6.770682148040638,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015794171522567546,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22943171560764314,
      "num_tokens": 154889781.0,
      "step": 74640
    },
    {
      "entropy": 5.792335271835327,
      "epoch": 6.7711357039187225,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015791408377844262,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21651365906000136,
      "num_tokens": 154899934.0,
      "step": 74645
    },
    {
      "entropy": 5.66039547920227,
      "epoch": 6.771589259796807,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001578864547525764,
      "loss": 4.853,
      "mean_token_accuracy": 0.22853411883115768,
      "num_tokens": 154911728.0,
      "step": 74650
    },
    {
      "entropy": 5.746025228500367,
      "epoch": 6.772042815674891,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015785882814864805,
      "loss": 4.886,
      "mean_token_accuracy": 0.22570523768663406,
      "num_tokens": 154922782.0,
      "step": 74655
    },
    {
      "entropy": 5.6574540615081785,
      "epoch": 6.772496371552975,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015783120396722903,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2436859279870987,
      "num_tokens": 154932408.0,
      "step": 74660
    },
    {
      "entropy": 5.685013294219971,
      "epoch": 6.77294992743106,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015780358220889056,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22790753692388535,
      "num_tokens": 154942432.0,
      "step": 74665
    },
    {
      "entropy": 5.636389541625976,
      "epoch": 6.773403483309144,
      "grad_norm": 1.5625,
      "learning_rate": 0.00015777596287420393,
      "loss": 4.7944,
      "mean_token_accuracy": 0.24161895662546157,
      "num_tokens": 154953610.0,
      "step": 74670
    },
    {
      "entropy": 5.685267972946167,
      "epoch": 6.773857039187228,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015774834596374022,
      "loss": 4.7694,
      "mean_token_accuracy": 0.24160154461860656,
      "num_tokens": 154963919.0,
      "step": 74675
    },
    {
      "entropy": 5.6866843700408936,
      "epoch": 6.774310595065312,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015772073147807063,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2288990020751953,
      "num_tokens": 154975298.0,
      "step": 74680
    },
    {
      "entropy": 5.770553493499756,
      "epoch": 6.774764150943396,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015769311941776616,
      "loss": 4.9048,
      "mean_token_accuracy": 0.23756162822246552,
      "num_tokens": 154985615.0,
      "step": 74685
    },
    {
      "entropy": 5.680081701278686,
      "epoch": 6.77521770682148,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015766550978339784,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2428151324391365,
      "num_tokens": 154996224.0,
      "step": 74690
    },
    {
      "entropy": 5.761427497863769,
      "epoch": 6.775671262699564,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015763790257553674,
      "loss": 4.8954,
      "mean_token_accuracy": 0.2229863688349724,
      "num_tokens": 155006615.0,
      "step": 74695
    },
    {
      "entropy": 5.67316837310791,
      "epoch": 6.776124818577649,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015761029779475368,
      "loss": 4.7458,
      "mean_token_accuracy": 0.2376384347677231,
      "num_tokens": 155017636.0,
      "step": 74700
    },
    {
      "entropy": 5.679613256454468,
      "epoch": 6.776578374455733,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001575826954416196,
      "loss": 4.8092,
      "mean_token_accuracy": 0.23214428573846818,
      "num_tokens": 155026942.0,
      "step": 74705
    },
    {
      "entropy": 5.681477117538452,
      "epoch": 6.777031930333817,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015755509551670534,
      "loss": 4.7821,
      "mean_token_accuracy": 0.23714470565319062,
      "num_tokens": 155037506.0,
      "step": 74710
    },
    {
      "entropy": 5.7212718486785885,
      "epoch": 6.777485486211901,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015752749802058159,
      "loss": 4.8145,
      "mean_token_accuracy": 0.23412743210792542,
      "num_tokens": 155047830.0,
      "step": 74715
    },
    {
      "entropy": 5.724223804473877,
      "epoch": 6.777939042089986,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001574999029538191,
      "loss": 4.8185,
      "mean_token_accuracy": 0.24008594453334808,
      "num_tokens": 155057397.0,
      "step": 74720
    },
    {
      "entropy": 5.723435878753662,
      "epoch": 6.77839259796807,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00015747231031698852,
      "loss": 4.8089,
      "mean_token_accuracy": 0.23400587290525438,
      "num_tokens": 155067448.0,
      "step": 74725
    },
    {
      "entropy": 5.744628429412842,
      "epoch": 6.778846153846154,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00015744472011066057,
      "loss": 4.8781,
      "mean_token_accuracy": 0.22964273095130922,
      "num_tokens": 155077852.0,
      "step": 74730
    },
    {
      "entropy": 5.632175731658935,
      "epoch": 6.7792997097242385,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015741713233540576,
      "loss": 4.7877,
      "mean_token_accuracy": 0.23912111967802047,
      "num_tokens": 155088498.0,
      "step": 74735
    },
    {
      "entropy": 5.725306987762451,
      "epoch": 6.779753265602322,
      "grad_norm": 1.578125,
      "learning_rate": 0.00015738954699179465,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22772260159254074,
      "num_tokens": 155099715.0,
      "step": 74740
    },
    {
      "entropy": 5.633829355239868,
      "epoch": 6.780206821480406,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015736196408039767,
      "loss": 4.785,
      "mean_token_accuracy": 0.2338012784719467,
      "num_tokens": 155110245.0,
      "step": 74745
    },
    {
      "entropy": 5.730634927749634,
      "epoch": 6.78066037735849,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015733438360178523,
      "loss": 4.8394,
      "mean_token_accuracy": 0.23362578749656676,
      "num_tokens": 155120905.0,
      "step": 74750
    },
    {
      "entropy": 5.712240695953369,
      "epoch": 6.781113933236575,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015730680555652767,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22887220978736877,
      "num_tokens": 155132586.0,
      "step": 74755
    },
    {
      "entropy": 5.674384832382202,
      "epoch": 6.781567489114659,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015727922994519542,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2277405872941017,
      "num_tokens": 155143197.0,
      "step": 74760
    },
    {
      "entropy": 5.769257688522339,
      "epoch": 6.782021044992743,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015725165676835868,
      "loss": 4.916,
      "mean_token_accuracy": 0.22349632084369658,
      "num_tokens": 155153928.0,
      "step": 74765
    },
    {
      "entropy": 5.7202962875366214,
      "epoch": 6.782474600870827,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015722408602658773,
      "loss": 4.8248,
      "mean_token_accuracy": 0.23128950148820876,
      "num_tokens": 155164177.0,
      "step": 74770
    },
    {
      "entropy": 5.739596223831176,
      "epoch": 6.782928156748912,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015719651772045273,
      "loss": 4.8588,
      "mean_token_accuracy": 0.23695608228445053,
      "num_tokens": 155175159.0,
      "step": 74775
    },
    {
      "entropy": 5.720144033432007,
      "epoch": 6.783381712626996,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001571689518505237,
      "loss": 4.8479,
      "mean_token_accuracy": 0.2286031499505043,
      "num_tokens": 155184704.0,
      "step": 74780
    },
    {
      "entropy": 5.794011020660401,
      "epoch": 6.783835268505079,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001571413884173707,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22933673709630967,
      "num_tokens": 155194747.0,
      "step": 74785
    },
    {
      "entropy": 5.71543254852295,
      "epoch": 6.7842888243831645,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001571138274215639,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2254225954413414,
      "num_tokens": 155205207.0,
      "step": 74790
    },
    {
      "entropy": 5.723735761642456,
      "epoch": 6.784742380261248,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015708626886367315,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23240772485733033,
      "num_tokens": 155215783.0,
      "step": 74795
    },
    {
      "entropy": 5.698229885101318,
      "epoch": 6.785195936139332,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015705871274426848,
      "loss": 4.845,
      "mean_token_accuracy": 0.23933880925178527,
      "num_tokens": 155226518.0,
      "step": 74800
    },
    {
      "entropy": 5.744544315338135,
      "epoch": 6.785649492017416,
      "grad_norm": 1.5625,
      "learning_rate": 0.00015703115906391964,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22866723388433458,
      "num_tokens": 155235841.0,
      "step": 74805
    },
    {
      "entropy": 5.704317331314087,
      "epoch": 6.786103047895501,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001570036078231965,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23373541831970215,
      "num_tokens": 155245906.0,
      "step": 74810
    },
    {
      "entropy": 5.733270311355591,
      "epoch": 6.786556603773585,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015697605902266883,
      "loss": 4.807,
      "mean_token_accuracy": 0.23272667378187178,
      "num_tokens": 155256704.0,
      "step": 74815
    },
    {
      "entropy": 5.681692266464234,
      "epoch": 6.787010159651669,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015694851266290623,
      "loss": 4.8554,
      "mean_token_accuracy": 0.23243319541215895,
      "num_tokens": 155266984.0,
      "step": 74820
    },
    {
      "entropy": 5.626868677139282,
      "epoch": 6.7874637155297535,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015692096874447855,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2398640349507332,
      "num_tokens": 155277783.0,
      "step": 74825
    },
    {
      "entropy": 5.6817371368408205,
      "epoch": 6.787917271407838,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015689342726795535,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23150929659605027,
      "num_tokens": 155288947.0,
      "step": 74830
    },
    {
      "entropy": 5.724450159072876,
      "epoch": 6.788370827285922,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001568658882339062,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2297436073422432,
      "num_tokens": 155299113.0,
      "step": 74835
    },
    {
      "entropy": 5.743064451217651,
      "epoch": 6.788824383164005,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001568383516429005,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23217587769031525,
      "num_tokens": 155310273.0,
      "step": 74840
    },
    {
      "entropy": 5.651137256622315,
      "epoch": 6.78927793904209,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015681081749550785,
      "loss": 4.7274,
      "mean_token_accuracy": 0.23544905930757523,
      "num_tokens": 155321346.0,
      "step": 74845
    },
    {
      "entropy": 5.716941976547242,
      "epoch": 6.789731494920174,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015678328579229754,
      "loss": 4.8711,
      "mean_token_accuracy": 0.22843815982341767,
      "num_tokens": 155332395.0,
      "step": 74850
    },
    {
      "entropy": 5.7082456111907955,
      "epoch": 6.790185050798258,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015675575653383906,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22809910476207734,
      "num_tokens": 155342467.0,
      "step": 74855
    },
    {
      "entropy": 5.6900200843811035,
      "epoch": 6.7906386066763424,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0001567282297207017,
      "loss": 4.822,
      "mean_token_accuracy": 0.23543685525655747,
      "num_tokens": 155352437.0,
      "step": 74860
    },
    {
      "entropy": 5.69362621307373,
      "epoch": 6.791092162554427,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015670070535345467,
      "loss": 4.7903,
      "mean_token_accuracy": 0.2393799141049385,
      "num_tokens": 155361894.0,
      "step": 74865
    },
    {
      "entropy": 5.693908786773681,
      "epoch": 6.791545718432511,
      "grad_norm": 1.578125,
      "learning_rate": 0.00015667318343266716,
      "loss": 4.855,
      "mean_token_accuracy": 0.22810473442077636,
      "num_tokens": 155373057.0,
      "step": 74870
    },
    {
      "entropy": 5.710803651809693,
      "epoch": 6.791999274310595,
      "grad_norm": 1.609375,
      "learning_rate": 0.00015664566395890846,
      "loss": 4.869,
      "mean_token_accuracy": 0.23382582366466523,
      "num_tokens": 155383233.0,
      "step": 74875
    },
    {
      "entropy": 5.677272319793701,
      "epoch": 6.7924528301886795,
      "grad_norm": 1.5,
      "learning_rate": 0.00015661814693274754,
      "loss": 4.8103,
      "mean_token_accuracy": 0.2318767264485359,
      "num_tokens": 155393256.0,
      "step": 74880
    },
    {
      "entropy": 5.6729011058807375,
      "epoch": 6.792906386066764,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001565906323547534,
      "loss": 4.8092,
      "mean_token_accuracy": 0.23668724149465561,
      "num_tokens": 155403197.0,
      "step": 74885
    },
    {
      "entropy": 5.711571979522705,
      "epoch": 6.793359941944848,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015656312022549527,
      "loss": 4.8679,
      "mean_token_accuracy": 0.2264749974012375,
      "num_tokens": 155413995.0,
      "step": 74890
    },
    {
      "entropy": 5.772419738769531,
      "epoch": 6.793813497822931,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015653561054554198,
      "loss": 4.9171,
      "mean_token_accuracy": 0.23169272094964982,
      "num_tokens": 155424806.0,
      "step": 74895
    },
    {
      "entropy": 5.752104234695435,
      "epoch": 6.794267053701016,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001565081033154625,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2264307588338852,
      "num_tokens": 155435813.0,
      "step": 74900
    },
    {
      "entropy": 5.786309909820557,
      "epoch": 6.7947206095791,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00015648059853582558,
      "loss": 4.8925,
      "mean_token_accuracy": 0.23357683122158052,
      "num_tokens": 155446159.0,
      "step": 74905
    },
    {
      "entropy": 5.7126898765563965,
      "epoch": 6.795174165457184,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001564530962072001,
      "loss": 4.8212,
      "mean_token_accuracy": 0.23068250715732574,
      "num_tokens": 155455773.0,
      "step": 74910
    },
    {
      "entropy": 5.74611988067627,
      "epoch": 6.7956277213352685,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001564255963301548,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2282474398612976,
      "num_tokens": 155466650.0,
      "step": 74915
    },
    {
      "entropy": 5.738385772705078,
      "epoch": 6.796081277213353,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015639809890525833,
      "loss": 4.8306,
      "mean_token_accuracy": 0.22874060720205308,
      "num_tokens": 155478676.0,
      "step": 74920
    },
    {
      "entropy": 5.644596958160401,
      "epoch": 6.796534833091437,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015637060393307946,
      "loss": 4.7846,
      "mean_token_accuracy": 0.24128990471363068,
      "num_tokens": 155489194.0,
      "step": 74925
    },
    {
      "entropy": 5.687523937225341,
      "epoch": 6.796988388969521,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015634311141418672,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23187336027622224,
      "num_tokens": 155500057.0,
      "step": 74930
    },
    {
      "entropy": 5.721893405914306,
      "epoch": 6.7974419448476056,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015631562134914867,
      "loss": 4.8154,
      "mean_token_accuracy": 0.2321080133318901,
      "num_tokens": 155510155.0,
      "step": 74935
    },
    {
      "entropy": 5.722184133529663,
      "epoch": 6.797895500725689,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015628813373853384,
      "loss": 4.7907,
      "mean_token_accuracy": 0.23295590728521348,
      "num_tokens": 155520153.0,
      "step": 74940
    },
    {
      "entropy": 5.702037572860718,
      "epoch": 6.798349056603773,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015626064858291066,
      "loss": 4.7711,
      "mean_token_accuracy": 0.22939137667417525,
      "num_tokens": 155529595.0,
      "step": 74945
    },
    {
      "entropy": 5.644517803192139,
      "epoch": 6.7988026124818575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001562331658828475,
      "loss": 4.7044,
      "mean_token_accuracy": 0.233280311524868,
      "num_tokens": 155540417.0,
      "step": 74950
    },
    {
      "entropy": 5.725898361206054,
      "epoch": 6.799256168359942,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015620568563891267,
      "loss": 4.8251,
      "mean_token_accuracy": 0.24143054485321044,
      "num_tokens": 155551219.0,
      "step": 74955
    },
    {
      "entropy": 5.634184169769287,
      "epoch": 6.799709724238026,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001561782078516746,
      "loss": 4.7192,
      "mean_token_accuracy": 0.24560523629188538,
      "num_tokens": 155560819.0,
      "step": 74960
    },
    {
      "entropy": 5.6422820568084715,
      "epoch": 6.80016328011611,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015615073252170147,
      "loss": 4.7939,
      "mean_token_accuracy": 0.23348014205694198,
      "num_tokens": 155571584.0,
      "step": 74965
    },
    {
      "entropy": 5.703877925872803,
      "epoch": 6.8006168359941945,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015612325964956152,
      "loss": 4.7863,
      "mean_token_accuracy": 0.2376110389828682,
      "num_tokens": 155581866.0,
      "step": 74970
    },
    {
      "entropy": 5.673033809661865,
      "epoch": 6.801070391872279,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001560957892358228,
      "loss": 4.7951,
      "mean_token_accuracy": 0.23914095610380173,
      "num_tokens": 155592133.0,
      "step": 74975
    },
    {
      "entropy": 5.749809885025025,
      "epoch": 6.801523947750363,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015606832128105348,
      "loss": 4.8212,
      "mean_token_accuracy": 0.23397316187620162,
      "num_tokens": 155601689.0,
      "step": 74980
    },
    {
      "entropy": 5.715540218353271,
      "epoch": 6.801977503628447,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015604085578582158,
      "loss": 4.8345,
      "mean_token_accuracy": 0.22911344319581986,
      "num_tokens": 155611908.0,
      "step": 74985
    },
    {
      "entropy": 5.657967567443848,
      "epoch": 6.802431059506532,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015601339275069504,
      "loss": 4.8682,
      "mean_token_accuracy": 0.23022680282592772,
      "num_tokens": 155623103.0,
      "step": 74990
    },
    {
      "entropy": 5.668922233581543,
      "epoch": 6.802884615384615,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001559859321762419,
      "loss": 4.8236,
      "mean_token_accuracy": 0.23551317304372787,
      "num_tokens": 155633350.0,
      "step": 74995
    },
    {
      "entropy": 5.688816261291504,
      "epoch": 6.803338171262699,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00015595847406303006,
      "loss": 4.7878,
      "mean_token_accuracy": 0.23920380622148513,
      "num_tokens": 155643264.0,
      "step": 75000
    },
    {
      "epoch": 6.803338171262699,
      "eval_entropy": 5.420817721914921,
      "eval_loss": 4.921440601348877,
      "eval_mean_token_accuracy": 0.23651875985525042,
      "eval_num_tokens": 155643264.0,
      "eval_runtime": 20.7215,
      "eval_samples_per_second": 1706.444,
      "eval_steps_per_second": 213.306,
      "step": 75000
    },
    {
      "entropy": 5.708639001846313,
      "epoch": 6.8037917271407835,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015593101841162725,
      "loss": 4.8261,
      "mean_token_accuracy": 0.23903507888317108,
      "num_tokens": 155652602.0,
      "step": 75005
    },
    {
      "entropy": 5.648255062103272,
      "epoch": 6.804245283018868,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015590356522260135,
      "loss": 4.7914,
      "mean_token_accuracy": 0.23777125179767608,
      "num_tokens": 155662954.0,
      "step": 75010
    },
    {
      "entropy": 5.7629907608032225,
      "epoch": 6.804698838896952,
      "grad_norm": 1.375,
      "learning_rate": 0.00015587611449652006,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2320044070482254,
      "num_tokens": 155674223.0,
      "step": 75015
    },
    {
      "entropy": 5.7992517948150635,
      "epoch": 6.805152394775036,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001558486662339511,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2275587484240532,
      "num_tokens": 155684909.0,
      "step": 75020
    },
    {
      "entropy": 5.746281242370605,
      "epoch": 6.805605950653121,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015582122043546198,
      "loss": 4.8671,
      "mean_token_accuracy": 0.23856054842472077,
      "num_tokens": 155695858.0,
      "step": 75025
    },
    {
      "entropy": 5.760446548461914,
      "epoch": 6.806059506531205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001557937771016205,
      "loss": 4.8579,
      "mean_token_accuracy": 0.230575992166996,
      "num_tokens": 155705952.0,
      "step": 75030
    },
    {
      "entropy": 5.731006336212158,
      "epoch": 6.806513062409289,
      "grad_norm": 1.375,
      "learning_rate": 0.00015576633623299408,
      "loss": 4.8129,
      "mean_token_accuracy": 0.23773254901170732,
      "num_tokens": 155715727.0,
      "step": 75035
    },
    {
      "entropy": 5.738781595230103,
      "epoch": 6.806966618287373,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015573889783015022,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2331650972366333,
      "num_tokens": 155725754.0,
      "step": 75040
    },
    {
      "entropy": 5.683665037155151,
      "epoch": 6.807420174165458,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015571146189365633,
      "loss": 4.7844,
      "mean_token_accuracy": 0.23978239446878433,
      "num_tokens": 155735643.0,
      "step": 75045
    },
    {
      "entropy": 5.685592412948608,
      "epoch": 6.807873730043541,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001556840284240798,
      "loss": 4.8044,
      "mean_token_accuracy": 0.23026364147663117,
      "num_tokens": 155746367.0,
      "step": 75050
    },
    {
      "entropy": 5.677387666702271,
      "epoch": 6.808327285921625,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001556565974219879,
      "loss": 4.7567,
      "mean_token_accuracy": 0.23718318492174148,
      "num_tokens": 155755849.0,
      "step": 75055
    },
    {
      "entropy": 5.663999938964844,
      "epoch": 6.8087808417997095,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015562916888794807,
      "loss": 4.8077,
      "mean_token_accuracy": 0.22933031767606735,
      "num_tokens": 155766643.0,
      "step": 75060
    },
    {
      "entropy": 5.690534162521362,
      "epoch": 6.809234397677794,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015560174282252742,
      "loss": 4.7939,
      "mean_token_accuracy": 0.23289039433002473,
      "num_tokens": 155776737.0,
      "step": 75065
    },
    {
      "entropy": 5.7105584144592285,
      "epoch": 6.809687953555878,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015557431922629317,
      "loss": 4.8072,
      "mean_token_accuracy": 0.2330017492175102,
      "num_tokens": 155786446.0,
      "step": 75070
    },
    {
      "entropy": 5.690123844146728,
      "epoch": 6.810141509433962,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015554689809981243,
      "loss": 4.7675,
      "mean_token_accuracy": 0.23297080993652344,
      "num_tokens": 155796118.0,
      "step": 75075
    },
    {
      "entropy": 5.6884297847747805,
      "epoch": 6.810595065312047,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015551947944365227,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2278666615486145,
      "num_tokens": 155806758.0,
      "step": 75080
    },
    {
      "entropy": 5.62774691581726,
      "epoch": 6.811048621190131,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015549206325837973,
      "loss": 4.7267,
      "mean_token_accuracy": 0.24223471134901048,
      "num_tokens": 155816510.0,
      "step": 75085
    },
    {
      "entropy": 5.639526414871216,
      "epoch": 6.811502177068215,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015546464954456168,
      "loss": 4.747,
      "mean_token_accuracy": 0.2469678983092308,
      "num_tokens": 155826124.0,
      "step": 75090
    },
    {
      "entropy": 5.687062168121338,
      "epoch": 6.8119557329462985,
      "grad_norm": 1.5,
      "learning_rate": 0.00015543723830276524,
      "loss": 4.7979,
      "mean_token_accuracy": 0.24077899158000945,
      "num_tokens": 155836151.0,
      "step": 75095
    },
    {
      "entropy": 5.6272759437561035,
      "epoch": 6.812409288824383,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015540982953355715,
      "loss": 4.8164,
      "mean_token_accuracy": 0.2321694537997246,
      "num_tokens": 155847032.0,
      "step": 75100
    },
    {
      "entropy": 5.7035094738006595,
      "epoch": 6.812862844702467,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015538242323750429,
      "loss": 4.8426,
      "mean_token_accuracy": 0.23265575766563415,
      "num_tokens": 155857459.0,
      "step": 75105
    },
    {
      "entropy": 5.694974279403686,
      "epoch": 6.813316400580551,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015535501941517341,
      "loss": 4.7511,
      "mean_token_accuracy": 0.23738006204366685,
      "num_tokens": 155867671.0,
      "step": 75110
    },
    {
      "entropy": 5.7538525581359865,
      "epoch": 6.813769956458636,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001553276180671312,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2244866147637367,
      "num_tokens": 155878593.0,
      "step": 75115
    },
    {
      "entropy": 5.707712697982788,
      "epoch": 6.81422351233672,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015530021919394433,
      "loss": 4.8196,
      "mean_token_accuracy": 0.241927932202816,
      "num_tokens": 155888437.0,
      "step": 75120
    },
    {
      "entropy": 5.650154304504395,
      "epoch": 6.814677068214804,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015527282279617936,
      "loss": 4.7964,
      "mean_token_accuracy": 0.24083514362573624,
      "num_tokens": 155898395.0,
      "step": 75125
    },
    {
      "entropy": 5.702820205688477,
      "epoch": 6.815130624092888,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015524542887440301,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22843299359083175,
      "num_tokens": 155909281.0,
      "step": 75130
    },
    {
      "entropy": 5.730334520339966,
      "epoch": 6.815584179970973,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015521803742918173,
      "loss": 4.8995,
      "mean_token_accuracy": 0.2299213692545891,
      "num_tokens": 155919940.0,
      "step": 75135
    },
    {
      "entropy": 5.695126152038574,
      "epoch": 6.816037735849057,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001551906484610819,
      "loss": 4.7697,
      "mean_token_accuracy": 0.2403179109096527,
      "num_tokens": 155929662.0,
      "step": 75140
    },
    {
      "entropy": 5.669596910476685,
      "epoch": 6.816491291727141,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015516326197067006,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2398814633488655,
      "num_tokens": 155940093.0,
      "step": 75145
    },
    {
      "entropy": 5.7351155281066895,
      "epoch": 6.816944847605225,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015513587795851246,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23417120277881623,
      "num_tokens": 155950151.0,
      "step": 75150
    },
    {
      "entropy": 5.651120519638061,
      "epoch": 6.817398403483309,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015510849642517546,
      "loss": 4.7822,
      "mean_token_accuracy": 0.24112894386053085,
      "num_tokens": 155960508.0,
      "step": 75155
    },
    {
      "entropy": 5.683395671844482,
      "epoch": 6.817851959361393,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001550811173712252,
      "loss": 4.8423,
      "mean_token_accuracy": 0.23922771066427231,
      "num_tokens": 155970314.0,
      "step": 75160
    },
    {
      "entropy": 5.6532440185546875,
      "epoch": 6.818305515239477,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015505374079722808,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23264157325029372,
      "num_tokens": 155980111.0,
      "step": 75165
    },
    {
      "entropy": 5.709686136245727,
      "epoch": 6.818759071117562,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015502636670375019,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23245463967323304,
      "num_tokens": 155990444.0,
      "step": 75170
    },
    {
      "entropy": 5.686948680877686,
      "epoch": 6.819212626995646,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015499899509135762,
      "loss": 4.7853,
      "mean_token_accuracy": 0.2396940231323242,
      "num_tokens": 156000146.0,
      "step": 75175
    },
    {
      "entropy": 5.734592485427856,
      "epoch": 6.81966618287373,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015497162596061638,
      "loss": 4.8436,
      "mean_token_accuracy": 0.23347198963165283,
      "num_tokens": 156010386.0,
      "step": 75180
    },
    {
      "entropy": 5.6370344161987305,
      "epoch": 6.820119738751814,
      "grad_norm": 1.5,
      "learning_rate": 0.0001549442593120925,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23855168372392654,
      "num_tokens": 156019718.0,
      "step": 75185
    },
    {
      "entropy": 5.6818183898925785,
      "epoch": 6.820573294629899,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015491689514635199,
      "loss": 4.828,
      "mean_token_accuracy": 0.23434765338897706,
      "num_tokens": 156029986.0,
      "step": 75190
    },
    {
      "entropy": 5.676862525939941,
      "epoch": 6.821026850507983,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015488953346396055,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2352064922451973,
      "num_tokens": 156041678.0,
      "step": 75195
    },
    {
      "entropy": 5.629822301864624,
      "epoch": 6.821480406386067,
      "grad_norm": 1.578125,
      "learning_rate": 0.00015486217426548428,
      "loss": 4.7422,
      "mean_token_accuracy": 0.24462963491678238,
      "num_tokens": 156051087.0,
      "step": 75200
    },
    {
      "entropy": 5.7481015682220455,
      "epoch": 6.821933962264151,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015483481755148887,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23501291424036025,
      "num_tokens": 156061609.0,
      "step": 75205
    },
    {
      "entropy": 5.678619003295898,
      "epoch": 6.822387518142235,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015480746332254005,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23621124625205994,
      "num_tokens": 156071501.0,
      "step": 75210
    },
    {
      "entropy": 5.704296016693116,
      "epoch": 6.822841074020319,
      "grad_norm": 1.5,
      "learning_rate": 0.00015478011157920348,
      "loss": 4.8614,
      "mean_token_accuracy": 0.22793442010879517,
      "num_tokens": 156082419.0,
      "step": 75215
    },
    {
      "entropy": 5.703660392761231,
      "epoch": 6.823294629898403,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015475276232204476,
      "loss": 4.8127,
      "mean_token_accuracy": 0.22805894017219544,
      "num_tokens": 156092686.0,
      "step": 75220
    },
    {
      "entropy": 5.719445848464966,
      "epoch": 6.823748185776488,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015472541555162968,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22712850123643874,
      "num_tokens": 156102938.0,
      "step": 75225
    },
    {
      "entropy": 5.735339879989624,
      "epoch": 6.824201741654572,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015469807126852362,
      "loss": 4.8493,
      "mean_token_accuracy": 0.23669242709875107,
      "num_tokens": 156112839.0,
      "step": 75230
    },
    {
      "entropy": 5.678485679626465,
      "epoch": 6.824655297532656,
      "grad_norm": 1.5625,
      "learning_rate": 0.00015467072947329208,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23599264919757842,
      "num_tokens": 156123266.0,
      "step": 75235
    },
    {
      "entropy": 5.708407592773438,
      "epoch": 6.8251088534107405,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015464339016650058,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23341472893953324,
      "num_tokens": 156133716.0,
      "step": 75240
    },
    {
      "entropy": 5.713318300247193,
      "epoch": 6.825562409288825,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015461605334871436,
      "loss": 4.8361,
      "mean_token_accuracy": 0.2315596804022789,
      "num_tokens": 156144937.0,
      "step": 75245
    },
    {
      "entropy": 5.634670877456665,
      "epoch": 6.826015965166908,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015458871902049892,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23703759163618088,
      "num_tokens": 156155733.0,
      "step": 75250
    },
    {
      "entropy": 5.676731109619141,
      "epoch": 6.826469521044992,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015456138718241932,
      "loss": 4.8113,
      "mean_token_accuracy": 0.2299705296754837,
      "num_tokens": 156167006.0,
      "step": 75255
    },
    {
      "entropy": 5.693284511566162,
      "epoch": 6.826923076923077,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015453405783504105,
      "loss": 4.781,
      "mean_token_accuracy": 0.24043135493993759,
      "num_tokens": 156176790.0,
      "step": 75260
    },
    {
      "entropy": 5.775340557098389,
      "epoch": 6.827376632801161,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015450673097892908,
      "loss": 4.9203,
      "mean_token_accuracy": 0.235049869120121,
      "num_tokens": 156187330.0,
      "step": 75265
    },
    {
      "entropy": 5.685671138763428,
      "epoch": 6.827830188679245,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001544794066146487,
      "loss": 4.851,
      "mean_token_accuracy": 0.2299023002386093,
      "num_tokens": 156197527.0,
      "step": 75270
    },
    {
      "entropy": 5.663446950912475,
      "epoch": 6.8282837445573294,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015445208474276488,
      "loss": 4.7587,
      "mean_token_accuracy": 0.24354366958141327,
      "num_tokens": 156206675.0,
      "step": 75275
    },
    {
      "entropy": 5.680635261535644,
      "epoch": 6.828737300435414,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015442476536384265,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23861982971429824,
      "num_tokens": 156218010.0,
      "step": 75280
    },
    {
      "entropy": 5.696334886550903,
      "epoch": 6.829190856313498,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015439744847844705,
      "loss": 4.7681,
      "mean_token_accuracy": 0.2422455295920372,
      "num_tokens": 156229123.0,
      "step": 75285
    },
    {
      "entropy": 5.731171989440918,
      "epoch": 6.829644412191582,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015437013408714278,
      "loss": 4.9182,
      "mean_token_accuracy": 0.23527497202157974,
      "num_tokens": 156239437.0,
      "step": 75290
    },
    {
      "entropy": 5.668341541290284,
      "epoch": 6.8300979680696665,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015434282219049504,
      "loss": 4.7806,
      "mean_token_accuracy": 0.23170819729566575,
      "num_tokens": 156250543.0,
      "step": 75295
    },
    {
      "entropy": 5.72125391960144,
      "epoch": 6.830551523947751,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015431551278906848,
      "loss": 4.8957,
      "mean_token_accuracy": 0.22463415414094925,
      "num_tokens": 156261250.0,
      "step": 75300
    },
    {
      "entropy": 5.717634820938111,
      "epoch": 6.831005079825834,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001542882058834278,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2342734917998314,
      "num_tokens": 156272062.0,
      "step": 75305
    },
    {
      "entropy": 5.796472692489624,
      "epoch": 6.831458635703918,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015426090147413783,
      "loss": 4.9618,
      "mean_token_accuracy": 0.22306664735078813,
      "num_tokens": 156282376.0,
      "step": 75310
    },
    {
      "entropy": 5.709507989883423,
      "epoch": 6.831912191582003,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015423359956176318,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23657817989587784,
      "num_tokens": 156292941.0,
      "step": 75315
    },
    {
      "entropy": 5.706334018707276,
      "epoch": 6.832365747460087,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015420630014686848,
      "loss": 4.7898,
      "mean_token_accuracy": 0.23587787449359893,
      "num_tokens": 156303257.0,
      "step": 75320
    },
    {
      "entropy": 5.5803345203399655,
      "epoch": 6.832819303338171,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015417900323001814,
      "loss": 4.6903,
      "mean_token_accuracy": 0.24890857487916945,
      "num_tokens": 156315608.0,
      "step": 75325
    },
    {
      "entropy": 5.737262916564942,
      "epoch": 6.8332728592162555,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00015415170881177693,
      "loss": 4.8547,
      "mean_token_accuracy": 0.23240405470132827,
      "num_tokens": 156324912.0,
      "step": 75330
    },
    {
      "entropy": 5.699205493927002,
      "epoch": 6.83372641509434,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015412441689270921,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23311815708875655,
      "num_tokens": 156335830.0,
      "step": 75335
    },
    {
      "entropy": 5.735593700408936,
      "epoch": 6.834179970972424,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015409712747337928,
      "loss": 4.9064,
      "mean_token_accuracy": 0.23111845105886458,
      "num_tokens": 156345974.0,
      "step": 75340
    },
    {
      "entropy": 5.692803525924683,
      "epoch": 6.834633526850508,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001540698405543516,
      "loss": 4.7608,
      "mean_token_accuracy": 0.24322600066661834,
      "num_tokens": 156356087.0,
      "step": 75345
    },
    {
      "entropy": 5.718534994125366,
      "epoch": 6.835087082728592,
      "grad_norm": 1.375,
      "learning_rate": 0.00015404255613619044,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23395222574472427,
      "num_tokens": 156367382.0,
      "step": 75350
    },
    {
      "entropy": 5.7608788967132565,
      "epoch": 6.835540638606677,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015401527421946001,
      "loss": 4.9002,
      "mean_token_accuracy": 0.23485763370990753,
      "num_tokens": 156378238.0,
      "step": 75355
    },
    {
      "entropy": 5.640186500549317,
      "epoch": 6.83599419448476,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015398799480472448,
      "loss": 4.7469,
      "mean_token_accuracy": 0.23846034109592437,
      "num_tokens": 156388485.0,
      "step": 75360
    },
    {
      "entropy": 5.6954234600067135,
      "epoch": 6.8364477503628445,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015396071789254814,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23045863658189775,
      "num_tokens": 156399256.0,
      "step": 75365
    },
    {
      "entropy": 5.561241149902344,
      "epoch": 6.836901306240929,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015393344348349502,
      "loss": 4.6515,
      "mean_token_accuracy": 0.2534958243370056,
      "num_tokens": 156409348.0,
      "step": 75370
    },
    {
      "entropy": 5.602859592437744,
      "epoch": 6.837354862119013,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015390617157812908,
      "loss": 4.7457,
      "mean_token_accuracy": 0.23919144868850709,
      "num_tokens": 156420391.0,
      "step": 75375
    },
    {
      "entropy": 5.649798965454101,
      "epoch": 6.837808417997097,
      "grad_norm": 1.640625,
      "learning_rate": 0.00015387890217701442,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2303858533501625,
      "num_tokens": 156430735.0,
      "step": 75380
    },
    {
      "entropy": 5.695679950714111,
      "epoch": 6.8382619738751815,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001538516352807149,
      "loss": 4.8331,
      "mean_token_accuracy": 0.23362144380807875,
      "num_tokens": 156441137.0,
      "step": 75385
    },
    {
      "entropy": 5.709078073501587,
      "epoch": 6.838715529753266,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001538243708897944,
      "loss": 4.7683,
      "mean_token_accuracy": 0.24039389491081237,
      "num_tokens": 156451092.0,
      "step": 75390
    },
    {
      "entropy": 5.68246488571167,
      "epoch": 6.83916908563135,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015379710900481682,
      "loss": 4.7677,
      "mean_token_accuracy": 0.24279021769762038,
      "num_tokens": 156461364.0,
      "step": 75395
    },
    {
      "entropy": 5.708893775939941,
      "epoch": 6.839622641509434,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015376984962634595,
      "loss": 4.8412,
      "mean_token_accuracy": 0.2271679535508156,
      "num_tokens": 156471423.0,
      "step": 75400
    },
    {
      "entropy": 5.7189287662506105,
      "epoch": 6.840076197387518,
      "grad_norm": 1.609375,
      "learning_rate": 0.00015374259275494545,
      "loss": 4.8096,
      "mean_token_accuracy": 0.23359890580177306,
      "num_tokens": 156480979.0,
      "step": 75405
    },
    {
      "entropy": 5.6259765625,
      "epoch": 6.840529753265602,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015371533839117905,
      "loss": 4.7374,
      "mean_token_accuracy": 0.24205704927444457,
      "num_tokens": 156490683.0,
      "step": 75410
    },
    {
      "entropy": 5.778942060470581,
      "epoch": 6.840983309143686,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015368808653561033,
      "loss": 4.9501,
      "mean_token_accuracy": 0.22678415477275848,
      "num_tokens": 156500948.0,
      "step": 75415
    },
    {
      "entropy": 5.725146341323852,
      "epoch": 6.8414368650217705,
      "grad_norm": 1.375,
      "learning_rate": 0.00015366083718880292,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23846079111099244,
      "num_tokens": 156510656.0,
      "step": 75420
    },
    {
      "entropy": 5.718677806854248,
      "epoch": 6.841890420899855,
      "grad_norm": 1.578125,
      "learning_rate": 0.00015363359035132024,
      "loss": 4.8133,
      "mean_token_accuracy": 0.2329955816268921,
      "num_tokens": 156520800.0,
      "step": 75425
    },
    {
      "entropy": 5.641876268386841,
      "epoch": 6.842343976777939,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015360634602372593,
      "loss": 4.7284,
      "mean_token_accuracy": 0.2286281853914261,
      "num_tokens": 156531375.0,
      "step": 75430
    },
    {
      "entropy": 5.643598461151123,
      "epoch": 6.842797532656023,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001535791042065833,
      "loss": 4.7443,
      "mean_token_accuracy": 0.23696034401655197,
      "num_tokens": 156541597.0,
      "step": 75435
    },
    {
      "entropy": 5.7348767757415775,
      "epoch": 6.843251088534108,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015355186490045576,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2280428558588028,
      "num_tokens": 156552369.0,
      "step": 75440
    },
    {
      "entropy": 5.722568225860596,
      "epoch": 6.843704644412192,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015352462810590655,
      "loss": 4.8571,
      "mean_token_accuracy": 0.23522895127534865,
      "num_tokens": 156563178.0,
      "step": 75445
    },
    {
      "entropy": 5.7066648483276365,
      "epoch": 6.844158200290276,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015349739382349904,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23423490673303604,
      "num_tokens": 156573306.0,
      "step": 75450
    },
    {
      "entropy": 5.677903079986573,
      "epoch": 6.84461175616836,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015347016205379639,
      "loss": 4.7869,
      "mean_token_accuracy": 0.2329905718564987,
      "num_tokens": 156584261.0,
      "step": 75455
    },
    {
      "entropy": 5.717812347412109,
      "epoch": 6.845065312046444,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015344293279736162,
      "loss": 4.777,
      "mean_token_accuracy": 0.23898203521966935,
      "num_tokens": 156594942.0,
      "step": 75460
    },
    {
      "entropy": 5.707562494277954,
      "epoch": 6.845518867924528,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001534157060547581,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23595524281263353,
      "num_tokens": 156605467.0,
      "step": 75465
    },
    {
      "entropy": 5.692091369628907,
      "epoch": 6.845972423802612,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015338848182654875,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23052549064159394,
      "num_tokens": 156615914.0,
      "step": 75470
    },
    {
      "entropy": 5.734007883071899,
      "epoch": 6.8464259796806965,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001533612601132966,
      "loss": 4.8076,
      "mean_token_accuracy": 0.2321406841278076,
      "num_tokens": 156627281.0,
      "step": 75475
    },
    {
      "entropy": 5.625699901580811,
      "epoch": 6.846879535558781,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001533340409155646,
      "loss": 4.742,
      "mean_token_accuracy": 0.2393789693713188,
      "num_tokens": 156638971.0,
      "step": 75480
    },
    {
      "entropy": 5.638016128540039,
      "epoch": 6.847333091436865,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001533068242339156,
      "loss": 4.7886,
      "mean_token_accuracy": 0.2323105052113533,
      "num_tokens": 156648905.0,
      "step": 75485
    },
    {
      "entropy": 5.697138071060181,
      "epoch": 6.847786647314949,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001532796100689125,
      "loss": 4.827,
      "mean_token_accuracy": 0.2355397269129753,
      "num_tokens": 156659554.0,
      "step": 75490
    },
    {
      "entropy": 5.66434416770935,
      "epoch": 6.848240203193034,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000153252398421118,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23207153826951982,
      "num_tokens": 156669281.0,
      "step": 75495
    },
    {
      "entropy": 5.740524482727051,
      "epoch": 6.848693759071118,
      "grad_norm": 1.5,
      "learning_rate": 0.000153225189291095,
      "loss": 4.8809,
      "mean_token_accuracy": 0.23121393620967864,
      "num_tokens": 156679239.0,
      "step": 75500
    },
    {
      "entropy": 5.6817680358886715,
      "epoch": 6.849147314949201,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00015319798267940615,
      "loss": 4.8186,
      "mean_token_accuracy": 0.2385571151971817,
      "num_tokens": 156689784.0,
      "step": 75505
    },
    {
      "entropy": 5.699314928054809,
      "epoch": 6.849600870827286,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015317077858661402,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23990172147750854,
      "num_tokens": 156700975.0,
      "step": 75510
    },
    {
      "entropy": 5.703538513183593,
      "epoch": 6.85005442670537,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015314357701328124,
      "loss": 4.8342,
      "mean_token_accuracy": 0.23627193421125411,
      "num_tokens": 156711117.0,
      "step": 75515
    },
    {
      "entropy": 5.762066507339478,
      "epoch": 6.850507982583454,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015311637795997036,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21864198446273803,
      "num_tokens": 156721505.0,
      "step": 75520
    },
    {
      "entropy": 5.690820121765137,
      "epoch": 6.850961538461538,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00015308918142724383,
      "loss": 4.8047,
      "mean_token_accuracy": 0.23950255960226058,
      "num_tokens": 156732101.0,
      "step": 75525
    },
    {
      "entropy": 5.719819068908691,
      "epoch": 6.851415094339623,
      "grad_norm": 1.453125,
      "learning_rate": 0.000153061987415664,
      "loss": 4.7894,
      "mean_token_accuracy": 0.23064806461334228,
      "num_tokens": 156742213.0,
      "step": 75530
    },
    {
      "entropy": 5.667103004455567,
      "epoch": 6.851868650217707,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015303479592579345,
      "loss": 4.7394,
      "mean_token_accuracy": 0.23954056203365326,
      "num_tokens": 156752283.0,
      "step": 75535
    },
    {
      "entropy": 5.728125381469726,
      "epoch": 6.852322206095791,
      "grad_norm": 1.65625,
      "learning_rate": 0.00015300760695819438,
      "loss": 4.7951,
      "mean_token_accuracy": 0.2365260526537895,
      "num_tokens": 156761765.0,
      "step": 75540
    },
    {
      "entropy": 5.649013233184815,
      "epoch": 6.852775761973875,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001529804205134291,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23723316937685013,
      "num_tokens": 156772303.0,
      "step": 75545
    },
    {
      "entropy": 5.69432487487793,
      "epoch": 6.85322931785196,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015295323659205985,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22771285474300385,
      "num_tokens": 156782053.0,
      "step": 75550
    },
    {
      "entropy": 5.635698413848877,
      "epoch": 6.853682873730044,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001529260551946487,
      "loss": 4.6679,
      "mean_token_accuracy": 0.25563654154539106,
      "num_tokens": 156792050.0,
      "step": 75555
    },
    {
      "entropy": 5.677253007888794,
      "epoch": 6.854136429608127,
      "grad_norm": 1.5,
      "learning_rate": 0.0001528988763217578,
      "loss": 4.7734,
      "mean_token_accuracy": 0.24023307114839554,
      "num_tokens": 156801678.0,
      "step": 75560
    },
    {
      "entropy": 5.73163595199585,
      "epoch": 6.854589985486212,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015287169997394933,
      "loss": 4.8339,
      "mean_token_accuracy": 0.229972206056118,
      "num_tokens": 156812058.0,
      "step": 75565
    },
    {
      "entropy": 5.631852388381958,
      "epoch": 6.855043541364296,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00015284452615178524,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2529259264469147,
      "num_tokens": 156822514.0,
      "step": 75570
    },
    {
      "entropy": 5.692941331863404,
      "epoch": 6.85549709724238,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015281735485582742,
      "loss": 4.8886,
      "mean_token_accuracy": 0.22547011077404022,
      "num_tokens": 156832553.0,
      "step": 75575
    },
    {
      "entropy": 5.754495334625244,
      "epoch": 6.855950653120464,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015279018608663789,
      "loss": 4.8663,
      "mean_token_accuracy": 0.22955463826656342,
      "num_tokens": 156842519.0,
      "step": 75580
    },
    {
      "entropy": 5.700035095214844,
      "epoch": 6.856404208998549,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015276301984477845,
      "loss": 4.8645,
      "mean_token_accuracy": 0.23234906494617463,
      "num_tokens": 156852662.0,
      "step": 75585
    },
    {
      "entropy": 5.732392740249634,
      "epoch": 6.856857764876633,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015273585613081086,
      "loss": 4.7975,
      "mean_token_accuracy": 0.2318106025457382,
      "num_tokens": 156862813.0,
      "step": 75590
    },
    {
      "entropy": 5.698159217834473,
      "epoch": 6.857311320754717,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015270869494529694,
      "loss": 4.8359,
      "mean_token_accuracy": 0.23860257714986802,
      "num_tokens": 156873923.0,
      "step": 75595
    },
    {
      "entropy": 5.697997379302978,
      "epoch": 6.857764876632801,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015268153628879836,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22110718488693237,
      "num_tokens": 156884206.0,
      "step": 75600
    },
    {
      "entropy": 5.676048040390015,
      "epoch": 6.858218432510886,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015265438016187684,
      "loss": 4.7704,
      "mean_token_accuracy": 0.238319531083107,
      "num_tokens": 156894604.0,
      "step": 75605
    },
    {
      "entropy": 5.716974925994873,
      "epoch": 6.85867198838897,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015262722656509388,
      "loss": 4.8067,
      "mean_token_accuracy": 0.23773904144763947,
      "num_tokens": 156904565.0,
      "step": 75610
    },
    {
      "entropy": 5.650501394271851,
      "epoch": 6.859125544267053,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015260007549901106,
      "loss": 4.7231,
      "mean_token_accuracy": 0.24530500173568726,
      "num_tokens": 156914127.0,
      "step": 75615
    },
    {
      "entropy": 5.721126222610474,
      "epoch": 6.859579100145138,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015257292696418986,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22241939306259156,
      "num_tokens": 156925102.0,
      "step": 75620
    },
    {
      "entropy": 5.709923362731933,
      "epoch": 6.860032656023222,
      "grad_norm": 1.640625,
      "learning_rate": 0.00015254578096119176,
      "loss": 4.7654,
      "mean_token_accuracy": 0.2367502197623253,
      "num_tokens": 156935207.0,
      "step": 75625
    },
    {
      "entropy": 5.6855724334716795,
      "epoch": 6.860486211901306,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001525186374905781,
      "loss": 4.8227,
      "mean_token_accuracy": 0.22979849129915236,
      "num_tokens": 156945733.0,
      "step": 75630
    },
    {
      "entropy": 5.712663269042968,
      "epoch": 6.86093976777939,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001524914965529102,
      "loss": 4.8124,
      "mean_token_accuracy": 0.24104027599096298,
      "num_tokens": 156955457.0,
      "step": 75635
    },
    {
      "entropy": 5.683501720428467,
      "epoch": 6.861393323657475,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015246435814874942,
      "loss": 4.8515,
      "mean_token_accuracy": 0.22904582768678666,
      "num_tokens": 156966005.0,
      "step": 75640
    },
    {
      "entropy": 5.676025009155273,
      "epoch": 6.861846879535559,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001524372222786569,
      "loss": 4.847,
      "mean_token_accuracy": 0.22790627777576447,
      "num_tokens": 156975264.0,
      "step": 75645
    },
    {
      "entropy": 5.673156070709228,
      "epoch": 6.862300435413643,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001524100889431939,
      "loss": 4.7578,
      "mean_token_accuracy": 0.236500284075737,
      "num_tokens": 156985482.0,
      "step": 75650
    },
    {
      "entropy": 5.675786781311035,
      "epoch": 6.8627539912917275,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015238295814292148,
      "loss": 4.7895,
      "mean_token_accuracy": 0.23383669853210448,
      "num_tokens": 156995554.0,
      "step": 75655
    },
    {
      "entropy": 5.741946506500244,
      "epoch": 6.863207547169811,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015235582987840063,
      "loss": 4.9105,
      "mean_token_accuracy": 0.23143766075372696,
      "num_tokens": 157005333.0,
      "step": 75660
    },
    {
      "entropy": 5.7616184711456295,
      "epoch": 6.863661103047895,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015232870415019258,
      "loss": 4.8675,
      "mean_token_accuracy": 0.22875334620475768,
      "num_tokens": 157014956.0,
      "step": 75665
    },
    {
      "entropy": 5.717011404037476,
      "epoch": 6.864114658925979,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015230158095885817,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2264579713344574,
      "num_tokens": 157024652.0,
      "step": 75670
    },
    {
      "entropy": 5.7231528759002686,
      "epoch": 6.864568214804064,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015227446030495833,
      "loss": 4.8944,
      "mean_token_accuracy": 0.22523243427276612,
      "num_tokens": 157035286.0,
      "step": 75675
    },
    {
      "entropy": 5.6673050880432125,
      "epoch": 6.865021770682148,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015224734218905394,
      "loss": 4.8529,
      "mean_token_accuracy": 0.23635275810956954,
      "num_tokens": 157045854.0,
      "step": 75680
    },
    {
      "entropy": 5.726788091659546,
      "epoch": 6.865475326560232,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001522202266117058,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22935620248317717,
      "num_tokens": 157055214.0,
      "step": 75685
    },
    {
      "entropy": 5.726490545272827,
      "epoch": 6.8659288824383164,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015219311357347458,
      "loss": 4.8327,
      "mean_token_accuracy": 0.23596496134996414,
      "num_tokens": 157065055.0,
      "step": 75690
    },
    {
      "entropy": 5.686109924316407,
      "epoch": 6.866382438316401,
      "grad_norm": 1.5,
      "learning_rate": 0.00015216600307492112,
      "loss": 4.7693,
      "mean_token_accuracy": 0.24171197712421416,
      "num_tokens": 157075344.0,
      "step": 75695
    },
    {
      "entropy": 5.69982795715332,
      "epoch": 6.866835994194485,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015213889511660607,
      "loss": 4.7991,
      "mean_token_accuracy": 0.23165389597415925,
      "num_tokens": 157085155.0,
      "step": 75700
    },
    {
      "entropy": 5.65440936088562,
      "epoch": 6.867289550072569,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001521117896990899,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2507082536816597,
      "num_tokens": 157095543.0,
      "step": 75705
    },
    {
      "entropy": 5.7361010074615475,
      "epoch": 6.8677431059506535,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001520846868229333,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22896915823221206,
      "num_tokens": 157106155.0,
      "step": 75710
    },
    {
      "entropy": 5.751904726028442,
      "epoch": 6.868196661828737,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015205758648869668,
      "loss": 4.9104,
      "mean_token_accuracy": 0.23263455778360367,
      "num_tokens": 157117105.0,
      "step": 75715
    },
    {
      "entropy": 5.753545951843262,
      "epoch": 6.868650217706821,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001520304886969405,
      "loss": 4.8946,
      "mean_token_accuracy": 0.23407551348209382,
      "num_tokens": 157127579.0,
      "step": 75720
    },
    {
      "entropy": 5.7259711742401125,
      "epoch": 6.869103773584905,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015200339344822506,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22973441928625107,
      "num_tokens": 157138533.0,
      "step": 75725
    },
    {
      "entropy": 5.726256799697876,
      "epoch": 6.86955732946299,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015197630074311087,
      "loss": 4.8521,
      "mean_token_accuracy": 0.23607059419155121,
      "num_tokens": 157148807.0,
      "step": 75730
    },
    {
      "entropy": 5.766498374938965,
      "epoch": 6.870010885341074,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015194921058215815,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23436038345098495,
      "num_tokens": 157159906.0,
      "step": 75735
    },
    {
      "entropy": 5.702524042129516,
      "epoch": 6.870464441219158,
      "grad_norm": 1.375,
      "learning_rate": 0.0001519221229659271,
      "loss": 4.8511,
      "mean_token_accuracy": 0.23303035646677017,
      "num_tokens": 157170935.0,
      "step": 75740
    },
    {
      "entropy": 5.7147563934326175,
      "epoch": 6.8709179970972425,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001518950378949779,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23695462197065353,
      "num_tokens": 157180962.0,
      "step": 75745
    },
    {
      "entropy": 5.690552186965943,
      "epoch": 6.871371552975327,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001518679553698707,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23941102176904677,
      "num_tokens": 157189617.0,
      "step": 75750
    },
    {
      "entropy": 5.7436281681060795,
      "epoch": 6.871825108853411,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001518408753911656,
      "loss": 4.8134,
      "mean_token_accuracy": 0.23836093097925187,
      "num_tokens": 157200553.0,
      "step": 75755
    },
    {
      "entropy": 5.743346405029297,
      "epoch": 6.872278664731495,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015181379795942243,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23218999803066254,
      "num_tokens": 157210888.0,
      "step": 75760
    },
    {
      "entropy": 5.701861429214477,
      "epoch": 6.87273222060958,
      "grad_norm": 1.5,
      "learning_rate": 0.0001517867230752014,
      "loss": 4.7822,
      "mean_token_accuracy": 0.2380447968840599,
      "num_tokens": 157221033.0,
      "step": 75765
    },
    {
      "entropy": 5.693667602539063,
      "epoch": 6.873185776487663,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015175965073906236,
      "loss": 4.8384,
      "mean_token_accuracy": 0.23118506669998168,
      "num_tokens": 157230958.0,
      "step": 75770
    },
    {
      "entropy": 5.735356044769287,
      "epoch": 6.873639332365747,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015173258095156514,
      "loss": 4.8823,
      "mean_token_accuracy": 0.23631784170866013,
      "num_tokens": 157241513.0,
      "step": 75775
    },
    {
      "entropy": 5.683316087722778,
      "epoch": 6.8740928882438315,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015170551371326957,
      "loss": 4.8063,
      "mean_token_accuracy": 0.237655171751976,
      "num_tokens": 157252296.0,
      "step": 75780
    },
    {
      "entropy": 5.575938558578491,
      "epoch": 6.874546444121916,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015167844902473537,
      "loss": 4.6497,
      "mean_token_accuracy": 0.25262352228164675,
      "num_tokens": 157262869.0,
      "step": 75785
    },
    {
      "entropy": 5.695075178146363,
      "epoch": 6.875,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001516513868865223,
      "loss": 4.8091,
      "mean_token_accuracy": 0.24555927217006684,
      "num_tokens": 157273636.0,
      "step": 75790
    },
    {
      "entropy": 5.657601737976075,
      "epoch": 6.875453555878084,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015162432729918986,
      "loss": 4.797,
      "mean_token_accuracy": 0.23031304627656937,
      "num_tokens": 157284105.0,
      "step": 75795
    },
    {
      "entropy": 5.670171308517456,
      "epoch": 6.8759071117561685,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001515972702632979,
      "loss": 4.8455,
      "mean_token_accuracy": 0.2326858252286911,
      "num_tokens": 157294722.0,
      "step": 75800
    },
    {
      "entropy": 5.670147466659546,
      "epoch": 6.876360667634253,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015157021577940583,
      "loss": 4.7592,
      "mean_token_accuracy": 0.24172329902648926,
      "num_tokens": 157304486.0,
      "step": 75805
    },
    {
      "entropy": 5.688837623596191,
      "epoch": 6.876814223512337,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015154316384807313,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22387246787548065,
      "num_tokens": 157315515.0,
      "step": 75810
    },
    {
      "entropy": 5.713851690292358,
      "epoch": 6.87726777939042,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001515161144698593,
      "loss": 4.7871,
      "mean_token_accuracy": 0.23638107478618622,
      "num_tokens": 157325333.0,
      "step": 75815
    },
    {
      "entropy": 5.7071123123168945,
      "epoch": 6.877721335268505,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015148906764532363,
      "loss": 4.8391,
      "mean_token_accuracy": 0.23174296021461488,
      "num_tokens": 157335444.0,
      "step": 75820
    },
    {
      "entropy": 5.645331716537475,
      "epoch": 6.878174891146589,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015146202337502557,
      "loss": 4.7687,
      "mean_token_accuracy": 0.24018481969833375,
      "num_tokens": 157346481.0,
      "step": 75825
    },
    {
      "entropy": 5.7194853782653805,
      "epoch": 6.878628447024673,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00015143498165952424,
      "loss": 4.787,
      "mean_token_accuracy": 0.23325071781873702,
      "num_tokens": 157355449.0,
      "step": 75830
    },
    {
      "entropy": 5.72223801612854,
      "epoch": 6.8790820029027575,
      "grad_norm": 1.5,
      "learning_rate": 0.00015140794249937905,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2360537424683571,
      "num_tokens": 157365417.0,
      "step": 75835
    },
    {
      "entropy": 5.653240156173706,
      "epoch": 6.879535558780842,
      "grad_norm": 1.375,
      "learning_rate": 0.00015138090589514913,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23357821851968766,
      "num_tokens": 157376436.0,
      "step": 75840
    },
    {
      "entropy": 5.687628316879272,
      "epoch": 6.879989114658926,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015135387184739355,
      "loss": 4.8679,
      "mean_token_accuracy": 0.23372991532087325,
      "num_tokens": 157386528.0,
      "step": 75845
    },
    {
      "entropy": 5.640972566604614,
      "epoch": 6.88044267053701,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015132684035667142,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23494026362895964,
      "num_tokens": 157398375.0,
      "step": 75850
    },
    {
      "entropy": 5.663111782073974,
      "epoch": 6.880896226415095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015129981142354178,
      "loss": 4.7399,
      "mean_token_accuracy": 0.2467467665672302,
      "num_tokens": 157409057.0,
      "step": 75855
    },
    {
      "entropy": 5.627550411224365,
      "epoch": 6.881349782293179,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015127278504856352,
      "loss": 4.7569,
      "mean_token_accuracy": 0.24075976461172105,
      "num_tokens": 157419070.0,
      "step": 75860
    },
    {
      "entropy": 5.713027477264404,
      "epoch": 6.881803338171263,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001512457612322955,
      "loss": 4.8272,
      "mean_token_accuracy": 0.23107607215642928,
      "num_tokens": 157430005.0,
      "step": 75865
    },
    {
      "entropy": 5.656521415710449,
      "epoch": 6.8822568940493465,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015121873997529678,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2308400884270668,
      "num_tokens": 157441016.0,
      "step": 75870
    },
    {
      "entropy": 5.691999197006226,
      "epoch": 6.882710449927431,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015119172127812608,
      "loss": 4.7559,
      "mean_token_accuracy": 0.2417181596159935,
      "num_tokens": 157450762.0,
      "step": 75875
    },
    {
      "entropy": 5.767321634292602,
      "epoch": 6.883164005805515,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00015116470514134212,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23465978354215622,
      "num_tokens": 157461021.0,
      "step": 75880
    },
    {
      "entropy": 5.687707805633545,
      "epoch": 6.883617561683599,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00015113769156550362,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22707845866680146,
      "num_tokens": 157470605.0,
      "step": 75885
    },
    {
      "entropy": 5.627413177490235,
      "epoch": 6.8840711175616836,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001511106805511692,
      "loss": 4.8178,
      "mean_token_accuracy": 0.23241351246833802,
      "num_tokens": 157481406.0,
      "step": 75890
    },
    {
      "entropy": 5.669136428833008,
      "epoch": 6.884524673439768,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001510836720988974,
      "loss": 4.7696,
      "mean_token_accuracy": 0.2324667289853096,
      "num_tokens": 157492411.0,
      "step": 75895
    },
    {
      "entropy": 5.672575235366821,
      "epoch": 6.884978229317852,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00015105666620924692,
      "loss": 4.7953,
      "mean_token_accuracy": 0.23559796810150146,
      "num_tokens": 157502869.0,
      "step": 75900
    },
    {
      "entropy": 5.731847524642944,
      "epoch": 6.885431785195936,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001510296628827762,
      "loss": 4.8541,
      "mean_token_accuracy": 0.2336750328540802,
      "num_tokens": 157513393.0,
      "step": 75905
    },
    {
      "entropy": 5.663247966766358,
      "epoch": 6.885885341074021,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001510026621200436,
      "loss": 4.7157,
      "mean_token_accuracy": 0.23812910169363022,
      "num_tokens": 157523271.0,
      "step": 75910
    },
    {
      "entropy": 5.707368278503418,
      "epoch": 6.886338896952105,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015097566392160755,
      "loss": 4.832,
      "mean_token_accuracy": 0.22935019731521605,
      "num_tokens": 157534004.0,
      "step": 75915
    },
    {
      "entropy": 5.64934458732605,
      "epoch": 6.886792452830189,
      "grad_norm": 1.5,
      "learning_rate": 0.0001509486682880264,
      "loss": 4.8151,
      "mean_token_accuracy": 0.2389421969652176,
      "num_tokens": 157545687.0,
      "step": 75920
    },
    {
      "entropy": 5.706874179840088,
      "epoch": 6.8872460087082725,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001509216752198584,
      "loss": 4.8219,
      "mean_token_accuracy": 0.2341713324189186,
      "num_tokens": 157556376.0,
      "step": 75925
    },
    {
      "entropy": 5.690069770812988,
      "epoch": 6.887699564586357,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001508946847176617,
      "loss": 4.8273,
      "mean_token_accuracy": 0.22940920442342758,
      "num_tokens": 157565887.0,
      "step": 75930
    },
    {
      "entropy": 5.676408338546753,
      "epoch": 6.888153120464441,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015086769678199457,
      "loss": 4.805,
      "mean_token_accuracy": 0.23708854615688324,
      "num_tokens": 157576092.0,
      "step": 75935
    },
    {
      "entropy": 5.651087284088135,
      "epoch": 6.888606676342525,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015084071141341514,
      "loss": 4.7563,
      "mean_token_accuracy": 0.24378986656665802,
      "num_tokens": 157585815.0,
      "step": 75940
    },
    {
      "entropy": 5.577186870574951,
      "epoch": 6.88906023222061,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001508137286124815,
      "loss": 4.7029,
      "mean_token_accuracy": 0.24882487207651138,
      "num_tokens": 157595600.0,
      "step": 75945
    },
    {
      "entropy": 5.643339776992798,
      "epoch": 6.889513788098694,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015078674837975154,
      "loss": 4.7276,
      "mean_token_accuracy": 0.24153047204017639,
      "num_tokens": 157605429.0,
      "step": 75950
    },
    {
      "entropy": 5.639279460906982,
      "epoch": 6.889967343976778,
      "grad_norm": 1.578125,
      "learning_rate": 0.00015075977071578322,
      "loss": 4.7562,
      "mean_token_accuracy": 0.23539571613073348,
      "num_tokens": 157616860.0,
      "step": 75955
    },
    {
      "entropy": 5.774731302261353,
      "epoch": 6.890420899854862,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015073279562113461,
      "loss": 4.9798,
      "mean_token_accuracy": 0.22430389374494553,
      "num_tokens": 157629547.0,
      "step": 75960
    },
    {
      "entropy": 5.735005140304565,
      "epoch": 6.890874455732947,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001507058230963633,
      "loss": 4.8453,
      "mean_token_accuracy": 0.22796317785978318,
      "num_tokens": 157639674.0,
      "step": 75965
    },
    {
      "entropy": 5.729006147384643,
      "epoch": 6.89132801161103,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015067885314202733,
      "loss": 4.8229,
      "mean_token_accuracy": 0.2375951200723648,
      "num_tokens": 157649944.0,
      "step": 75970
    },
    {
      "entropy": 5.749359512329102,
      "epoch": 6.891781567489114,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015065188575868438,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2331776425242424,
      "num_tokens": 157660948.0,
      "step": 75975
    },
    {
      "entropy": 5.7819517135620115,
      "epoch": 6.892235123367199,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015062492094689215,
      "loss": 4.8865,
      "mean_token_accuracy": 0.22995537668466567,
      "num_tokens": 157672228.0,
      "step": 75980
    },
    {
      "entropy": 5.678942251205444,
      "epoch": 6.892688679245283,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015059795870720822,
      "loss": 4.779,
      "mean_token_accuracy": 0.2362314760684967,
      "num_tokens": 157683578.0,
      "step": 75985
    },
    {
      "entropy": 5.60044150352478,
      "epoch": 6.893142235123367,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015057099904019016,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2504554569721222,
      "num_tokens": 157694046.0,
      "step": 75990
    },
    {
      "entropy": 5.68962140083313,
      "epoch": 6.893595791001451,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001505440419463956,
      "loss": 4.7786,
      "mean_token_accuracy": 0.2392241820693016,
      "num_tokens": 157703586.0,
      "step": 75995
    },
    {
      "entropy": 5.763845920562744,
      "epoch": 6.894049346879536,
      "grad_norm": 1.5,
      "learning_rate": 0.00015051708742638186,
      "loss": 4.8552,
      "mean_token_accuracy": 0.24247452169656752,
      "num_tokens": 157714340.0,
      "step": 76000
    },
    {
      "entropy": 5.67534384727478,
      "epoch": 6.89450290275762,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001504901354807066,
      "loss": 4.8668,
      "mean_token_accuracy": 0.2330644726753235,
      "num_tokens": 157725044.0,
      "step": 76005
    },
    {
      "entropy": 5.706527137756348,
      "epoch": 6.894956458635704,
      "grad_norm": 1.59375,
      "learning_rate": 0.00015046318610992704,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22533582150936127,
      "num_tokens": 157735103.0,
      "step": 76010
    },
    {
      "entropy": 5.684224605560303,
      "epoch": 6.895410014513788,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001504362393146005,
      "loss": 4.7875,
      "mean_token_accuracy": 0.2365830957889557,
      "num_tokens": 157744545.0,
      "step": 76015
    },
    {
      "entropy": 5.683056116104126,
      "epoch": 6.895863570391873,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015040929509528434,
      "loss": 4.6959,
      "mean_token_accuracy": 0.24879397451877594,
      "num_tokens": 157755139.0,
      "step": 76020
    },
    {
      "entropy": 5.778449583053589,
      "epoch": 6.896317126269956,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015038235345253562,
      "loss": 4.8852,
      "mean_token_accuracy": 0.226531545817852,
      "num_tokens": 157765373.0,
      "step": 76025
    },
    {
      "entropy": 5.707099914550781,
      "epoch": 6.89677068214804,
      "grad_norm": 1.59375,
      "learning_rate": 0.00015035541438691163,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23300256878137587,
      "num_tokens": 157775172.0,
      "step": 76030
    },
    {
      "entropy": 5.728142976760864,
      "epoch": 6.897224238026125,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015032847789896937,
      "loss": 4.9364,
      "mean_token_accuracy": 0.22214571982622147,
      "num_tokens": 157785003.0,
      "step": 76035
    },
    {
      "entropy": 5.7248101234436035,
      "epoch": 6.897677793904209,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015030154398926602,
      "loss": 4.789,
      "mean_token_accuracy": 0.2436317801475525,
      "num_tokens": 157794413.0,
      "step": 76040
    },
    {
      "entropy": 5.718546628952026,
      "epoch": 6.898131349782293,
      "grad_norm": 1.5,
      "learning_rate": 0.0001502746126583585,
      "loss": 4.881,
      "mean_token_accuracy": 0.23377985507249832,
      "num_tokens": 157804980.0,
      "step": 76045
    },
    {
      "entropy": 5.677448844909668,
      "epoch": 6.898584905660377,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015024768390680383,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23791811913251876,
      "num_tokens": 157815373.0,
      "step": 76050
    },
    {
      "entropy": 5.674863290786743,
      "epoch": 6.899038461538462,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015022075773515886,
      "loss": 4.8258,
      "mean_token_accuracy": 0.2324338138103485,
      "num_tokens": 157825890.0,
      "step": 76055
    },
    {
      "entropy": 5.6822192668914795,
      "epoch": 6.899492017416546,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001501938341439803,
      "loss": 4.8834,
      "mean_token_accuracy": 0.2267366498708725,
      "num_tokens": 157836714.0,
      "step": 76060
    },
    {
      "entropy": 5.6985588550567625,
      "epoch": 6.89994557329463,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015016691313382514,
      "loss": 4.7688,
      "mean_token_accuracy": 0.23697636574506759,
      "num_tokens": 157845702.0,
      "step": 76065
    },
    {
      "entropy": 5.6694519996643065,
      "epoch": 6.900399129172714,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015013999470524998,
      "loss": 4.787,
      "mean_token_accuracy": 0.23704492747783662,
      "num_tokens": 157856210.0,
      "step": 76070
    },
    {
      "entropy": 5.640703105926514,
      "epoch": 6.900852685050799,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001501130788588116,
      "loss": 4.7118,
      "mean_token_accuracy": 0.24408518224954606,
      "num_tokens": 157865899.0,
      "step": 76075
    },
    {
      "entropy": 5.633013439178467,
      "epoch": 6.901306240928882,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015008616559506653,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23861388564109803,
      "num_tokens": 157875564.0,
      "step": 76080
    },
    {
      "entropy": 5.688043260574341,
      "epoch": 6.901759796806966,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001500592549145714,
      "loss": 4.7791,
      "mean_token_accuracy": 0.23313928097486497,
      "num_tokens": 157886634.0,
      "step": 76085
    },
    {
      "entropy": 5.707588005065918,
      "epoch": 6.902213352685051,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015003234681788274,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2334731325507164,
      "num_tokens": 157898104.0,
      "step": 76090
    },
    {
      "entropy": 5.733751392364502,
      "epoch": 6.902666908563135,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015000544130555691,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23385411649942398,
      "num_tokens": 157908400.0,
      "step": 76095
    },
    {
      "entropy": 5.6789412021636965,
      "epoch": 6.903120464441219,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00014997853837815046,
      "loss": 4.805,
      "mean_token_accuracy": 0.23200196027755737,
      "num_tokens": 157917568.0,
      "step": 76100
    },
    {
      "entropy": 5.6961876392364506,
      "epoch": 6.9035740203193035,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001499516380362197,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23469828963279724,
      "num_tokens": 157928546.0,
      "step": 76105
    },
    {
      "entropy": 5.709724092483521,
      "epoch": 6.904027576197388,
      "grad_norm": 1.65625,
      "learning_rate": 0.000149924740280321,
      "loss": 4.7796,
      "mean_token_accuracy": 0.23202790468931198,
      "num_tokens": 157939137.0,
      "step": 76110
    },
    {
      "entropy": 5.689351892471313,
      "epoch": 6.904481132075472,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014989784511101046,
      "loss": 4.7886,
      "mean_token_accuracy": 0.23624926209449768,
      "num_tokens": 157949460.0,
      "step": 76115
    },
    {
      "entropy": 5.6939929008483885,
      "epoch": 6.904934687953556,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014987095252884437,
      "loss": 4.7439,
      "mean_token_accuracy": 0.24244786947965621,
      "num_tokens": 157959163.0,
      "step": 76120
    },
    {
      "entropy": 5.754985666275024,
      "epoch": 6.90538824383164,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001498440625343789,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22283972948789596,
      "num_tokens": 157968911.0,
      "step": 76125
    },
    {
      "entropy": 5.674752998352051,
      "epoch": 6.905841799709724,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014981717512817,
      "loss": 4.8452,
      "mean_token_accuracy": 0.22574277520179747,
      "num_tokens": 157979252.0,
      "step": 76130
    },
    {
      "entropy": 5.711921262741089,
      "epoch": 6.906295355587808,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014979029031077392,
      "loss": 4.8266,
      "mean_token_accuracy": 0.23306595385074616,
      "num_tokens": 157990721.0,
      "step": 76135
    },
    {
      "entropy": 5.660438299179077,
      "epoch": 6.906748911465892,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014976340808274655,
      "loss": 4.7616,
      "mean_token_accuracy": 0.24342960715293885,
      "num_tokens": 158000555.0,
      "step": 76140
    },
    {
      "entropy": 5.756989479064941,
      "epoch": 6.907202467343977,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014973652844464382,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22138986140489578,
      "num_tokens": 158011175.0,
      "step": 76145
    },
    {
      "entropy": 5.697189474105835,
      "epoch": 6.907656023222061,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001497096513970216,
      "loss": 4.732,
      "mean_token_accuracy": 0.2516694128513336,
      "num_tokens": 158021598.0,
      "step": 76150
    },
    {
      "entropy": 5.667864608764648,
      "epoch": 6.908109579100145,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001496827769404357,
      "loss": 4.7729,
      "mean_token_accuracy": 0.23864722102880478,
      "num_tokens": 158031810.0,
      "step": 76155
    },
    {
      "entropy": 5.7634093284606935,
      "epoch": 6.9085631349782295,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014965590507544192,
      "loss": 4.847,
      "mean_token_accuracy": 0.2409600079059601,
      "num_tokens": 158042917.0,
      "step": 76160
    },
    {
      "entropy": 5.682668256759643,
      "epoch": 6.909016690856314,
      "grad_norm": 1.5,
      "learning_rate": 0.00014962903580259586,
      "loss": 4.7543,
      "mean_token_accuracy": 0.23359209150075913,
      "num_tokens": 158053222.0,
      "step": 76165
    },
    {
      "entropy": 5.749226427078247,
      "epoch": 6.909470246734398,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014960216912245338,
      "loss": 4.873,
      "mean_token_accuracy": 0.2314084306359291,
      "num_tokens": 158062700.0,
      "step": 76170
    },
    {
      "entropy": 5.68459997177124,
      "epoch": 6.909923802612482,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014957530503556998,
      "loss": 4.7893,
      "mean_token_accuracy": 0.2405149146914482,
      "num_tokens": 158072486.0,
      "step": 76175
    },
    {
      "entropy": 5.620491933822632,
      "epoch": 6.910377358490566,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014954844354250128,
      "loss": 4.7276,
      "mean_token_accuracy": 0.24873732179403304,
      "num_tokens": 158082105.0,
      "step": 76180
    },
    {
      "entropy": 5.697792100906372,
      "epoch": 6.91083091436865,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00014952158464380272,
      "loss": 4.8291,
      "mean_token_accuracy": 0.24105567038059234,
      "num_tokens": 158092355.0,
      "step": 76185
    },
    {
      "entropy": 5.728909111022949,
      "epoch": 6.911284470246734,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014949472834002974,
      "loss": 4.8918,
      "mean_token_accuracy": 0.22808912247419358,
      "num_tokens": 158101962.0,
      "step": 76190
    },
    {
      "entropy": 5.7099309921264645,
      "epoch": 6.9117380261248185,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014946787463173772,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2256258472800255,
      "num_tokens": 158112256.0,
      "step": 76195
    },
    {
      "entropy": 5.673152112960816,
      "epoch": 6.912191582002903,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001494410235194821,
      "loss": 4.8304,
      "mean_token_accuracy": 0.23183291852474214,
      "num_tokens": 158122573.0,
      "step": 76200
    },
    {
      "entropy": 5.6773933410644535,
      "epoch": 6.912645137880987,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0001494141750038181,
      "loss": 4.7409,
      "mean_token_accuracy": 0.23788266777992248,
      "num_tokens": 158132974.0,
      "step": 76205
    },
    {
      "entropy": 5.743388080596924,
      "epoch": 6.913098693759071,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014938732908530096,
      "loss": 4.8403,
      "mean_token_accuracy": 0.23754567950963973,
      "num_tokens": 158144069.0,
      "step": 76210
    },
    {
      "entropy": 5.640921545028687,
      "epoch": 6.9135522496371555,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014936048576448592,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23870235830545425,
      "num_tokens": 158154609.0,
      "step": 76215
    },
    {
      "entropy": 5.6683728218078615,
      "epoch": 6.91400580551524,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014933364504192797,
      "loss": 4.8105,
      "mean_token_accuracy": 0.2312212437391281,
      "num_tokens": 158165248.0,
      "step": 76220
    },
    {
      "entropy": 5.739650774002075,
      "epoch": 6.914459361393323,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014930680691818232,
      "loss": 4.9294,
      "mean_token_accuracy": 0.21923813223838806,
      "num_tokens": 158175385.0,
      "step": 76225
    },
    {
      "entropy": 5.691922760009765,
      "epoch": 6.914912917271408,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014927997139380385,
      "loss": 4.7967,
      "mean_token_accuracy": 0.2252727657556534,
      "num_tokens": 158186513.0,
      "step": 76230
    },
    {
      "entropy": 5.676064777374267,
      "epoch": 6.915366473149492,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014925313846934768,
      "loss": 4.8066,
      "mean_token_accuracy": 0.22816089391708375,
      "num_tokens": 158197050.0,
      "step": 76235
    },
    {
      "entropy": 5.777716398239136,
      "epoch": 6.915820029027576,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014922630814536865,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2284727767109871,
      "num_tokens": 158207341.0,
      "step": 76240
    },
    {
      "entropy": 5.736528158187866,
      "epoch": 6.91627358490566,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001491994804224216,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23802590519189834,
      "num_tokens": 158217088.0,
      "step": 76245
    },
    {
      "entropy": 5.774774694442749,
      "epoch": 6.9167271407837445,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014917265530106137,
      "loss": 4.9075,
      "mean_token_accuracy": 0.23363739252090454,
      "num_tokens": 158227841.0,
      "step": 76250
    },
    {
      "entropy": 5.6891858100891115,
      "epoch": 6.917180696661829,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001491458327818427,
      "loss": 4.8302,
      "mean_token_accuracy": 0.2287922188639641,
      "num_tokens": 158238540.0,
      "step": 76255
    },
    {
      "entropy": 5.692145299911499,
      "epoch": 6.917634252539913,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014911901286532027,
      "loss": 4.835,
      "mean_token_accuracy": 0.23702218979597092,
      "num_tokens": 158248563.0,
      "step": 76260
    },
    {
      "entropy": 5.683225297927857,
      "epoch": 6.918087808417997,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014909219555204862,
      "loss": 4.8054,
      "mean_token_accuracy": 0.23356909155845643,
      "num_tokens": 158259100.0,
      "step": 76265
    },
    {
      "entropy": 5.68816442489624,
      "epoch": 6.918541364296082,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014906538084258258,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22586294412612914,
      "num_tokens": 158269506.0,
      "step": 76270
    },
    {
      "entropy": 5.76298155784607,
      "epoch": 6.918994920174166,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014903856873747654,
      "loss": 4.8377,
      "mean_token_accuracy": 0.23851780444383622,
      "num_tokens": 158279675.0,
      "step": 76275
    },
    {
      "entropy": 5.699305200576783,
      "epoch": 6.919448476052249,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000149011759237285,
      "loss": 4.7579,
      "mean_token_accuracy": 0.23850862234830855,
      "num_tokens": 158289803.0,
      "step": 76280
    },
    {
      "entropy": 5.763399744033814,
      "epoch": 6.9199020319303335,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014898495234256238,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22543326020240784,
      "num_tokens": 158300453.0,
      "step": 76285
    },
    {
      "entropy": 5.675444221496582,
      "epoch": 6.920355587808418,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014895814805386307,
      "loss": 4.7352,
      "mean_token_accuracy": 0.23817292600870132,
      "num_tokens": 158310189.0,
      "step": 76290
    },
    {
      "entropy": 5.731135416030884,
      "epoch": 6.920809143686502,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014893134637174136,
      "loss": 4.8446,
      "mean_token_accuracy": 0.232140351831913,
      "num_tokens": 158321507.0,
      "step": 76295
    },
    {
      "entropy": 5.68493332862854,
      "epoch": 6.921262699564586,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014890454729675145,
      "loss": 4.7941,
      "mean_token_accuracy": 0.23659005463123323,
      "num_tokens": 158331524.0,
      "step": 76300
    },
    {
      "entropy": 5.73561110496521,
      "epoch": 6.9217162554426706,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00014887775082944772,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23569656908512115,
      "num_tokens": 158341582.0,
      "step": 76305
    },
    {
      "entropy": 5.704713296890259,
      "epoch": 6.922169811320755,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014885095697038424,
      "loss": 4.7875,
      "mean_token_accuracy": 0.2343028962612152,
      "num_tokens": 158351752.0,
      "step": 76310
    },
    {
      "entropy": 5.715418338775635,
      "epoch": 6.922623367198839,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014882416572011516,
      "loss": 4.888,
      "mean_token_accuracy": 0.22414685487747193,
      "num_tokens": 158362144.0,
      "step": 76315
    },
    {
      "entropy": 5.77678804397583,
      "epoch": 6.923076923076923,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014879737707919444,
      "loss": 4.9164,
      "mean_token_accuracy": 0.22367384880781174,
      "num_tokens": 158372942.0,
      "step": 76320
    },
    {
      "entropy": 5.702429866790771,
      "epoch": 6.923530478955008,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014877059104817614,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2351216122508049,
      "num_tokens": 158383929.0,
      "step": 76325
    },
    {
      "entropy": 5.7126220703125,
      "epoch": 6.923984034833092,
      "grad_norm": 1.6875,
      "learning_rate": 0.00014874380762761418,
      "loss": 4.8183,
      "mean_token_accuracy": 0.24049172699451446,
      "num_tokens": 158394240.0,
      "step": 76330
    },
    {
      "entropy": 5.652669954299927,
      "epoch": 6.924437590711175,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014871702681806237,
      "loss": 4.8326,
      "mean_token_accuracy": 0.23282129615545272,
      "num_tokens": 158404973.0,
      "step": 76335
    },
    {
      "entropy": 5.70833191871643,
      "epoch": 6.9248911465892595,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00014869024862007468,
      "loss": 4.8563,
      "mean_token_accuracy": 0.23471540361642837,
      "num_tokens": 158415321.0,
      "step": 76340
    },
    {
      "entropy": 5.825147724151611,
      "epoch": 6.925344702467344,
      "grad_norm": 1.375,
      "learning_rate": 0.0001486634730342049,
      "loss": 4.9842,
      "mean_token_accuracy": 0.22692115753889083,
      "num_tokens": 158427725.0,
      "step": 76345
    },
    {
      "entropy": 5.69606647491455,
      "epoch": 6.925798258345428,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014863670006100665,
      "loss": 4.7956,
      "mean_token_accuracy": 0.24295108020305634,
      "num_tokens": 158439629.0,
      "step": 76350
    },
    {
      "entropy": 5.688523387908935,
      "epoch": 6.926251814223512,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014860992970103365,
      "loss": 4.7603,
      "mean_token_accuracy": 0.23783452212810516,
      "num_tokens": 158449160.0,
      "step": 76355
    },
    {
      "entropy": 5.657704401016235,
      "epoch": 6.926705370101597,
      "grad_norm": 1.375,
      "learning_rate": 0.0001485831619548395,
      "loss": 4.7558,
      "mean_token_accuracy": 0.2344096690416336,
      "num_tokens": 158459786.0,
      "step": 76360
    },
    {
      "entropy": 5.7135124683380125,
      "epoch": 6.927158925979681,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001485563968229778,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23299406468868256,
      "num_tokens": 158470243.0,
      "step": 76365
    },
    {
      "entropy": 5.666988325119019,
      "epoch": 6.927612481857765,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001485296343060019,
      "loss": 4.7711,
      "mean_token_accuracy": 0.23287459909915925,
      "num_tokens": 158481544.0,
      "step": 76370
    },
    {
      "entropy": 5.700666189193726,
      "epoch": 6.928066037735849,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014850287440446554,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23242850303649903,
      "num_tokens": 158491861.0,
      "step": 76375
    },
    {
      "entropy": 5.665220260620117,
      "epoch": 6.928519593613933,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014847611711892196,
      "loss": 4.8425,
      "mean_token_accuracy": 0.2277361586689949,
      "num_tokens": 158501711.0,
      "step": 76380
    },
    {
      "entropy": 5.70940842628479,
      "epoch": 6.928973149492017,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014844936244992447,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22659782618284224,
      "num_tokens": 158511297.0,
      "step": 76385
    },
    {
      "entropy": 5.6767822265625,
      "epoch": 6.929426705370101,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014842261039802645,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24181401878595352,
      "num_tokens": 158521660.0,
      "step": 76390
    },
    {
      "entropy": 5.78272852897644,
      "epoch": 6.929880261248186,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001483958609637811,
      "loss": 4.8892,
      "mean_token_accuracy": 0.23276086300611495,
      "num_tokens": 158531027.0,
      "step": 76395
    },
    {
      "entropy": 5.681257343292236,
      "epoch": 6.93033381712627,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014836911414774155,
      "loss": 4.8098,
      "mean_token_accuracy": 0.23330843448638916,
      "num_tokens": 158542970.0,
      "step": 76400
    },
    {
      "entropy": 5.6810160160064695,
      "epoch": 6.930787373004354,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014834236995046104,
      "loss": 4.772,
      "mean_token_accuracy": 0.23482905924320222,
      "num_tokens": 158554244.0,
      "step": 76405
    },
    {
      "entropy": 5.711627197265625,
      "epoch": 6.931240928882438,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001483156283724926,
      "loss": 4.7947,
      "mean_token_accuracy": 0.2346203938126564,
      "num_tokens": 158563480.0,
      "step": 76410
    },
    {
      "entropy": 5.686605978012085,
      "epoch": 6.931694484760523,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014828888941438928,
      "loss": 4.7695,
      "mean_token_accuracy": 0.23709930032491683,
      "num_tokens": 158572366.0,
      "step": 76415
    },
    {
      "entropy": 5.621817636489868,
      "epoch": 6.932148040638607,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014826215307670398,
      "loss": 4.7232,
      "mean_token_accuracy": 0.24462803453207016,
      "num_tokens": 158583040.0,
      "step": 76420
    },
    {
      "entropy": 5.6305724620819095,
      "epoch": 6.932601596516691,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014823541935998964,
      "loss": 4.7814,
      "mean_token_accuracy": 0.24315789192914963,
      "num_tokens": 158593455.0,
      "step": 76425
    },
    {
      "entropy": 5.689174509048462,
      "epoch": 6.933055152394775,
      "grad_norm": 1.640625,
      "learning_rate": 0.00014820868826479917,
      "loss": 4.8345,
      "mean_token_accuracy": 0.22983923107385634,
      "num_tokens": 158603841.0,
      "step": 76430
    },
    {
      "entropy": 5.736449432373047,
      "epoch": 6.933508708272859,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014818195979168525,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22880935668945312,
      "num_tokens": 158614428.0,
      "step": 76435
    },
    {
      "entropy": 5.721867322921753,
      "epoch": 6.933962264150943,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001481552339412008,
      "loss": 4.888,
      "mean_token_accuracy": 0.22532823085784912,
      "num_tokens": 158624763.0,
      "step": 76440
    },
    {
      "entropy": 5.737617921829224,
      "epoch": 6.934415820029027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014812851071389837,
      "loss": 4.8833,
      "mean_token_accuracy": 0.222005595266819,
      "num_tokens": 158634821.0,
      "step": 76445
    },
    {
      "entropy": 5.751344156265259,
      "epoch": 6.934869375907112,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014810179011033077,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2379043310880661,
      "num_tokens": 158645880.0,
      "step": 76450
    },
    {
      "entropy": 5.778971576690674,
      "epoch": 6.935322931785196,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014807507213105038,
      "loss": 4.8798,
      "mean_token_accuracy": 0.2414841189980507,
      "num_tokens": 158656176.0,
      "step": 76455
    },
    {
      "entropy": 5.731155729293823,
      "epoch": 6.93577648766328,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001480483567766099,
      "loss": 4.8264,
      "mean_token_accuracy": 0.23419745415449142,
      "num_tokens": 158666123.0,
      "step": 76460
    },
    {
      "entropy": 5.673603868484497,
      "epoch": 6.936230043541364,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001480216440475618,
      "loss": 4.7793,
      "mean_token_accuracy": 0.24153043031692506,
      "num_tokens": 158676431.0,
      "step": 76465
    },
    {
      "entropy": 5.620188617706299,
      "epoch": 6.936683599419449,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001479949339444583,
      "loss": 4.7378,
      "mean_token_accuracy": 0.2432994082570076,
      "num_tokens": 158685993.0,
      "step": 76470
    },
    {
      "entropy": 5.647276306152344,
      "epoch": 6.937137155297533,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014796822646785204,
      "loss": 4.753,
      "mean_token_accuracy": 0.2427660897374153,
      "num_tokens": 158696150.0,
      "step": 76475
    },
    {
      "entropy": 5.724577522277832,
      "epoch": 6.937590711175617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014794152161829523,
      "loss": 4.9305,
      "mean_token_accuracy": 0.22326965034008026,
      "num_tokens": 158707337.0,
      "step": 76480
    },
    {
      "entropy": 5.613260412216187,
      "epoch": 6.9380442670537015,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014791481939634022,
      "loss": 4.7367,
      "mean_token_accuracy": 0.2360865816473961,
      "num_tokens": 158717436.0,
      "step": 76485
    },
    {
      "entropy": 5.712132692337036,
      "epoch": 6.938497822931785,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00014788811980253906,
      "loss": 4.8213,
      "mean_token_accuracy": 0.2310455173254013,
      "num_tokens": 158727367.0,
      "step": 76490
    },
    {
      "entropy": 5.6974365234375,
      "epoch": 6.938951378809869,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014786142283744385,
      "loss": 4.8972,
      "mean_token_accuracy": 0.22981067895889282,
      "num_tokens": 158738276.0,
      "step": 76495
    },
    {
      "entropy": 5.701214218139649,
      "epoch": 6.939404934687953,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014783472850160695,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24211081266403198,
      "num_tokens": 158749035.0,
      "step": 76500
    },
    {
      "entropy": 5.725986814498901,
      "epoch": 6.939858490566038,
      "grad_norm": 1.5,
      "learning_rate": 0.00014780803679558026,
      "loss": 4.8447,
      "mean_token_accuracy": 0.23297507613897322,
      "num_tokens": 158759179.0,
      "step": 76505
    },
    {
      "entropy": 5.684626293182373,
      "epoch": 6.940312046444122,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014778134771991575,
      "loss": 4.8036,
      "mean_token_accuracy": 0.24483146518468857,
      "num_tokens": 158768692.0,
      "step": 76510
    },
    {
      "entropy": 5.675064754486084,
      "epoch": 6.940765602322206,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001477546612751654,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23931947648525237,
      "num_tokens": 158779801.0,
      "step": 76515
    },
    {
      "entropy": 5.628626918792724,
      "epoch": 6.9412191582002905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001477279774618811,
      "loss": 4.7693,
      "mean_token_accuracy": 0.23320643603801727,
      "num_tokens": 158789312.0,
      "step": 76520
    },
    {
      "entropy": 5.703303670883178,
      "epoch": 6.941672714078375,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014770129628061463,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22792602628469466,
      "num_tokens": 158799704.0,
      "step": 76525
    },
    {
      "entropy": 5.75102858543396,
      "epoch": 6.942126269956459,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001476746177319177,
      "loss": 4.8948,
      "mean_token_accuracy": 0.23614937961101531,
      "num_tokens": 158810050.0,
      "step": 76530
    },
    {
      "entropy": 5.679186010360718,
      "epoch": 6.942579825834542,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014764794181634222,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23621614575386046,
      "num_tokens": 158819770.0,
      "step": 76535
    },
    {
      "entropy": 5.756739807128906,
      "epoch": 6.943033381712627,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001476212685344398,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2316071406006813,
      "num_tokens": 158830588.0,
      "step": 76540
    },
    {
      "entropy": 5.676826620101929,
      "epoch": 6.943486937590711,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014759459788676193,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23100883066654204,
      "num_tokens": 158840191.0,
      "step": 76545
    },
    {
      "entropy": 5.71349196434021,
      "epoch": 6.943940493468795,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014756792987386032,
      "loss": 4.8397,
      "mean_token_accuracy": 0.2332492619752884,
      "num_tokens": 158850576.0,
      "step": 76550
    },
    {
      "entropy": 5.707177400588989,
      "epoch": 6.944394049346879,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014754126449628637,
      "loss": 4.8444,
      "mean_token_accuracy": 0.22816979736089707,
      "num_tokens": 158860406.0,
      "step": 76555
    },
    {
      "entropy": 5.701448822021485,
      "epoch": 6.944847605224964,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014751460175459153,
      "loss": 4.8417,
      "mean_token_accuracy": 0.23509809672832488,
      "num_tokens": 158870220.0,
      "step": 76560
    },
    {
      "entropy": 5.693501138687134,
      "epoch": 6.945301161103048,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001474879416493271,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22226351201534272,
      "num_tokens": 158880910.0,
      "step": 76565
    },
    {
      "entropy": 5.697499465942383,
      "epoch": 6.945754716981132,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014746128418104468,
      "loss": 4.8395,
      "mean_token_accuracy": 0.23064291179180146,
      "num_tokens": 158891782.0,
      "step": 76570
    },
    {
      "entropy": 5.766556644439698,
      "epoch": 6.9462082728592165,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014743462935029537,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2293320044875145,
      "num_tokens": 158901136.0,
      "step": 76575
    },
    {
      "entropy": 5.699216604232788,
      "epoch": 6.946661828737301,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00014740797715763043,
      "loss": 4.7357,
      "mean_token_accuracy": 0.24016833752393724,
      "num_tokens": 158910661.0,
      "step": 76580
    },
    {
      "entropy": 5.730390167236328,
      "epoch": 6.947115384615385,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014738132760360107,
      "loss": 4.811,
      "mean_token_accuracy": 0.23564220517873763,
      "num_tokens": 158920432.0,
      "step": 76585
    },
    {
      "entropy": 5.673194694519043,
      "epoch": 6.947568940493468,
      "grad_norm": 1.375,
      "learning_rate": 0.00014735468068875836,
      "loss": 4.7777,
      "mean_token_accuracy": 0.23122416138648988,
      "num_tokens": 158930481.0,
      "step": 76590
    },
    {
      "entropy": 5.6840739250183105,
      "epoch": 6.948022496371553,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014732803641365333,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22776216715574266,
      "num_tokens": 158940849.0,
      "step": 76595
    },
    {
      "entropy": 5.6503355503082275,
      "epoch": 6.948476052249637,
      "grad_norm": 1.4375,
      "learning_rate": 0.000147301394778837,
      "loss": 4.8024,
      "mean_token_accuracy": 0.23490055054426193,
      "num_tokens": 158951811.0,
      "step": 76600
    },
    {
      "entropy": 5.684177207946777,
      "epoch": 6.948929608127721,
      "grad_norm": 1.375,
      "learning_rate": 0.00014727475578486045,
      "loss": 4.8263,
      "mean_token_accuracy": 0.23089248836040496,
      "num_tokens": 158962620.0,
      "step": 76605
    },
    {
      "entropy": 5.6868492603302006,
      "epoch": 6.9493831640058055,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001472481194322745,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23908979296684266,
      "num_tokens": 158972711.0,
      "step": 76610
    },
    {
      "entropy": 5.653559160232544,
      "epoch": 6.94983671988389,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00014722148572163,
      "loss": 4.7611,
      "mean_token_accuracy": 0.24093095511198043,
      "num_tokens": 158982853.0,
      "step": 76615
    },
    {
      "entropy": 5.64142255783081,
      "epoch": 6.950290275761974,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014719485465347775,
      "loss": 4.8017,
      "mean_token_accuracy": 0.2382335975766182,
      "num_tokens": 158993476.0,
      "step": 76620
    },
    {
      "entropy": 5.717353582382202,
      "epoch": 6.950743831640058,
      "grad_norm": 1.75,
      "learning_rate": 0.00014716822622836846,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22858446836471558,
      "num_tokens": 159003940.0,
      "step": 76625
    },
    {
      "entropy": 5.710549974441529,
      "epoch": 6.9511973875181425,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001471416004468528,
      "loss": 4.765,
      "mean_token_accuracy": 0.23968655318021775,
      "num_tokens": 159013196.0,
      "step": 76630
    },
    {
      "entropy": 5.683499383926391,
      "epoch": 6.951650943396227,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00014711497730948136,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23193038552999495,
      "num_tokens": 159023726.0,
      "step": 76635
    },
    {
      "entropy": 5.669155025482178,
      "epoch": 6.952104499274311,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001470883568168049,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2358968064188957,
      "num_tokens": 159033929.0,
      "step": 76640
    },
    {
      "entropy": 5.77858715057373,
      "epoch": 6.9525580551523944,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014706173896937374,
      "loss": 4.9379,
      "mean_token_accuracy": 0.22115369737148285,
      "num_tokens": 159044461.0,
      "step": 76645
    },
    {
      "entropy": 5.703509902954101,
      "epoch": 6.953011611030479,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014703512376773847,
      "loss": 4.808,
      "mean_token_accuracy": 0.2346939578652382,
      "num_tokens": 159054541.0,
      "step": 76650
    },
    {
      "entropy": 5.715353631973267,
      "epoch": 6.953465166908563,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014700851121244945,
      "loss": 4.8254,
      "mean_token_accuracy": 0.2342309460043907,
      "num_tokens": 159064916.0,
      "step": 76655
    },
    {
      "entropy": 5.638476848602295,
      "epoch": 6.953918722786647,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014698190130405702,
      "loss": 4.817,
      "mean_token_accuracy": 0.23308888524770738,
      "num_tokens": 159076567.0,
      "step": 76660
    },
    {
      "entropy": 5.720667886734009,
      "epoch": 6.9543722786647315,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014695529404311152,
      "loss": 4.8646,
      "mean_token_accuracy": 0.2270863473415375,
      "num_tokens": 159087438.0,
      "step": 76665
    },
    {
      "entropy": 5.603707218170166,
      "epoch": 6.954825834542816,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014692868943016308,
      "loss": 4.6885,
      "mean_token_accuracy": 0.24973445683717727,
      "num_tokens": 159097390.0,
      "step": 76670
    },
    {
      "entropy": 5.732049560546875,
      "epoch": 6.9552793904209,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014690208746576203,
      "loss": 4.9326,
      "mean_token_accuracy": 0.22323010712862015,
      "num_tokens": 159108709.0,
      "step": 76675
    },
    {
      "entropy": 5.737214660644531,
      "epoch": 6.955732946298984,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014687548815045848,
      "loss": 4.829,
      "mean_token_accuracy": 0.2311860889196396,
      "num_tokens": 159119603.0,
      "step": 76680
    },
    {
      "entropy": 5.775939893722534,
      "epoch": 6.956186502177069,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014684889148480247,
      "loss": 4.8706,
      "mean_token_accuracy": 0.22807177007198334,
      "num_tokens": 159131064.0,
      "step": 76685
    },
    {
      "entropy": 5.687621784210205,
      "epoch": 6.956640058055152,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014682229746934406,
      "loss": 4.817,
      "mean_token_accuracy": 0.23079476058483123,
      "num_tokens": 159142306.0,
      "step": 76690
    },
    {
      "entropy": 5.673066329956055,
      "epoch": 6.957093613933236,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014679570610463317,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23792658299207686,
      "num_tokens": 159153192.0,
      "step": 76695
    },
    {
      "entropy": 5.608816385269165,
      "epoch": 6.9575471698113205,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014676911739121967,
      "loss": 4.7048,
      "mean_token_accuracy": 0.23769812434911727,
      "num_tokens": 159164349.0,
      "step": 76700
    },
    {
      "entropy": 5.672329092025757,
      "epoch": 6.958000725689405,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001467425313296536,
      "loss": 4.794,
      "mean_token_accuracy": 0.23494291454553604,
      "num_tokens": 159175220.0,
      "step": 76705
    },
    {
      "entropy": 5.701153087615967,
      "epoch": 6.958454281567489,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001467159479204847,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22357545047998428,
      "num_tokens": 159185848.0,
      "step": 76710
    },
    {
      "entropy": 5.640890598297119,
      "epoch": 6.958907837445573,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014668936716426262,
      "loss": 4.7668,
      "mean_token_accuracy": 0.23714106827974318,
      "num_tokens": 159196769.0,
      "step": 76715
    },
    {
      "entropy": 5.715071201324463,
      "epoch": 6.9593613933236576,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014666278906153718,
      "loss": 4.8683,
      "mean_token_accuracy": 0.23400251120328902,
      "num_tokens": 159207268.0,
      "step": 76720
    },
    {
      "entropy": 5.628355360031128,
      "epoch": 6.959814949201742,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00014663621361285796,
      "loss": 4.7729,
      "mean_token_accuracy": 0.25086259841918945,
      "num_tokens": 159217726.0,
      "step": 76725
    },
    {
      "entropy": 5.703917837142944,
      "epoch": 6.960268505079826,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001466096408187745,
      "loss": 4.7783,
      "mean_token_accuracy": 0.23624112606048583,
      "num_tokens": 159226336.0,
      "step": 76730
    },
    {
      "entropy": 5.724439668655395,
      "epoch": 6.96072206095791,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014658307067983635,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23477787673473358,
      "num_tokens": 159235925.0,
      "step": 76735
    },
    {
      "entropy": 5.702263069152832,
      "epoch": 6.961175616835995,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014655650319659311,
      "loss": 4.8494,
      "mean_token_accuracy": 0.22888152450323104,
      "num_tokens": 159246788.0,
      "step": 76740
    },
    {
      "entropy": 5.601063251495361,
      "epoch": 6.961629172714078,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001465299383695941,
      "loss": 4.7013,
      "mean_token_accuracy": 0.24817852377891542,
      "num_tokens": 159257453.0,
      "step": 76745
    },
    {
      "entropy": 5.723426818847656,
      "epoch": 6.962082728592162,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014650337619938873,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2240820825099945,
      "num_tokens": 159268037.0,
      "step": 76750
    },
    {
      "entropy": 5.726772737503052,
      "epoch": 6.9625362844702465,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014647681668652622,
      "loss": 4.9298,
      "mean_token_accuracy": 0.22432485669851304,
      "num_tokens": 159279449.0,
      "step": 76755
    },
    {
      "entropy": 5.644286632537842,
      "epoch": 6.962989840348331,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014645025983155597,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23646050393581391,
      "num_tokens": 159290383.0,
      "step": 76760
    },
    {
      "entropy": 5.705339622497559,
      "epoch": 6.963443396226415,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014642370563502704,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23833815455436708,
      "num_tokens": 159300426.0,
      "step": 76765
    },
    {
      "entropy": 5.599971342086792,
      "epoch": 6.963896952104499,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014639715409748858,
      "loss": 4.7087,
      "mean_token_accuracy": 0.24500664174556733,
      "num_tokens": 159311642.0,
      "step": 76770
    },
    {
      "entropy": 5.695181226730346,
      "epoch": 6.964350507982584,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014637060521948984,
      "loss": 4.8546,
      "mean_token_accuracy": 0.23326488733291625,
      "num_tokens": 159321907.0,
      "step": 76775
    },
    {
      "entropy": 5.667880392074585,
      "epoch": 6.964804063860668,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014634405900157977,
      "loss": 4.8493,
      "mean_token_accuracy": 0.22957399785518645,
      "num_tokens": 159332249.0,
      "step": 76780
    },
    {
      "entropy": 5.711263275146484,
      "epoch": 6.965257619738752,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014631751544430733,
      "loss": 4.8758,
      "mean_token_accuracy": 0.2270182564854622,
      "num_tokens": 159342099.0,
      "step": 76785
    },
    {
      "entropy": 5.683055210113525,
      "epoch": 6.9657111756168355,
      "grad_norm": 1.609375,
      "learning_rate": 0.00014629097454822143,
      "loss": 4.7465,
      "mean_token_accuracy": 0.241212959587574,
      "num_tokens": 159352249.0,
      "step": 76790
    },
    {
      "entropy": 5.7030140399932865,
      "epoch": 6.966164731494921,
      "grad_norm": 1.5,
      "learning_rate": 0.00014626443631387105,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2242736965417862,
      "num_tokens": 159362967.0,
      "step": 76795
    },
    {
      "entropy": 5.680172681808472,
      "epoch": 6.966618287373004,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014623790074180486,
      "loss": 4.7501,
      "mean_token_accuracy": 0.23519307523965835,
      "num_tokens": 159372943.0,
      "step": 76800
    },
    {
      "entropy": 5.71634202003479,
      "epoch": 6.967071843251088,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014621136783257165,
      "loss": 4.916,
      "mean_token_accuracy": 0.2322198659181595,
      "num_tokens": 159383669.0,
      "step": 76805
    },
    {
      "entropy": 5.7375099658966064,
      "epoch": 6.967525399129173,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014618483758672027,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2285226583480835,
      "num_tokens": 159394681.0,
      "step": 76810
    },
    {
      "entropy": 5.72533164024353,
      "epoch": 6.967978955007257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014615831000479922,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23087752163410186,
      "num_tokens": 159404732.0,
      "step": 76815
    },
    {
      "entropy": 5.7158221244812015,
      "epoch": 6.968432510885341,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014613178508735719,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23283831775188446,
      "num_tokens": 159416016.0,
      "step": 76820
    },
    {
      "entropy": 5.724056386947632,
      "epoch": 6.968886066763425,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001461052628349427,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22813523262739183,
      "num_tokens": 159427294.0,
      "step": 76825
    },
    {
      "entropy": 5.73108286857605,
      "epoch": 6.96933962264151,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001460787432481042,
      "loss": 4.8973,
      "mean_token_accuracy": 0.23326969891786575,
      "num_tokens": 159438394.0,
      "step": 76830
    },
    {
      "entropy": 5.723553609848023,
      "epoch": 6.969793178519594,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014605222632739016,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23744205087423326,
      "num_tokens": 159449052.0,
      "step": 76835
    },
    {
      "entropy": 5.705357646942138,
      "epoch": 6.970246734397678,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014602571207334884,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23834919184446335,
      "num_tokens": 159458960.0,
      "step": 76840
    },
    {
      "entropy": 5.672084665298462,
      "epoch": 6.9707002902757615,
      "grad_norm": 1.5,
      "learning_rate": 0.0001459992004865287,
      "loss": 4.7802,
      "mean_token_accuracy": 0.24325588643550872,
      "num_tokens": 159468936.0,
      "step": 76845
    },
    {
      "entropy": 5.676969051361084,
      "epoch": 6.971153846153846,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014597269156747805,
      "loss": 4.8423,
      "mean_token_accuracy": 0.23092751950025558,
      "num_tokens": 159478428.0,
      "step": 76850
    },
    {
      "entropy": 5.765127038955688,
      "epoch": 6.97160740203193,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014594618531674498,
      "loss": 4.9183,
      "mean_token_accuracy": 0.21732520908117295,
      "num_tokens": 159489876.0,
      "step": 76855
    },
    {
      "entropy": 5.767058515548706,
      "epoch": 6.972060957910014,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001459196817348777,
      "loss": 4.9545,
      "mean_token_accuracy": 0.22401095181703568,
      "num_tokens": 159500507.0,
      "step": 76860
    },
    {
      "entropy": 5.702939844131469,
      "epoch": 6.972514513788099,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014589318082242425,
      "loss": 4.7633,
      "mean_token_accuracy": 0.2412579208612442,
      "num_tokens": 159510299.0,
      "step": 76865
    },
    {
      "entropy": 5.681483602523803,
      "epoch": 6.972968069666183,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014586668257993276,
      "loss": 4.7782,
      "mean_token_accuracy": 0.23796294033527374,
      "num_tokens": 159520699.0,
      "step": 76870
    },
    {
      "entropy": 5.7198868751525875,
      "epoch": 6.973421625544267,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001458401870079511,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23405031114816666,
      "num_tokens": 159531413.0,
      "step": 76875
    },
    {
      "entropy": 5.627262115478516,
      "epoch": 6.973875181422351,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014581369410702734,
      "loss": 4.7151,
      "mean_token_accuracy": 0.23565645515918732,
      "num_tokens": 159542065.0,
      "step": 76880
    },
    {
      "entropy": 5.671534204483033,
      "epoch": 6.974328737300436,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014578720387770932,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23621074110269547,
      "num_tokens": 159552074.0,
      "step": 76885
    },
    {
      "entropy": 5.674719142913818,
      "epoch": 6.97478229317852,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014576071632054489,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23314507603645324,
      "num_tokens": 159563228.0,
      "step": 76890
    },
    {
      "entropy": 5.735728359222412,
      "epoch": 6.975235849056604,
      "grad_norm": 1.5,
      "learning_rate": 0.0001457342314360817,
      "loss": 4.8728,
      "mean_token_accuracy": 0.2343612626194954,
      "num_tokens": 159573997.0,
      "step": 76895
    },
    {
      "entropy": 5.702533388137818,
      "epoch": 6.975689404934688,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014570774922486763,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23464981615543365,
      "num_tokens": 159584418.0,
      "step": 76900
    },
    {
      "entropy": 5.751979541778565,
      "epoch": 6.976142960812772,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014568126968745012,
      "loss": 4.8502,
      "mean_token_accuracy": 0.2317182704806328,
      "num_tokens": 159595369.0,
      "step": 76905
    },
    {
      "entropy": 5.7295530319213865,
      "epoch": 6.976596516690856,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001456547928243771,
      "loss": 4.854,
      "mean_token_accuracy": 0.2264639139175415,
      "num_tokens": 159606568.0,
      "step": 76910
    },
    {
      "entropy": 5.612880897521973,
      "epoch": 6.97705007256894,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001456283186361958,
      "loss": 4.7494,
      "mean_token_accuracy": 0.24094678163528443,
      "num_tokens": 159616966.0,
      "step": 76915
    },
    {
      "entropy": 5.678699111938476,
      "epoch": 6.977503628447025,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001456018471234539,
      "loss": 4.7734,
      "mean_token_accuracy": 0.24948033988475798,
      "num_tokens": 159626450.0,
      "step": 76920
    },
    {
      "entropy": 5.779861831665039,
      "epoch": 6.977957184325109,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014557537828669873,
      "loss": 4.9104,
      "mean_token_accuracy": 0.23437164723873138,
      "num_tokens": 159636914.0,
      "step": 76925
    },
    {
      "entropy": 5.652892827987671,
      "epoch": 6.978410740203193,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014554891212647774,
      "loss": 4.827,
      "mean_token_accuracy": 0.23455643206834792,
      "num_tokens": 159648225.0,
      "step": 76930
    },
    {
      "entropy": 5.650096321105957,
      "epoch": 6.9788642960812775,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014552244864333814,
      "loss": 4.7331,
      "mean_token_accuracy": 0.24315284490585326,
      "num_tokens": 159658761.0,
      "step": 76935
    },
    {
      "entropy": 5.73770580291748,
      "epoch": 6.979317851959362,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014549598783782737,
      "loss": 4.8205,
      "mean_token_accuracy": 0.2282203271985054,
      "num_tokens": 159669176.0,
      "step": 76940
    },
    {
      "entropy": 5.699492692947388,
      "epoch": 6.979771407837445,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001454695297104926,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2268879383802414,
      "num_tokens": 159678720.0,
      "step": 76945
    },
    {
      "entropy": 5.715237426757812,
      "epoch": 6.98022496371553,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014544307426188094,
      "loss": 4.8595,
      "mean_token_accuracy": 0.23659369200468064,
      "num_tokens": 159687854.0,
      "step": 76950
    },
    {
      "entropy": 5.699179315567017,
      "epoch": 6.980678519593614,
      "grad_norm": 1.671875,
      "learning_rate": 0.00014541662149253953,
      "loss": 4.8281,
      "mean_token_accuracy": 0.2347450241446495,
      "num_tokens": 159697788.0,
      "step": 76955
    },
    {
      "entropy": 5.744288206100464,
      "epoch": 6.981132075471698,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001453901714030154,
      "loss": 4.8637,
      "mean_token_accuracy": 0.23207309544086457,
      "num_tokens": 159708050.0,
      "step": 76960
    },
    {
      "entropy": 5.592914485931397,
      "epoch": 6.981585631349782,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014536372399385556,
      "loss": 4.7032,
      "mean_token_accuracy": 0.24270872324705123,
      "num_tokens": 159718797.0,
      "step": 76965
    },
    {
      "entropy": 5.63022255897522,
      "epoch": 6.982039187227866,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014533727926560686,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2283049240708351,
      "num_tokens": 159728574.0,
      "step": 76970
    },
    {
      "entropy": 5.676463031768799,
      "epoch": 6.982492743105951,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014531083721881639,
      "loss": 4.8361,
      "mean_token_accuracy": 0.233115117251873,
      "num_tokens": 159739315.0,
      "step": 76975
    },
    {
      "entropy": 5.668244218826294,
      "epoch": 6.982946298984035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001452843978540308,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22862468957901,
      "num_tokens": 159750306.0,
      "step": 76980
    },
    {
      "entropy": 5.686702013015747,
      "epoch": 6.983399854862119,
      "grad_norm": 1.609375,
      "learning_rate": 0.00014525796117179692,
      "loss": 4.817,
      "mean_token_accuracy": 0.23278266787528992,
      "num_tokens": 159760396.0,
      "step": 76985
    },
    {
      "entropy": 5.7283789157867435,
      "epoch": 6.9838534107402035,
      "grad_norm": 1.578125,
      "learning_rate": 0.00014523152717266146,
      "loss": 4.8168,
      "mean_token_accuracy": 0.23116419315338135,
      "num_tokens": 159770781.0,
      "step": 76990
    },
    {
      "entropy": 5.71638765335083,
      "epoch": 6.984306966618288,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001452050958571711,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23398715555667876,
      "num_tokens": 159780379.0,
      "step": 76995
    },
    {
      "entropy": 5.747039699554444,
      "epoch": 6.984760522496371,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014517866722587238,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22499655187129974,
      "num_tokens": 159790949.0,
      "step": 77000
    },
    {
      "entropy": 5.706114053726196,
      "epoch": 6.985214078374455,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014515224127931197,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22878055572509765,
      "num_tokens": 159801398.0,
      "step": 77005
    },
    {
      "entropy": 5.708946704864502,
      "epoch": 6.98566763425254,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00014512581801803633,
      "loss": 4.8883,
      "mean_token_accuracy": 0.23182636946439744,
      "num_tokens": 159811552.0,
      "step": 77010
    },
    {
      "entropy": 5.7637533187866214,
      "epoch": 6.986121190130624,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014509939744259182,
      "loss": 4.8781,
      "mean_token_accuracy": 0.22669627666473388,
      "num_tokens": 159822204.0,
      "step": 77015
    },
    {
      "entropy": 5.687879610061645,
      "epoch": 6.986574746008708,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014507297955352488,
      "loss": 4.8177,
      "mean_token_accuracy": 0.23330639749765397,
      "num_tokens": 159832074.0,
      "step": 77020
    },
    {
      "entropy": 5.7152375221252445,
      "epoch": 6.9870283018867925,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014504656435138186,
      "loss": 4.7828,
      "mean_token_accuracy": 0.23973934054374696,
      "num_tokens": 159841721.0,
      "step": 77025
    },
    {
      "entropy": 5.69192590713501,
      "epoch": 6.987481857764877,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014502015183670897,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23935825228691102,
      "num_tokens": 159851963.0,
      "step": 77030
    },
    {
      "entropy": 5.731685590744019,
      "epoch": 6.987935413642961,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014499374201005239,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2328728973865509,
      "num_tokens": 159862345.0,
      "step": 77035
    },
    {
      "entropy": 5.706037092208862,
      "epoch": 6.988388969521045,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014496733487195845,
      "loss": 4.8403,
      "mean_token_accuracy": 0.23425789624452592,
      "num_tokens": 159873463.0,
      "step": 77040
    },
    {
      "entropy": 5.6950784683227536,
      "epoch": 6.9888425253991295,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014494093042297313,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23404417484998702,
      "num_tokens": 159884405.0,
      "step": 77045
    },
    {
      "entropy": 5.710197401046753,
      "epoch": 6.989296081277214,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014491452866364254,
      "loss": 4.8102,
      "mean_token_accuracy": 0.24009515941143036,
      "num_tokens": 159895226.0,
      "step": 77050
    },
    {
      "entropy": 5.689485645294189,
      "epoch": 6.989749637155297,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014488812959451264,
      "loss": 4.8218,
      "mean_token_accuracy": 0.23055947870016097,
      "num_tokens": 159905968.0,
      "step": 77055
    },
    {
      "entropy": 5.6871014595031735,
      "epoch": 6.9902031930333814,
      "grad_norm": 1.375,
      "learning_rate": 0.0001448617332161294,
      "loss": 4.8334,
      "mean_token_accuracy": 0.22408517599105834,
      "num_tokens": 159916251.0,
      "step": 77060
    },
    {
      "entropy": 5.656939935684204,
      "epoch": 6.990656748911466,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00014483533952903865,
      "loss": 4.7703,
      "mean_token_accuracy": 0.2392516851425171,
      "num_tokens": 159926867.0,
      "step": 77065
    },
    {
      "entropy": 5.716985034942627,
      "epoch": 6.99111030478955,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014480894853378617,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22502606809139253,
      "num_tokens": 159937489.0,
      "step": 77070
    },
    {
      "entropy": 5.662891817092896,
      "epoch": 6.991563860667634,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001447825602309179,
      "loss": 4.832,
      "mean_token_accuracy": 0.22929843366146088,
      "num_tokens": 159947190.0,
      "step": 77075
    },
    {
      "entropy": 5.689303922653198,
      "epoch": 6.9920174165457185,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014475617462097944,
      "loss": 4.7911,
      "mean_token_accuracy": 0.2344690144062042,
      "num_tokens": 159956757.0,
      "step": 77080
    },
    {
      "entropy": 5.6905721664428714,
      "epoch": 6.992470972423803,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014472979170451655,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23156612366437912,
      "num_tokens": 159967040.0,
      "step": 77085
    },
    {
      "entropy": 5.757380104064941,
      "epoch": 6.992924528301887,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014470341148207472,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2288462206721306,
      "num_tokens": 159977181.0,
      "step": 77090
    },
    {
      "entropy": 5.755026578903198,
      "epoch": 6.993378084179971,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00014467703395419952,
      "loss": 4.9017,
      "mean_token_accuracy": 0.23280181139707565,
      "num_tokens": 159986900.0,
      "step": 77095
    },
    {
      "entropy": 5.636216449737549,
      "epoch": 6.993831640058055,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014465065912143649,
      "loss": 4.7381,
      "mean_token_accuracy": 0.23681196570396423,
      "num_tokens": 159997165.0,
      "step": 77100
    },
    {
      "entropy": 5.6744036197662355,
      "epoch": 6.994285195936139,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000144624286984331,
      "loss": 4.7791,
      "mean_token_accuracy": 0.2277887850999832,
      "num_tokens": 160008283.0,
      "step": 77105
    },
    {
      "entropy": 5.672108125686646,
      "epoch": 6.994738751814223,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001445979175434285,
      "loss": 4.7446,
      "mean_token_accuracy": 0.23706294000148773,
      "num_tokens": 160017548.0,
      "step": 77110
    },
    {
      "entropy": 5.630478429794311,
      "epoch": 6.9951923076923075,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014457155079927432,
      "loss": 4.733,
      "mean_token_accuracy": 0.23587964475154877,
      "num_tokens": 160027472.0,
      "step": 77115
    },
    {
      "entropy": 5.686278581619263,
      "epoch": 6.995645863570392,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014454518675241374,
      "loss": 4.7969,
      "mean_token_accuracy": 0.2376437097787857,
      "num_tokens": 160037813.0,
      "step": 77120
    },
    {
      "entropy": 5.651417779922485,
      "epoch": 6.996099419448476,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014451882540339188,
      "loss": 4.7702,
      "mean_token_accuracy": 0.23536409735679625,
      "num_tokens": 160048030.0,
      "step": 77125
    },
    {
      "entropy": 5.7060840129852295,
      "epoch": 6.99655297532656,
      "grad_norm": 1.59375,
      "learning_rate": 0.00014449246675275399,
      "loss": 4.8309,
      "mean_token_accuracy": 0.230994114279747,
      "num_tokens": 160058735.0,
      "step": 77130
    },
    {
      "entropy": 5.704449272155761,
      "epoch": 6.9970065312046446,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014446611080104515,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23592915534973144,
      "num_tokens": 160068922.0,
      "step": 77135
    },
    {
      "entropy": 5.681045722961426,
      "epoch": 6.997460087082729,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014443975754881033,
      "loss": 4.7709,
      "mean_token_accuracy": 0.2425166606903076,
      "num_tokens": 160079672.0,
      "step": 77140
    },
    {
      "entropy": 5.74401969909668,
      "epoch": 6.997913642960813,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014441340699659466,
      "loss": 4.9556,
      "mean_token_accuracy": 0.21655215173959733,
      "num_tokens": 160090709.0,
      "step": 77145
    },
    {
      "entropy": 5.7140968322753904,
      "epoch": 6.998367198838897,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000144387059144943,
      "loss": 4.8526,
      "mean_token_accuracy": 0.23426254987716674,
      "num_tokens": 160101651.0,
      "step": 77150
    },
    {
      "entropy": 5.665977954864502,
      "epoch": 6.998820754716981,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014436071399440026,
      "loss": 4.8312,
      "mean_token_accuracy": 0.2310844913125038,
      "num_tokens": 160112009.0,
      "step": 77155
    },
    {
      "entropy": 5.75609827041626,
      "epoch": 6.999274310595065,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014433437154551124,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23846462815999986,
      "num_tokens": 160122594.0,
      "step": 77160
    },
    {
      "entropy": 5.737393712997436,
      "epoch": 6.999727866473149,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001443080317988207,
      "loss": 4.9024,
      "mean_token_accuracy": 0.23273326605558395,
      "num_tokens": 160134872.0,
      "step": 77165
    },
    {
      "entropy": 5.701623058319091,
      "epoch": 7.0001814223512335,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001442816947548734,
      "loss": 4.8101,
      "mean_token_accuracy": 0.23883077949285508,
      "num_tokens": 160144387.0,
      "step": 77170
    },
    {
      "entropy": 5.701907110214234,
      "epoch": 7.000634978229318,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00014425536041421386,
      "loss": 4.7841,
      "mean_token_accuracy": 0.24601631611585617,
      "num_tokens": 160155346.0,
      "step": 77175
    },
    {
      "entropy": 5.742257738113404,
      "epoch": 7.001088534107402,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014422902877738686,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2364062935113907,
      "num_tokens": 160166865.0,
      "step": 77180
    },
    {
      "entropy": 5.657144784927368,
      "epoch": 7.001542089985486,
      "grad_norm": 1.375,
      "learning_rate": 0.0001442026998449369,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23891108781099318,
      "num_tokens": 160177385.0,
      "step": 77185
    },
    {
      "entropy": 5.701470899581909,
      "epoch": 7.001995645863571,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001441763736174084,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23393601030111313,
      "num_tokens": 160187631.0,
      "step": 77190
    },
    {
      "entropy": 5.712139749526978,
      "epoch": 7.002449201741655,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014415005009534587,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2266813412308693,
      "num_tokens": 160197188.0,
      "step": 77195
    },
    {
      "entropy": 5.76731686592102,
      "epoch": 7.002902757619739,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014412372927929363,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2290873944759369,
      "num_tokens": 160207598.0,
      "step": 77200
    },
    {
      "entropy": 5.749020004272461,
      "epoch": 7.0033563134978225,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014409741116979592,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23694383651018142,
      "num_tokens": 160219106.0,
      "step": 77205
    },
    {
      "entropy": 5.6265359878540036,
      "epoch": 7.003809869375907,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014407109576739722,
      "loss": 4.7194,
      "mean_token_accuracy": 0.24236143082380296,
      "num_tokens": 160229027.0,
      "step": 77210
    },
    {
      "entropy": 5.687509679794312,
      "epoch": 7.004263425253991,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014404478307264158,
      "loss": 4.7867,
      "mean_token_accuracy": 0.23127447068691254,
      "num_tokens": 160239606.0,
      "step": 77215
    },
    {
      "entropy": 5.709599542617798,
      "epoch": 7.004716981132075,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014401847308607323,
      "loss": 4.7866,
      "mean_token_accuracy": 0.23735120445489882,
      "num_tokens": 160250182.0,
      "step": 77220
    },
    {
      "entropy": 5.719716548919678,
      "epoch": 7.00517053701016,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001439921658082362,
      "loss": 4.8815,
      "mean_token_accuracy": 0.2329138159751892,
      "num_tokens": 160260028.0,
      "step": 77225
    },
    {
      "entropy": 5.692617177963257,
      "epoch": 7.005624092888244,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014396586123967465,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2314023643732071,
      "num_tokens": 160269903.0,
      "step": 77230
    },
    {
      "entropy": 5.626556873321533,
      "epoch": 7.006077648766328,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00014393955938093238,
      "loss": 4.6868,
      "mean_token_accuracy": 0.24092392772436141,
      "num_tokens": 160280051.0,
      "step": 77235
    },
    {
      "entropy": 5.684554672241211,
      "epoch": 7.006531204644412,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001439132602325534,
      "loss": 4.8043,
      "mean_token_accuracy": 0.2386559396982193,
      "num_tokens": 160290110.0,
      "step": 77240
    },
    {
      "entropy": 5.681786441802979,
      "epoch": 7.006984760522497,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001438869637950817,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2392431914806366,
      "num_tokens": 160300118.0,
      "step": 77245
    },
    {
      "entropy": 5.733400011062622,
      "epoch": 7.007438316400581,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014386067006906094,
      "loss": 4.8684,
      "mean_token_accuracy": 0.23305837512016297,
      "num_tokens": 160309035.0,
      "step": 77250
    },
    {
      "entropy": 5.714116430282592,
      "epoch": 7.007891872278665,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000143834379055035,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23599157929420472,
      "num_tokens": 160318547.0,
      "step": 77255
    },
    {
      "entropy": 5.679661273956299,
      "epoch": 7.0083454281567485,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001438080907535475,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22858427315950394,
      "num_tokens": 160327846.0,
      "step": 77260
    },
    {
      "entropy": 5.692919635772705,
      "epoch": 7.008798984034833,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014378180516514208,
      "loss": 4.7656,
      "mean_token_accuracy": 0.23855527937412263,
      "num_tokens": 160337890.0,
      "step": 77265
    },
    {
      "entropy": 5.697426223754883,
      "epoch": 7.009252539912917,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014375552229036238,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23667207062244416,
      "num_tokens": 160348932.0,
      "step": 77270
    },
    {
      "entropy": 5.698075819015503,
      "epoch": 7.009706095791001,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001437292421297519,
      "loss": 4.7768,
      "mean_token_accuracy": 0.23590523302555083,
      "num_tokens": 160358919.0,
      "step": 77275
    },
    {
      "entropy": 5.671058607101441,
      "epoch": 7.010159651669086,
      "grad_norm": 1.5,
      "learning_rate": 0.00014370296468385415,
      "loss": 4.7876,
      "mean_token_accuracy": 0.23118586242198944,
      "num_tokens": 160369418.0,
      "step": 77280
    },
    {
      "entropy": 5.676950693130493,
      "epoch": 7.01061320754717,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001436766899532126,
      "loss": 4.7356,
      "mean_token_accuracy": 0.24349166601896285,
      "num_tokens": 160379003.0,
      "step": 77285
    },
    {
      "entropy": 5.652704668045044,
      "epoch": 7.011066763425254,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014365041793837057,
      "loss": 4.7863,
      "mean_token_accuracy": 0.23529667556285858,
      "num_tokens": 160389784.0,
      "step": 77290
    },
    {
      "entropy": 5.721407175064087,
      "epoch": 7.011520319303338,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014362414863987135,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22506805509328842,
      "num_tokens": 160400678.0,
      "step": 77295
    },
    {
      "entropy": 5.704273748397827,
      "epoch": 7.011973875181423,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001435978820582582,
      "loss": 4.8423,
      "mean_token_accuracy": 0.23029846847057342,
      "num_tokens": 160412607.0,
      "step": 77300
    },
    {
      "entropy": 5.663089227676392,
      "epoch": 7.012427431059507,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014357161819407435,
      "loss": 4.764,
      "mean_token_accuracy": 0.2360628843307495,
      "num_tokens": 160422472.0,
      "step": 77305
    },
    {
      "entropy": 5.716339111328125,
      "epoch": 7.01288098693759,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014354535704786287,
      "loss": 4.7818,
      "mean_token_accuracy": 0.23729650527238846,
      "num_tokens": 160431990.0,
      "step": 77310
    },
    {
      "entropy": 5.666554164886475,
      "epoch": 7.013334542815675,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014351909862016693,
      "loss": 4.7623,
      "mean_token_accuracy": 0.22992208749055862,
      "num_tokens": 160441719.0,
      "step": 77315
    },
    {
      "entropy": 5.598827123641968,
      "epoch": 7.013788098693759,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014349284291152958,
      "loss": 4.741,
      "mean_token_accuracy": 0.2415900006890297,
      "num_tokens": 160452501.0,
      "step": 77320
    },
    {
      "entropy": 5.630956649780273,
      "epoch": 7.014241654571843,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014346658992249372,
      "loss": 4.7228,
      "mean_token_accuracy": 0.24788628667593002,
      "num_tokens": 160462542.0,
      "step": 77325
    },
    {
      "entropy": 5.643096780776977,
      "epoch": 7.014695210449927,
      "grad_norm": 1.6875,
      "learning_rate": 0.00014344033965360233,
      "loss": 4.7767,
      "mean_token_accuracy": 0.23530797958374022,
      "num_tokens": 160472897.0,
      "step": 77330
    },
    {
      "entropy": 5.672302913665772,
      "epoch": 7.015148766328012,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014341409210539825,
      "loss": 4.7828,
      "mean_token_accuracy": 0.24117082804441453,
      "num_tokens": 160483152.0,
      "step": 77335
    },
    {
      "entropy": 5.708800983428955,
      "epoch": 7.015602322206096,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014338784727842426,
      "loss": 4.8156,
      "mean_token_accuracy": 0.22933336794376374,
      "num_tokens": 160492445.0,
      "step": 77340
    },
    {
      "entropy": 5.663249778747558,
      "epoch": 7.01605587808418,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001433616051732231,
      "loss": 4.8149,
      "mean_token_accuracy": 0.23945230692625047,
      "num_tokens": 160502012.0,
      "step": 77345
    },
    {
      "entropy": 5.684179878234863,
      "epoch": 7.0165094339622645,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014333536579033752,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23440199345350266,
      "num_tokens": 160512364.0,
      "step": 77350
    },
    {
      "entropy": 5.722099161148071,
      "epoch": 7.016962989840349,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00014330912913031014,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23646466434001923,
      "num_tokens": 160521971.0,
      "step": 77355
    },
    {
      "entropy": 5.655198431015014,
      "epoch": 7.017416545718432,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001432828951936835,
      "loss": 4.7239,
      "mean_token_accuracy": 0.2454753890633583,
      "num_tokens": 160532965.0,
      "step": 77360
    },
    {
      "entropy": 5.648685598373413,
      "epoch": 7.017870101596516,
      "grad_norm": 1.578125,
      "learning_rate": 0.00014325666398100017,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22811904847621917,
      "num_tokens": 160542073.0,
      "step": 77365
    },
    {
      "entropy": 5.635767269134521,
      "epoch": 7.018323657474601,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014323043549280251,
      "loss": 4.7146,
      "mean_token_accuracy": 0.24638464152812958,
      "num_tokens": 160552198.0,
      "step": 77370
    },
    {
      "entropy": 5.678537130355835,
      "epoch": 7.018777213352685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001432042097296331,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23445666581392288,
      "num_tokens": 160562935.0,
      "step": 77375
    },
    {
      "entropy": 5.699514007568359,
      "epoch": 7.019230769230769,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014317798669203426,
      "loss": 4.8223,
      "mean_token_accuracy": 0.23913816958665848,
      "num_tokens": 160573098.0,
      "step": 77380
    },
    {
      "entropy": 5.678809881210327,
      "epoch": 7.019684325108853,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001431517663805482,
      "loss": 4.7719,
      "mean_token_accuracy": 0.24167252331972122,
      "num_tokens": 160582406.0,
      "step": 77385
    },
    {
      "entropy": 5.623688554763794,
      "epoch": 7.020137880986938,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001431255487957172,
      "loss": 4.7356,
      "mean_token_accuracy": 0.23786797374486923,
      "num_tokens": 160592502.0,
      "step": 77390
    },
    {
      "entropy": 5.678639459609985,
      "epoch": 7.020591436865022,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001430993339380835,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2325267255306244,
      "num_tokens": 160603447.0,
      "step": 77395
    },
    {
      "entropy": 5.6636182308197025,
      "epoch": 7.021044992743106,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014307312180818915,
      "loss": 4.765,
      "mean_token_accuracy": 0.23901814222335815,
      "num_tokens": 160612929.0,
      "step": 77400
    },
    {
      "entropy": 5.68871750831604,
      "epoch": 7.0214985486211905,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014304691240657624,
      "loss": 4.7758,
      "mean_token_accuracy": 0.23984303772449495,
      "num_tokens": 160623828.0,
      "step": 77405
    },
    {
      "entropy": 5.65784797668457,
      "epoch": 7.021952104499274,
      "grad_norm": 1.5,
      "learning_rate": 0.00014302070573378684,
      "loss": 4.8438,
      "mean_token_accuracy": 0.23056739717721939,
      "num_tokens": 160633343.0,
      "step": 77410
    },
    {
      "entropy": 5.681125640869141,
      "epoch": 7.022405660377358,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014299450179036286,
      "loss": 4.7969,
      "mean_token_accuracy": 0.2409711718559265,
      "num_tokens": 160644931.0,
      "step": 77415
    },
    {
      "entropy": 5.723360538482666,
      "epoch": 7.022859216255442,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00014296830057684623,
      "loss": 4.7967,
      "mean_token_accuracy": 0.24143570959568023,
      "num_tokens": 160654825.0,
      "step": 77420
    },
    {
      "entropy": 5.6248876571655275,
      "epoch": 7.023312772133527,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001429421020937788,
      "loss": 4.7295,
      "mean_token_accuracy": 0.2391984924674034,
      "num_tokens": 160664835.0,
      "step": 77425
    },
    {
      "entropy": 5.671863174438476,
      "epoch": 7.023766328011611,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014291590634170237,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23112652599811553,
      "num_tokens": 160674968.0,
      "step": 77430
    },
    {
      "entropy": 5.624582386016845,
      "epoch": 7.024219883889695,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014288971332115864,
      "loss": 4.7193,
      "mean_token_accuracy": 0.24824371486902236,
      "num_tokens": 160685840.0,
      "step": 77435
    },
    {
      "entropy": 5.621169090270996,
      "epoch": 7.0246734397677795,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014286352303268923,
      "loss": 4.7141,
      "mean_token_accuracy": 0.250179223716259,
      "num_tokens": 160695743.0,
      "step": 77440
    },
    {
      "entropy": 5.665004301071167,
      "epoch": 7.025126995645864,
      "grad_norm": 1.578125,
      "learning_rate": 0.00014283733547683591,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23876732885837554,
      "num_tokens": 160706189.0,
      "step": 77445
    },
    {
      "entropy": 5.6345268249511715,
      "epoch": 7.025580551523948,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001428111506541402,
      "loss": 4.7421,
      "mean_token_accuracy": 0.2402929350733757,
      "num_tokens": 160716944.0,
      "step": 77450
    },
    {
      "entropy": 5.752272748947144,
      "epoch": 7.026034107402032,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014278496856514353,
      "loss": 4.8838,
      "mean_token_accuracy": 0.23381978422403335,
      "num_tokens": 160727552.0,
      "step": 77455
    },
    {
      "entropy": 5.698472738265991,
      "epoch": 7.0264876632801165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014275878921038744,
      "loss": 4.7678,
      "mean_token_accuracy": 0.24322204738855363,
      "num_tokens": 160738050.0,
      "step": 77460
    },
    {
      "entropy": 5.714723110198975,
      "epoch": 7.0269412191582,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014273261259041325,
      "loss": 4.798,
      "mean_token_accuracy": 0.2365357458591461,
      "num_tokens": 160748757.0,
      "step": 77465
    },
    {
      "entropy": 5.72215461730957,
      "epoch": 7.027394775036284,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014270643870576238,
      "loss": 4.8253,
      "mean_token_accuracy": 0.23214073926210405,
      "num_tokens": 160759570.0,
      "step": 77470
    },
    {
      "entropy": 5.743870210647583,
      "epoch": 7.0278483309143684,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014268026755697593,
      "loss": 4.8925,
      "mean_token_accuracy": 0.22701337188482285,
      "num_tokens": 160769314.0,
      "step": 77475
    },
    {
      "entropy": 5.691839694976807,
      "epoch": 7.028301886792453,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001426540991445954,
      "loss": 4.833,
      "mean_token_accuracy": 0.23227038830518723,
      "num_tokens": 160780024.0,
      "step": 77480
    },
    {
      "entropy": 5.675723886489868,
      "epoch": 7.028755442670537,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014262793346916185,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23023294657468796,
      "num_tokens": 160791972.0,
      "step": 77485
    },
    {
      "entropy": 5.672899675369263,
      "epoch": 7.029208998548621,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014260177053121632,
      "loss": 4.808,
      "mean_token_accuracy": 0.23141637295484543,
      "num_tokens": 160802873.0,
      "step": 77490
    },
    {
      "entropy": 5.707842254638672,
      "epoch": 7.0296625544267055,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014257561033129989,
      "loss": 4.7951,
      "mean_token_accuracy": 0.23493487983942032,
      "num_tokens": 160813192.0,
      "step": 77495
    },
    {
      "entropy": 5.695546340942383,
      "epoch": 7.03011611030479,
      "grad_norm": 1.609375,
      "learning_rate": 0.00014254945286995362,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23933606445789338,
      "num_tokens": 160822621.0,
      "step": 77500
    },
    {
      "entropy": 5.679905986785888,
      "epoch": 7.030569666182874,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014252329814771835,
      "loss": 4.7644,
      "mean_token_accuracy": 0.23704007714986802,
      "num_tokens": 160832440.0,
      "step": 77505
    },
    {
      "entropy": 5.734280633926391,
      "epoch": 7.031023222060958,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001424971461651351,
      "loss": 4.885,
      "mean_token_accuracy": 0.22561935037374498,
      "num_tokens": 160844185.0,
      "step": 77510
    },
    {
      "entropy": 5.675003147125244,
      "epoch": 7.031476777939042,
      "grad_norm": 1.5,
      "learning_rate": 0.00014247099692274462,
      "loss": 4.7927,
      "mean_token_accuracy": 0.23213668763637543,
      "num_tokens": 160855391.0,
      "step": 77515
    },
    {
      "entropy": 5.685036706924438,
      "epoch": 7.031930333817126,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001424448504210877,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2370639279484749,
      "num_tokens": 160866774.0,
      "step": 77520
    },
    {
      "entropy": 5.708306980133057,
      "epoch": 7.03238388969521,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014241870666070505,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2303096681833267,
      "num_tokens": 160878197.0,
      "step": 77525
    },
    {
      "entropy": 5.716992330551148,
      "epoch": 7.0328374455732945,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014239256564213738,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2306454822421074,
      "num_tokens": 160889579.0,
      "step": 77530
    },
    {
      "entropy": 5.768022060394287,
      "epoch": 7.033291001451379,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001423664273659252,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23441691994667052,
      "num_tokens": 160900200.0,
      "step": 77535
    },
    {
      "entropy": 5.715114402770996,
      "epoch": 7.033744557329463,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014234029183260905,
      "loss": 4.7509,
      "mean_token_accuracy": 0.2340504691004753,
      "num_tokens": 160909668.0,
      "step": 77540
    },
    {
      "entropy": 5.747791481018067,
      "epoch": 7.034198113207547,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014231415904272947,
      "loss": 4.8403,
      "mean_token_accuracy": 0.24072232842445374,
      "num_tokens": 160920058.0,
      "step": 77545
    },
    {
      "entropy": 5.7355485439300535,
      "epoch": 7.0346516690856316,
      "grad_norm": 1.640625,
      "learning_rate": 0.00014228802899682696,
      "loss": 4.8495,
      "mean_token_accuracy": 0.24079643338918685,
      "num_tokens": 160929539.0,
      "step": 77550
    },
    {
      "entropy": 5.66005048751831,
      "epoch": 7.035105224963716,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014226190169544184,
      "loss": 4.7541,
      "mean_token_accuracy": 0.23431912958621978,
      "num_tokens": 160939654.0,
      "step": 77555
    },
    {
      "entropy": 5.652992200851441,
      "epoch": 7.0355587808418,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001422357771391144,
      "loss": 4.6829,
      "mean_token_accuracy": 0.24846530854701995,
      "num_tokens": 160949234.0,
      "step": 77560
    },
    {
      "entropy": 5.708023500442505,
      "epoch": 7.0360123367198835,
      "grad_norm": 1.65625,
      "learning_rate": 0.00014220965532838492,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2263159766793251,
      "num_tokens": 160960502.0,
      "step": 77565
    },
    {
      "entropy": 5.698748683929443,
      "epoch": 7.036465892597968,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014218353626379362,
      "loss": 4.805,
      "mean_token_accuracy": 0.24109925031661988,
      "num_tokens": 160970967.0,
      "step": 77570
    },
    {
      "entropy": 5.729002666473389,
      "epoch": 7.036919448476052,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014215741994588055,
      "loss": 4.828,
      "mean_token_accuracy": 0.24100825935602188,
      "num_tokens": 160980353.0,
      "step": 77575
    },
    {
      "entropy": 5.727004194259644,
      "epoch": 7.037373004354136,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014213130637518595,
      "loss": 4.8333,
      "mean_token_accuracy": 0.2343871533870697,
      "num_tokens": 160991385.0,
      "step": 77580
    },
    {
      "entropy": 5.665896987915039,
      "epoch": 7.0378265602322205,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014210519555224983,
      "loss": 4.7863,
      "mean_token_accuracy": 0.24125753939151764,
      "num_tokens": 161001764.0,
      "step": 77585
    },
    {
      "entropy": 5.727365016937256,
      "epoch": 7.038280116110305,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001420790874776121,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23076273053884505,
      "num_tokens": 161012173.0,
      "step": 77590
    },
    {
      "entropy": 5.685595703125,
      "epoch": 7.038733671988389,
      "grad_norm": 1.375,
      "learning_rate": 0.00014205298215181272,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2401510640978813,
      "num_tokens": 161022846.0,
      "step": 77595
    },
    {
      "entropy": 5.721949672698974,
      "epoch": 7.039187227866473,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014202687957539154,
      "loss": 4.8426,
      "mean_token_accuracy": 0.2376594677567482,
      "num_tokens": 161032469.0,
      "step": 77600
    },
    {
      "entropy": 5.785971736907959,
      "epoch": 7.039640783744558,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014200077974888838,
      "loss": 4.8815,
      "mean_token_accuracy": 0.23117593824863433,
      "num_tokens": 161042289.0,
      "step": 77605
    },
    {
      "entropy": 5.737528705596924,
      "epoch": 7.040094339622642,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001419746826728429,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22783035039901733,
      "num_tokens": 161052698.0,
      "step": 77610
    },
    {
      "entropy": 5.684416913986206,
      "epoch": 7.040547895500726,
      "grad_norm": 1.46875,
      "learning_rate": 0.000141948588347795,
      "loss": 4.8143,
      "mean_token_accuracy": 0.2340683862566948,
      "num_tokens": 161062592.0,
      "step": 77615
    },
    {
      "entropy": 5.644811344146729,
      "epoch": 7.0410014513788095,
      "grad_norm": 1.609375,
      "learning_rate": 0.00014192249677428414,
      "loss": 4.7817,
      "mean_token_accuracy": 0.23860862851142883,
      "num_tokens": 161073494.0,
      "step": 77620
    },
    {
      "entropy": 5.683847522735595,
      "epoch": 7.041455007256894,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00014189640795285002,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2356242060661316,
      "num_tokens": 161083723.0,
      "step": 77625
    },
    {
      "entropy": 5.763135480880737,
      "epoch": 7.041908563134978,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014187032188403204,
      "loss": 4.9212,
      "mean_token_accuracy": 0.22729863226413727,
      "num_tokens": 161093505.0,
      "step": 77630
    },
    {
      "entropy": 5.7470461368560795,
      "epoch": 7.042362119013062,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014184423856836978,
      "loss": 4.8246,
      "mean_token_accuracy": 0.2360954090952873,
      "num_tokens": 161104223.0,
      "step": 77635
    },
    {
      "entropy": 5.774609565734863,
      "epoch": 7.042815674891147,
      "grad_norm": 1.640625,
      "learning_rate": 0.00014181815800640258,
      "loss": 4.8756,
      "mean_token_accuracy": 0.2262670084834099,
      "num_tokens": 161113889.0,
      "step": 77640
    },
    {
      "entropy": 5.74036431312561,
      "epoch": 7.043269230769231,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014179208019866974,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2365077719092369,
      "num_tokens": 161124389.0,
      "step": 77645
    },
    {
      "entropy": 5.753848695755005,
      "epoch": 7.043722786647315,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001417660051457107,
      "loss": 4.8476,
      "mean_token_accuracy": 0.23401230573654175,
      "num_tokens": 161134968.0,
      "step": 77650
    },
    {
      "entropy": 5.724943447113037,
      "epoch": 7.044176342525399,
      "grad_norm": 1.5,
      "learning_rate": 0.0001417399328480646,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23329166918992997,
      "num_tokens": 161145749.0,
      "step": 77655
    },
    {
      "entropy": 5.758871746063233,
      "epoch": 7.044629898403484,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00014171386330627067,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2371405005455017,
      "num_tokens": 161156418.0,
      "step": 77660
    },
    {
      "entropy": 5.674684286117554,
      "epoch": 7.045083454281568,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014168779652086805,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2372312381863594,
      "num_tokens": 161165802.0,
      "step": 77665
    },
    {
      "entropy": 5.726112699508667,
      "epoch": 7.045537010159651,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014166173249239572,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22891789823770523,
      "num_tokens": 161175338.0,
      "step": 77670
    },
    {
      "entropy": 5.689297008514404,
      "epoch": 7.0459905660377355,
      "grad_norm": 1.71875,
      "learning_rate": 0.00014163567122139277,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2414952963590622,
      "num_tokens": 161185716.0,
      "step": 77675
    },
    {
      "entropy": 5.67293348312378,
      "epoch": 7.04644412191582,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000141609612708398,
      "loss": 4.7914,
      "mean_token_accuracy": 0.24398335069417953,
      "num_tokens": 161195212.0,
      "step": 77680
    },
    {
      "entropy": 5.749218082427978,
      "epoch": 7.046897677793904,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014158355695395055,
      "loss": 4.8069,
      "mean_token_accuracy": 0.241521018743515,
      "num_tokens": 161205636.0,
      "step": 77685
    },
    {
      "entropy": 5.749776983261109,
      "epoch": 7.047351233671988,
      "grad_norm": 1.375,
      "learning_rate": 0.00014155750395858913,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22985814809799193,
      "num_tokens": 161217473.0,
      "step": 77690
    },
    {
      "entropy": 5.652680826187134,
      "epoch": 7.047804789550073,
      "grad_norm": 1.5,
      "learning_rate": 0.00014153145372285253,
      "loss": 4.7041,
      "mean_token_accuracy": 0.24240962713956832,
      "num_tokens": 161227204.0,
      "step": 77695
    },
    {
      "entropy": 5.774477481842041,
      "epoch": 7.048258345428157,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001415054062472795,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22156137675046922,
      "num_tokens": 161237019.0,
      "step": 77700
    },
    {
      "entropy": 5.692649221420288,
      "epoch": 7.048711901306241,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014147936153240866,
      "loss": 4.7417,
      "mean_token_accuracy": 0.2383836254477501,
      "num_tokens": 161246822.0,
      "step": 77705
    },
    {
      "entropy": 5.758398103713989,
      "epoch": 7.049165457184325,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014145331957877855,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23465315252542496,
      "num_tokens": 161256352.0,
      "step": 77710
    },
    {
      "entropy": 5.716844797134399,
      "epoch": 7.04961901306241,
      "grad_norm": 1.59375,
      "learning_rate": 0.00014142728038692794,
      "loss": 4.847,
      "mean_token_accuracy": 0.23247073739767074,
      "num_tokens": 161265670.0,
      "step": 77715
    },
    {
      "entropy": 5.730145406723023,
      "epoch": 7.050072568940493,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001414012439573952,
      "loss": 4.8687,
      "mean_token_accuracy": 0.23047761023044586,
      "num_tokens": 161274774.0,
      "step": 77720
    },
    {
      "entropy": 5.697854042053223,
      "epoch": 7.050526124818577,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001413752102907188,
      "loss": 4.8171,
      "mean_token_accuracy": 0.22894108295440674,
      "num_tokens": 161286265.0,
      "step": 77725
    },
    {
      "entropy": 5.7158760070800785,
      "epoch": 7.050979680696662,
      "grad_norm": 1.5,
      "learning_rate": 0.00014134917938743707,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2380012646317482,
      "num_tokens": 161296666.0,
      "step": 77730
    },
    {
      "entropy": 5.66181526184082,
      "epoch": 7.051433236574746,
      "grad_norm": 1.640625,
      "learning_rate": 0.00014132315124808838,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23261355757713317,
      "num_tokens": 161306755.0,
      "step": 77735
    },
    {
      "entropy": 5.689394617080689,
      "epoch": 7.05188679245283,
      "grad_norm": 1.515625,
      "learning_rate": 0.000141297125873211,
      "loss": 4.8084,
      "mean_token_accuracy": 0.2325386017560959,
      "num_tokens": 161316919.0,
      "step": 77740
    },
    {
      "entropy": 5.713862085342408,
      "epoch": 7.052340348330914,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00014127110326334302,
      "loss": 4.7821,
      "mean_token_accuracy": 0.24130775779485703,
      "num_tokens": 161326130.0,
      "step": 77745
    },
    {
      "entropy": 5.7082561492919925,
      "epoch": 7.052793904208999,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014124508341902282,
      "loss": 4.7901,
      "mean_token_accuracy": 0.24860885888338088,
      "num_tokens": 161337409.0,
      "step": 77750
    },
    {
      "entropy": 5.70789065361023,
      "epoch": 7.053247460087083,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014121906634078835,
      "loss": 4.7766,
      "mean_token_accuracy": 0.2429693728685379,
      "num_tokens": 161347451.0,
      "step": 77755
    },
    {
      "entropy": 5.703413724899292,
      "epoch": 7.053701015965167,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001411930520291777,
      "loss": 4.7562,
      "mean_token_accuracy": 0.24931972920894624,
      "num_tokens": 161357009.0,
      "step": 77760
    },
    {
      "entropy": 5.700445175170898,
      "epoch": 7.0541545718432515,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001411670404847289,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23237192630767822,
      "num_tokens": 161366892.0,
      "step": 77765
    },
    {
      "entropy": 5.680361366271972,
      "epoch": 7.054608127721335,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001411410317079797,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23137222081422806,
      "num_tokens": 161377913.0,
      "step": 77770
    },
    {
      "entropy": 5.682507133483886,
      "epoch": 7.055061683599419,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014111502569946806,
      "loss": 4.7404,
      "mean_token_accuracy": 0.24555578380823134,
      "num_tokens": 161388565.0,
      "step": 77775
    },
    {
      "entropy": 5.729126596450806,
      "epoch": 7.055515239477503,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014108902245973183,
      "loss": 4.8096,
      "mean_token_accuracy": 0.24389982223510742,
      "num_tokens": 161397830.0,
      "step": 77780
    },
    {
      "entropy": 5.579501008987426,
      "epoch": 7.055968795355588,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014106302198930879,
      "loss": 4.666,
      "mean_token_accuracy": 0.24638115018606185,
      "num_tokens": 161408396.0,
      "step": 77785
    },
    {
      "entropy": 5.660636568069458,
      "epoch": 7.056422351233672,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014103702428873655,
      "loss": 4.835,
      "mean_token_accuracy": 0.23018980920314788,
      "num_tokens": 161418437.0,
      "step": 77790
    },
    {
      "entropy": 5.755228662490845,
      "epoch": 7.056875907111756,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014101102935855278,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22069515734910966,
      "num_tokens": 161428771.0,
      "step": 77795
    },
    {
      "entropy": 5.669010019302368,
      "epoch": 7.05732946298984,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001409850371992951,
      "loss": 4.6992,
      "mean_token_accuracy": 0.23725765943527222,
      "num_tokens": 161438604.0,
      "step": 77800
    },
    {
      "entropy": 5.695471525192261,
      "epoch": 7.057783018867925,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014095904781150095,
      "loss": 4.7343,
      "mean_token_accuracy": 0.239231838285923,
      "num_tokens": 161448045.0,
      "step": 77805
    },
    {
      "entropy": 5.748492670059204,
      "epoch": 7.058236574746009,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001409330611957078,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22900084257125855,
      "num_tokens": 161459224.0,
      "step": 77810
    },
    {
      "entropy": 5.6469566345214846,
      "epoch": 7.058690130624093,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001409070773524532,
      "loss": 4.7623,
      "mean_token_accuracy": 0.24500320851802826,
      "num_tokens": 161470338.0,
      "step": 77815
    },
    {
      "entropy": 5.709521293640137,
      "epoch": 7.0591436865021775,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014088109628227443,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23860646933317184,
      "num_tokens": 161480002.0,
      "step": 77820
    },
    {
      "entropy": 5.642429828643799,
      "epoch": 7.059597242380261,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014085511798570872,
      "loss": 4.742,
      "mean_token_accuracy": 0.23718296587467194,
      "num_tokens": 161490544.0,
      "step": 77825
    },
    {
      "entropy": 5.663790082931518,
      "epoch": 7.060050798258345,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014082914246329336,
      "loss": 4.844,
      "mean_token_accuracy": 0.23716705441474914,
      "num_tokens": 161501270.0,
      "step": 77830
    },
    {
      "entropy": 5.716796398162842,
      "epoch": 7.060504354136429,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014080316971556555,
      "loss": 4.8032,
      "mean_token_accuracy": 0.2361623615026474,
      "num_tokens": 161512576.0,
      "step": 77835
    },
    {
      "entropy": 5.693127584457398,
      "epoch": 7.060957910014514,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001407771997430623,
      "loss": 4.7456,
      "mean_token_accuracy": 0.24365614950656891,
      "num_tokens": 161523430.0,
      "step": 77840
    },
    {
      "entropy": 5.698119688034057,
      "epoch": 7.061411465892598,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014075123254632082,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23366989940404892,
      "num_tokens": 161534339.0,
      "step": 77845
    },
    {
      "entropy": 5.659157800674438,
      "epoch": 7.061865021770682,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001407252681258781,
      "loss": 4.7776,
      "mean_token_accuracy": 0.2336810275912285,
      "num_tokens": 161545120.0,
      "step": 77850
    },
    {
      "entropy": 5.702405500411987,
      "epoch": 7.0623185776487665,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014069930648227103,
      "loss": 4.8622,
      "mean_token_accuracy": 0.229327954351902,
      "num_tokens": 161555587.0,
      "step": 77855
    },
    {
      "entropy": 5.7583776950836185,
      "epoch": 7.062772133526851,
      "grad_norm": 1.5,
      "learning_rate": 0.00014067334761603653,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23644064515829086,
      "num_tokens": 161565929.0,
      "step": 77860
    },
    {
      "entropy": 5.7755194187164305,
      "epoch": 7.063225689404935,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014064739152771144,
      "loss": 4.9068,
      "mean_token_accuracy": 0.22549049258232118,
      "num_tokens": 161577037.0,
      "step": 77865
    },
    {
      "entropy": 5.746003580093384,
      "epoch": 7.063679245283019,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014062143821783253,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23587419390678405,
      "num_tokens": 161588055.0,
      "step": 77870
    },
    {
      "entropy": 5.626426267623901,
      "epoch": 7.064132801161103,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001405954876869364,
      "loss": 4.7376,
      "mean_token_accuracy": 0.24385998994112015,
      "num_tokens": 161597629.0,
      "step": 77875
    },
    {
      "entropy": 5.6942414283752445,
      "epoch": 7.064586357039187,
      "grad_norm": 1.578125,
      "learning_rate": 0.00014056953993555997,
      "loss": 4.8047,
      "mean_token_accuracy": 0.23590844720602036,
      "num_tokens": 161608191.0,
      "step": 77880
    },
    {
      "entropy": 5.663621044158935,
      "epoch": 7.065039912917271,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014054359496423969,
      "loss": 4.7433,
      "mean_token_accuracy": 0.2386705309152603,
      "num_tokens": 161618155.0,
      "step": 77885
    },
    {
      "entropy": 5.634976053237915,
      "epoch": 7.0654934687953554,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014051765277351208,
      "loss": 4.7112,
      "mean_token_accuracy": 0.24135844260454178,
      "num_tokens": 161628588.0,
      "step": 77890
    },
    {
      "entropy": 5.6866051197052006,
      "epoch": 7.06594702467344,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014049171336391375,
      "loss": 4.7581,
      "mean_token_accuracy": 0.23503522276878358,
      "num_tokens": 161638791.0,
      "step": 77895
    },
    {
      "entropy": 5.785683917999267,
      "epoch": 7.066400580551524,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014046577673598105,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22820158153772355,
      "num_tokens": 161649869.0,
      "step": 77900
    },
    {
      "entropy": 5.7186863899230955,
      "epoch": 7.066854136429608,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014043984289025037,
      "loss": 4.7943,
      "mean_token_accuracy": 0.2394929513335228,
      "num_tokens": 161659875.0,
      "step": 77905
    },
    {
      "entropy": 5.6892448425292965,
      "epoch": 7.0673076923076925,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014041391182725794,
      "loss": 4.8019,
      "mean_token_accuracy": 0.2341044530272484,
      "num_tokens": 161670533.0,
      "step": 77910
    },
    {
      "entropy": 5.719005727767945,
      "epoch": 7.067761248185777,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001403879835475402,
      "loss": 4.809,
      "mean_token_accuracy": 0.23558745235204698,
      "num_tokens": 161680324.0,
      "step": 77915
    },
    {
      "entropy": 5.6739012718200685,
      "epoch": 7.068214804063861,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014036205805163324,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23582984805107116,
      "num_tokens": 161690811.0,
      "step": 77920
    },
    {
      "entropy": 5.707458114624023,
      "epoch": 7.068668359941944,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014033613534007323,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2221420392394066,
      "num_tokens": 161702606.0,
      "step": 77925
    },
    {
      "entropy": 5.736624002456665,
      "epoch": 7.069121915820029,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014031021541339628,
      "loss": 4.8223,
      "mean_token_accuracy": 0.23280966132879258,
      "num_tokens": 161714210.0,
      "step": 77930
    },
    {
      "entropy": 5.760966062545776,
      "epoch": 7.069575471698113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001402842982721384,
      "loss": 4.8815,
      "mean_token_accuracy": 0.2334542989730835,
      "num_tokens": 161724163.0,
      "step": 77935
    },
    {
      "entropy": 5.68834228515625,
      "epoch": 7.070029027576197,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014025838391683554,
      "loss": 4.8219,
      "mean_token_accuracy": 0.24172732084989548,
      "num_tokens": 161733792.0,
      "step": 77940
    },
    {
      "entropy": 5.653335523605347,
      "epoch": 7.0704825834542815,
      "grad_norm": 1.5,
      "learning_rate": 0.00014023247234802358,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24111554473638536,
      "num_tokens": 161744462.0,
      "step": 77945
    },
    {
      "entropy": 5.688224601745605,
      "epoch": 7.070936139332366,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00014020656356623852,
      "loss": 4.7996,
      "mean_token_accuracy": 0.24087101817131043,
      "num_tokens": 161754157.0,
      "step": 77950
    },
    {
      "entropy": 5.698153877258301,
      "epoch": 7.07138969521045,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014018065757201605,
      "loss": 4.7773,
      "mean_token_accuracy": 0.24095641523599626,
      "num_tokens": 161763729.0,
      "step": 77955
    },
    {
      "entropy": 5.659885025024414,
      "epoch": 7.071843251088534,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001401547543658919,
      "loss": 4.7744,
      "mean_token_accuracy": 0.23946926295757293,
      "num_tokens": 161774281.0,
      "step": 77960
    },
    {
      "entropy": 5.691292333602905,
      "epoch": 7.0722968069666186,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014012885394840187,
      "loss": 4.8681,
      "mean_token_accuracy": 0.23319371193647384,
      "num_tokens": 161784521.0,
      "step": 77965
    },
    {
      "entropy": 5.6896415710449215,
      "epoch": 7.072750362844703,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014010295632008147,
      "loss": 4.798,
      "mean_token_accuracy": 0.22934657484292983,
      "num_tokens": 161794090.0,
      "step": 77970
    },
    {
      "entropy": 5.719210433959961,
      "epoch": 7.073203918722786,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014007706148146633,
      "loss": 4.8715,
      "mean_token_accuracy": 0.22994520217180253,
      "num_tokens": 161805301.0,
      "step": 77975
    },
    {
      "entropy": 5.7493979930877686,
      "epoch": 7.0736574746008705,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014005116943309188,
      "loss": 4.8945,
      "mean_token_accuracy": 0.22594667971134186,
      "num_tokens": 161816387.0,
      "step": 77980
    },
    {
      "entropy": 5.758913278579712,
      "epoch": 7.074111030478955,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014002528017549365,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23022732585668565,
      "num_tokens": 161829062.0,
      "step": 77985
    },
    {
      "entropy": 5.733099842071534,
      "epoch": 7.074564586357039,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013999939370920705,
      "loss": 4.7911,
      "mean_token_accuracy": 0.23329598158597947,
      "num_tokens": 161839865.0,
      "step": 77990
    },
    {
      "entropy": 5.745138835906983,
      "epoch": 7.075018142235123,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001399735100347674,
      "loss": 4.911,
      "mean_token_accuracy": 0.22432901710271835,
      "num_tokens": 161850199.0,
      "step": 77995
    },
    {
      "entropy": 5.683888578414917,
      "epoch": 7.0754716981132075,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013994762915270999,
      "loss": 4.7457,
      "mean_token_accuracy": 0.24240129292011262,
      "num_tokens": 161861149.0,
      "step": 78000
    },
    {
      "epoch": 7.0754716981132075,
      "eval_entropy": 5.414719475448401,
      "eval_loss": 4.919704437255859,
      "eval_mean_token_accuracy": 0.23715166505341886,
      "eval_num_tokens": 161861149.0,
      "eval_runtime": 20.6531,
      "eval_samples_per_second": 1712.093,
      "eval_steps_per_second": 214.012,
      "step": 78000
    },
    {
      "entropy": 5.717284870147705,
      "epoch": 7.075925253991292,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013992175106356997,
      "loss": 4.7902,
      "mean_token_accuracy": 0.23687745332717897,
      "num_tokens": 161871364.0,
      "step": 78005
    },
    {
      "entropy": 5.71596212387085,
      "epoch": 7.076378809869376,
      "grad_norm": 1.625,
      "learning_rate": 0.0001398958757678826,
      "loss": 4.846,
      "mean_token_accuracy": 0.23393018543720245,
      "num_tokens": 161882853.0,
      "step": 78010
    },
    {
      "entropy": 5.745302820205689,
      "epoch": 7.07683236574746,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013987000326618287,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2307598665356636,
      "num_tokens": 161894785.0,
      "step": 78015
    },
    {
      "entropy": 5.642197895050049,
      "epoch": 7.077285921625545,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013984413355900598,
      "loss": 4.7129,
      "mean_token_accuracy": 0.24096719026565552,
      "num_tokens": 161905624.0,
      "step": 78020
    },
    {
      "entropy": 5.740350246429443,
      "epoch": 7.077739477503629,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013981826664688686,
      "loss": 4.815,
      "mean_token_accuracy": 0.2353294461965561,
      "num_tokens": 161915614.0,
      "step": 78025
    },
    {
      "entropy": 5.6957300186157225,
      "epoch": 7.078193033381712,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001397924025303604,
      "loss": 4.868,
      "mean_token_accuracy": 0.2349262610077858,
      "num_tokens": 161925999.0,
      "step": 78030
    },
    {
      "entropy": 5.688728761672974,
      "epoch": 7.0786465892597965,
      "grad_norm": 1.734375,
      "learning_rate": 0.00013976654120996152,
      "loss": 4.7898,
      "mean_token_accuracy": 0.23496053665876387,
      "num_tokens": 161935296.0,
      "step": 78035
    },
    {
      "entropy": 5.686034870147705,
      "epoch": 7.079100145137881,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013974068268622506,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23239739537239074,
      "num_tokens": 161945567.0,
      "step": 78040
    },
    {
      "entropy": 5.672847509384155,
      "epoch": 7.079553701015965,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013971482695968564,
      "loss": 4.7413,
      "mean_token_accuracy": 0.2366655722260475,
      "num_tokens": 161955761.0,
      "step": 78045
    },
    {
      "entropy": 5.667268991470337,
      "epoch": 7.080007256894049,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013968897403087815,
      "loss": 4.8015,
      "mean_token_accuracy": 0.23709950745105743,
      "num_tokens": 161965150.0,
      "step": 78050
    },
    {
      "entropy": 5.671409320831299,
      "epoch": 7.080460812772134,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001396631239003372,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23714449852705002,
      "num_tokens": 161975591.0,
      "step": 78055
    },
    {
      "entropy": 5.703789710998535,
      "epoch": 7.080914368650218,
      "grad_norm": 1.5,
      "learning_rate": 0.00013963727656859728,
      "loss": 4.911,
      "mean_token_accuracy": 0.22887856215238572,
      "num_tokens": 161986605.0,
      "step": 78060
    },
    {
      "entropy": 5.760721492767334,
      "epoch": 7.081367924528302,
      "grad_norm": 1.453125,
      "learning_rate": 0.000139611432036193,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22986616790294648,
      "num_tokens": 161996104.0,
      "step": 78065
    },
    {
      "entropy": 5.649033641815185,
      "epoch": 7.081821480406386,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001395855903036588,
      "loss": 4.739,
      "mean_token_accuracy": 0.23863118141889572,
      "num_tokens": 162006068.0,
      "step": 78070
    },
    {
      "entropy": 5.803308725357056,
      "epoch": 7.082275036284471,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013955975137152916,
      "loss": 4.8953,
      "mean_token_accuracy": 0.23217888325452804,
      "num_tokens": 162018491.0,
      "step": 78075
    },
    {
      "entropy": 5.751314258575439,
      "epoch": 7.082728592162554,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001395339152403382,
      "loss": 4.8996,
      "mean_token_accuracy": 0.23114594370126723,
      "num_tokens": 162028261.0,
      "step": 78080
    },
    {
      "entropy": 5.684852933883667,
      "epoch": 7.083182148040638,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013950808191062053,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23515222817659379,
      "num_tokens": 162038994.0,
      "step": 78085
    },
    {
      "entropy": 5.630716943740845,
      "epoch": 7.0836357039187225,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013948225138291027,
      "loss": 4.708,
      "mean_token_accuracy": 0.2395515188574791,
      "num_tokens": 162049842.0,
      "step": 78090
    },
    {
      "entropy": 5.706476449966431,
      "epoch": 7.084089259796807,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013945642365774153,
      "loss": 4.7673,
      "mean_token_accuracy": 0.2337488755583763,
      "num_tokens": 162059781.0,
      "step": 78095
    },
    {
      "entropy": 5.738516759872437,
      "epoch": 7.084542815674891,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013943059873564853,
      "loss": 4.8726,
      "mean_token_accuracy": 0.23159826844930648,
      "num_tokens": 162069555.0,
      "step": 78100
    },
    {
      "entropy": 5.727409601211548,
      "epoch": 7.084996371552975,
      "grad_norm": 1.5,
      "learning_rate": 0.00013940477661716528,
      "loss": 4.7708,
      "mean_token_accuracy": 0.23534568548202514,
      "num_tokens": 162079435.0,
      "step": 78105
    },
    {
      "entropy": 5.702721977233887,
      "epoch": 7.08544992743106,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013937895730282583,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23549092561006546,
      "num_tokens": 162089086.0,
      "step": 78110
    },
    {
      "entropy": 5.681752443313599,
      "epoch": 7.085903483309144,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013935314079316396,
      "loss": 4.7973,
      "mean_token_accuracy": 0.23943838775157927,
      "num_tokens": 162099281.0,
      "step": 78115
    },
    {
      "entropy": 5.640607738494873,
      "epoch": 7.086357039187228,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013932732708871385,
      "loss": 4.7817,
      "mean_token_accuracy": 0.22854330241680146,
      "num_tokens": 162109588.0,
      "step": 78120
    },
    {
      "entropy": 5.680173778533936,
      "epoch": 7.086810595065312,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013930151619000915,
      "loss": 4.7478,
      "mean_token_accuracy": 0.23626337200403214,
      "num_tokens": 162121250.0,
      "step": 78125
    },
    {
      "entropy": 5.763604354858399,
      "epoch": 7.087264150943396,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013927570809758373,
      "loss": 4.7886,
      "mean_token_accuracy": 0.23788794577121736,
      "num_tokens": 162130835.0,
      "step": 78130
    },
    {
      "entropy": 5.684652423858642,
      "epoch": 7.08771770682148,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013924990281197119,
      "loss": 4.7932,
      "mean_token_accuracy": 0.24051551818847655,
      "num_tokens": 162141371.0,
      "step": 78135
    },
    {
      "entropy": 5.626433372497559,
      "epoch": 7.088171262699564,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013922410033370526,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2416353479027748,
      "num_tokens": 162151777.0,
      "step": 78140
    },
    {
      "entropy": 5.674720287322998,
      "epoch": 7.088624818577649,
      "grad_norm": 1.609375,
      "learning_rate": 0.00013919830066331956,
      "loss": 4.7695,
      "mean_token_accuracy": 0.23487507104873656,
      "num_tokens": 162161829.0,
      "step": 78145
    },
    {
      "entropy": 5.660694026947022,
      "epoch": 7.089078374455733,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00013917250380134752,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23989956676959992,
      "num_tokens": 162171721.0,
      "step": 78150
    },
    {
      "entropy": 5.66619029045105,
      "epoch": 7.089531930333817,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013914670974832282,
      "loss": 4.7607,
      "mean_token_accuracy": 0.24476213455200196,
      "num_tokens": 162182178.0,
      "step": 78155
    },
    {
      "entropy": 5.745655488967896,
      "epoch": 7.089985486211901,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013912091850477873,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2314743474125862,
      "num_tokens": 162193167.0,
      "step": 78160
    },
    {
      "entropy": 5.687080907821655,
      "epoch": 7.090439042089986,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001390951300712487,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23543713241815567,
      "num_tokens": 162202576.0,
      "step": 78165
    },
    {
      "entropy": 5.657345867156982,
      "epoch": 7.09089259796807,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013906934444826602,
      "loss": 4.7786,
      "mean_token_accuracy": 0.23854015320539473,
      "num_tokens": 162213819.0,
      "step": 78170
    },
    {
      "entropy": 5.696435499191284,
      "epoch": 7.091346153846154,
      "grad_norm": 1.75,
      "learning_rate": 0.00013904356163636395,
      "loss": 4.7952,
      "mean_token_accuracy": 0.24311360567808152,
      "num_tokens": 162225139.0,
      "step": 78175
    },
    {
      "entropy": 5.706130170822144,
      "epoch": 7.0917997097242385,
      "grad_norm": 1.59375,
      "learning_rate": 0.00013901778163607565,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23226025104522705,
      "num_tokens": 162236477.0,
      "step": 78180
    },
    {
      "entropy": 5.656276607513428,
      "epoch": 7.092253265602322,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013899200444793418,
      "loss": 4.7542,
      "mean_token_accuracy": 0.23909735679626465,
      "num_tokens": 162246923.0,
      "step": 78185
    },
    {
      "entropy": 5.71364803314209,
      "epoch": 7.092706821480406,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001389662300724729,
      "loss": 4.8413,
      "mean_token_accuracy": 0.23205020278692245,
      "num_tokens": 162258704.0,
      "step": 78190
    },
    {
      "entropy": 5.639937543869019,
      "epoch": 7.09316037735849,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013894045851022455,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23397423177957535,
      "num_tokens": 162269217.0,
      "step": 78195
    },
    {
      "entropy": 5.720947313308716,
      "epoch": 7.093613933236575,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013891468976172218,
      "loss": 4.8178,
      "mean_token_accuracy": 0.2401677429676056,
      "num_tokens": 162279392.0,
      "step": 78200
    },
    {
      "entropy": 5.739533805847168,
      "epoch": 7.094067489114659,
      "grad_norm": 1.5,
      "learning_rate": 0.00013888892382749873,
      "loss": 4.8923,
      "mean_token_accuracy": 0.23174362927675246,
      "num_tokens": 162291217.0,
      "step": 78205
    },
    {
      "entropy": 5.724644422531128,
      "epoch": 7.094521044992743,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013886316070808698,
      "loss": 4.7405,
      "mean_token_accuracy": 0.23980239778757095,
      "num_tokens": 162301743.0,
      "step": 78210
    },
    {
      "entropy": 5.7546182632446286,
      "epoch": 7.094974600870827,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013883740040401975,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23558233231306075,
      "num_tokens": 162311522.0,
      "step": 78215
    },
    {
      "entropy": 5.74851655960083,
      "epoch": 7.095428156748912,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013881164291582984,
      "loss": 4.7852,
      "mean_token_accuracy": 0.23888550102710723,
      "num_tokens": 162322758.0,
      "step": 78220
    },
    {
      "entropy": 5.741287708282471,
      "epoch": 7.095881712626996,
      "grad_norm": 1.59375,
      "learning_rate": 0.00013878588824404984,
      "loss": 4.8547,
      "mean_token_accuracy": 0.224819116294384,
      "num_tokens": 162332941.0,
      "step": 78225
    },
    {
      "entropy": 5.776361179351807,
      "epoch": 7.09633526850508,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013876013638921238,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2263280600309372,
      "num_tokens": 162343241.0,
      "step": 78230
    },
    {
      "entropy": 5.773040533065796,
      "epoch": 7.096788824383164,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013873438735185,
      "loss": 4.9026,
      "mean_token_accuracy": 0.2251865014433861,
      "num_tokens": 162353244.0,
      "step": 78235
    },
    {
      "entropy": 5.625294637680054,
      "epoch": 7.097242380261248,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013870864113249527,
      "loss": 4.6925,
      "mean_token_accuracy": 0.24402712136507035,
      "num_tokens": 162363086.0,
      "step": 78240
    },
    {
      "entropy": 5.600286960601807,
      "epoch": 7.097695936139332,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013868289773168044,
      "loss": 4.7401,
      "mean_token_accuracy": 0.2469801589846611,
      "num_tokens": 162373868.0,
      "step": 78245
    },
    {
      "entropy": 5.769382190704346,
      "epoch": 7.098149492017416,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013865715714993812,
      "loss": 4.9259,
      "mean_token_accuracy": 0.2289595425128937,
      "num_tokens": 162383687.0,
      "step": 78250
    },
    {
      "entropy": 5.729807138442993,
      "epoch": 7.098603047895501,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013863141938780054,
      "loss": 4.9067,
      "mean_token_accuracy": 0.23208791315555571,
      "num_tokens": 162394449.0,
      "step": 78255
    },
    {
      "entropy": 5.7907304763793945,
      "epoch": 7.099056603773585,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013860568444579996,
      "loss": 4.8533,
      "mean_token_accuracy": 0.2272571012377739,
      "num_tokens": 162404266.0,
      "step": 78260
    },
    {
      "entropy": 5.660666799545288,
      "epoch": 7.099510159651669,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013857995232446858,
      "loss": 4.763,
      "mean_token_accuracy": 0.24718033224344255,
      "num_tokens": 162415865.0,
      "step": 78265
    },
    {
      "entropy": 5.675036382675171,
      "epoch": 7.0999637155297535,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013855422302433851,
      "loss": 4.7156,
      "mean_token_accuracy": 0.2436263382434845,
      "num_tokens": 162426130.0,
      "step": 78270
    },
    {
      "entropy": 5.662207078933716,
      "epoch": 7.100417271407838,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013852849654594196,
      "loss": 4.7634,
      "mean_token_accuracy": 0.23657141923904418,
      "num_tokens": 162436069.0,
      "step": 78275
    },
    {
      "entropy": 5.747991323471069,
      "epoch": 7.100870827285922,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013850277288981073,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23640093058347703,
      "num_tokens": 162446080.0,
      "step": 78280
    },
    {
      "entropy": 5.608630561828614,
      "epoch": 7.101324383164005,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013847705205647704,
      "loss": 4.7418,
      "mean_token_accuracy": 0.244464735686779,
      "num_tokens": 162456380.0,
      "step": 78285
    },
    {
      "entropy": 5.726245880126953,
      "epoch": 7.10177793904209,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013845133404647269,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23136286288499833,
      "num_tokens": 162465848.0,
      "step": 78290
    },
    {
      "entropy": 5.685933303833008,
      "epoch": 7.102231494920174,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001384256188603296,
      "loss": 4.7501,
      "mean_token_accuracy": 0.24542141556739808,
      "num_tokens": 162476193.0,
      "step": 78295
    },
    {
      "entropy": 5.720505571365356,
      "epoch": 7.102685050798258,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013839990649857948,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23716007322072982,
      "num_tokens": 162486137.0,
      "step": 78300
    },
    {
      "entropy": 5.637034511566162,
      "epoch": 7.1031386066763424,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013837419696175406,
      "loss": 4.7617,
      "mean_token_accuracy": 0.23701056241989135,
      "num_tokens": 162496421.0,
      "step": 78305
    },
    {
      "entropy": 5.663493204116821,
      "epoch": 7.103592162554427,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013834849025038515,
      "loss": 4.819,
      "mean_token_accuracy": 0.23713141232728957,
      "num_tokens": 162506905.0,
      "step": 78310
    },
    {
      "entropy": 5.665891933441162,
      "epoch": 7.104045718432511,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013832278636500414,
      "loss": 4.8037,
      "mean_token_accuracy": 0.23612144142389296,
      "num_tokens": 162517322.0,
      "step": 78315
    },
    {
      "entropy": 5.668602848052979,
      "epoch": 7.104499274310595,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013829708530614285,
      "loss": 4.7751,
      "mean_token_accuracy": 0.23214294761419296,
      "num_tokens": 162528673.0,
      "step": 78320
    },
    {
      "entropy": 5.640815019607544,
      "epoch": 7.1049528301886795,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001382713870743327,
      "loss": 4.7042,
      "mean_token_accuracy": 0.24714667201042176,
      "num_tokens": 162537634.0,
      "step": 78325
    },
    {
      "entropy": 5.680509996414185,
      "epoch": 7.105406386066764,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00013824569167010504,
      "loss": 4.7933,
      "mean_token_accuracy": 0.23693906217813493,
      "num_tokens": 162547278.0,
      "step": 78330
    },
    {
      "entropy": 5.674365091323852,
      "epoch": 7.105859941944848,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013821999909399135,
      "loss": 4.7891,
      "mean_token_accuracy": 0.23529691845178605,
      "num_tokens": 162558158.0,
      "step": 78335
    },
    {
      "entropy": 5.6172527313232425,
      "epoch": 7.106313497822931,
      "grad_norm": 1.640625,
      "learning_rate": 0.00013819430934652294,
      "loss": 4.7275,
      "mean_token_accuracy": 0.2440507233142853,
      "num_tokens": 162568092.0,
      "step": 78340
    },
    {
      "entropy": 5.755342674255371,
      "epoch": 7.106767053701016,
      "grad_norm": 1.609375,
      "learning_rate": 0.00013816862242823098,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2317149221897125,
      "num_tokens": 162578862.0,
      "step": 78345
    },
    {
      "entropy": 5.719938087463379,
      "epoch": 7.1072206095791,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001381429383396469,
      "loss": 4.7566,
      "mean_token_accuracy": 0.23803003281354904,
      "num_tokens": 162588826.0,
      "step": 78350
    },
    {
      "entropy": 5.672025156021118,
      "epoch": 7.107674165457184,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001381172570813017,
      "loss": 4.7359,
      "mean_token_accuracy": 0.24582240879535674,
      "num_tokens": 162599535.0,
      "step": 78355
    },
    {
      "entropy": 5.649727058410645,
      "epoch": 7.1081277213352685,
      "grad_norm": 1.25,
      "learning_rate": 0.00013809157865372651,
      "loss": 4.7816,
      "mean_token_accuracy": 0.230621038377285,
      "num_tokens": 162611305.0,
      "step": 78360
    },
    {
      "entropy": 5.697272872924804,
      "epoch": 7.108581277213353,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013806590305745234,
      "loss": 4.7953,
      "mean_token_accuracy": 0.23973355889320375,
      "num_tokens": 162622528.0,
      "step": 78365
    },
    {
      "entropy": 5.664434337615967,
      "epoch": 7.109034833091437,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013804023029301022,
      "loss": 4.7214,
      "mean_token_accuracy": 0.23955574482679368,
      "num_tokens": 162632346.0,
      "step": 78370
    },
    {
      "entropy": 5.704129648208618,
      "epoch": 7.109488388969521,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013801456036093102,
      "loss": 4.7698,
      "mean_token_accuracy": 0.2404221400618553,
      "num_tokens": 162642476.0,
      "step": 78375
    },
    {
      "entropy": 5.689782285690308,
      "epoch": 7.1099419448476056,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013798889326174554,
      "loss": 4.8459,
      "mean_token_accuracy": 0.23367176800966263,
      "num_tokens": 162653742.0,
      "step": 78380
    },
    {
      "entropy": 5.775472211837768,
      "epoch": 7.11039550072569,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013796322899598475,
      "loss": 4.9243,
      "mean_token_accuracy": 0.22742239981889725,
      "num_tokens": 162664702.0,
      "step": 78385
    },
    {
      "entropy": 5.750144672393799,
      "epoch": 7.110849056603773,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001379375675641793,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2288101464509964,
      "num_tokens": 162675340.0,
      "step": 78390
    },
    {
      "entropy": 5.724632596969604,
      "epoch": 7.1113026124818575,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001379119089668599,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2318024218082428,
      "num_tokens": 162686264.0,
      "step": 78395
    },
    {
      "entropy": 5.672115612030029,
      "epoch": 7.111756168359942,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00013788625320455716,
      "loss": 4.7076,
      "mean_token_accuracy": 0.24639234840869903,
      "num_tokens": 162696379.0,
      "step": 78400
    },
    {
      "entropy": 5.723863887786865,
      "epoch": 7.112209724238026,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013786060027780163,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2258755758404732,
      "num_tokens": 162706991.0,
      "step": 78405
    },
    {
      "entropy": 5.682411432266235,
      "epoch": 7.11266328011611,
      "grad_norm": 1.5,
      "learning_rate": 0.00013783495018712384,
      "loss": 4.754,
      "mean_token_accuracy": 0.24017549455165862,
      "num_tokens": 162717152.0,
      "step": 78410
    },
    {
      "entropy": 5.6687287330627445,
      "epoch": 7.1131168359941945,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001378093029330541,
      "loss": 4.8009,
      "mean_token_accuracy": 0.24261026084423065,
      "num_tokens": 162727579.0,
      "step": 78415
    },
    {
      "entropy": 5.6963366031646725,
      "epoch": 7.113570391872279,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013778365851612308,
      "loss": 4.836,
      "mean_token_accuracy": 0.23571523278951645,
      "num_tokens": 162738878.0,
      "step": 78420
    },
    {
      "entropy": 5.762719011306762,
      "epoch": 7.114023947750363,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013775801693686096,
      "loss": 4.8683,
      "mean_token_accuracy": 0.22899436950683594,
      "num_tokens": 162748902.0,
      "step": 78425
    },
    {
      "entropy": 5.717480754852295,
      "epoch": 7.114477503628447,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013773237819579798,
      "loss": 4.8356,
      "mean_token_accuracy": 0.23343852758407593,
      "num_tokens": 162759109.0,
      "step": 78430
    },
    {
      "entropy": 5.7489259243011475,
      "epoch": 7.114931059506532,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00013770674229346443,
      "loss": 4.8623,
      "mean_token_accuracy": 0.23275235593318938,
      "num_tokens": 162769901.0,
      "step": 78435
    },
    {
      "entropy": 5.743864631652832,
      "epoch": 7.115384615384615,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013768110923039045,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2317579358816147,
      "num_tokens": 162780056.0,
      "step": 78440
    },
    {
      "entropy": 5.72549204826355,
      "epoch": 7.115838171262699,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013765547900710612,
      "loss": 4.7816,
      "mean_token_accuracy": 0.2415844514966011,
      "num_tokens": 162790856.0,
      "step": 78445
    },
    {
      "entropy": 5.688048601150513,
      "epoch": 7.1162917271407835,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013762985162414138,
      "loss": 4.7663,
      "mean_token_accuracy": 0.23738037645816804,
      "num_tokens": 162801374.0,
      "step": 78450
    },
    {
      "entropy": 5.6308746337890625,
      "epoch": 7.116745283018868,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001376042270820264,
      "loss": 4.7491,
      "mean_token_accuracy": 0.24099978059530258,
      "num_tokens": 162811724.0,
      "step": 78455
    },
    {
      "entropy": 5.785648393630981,
      "epoch": 7.117198838896952,
      "grad_norm": 1.6875,
      "learning_rate": 0.000137578605381291,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23720424622297287,
      "num_tokens": 162822001.0,
      "step": 78460
    },
    {
      "entropy": 5.710377883911133,
      "epoch": 7.117652394775036,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001375529865224651,
      "loss": 4.7748,
      "mean_token_accuracy": 0.24000790119171142,
      "num_tokens": 162831088.0,
      "step": 78465
    },
    {
      "entropy": 5.669072437286377,
      "epoch": 7.118105950653121,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013752737050607848,
      "loss": 4.7329,
      "mean_token_accuracy": 0.24524120539426802,
      "num_tokens": 162841362.0,
      "step": 78470
    },
    {
      "entropy": 5.75832462310791,
      "epoch": 7.118559506531205,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013750175733266085,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23401635736227036,
      "num_tokens": 162851802.0,
      "step": 78475
    },
    {
      "entropy": 5.760273504257202,
      "epoch": 7.119013062409289,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001374761470027419,
      "loss": 4.8665,
      "mean_token_accuracy": 0.2301321506500244,
      "num_tokens": 162863513.0,
      "step": 78480
    },
    {
      "entropy": 5.698005676269531,
      "epoch": 7.119466618287373,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013745053951685118,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23484373986721038,
      "num_tokens": 162874222.0,
      "step": 78485
    },
    {
      "entropy": 5.727429628372192,
      "epoch": 7.119920174165457,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001374249348755185,
      "loss": 4.8791,
      "mean_token_accuracy": 0.2321878805756569,
      "num_tokens": 162885787.0,
      "step": 78490
    },
    {
      "entropy": 5.726520109176636,
      "epoch": 7.120373730043541,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013739933307927316,
      "loss": 4.8604,
      "mean_token_accuracy": 0.23437925577163696,
      "num_tokens": 162896770.0,
      "step": 78495
    },
    {
      "entropy": 5.66401104927063,
      "epoch": 7.120827285921625,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001373737341286447,
      "loss": 4.769,
      "mean_token_accuracy": 0.24607294648885727,
      "num_tokens": 162907102.0,
      "step": 78500
    },
    {
      "entropy": 5.658448600769043,
      "epoch": 7.1212808417997095,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013734813802416247,
      "loss": 4.8012,
      "mean_token_accuracy": 0.23617451936006545,
      "num_tokens": 162917669.0,
      "step": 78505
    },
    {
      "entropy": 5.774876689910888,
      "epoch": 7.121734397677794,
      "grad_norm": 1.375,
      "learning_rate": 0.00013732254476635586,
      "loss": 4.8579,
      "mean_token_accuracy": 0.2336057171225548,
      "num_tokens": 162927796.0,
      "step": 78510
    },
    {
      "entropy": 5.748327350616455,
      "epoch": 7.122187953555878,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013729695435575406,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2268838033080101,
      "num_tokens": 162937891.0,
      "step": 78515
    },
    {
      "entropy": 5.701828908920288,
      "epoch": 7.122641509433962,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013727136679288633,
      "loss": 4.7619,
      "mean_token_accuracy": 0.23999157547950745,
      "num_tokens": 162948444.0,
      "step": 78520
    },
    {
      "entropy": 5.630615282058716,
      "epoch": 7.123095065312047,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013724578207828183,
      "loss": 4.7408,
      "mean_token_accuracy": 0.24351608455181123,
      "num_tokens": 162958358.0,
      "step": 78525
    },
    {
      "entropy": 5.67740535736084,
      "epoch": 7.123548621190131,
      "grad_norm": 1.59375,
      "learning_rate": 0.00013722020021246972,
      "loss": 4.8061,
      "mean_token_accuracy": 0.23528604358434677,
      "num_tokens": 162968031.0,
      "step": 78530
    },
    {
      "entropy": 5.670395374298096,
      "epoch": 7.124002177068215,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013719462119597895,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23732685595750808,
      "num_tokens": 162979539.0,
      "step": 78535
    },
    {
      "entropy": 5.68921275138855,
      "epoch": 7.124455732946299,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013716904502933852,
      "loss": 4.7946,
      "mean_token_accuracy": 0.2393954187631607,
      "num_tokens": 162991010.0,
      "step": 78540
    },
    {
      "entropy": 5.671347332000733,
      "epoch": 7.124909288824383,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001371434717130774,
      "loss": 4.779,
      "mean_token_accuracy": 0.23508177995681762,
      "num_tokens": 163001807.0,
      "step": 78545
    },
    {
      "entropy": 5.670802879333496,
      "epoch": 7.125362844702467,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013711790124772434,
      "loss": 4.6899,
      "mean_token_accuracy": 0.238978511095047,
      "num_tokens": 163011221.0,
      "step": 78550
    },
    {
      "entropy": 5.6378716945648195,
      "epoch": 7.125816400580551,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013709233363380825,
      "loss": 4.7791,
      "mean_token_accuracy": 0.23808763772249222,
      "num_tokens": 163021997.0,
      "step": 78555
    },
    {
      "entropy": 5.659945487976074,
      "epoch": 7.126269956458636,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00013706676887185787,
      "loss": 4.7958,
      "mean_token_accuracy": 0.23631325215101243,
      "num_tokens": 163032491.0,
      "step": 78560
    },
    {
      "entropy": 5.700782108306885,
      "epoch": 7.12672351233672,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013704120696240185,
      "loss": 4.7984,
      "mean_token_accuracy": 0.233638995885849,
      "num_tokens": 163042808.0,
      "step": 78565
    },
    {
      "entropy": 5.741146421432495,
      "epoch": 7.127177068214804,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013701564790596885,
      "loss": 4.8384,
      "mean_token_accuracy": 0.24697252362966537,
      "num_tokens": 163053896.0,
      "step": 78570
    },
    {
      "entropy": 5.641408157348633,
      "epoch": 7.127630624092888,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013699009170308737,
      "loss": 4.7608,
      "mean_token_accuracy": 0.2468437135219574,
      "num_tokens": 163064229.0,
      "step": 78575
    },
    {
      "entropy": 5.690195369720459,
      "epoch": 7.128084179970973,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000136964538354286,
      "loss": 4.7756,
      "mean_token_accuracy": 0.23230016231536865,
      "num_tokens": 163074793.0,
      "step": 78580
    },
    {
      "entropy": 5.749879360198975,
      "epoch": 7.128537735849057,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013693898786009307,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2293240800499916,
      "num_tokens": 163084971.0,
      "step": 78585
    },
    {
      "entropy": 5.7395713329315186,
      "epoch": 7.128991291727141,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013691344022103712,
      "loss": 4.8715,
      "mean_token_accuracy": 0.23372411280870437,
      "num_tokens": 163095379.0,
      "step": 78590
    },
    {
      "entropy": 5.800705194473267,
      "epoch": 7.129444847605225,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013688789543764643,
      "loss": 4.8783,
      "mean_token_accuracy": 0.24076673090457917,
      "num_tokens": 163105433.0,
      "step": 78595
    },
    {
      "entropy": 5.72156252861023,
      "epoch": 7.129898403483309,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013686235351044923,
      "loss": 4.8591,
      "mean_token_accuracy": 0.2358464077115059,
      "num_tokens": 163116704.0,
      "step": 78600
    },
    {
      "entropy": 5.727947568893432,
      "epoch": 7.130351959361393,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001368368144399738,
      "loss": 4.8501,
      "mean_token_accuracy": 0.23423684388399124,
      "num_tokens": 163126971.0,
      "step": 78605
    },
    {
      "entropy": 5.76754412651062,
      "epoch": 7.130805515239477,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001368112782267482,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2256955921649933,
      "num_tokens": 163137625.0,
      "step": 78610
    },
    {
      "entropy": 5.617865085601807,
      "epoch": 7.131259071117562,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013678574487130063,
      "loss": 4.7555,
      "mean_token_accuracy": 0.24189618825912476,
      "num_tokens": 163147550.0,
      "step": 78615
    },
    {
      "entropy": 5.703057527542114,
      "epoch": 7.131712626995646,
      "grad_norm": 1.5,
      "learning_rate": 0.0001367602143741591,
      "loss": 4.8004,
      "mean_token_accuracy": 0.2357431173324585,
      "num_tokens": 163157794.0,
      "step": 78620
    },
    {
      "entropy": 5.72846941947937,
      "epoch": 7.13216618287373,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013673468673585152,
      "loss": 4.8219,
      "mean_token_accuracy": 0.23803635239601134,
      "num_tokens": 163167792.0,
      "step": 78625
    },
    {
      "entropy": 5.747253322601319,
      "epoch": 7.132619738751814,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00013670916195690586,
      "loss": 4.818,
      "mean_token_accuracy": 0.2323000431060791,
      "num_tokens": 163177406.0,
      "step": 78630
    },
    {
      "entropy": 5.632692289352417,
      "epoch": 7.133073294629899,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013668364003784994,
      "loss": 4.753,
      "mean_token_accuracy": 0.23754142969846725,
      "num_tokens": 163188337.0,
      "step": 78635
    },
    {
      "entropy": 5.645767116546631,
      "epoch": 7.133526850507983,
      "grad_norm": 1.59375,
      "learning_rate": 0.00013665812097921165,
      "loss": 4.7498,
      "mean_token_accuracy": 0.23659437745809556,
      "num_tokens": 163198704.0,
      "step": 78640
    },
    {
      "entropy": 5.663324594497681,
      "epoch": 7.133980406386066,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013663260478151857,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23744633942842483,
      "num_tokens": 163208214.0,
      "step": 78645
    },
    {
      "entropy": 5.6516608715057375,
      "epoch": 7.134433962264151,
      "grad_norm": 1.5,
      "learning_rate": 0.00013660709144529853,
      "loss": 4.7874,
      "mean_token_accuracy": 0.23707669973373413,
      "num_tokens": 163218777.0,
      "step": 78650
    },
    {
      "entropy": 5.7190718173980715,
      "epoch": 7.134887518142235,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013658158097107915,
      "loss": 4.7836,
      "mean_token_accuracy": 0.2393653079867363,
      "num_tokens": 163228530.0,
      "step": 78655
    },
    {
      "entropy": 5.682757568359375,
      "epoch": 7.135341074020319,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001365560733593879,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2434580519795418,
      "num_tokens": 163238676.0,
      "step": 78660
    },
    {
      "entropy": 5.639494180679321,
      "epoch": 7.135794629898403,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013653056861075242,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23659365326166154,
      "num_tokens": 163248444.0,
      "step": 78665
    },
    {
      "entropy": 5.705776882171631,
      "epoch": 7.136248185776488,
      "grad_norm": 1.375,
      "learning_rate": 0.00013650506672569998,
      "loss": 4.8446,
      "mean_token_accuracy": 0.23445503711700438,
      "num_tokens": 163258502.0,
      "step": 78670
    },
    {
      "entropy": 5.7367956161499025,
      "epoch": 7.136701741654572,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013647956770475807,
      "loss": 4.9003,
      "mean_token_accuracy": 0.2319065362215042,
      "num_tokens": 163269128.0,
      "step": 78675
    },
    {
      "entropy": 5.693265056610107,
      "epoch": 7.137155297532656,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013645407154845397,
      "loss": 4.7992,
      "mean_token_accuracy": 0.23168852925300598,
      "num_tokens": 163279770.0,
      "step": 78680
    },
    {
      "entropy": 5.697032833099366,
      "epoch": 7.1376088534107405,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000136428578257315,
      "loss": 4.8658,
      "mean_token_accuracy": 0.23238180577754974,
      "num_tokens": 163290249.0,
      "step": 78685
    },
    {
      "entropy": 5.69793381690979,
      "epoch": 7.138062409288825,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013640308783186835,
      "loss": 4.8473,
      "mean_token_accuracy": 0.22776590883731843,
      "num_tokens": 163301030.0,
      "step": 78690
    },
    {
      "entropy": 5.7734137058258055,
      "epoch": 7.138515965166908,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013637760027264122,
      "loss": 4.8977,
      "mean_token_accuracy": 0.2330290198326111,
      "num_tokens": 163312253.0,
      "step": 78695
    },
    {
      "entropy": 5.672278261184692,
      "epoch": 7.138969521044992,
      "grad_norm": 1.609375,
      "learning_rate": 0.00013635211558016055,
      "loss": 4.7257,
      "mean_token_accuracy": 0.24367477893829345,
      "num_tokens": 163322292.0,
      "step": 78700
    },
    {
      "entropy": 5.644926309585571,
      "epoch": 7.139423076923077,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013632663375495353,
      "loss": 4.7583,
      "mean_token_accuracy": 0.24181460291147233,
      "num_tokens": 163332984.0,
      "step": 78705
    },
    {
      "entropy": 5.707445812225342,
      "epoch": 7.139876632801161,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013630115479754704,
      "loss": 4.799,
      "mean_token_accuracy": 0.23436026126146317,
      "num_tokens": 163343053.0,
      "step": 78710
    },
    {
      "entropy": 5.659147167205811,
      "epoch": 7.140330188679245,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013627567870846795,
      "loss": 4.7858,
      "mean_token_accuracy": 0.2419064998626709,
      "num_tokens": 163353529.0,
      "step": 78715
    },
    {
      "entropy": 5.695244359970093,
      "epoch": 7.1407837445573294,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013625020548824323,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2379578471183777,
      "num_tokens": 163363032.0,
      "step": 78720
    },
    {
      "entropy": 5.675002384185791,
      "epoch": 7.141237300435414,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013622473513739963,
      "loss": 4.7578,
      "mean_token_accuracy": 0.23051732629537583,
      "num_tokens": 163373680.0,
      "step": 78725
    },
    {
      "entropy": 5.655147933959961,
      "epoch": 7.141690856313498,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013619926765646385,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2453450709581375,
      "num_tokens": 163383289.0,
      "step": 78730
    },
    {
      "entropy": 5.629238891601562,
      "epoch": 7.142144412191582,
      "grad_norm": 1.375,
      "learning_rate": 0.0001361738030459626,
      "loss": 4.7474,
      "mean_token_accuracy": 0.23884391337633132,
      "num_tokens": 163393385.0,
      "step": 78735
    },
    {
      "entropy": 5.688266086578369,
      "epoch": 7.1425979680696665,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001361483413064225,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2327102854847908,
      "num_tokens": 163404043.0,
      "step": 78740
    },
    {
      "entropy": 5.64171404838562,
      "epoch": 7.143051523947751,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001361228824383701,
      "loss": 4.7674,
      "mean_token_accuracy": 0.24687813669443132,
      "num_tokens": 163414737.0,
      "step": 78745
    },
    {
      "entropy": 5.636774921417237,
      "epoch": 7.143505079825834,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013609742644233176,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2369693011045456,
      "num_tokens": 163424651.0,
      "step": 78750
    },
    {
      "entropy": 5.621761703491211,
      "epoch": 7.143958635703918,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013607197331883415,
      "loss": 4.7575,
      "mean_token_accuracy": 0.24113547205924987,
      "num_tokens": 163434651.0,
      "step": 78755
    },
    {
      "entropy": 5.716899633407593,
      "epoch": 7.144412191582003,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013604652306840347,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23415762931108475,
      "num_tokens": 163445867.0,
      "step": 78760
    },
    {
      "entropy": 5.7822174549102785,
      "epoch": 7.144865747460087,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013602107569156616,
      "loss": 4.9127,
      "mean_token_accuracy": 0.23250926285982132,
      "num_tokens": 163455914.0,
      "step": 78765
    },
    {
      "entropy": 5.721843147277832,
      "epoch": 7.145319303338171,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013599563118884842,
      "loss": 4.8583,
      "mean_token_accuracy": 0.2323367178440094,
      "num_tokens": 163465394.0,
      "step": 78770
    },
    {
      "entropy": 5.68798828125,
      "epoch": 7.1457728592162555,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013597018956077646,
      "loss": 4.7846,
      "mean_token_accuracy": 0.2383976846933365,
      "num_tokens": 163476221.0,
      "step": 78775
    },
    {
      "entropy": 5.6364830493927,
      "epoch": 7.14622641509434,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013594475080787639,
      "loss": 4.7456,
      "mean_token_accuracy": 0.23997930139303209,
      "num_tokens": 163485568.0,
      "step": 78780
    },
    {
      "entropy": 5.741697931289673,
      "epoch": 7.146679970972424,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001359193149306742,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22710282057523729,
      "num_tokens": 163495862.0,
      "step": 78785
    },
    {
      "entropy": 5.666263008117676,
      "epoch": 7.147133526850508,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001358938819296961,
      "loss": 4.7172,
      "mean_token_accuracy": 0.23915060609579086,
      "num_tokens": 163505446.0,
      "step": 78790
    },
    {
      "entropy": 5.688984537124634,
      "epoch": 7.1475870827285926,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013586845180546802,
      "loss": 4.8541,
      "mean_token_accuracy": 0.23102352619171143,
      "num_tokens": 163515318.0,
      "step": 78795
    },
    {
      "entropy": 5.765437507629395,
      "epoch": 7.148040638606676,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013584302455851578,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23781955540180205,
      "num_tokens": 163525466.0,
      "step": 78800
    },
    {
      "entropy": 5.693397760391235,
      "epoch": 7.14849419448476,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001358176001893653,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23239183574914932,
      "num_tokens": 163535530.0,
      "step": 78805
    },
    {
      "entropy": 5.766490840911866,
      "epoch": 7.1489477503628445,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013579217869854224,
      "loss": 4.8647,
      "mean_token_accuracy": 0.23449540436267852,
      "num_tokens": 163545661.0,
      "step": 78810
    },
    {
      "entropy": 5.757443761825561,
      "epoch": 7.149401306240929,
      "grad_norm": 1.5,
      "learning_rate": 0.00013576676008657243,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23723236918449403,
      "num_tokens": 163555067.0,
      "step": 78815
    },
    {
      "entropy": 5.608670711517334,
      "epoch": 7.149854862119013,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001357413443539814,
      "loss": 4.7207,
      "mean_token_accuracy": 0.23969468474388123,
      "num_tokens": 163565399.0,
      "step": 78820
    },
    {
      "entropy": 5.625225782394409,
      "epoch": 7.150308417997097,
      "grad_norm": 1.40625,
      "learning_rate": 0.000135715931501295,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2470349609851837,
      "num_tokens": 163575742.0,
      "step": 78825
    },
    {
      "entropy": 5.599849653244019,
      "epoch": 7.1507619738751815,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001356905215290386,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24242496490478516,
      "num_tokens": 163586804.0,
      "step": 78830
    },
    {
      "entropy": 5.669619417190551,
      "epoch": 7.151215529753266,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013566511443773772,
      "loss": 4.7725,
      "mean_token_accuracy": 0.24070354402065278,
      "num_tokens": 163597019.0,
      "step": 78835
    },
    {
      "entropy": 5.6734953880310055,
      "epoch": 7.15166908563135,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013563971022791777,
      "loss": 4.7702,
      "mean_token_accuracy": 0.23788592964410782,
      "num_tokens": 163608384.0,
      "step": 78840
    },
    {
      "entropy": 5.73871340751648,
      "epoch": 7.152122641509434,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001356143089001041,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2194761574268341,
      "num_tokens": 163620035.0,
      "step": 78845
    },
    {
      "entropy": 5.703377389907837,
      "epoch": 7.152576197387518,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00013558891045482198,
      "loss": 4.7667,
      "mean_token_accuracy": 0.23546853810548782,
      "num_tokens": 163629915.0,
      "step": 78850
    },
    {
      "entropy": 5.718580341339111,
      "epoch": 7.153029753265602,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001355635148925968,
      "loss": 4.8378,
      "mean_token_accuracy": 0.23207762688398362,
      "num_tokens": 163642253.0,
      "step": 78855
    },
    {
      "entropy": 5.71967625617981,
      "epoch": 7.153483309143686,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013553812221395368,
      "loss": 4.757,
      "mean_token_accuracy": 0.2403762772679329,
      "num_tokens": 163652418.0,
      "step": 78860
    },
    {
      "entropy": 5.7444366931915285,
      "epoch": 7.1539368650217705,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013551273241941772,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23058467507362365,
      "num_tokens": 163662070.0,
      "step": 78865
    },
    {
      "entropy": 5.726290655136109,
      "epoch": 7.154390420899855,
      "grad_norm": 1.671875,
      "learning_rate": 0.000135487345509514,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23976608961820603,
      "num_tokens": 163671175.0,
      "step": 78870
    },
    {
      "entropy": 5.712575387954712,
      "epoch": 7.154843976777939,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013546196148476753,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23111357539892197,
      "num_tokens": 163681787.0,
      "step": 78875
    },
    {
      "entropy": 5.745876359939575,
      "epoch": 7.155297532656023,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013543658034570328,
      "loss": 4.8724,
      "mean_token_accuracy": 0.23381874114274978,
      "num_tokens": 163692642.0,
      "step": 78880
    },
    {
      "entropy": 5.715390872955322,
      "epoch": 7.155751088534108,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013541120209284595,
      "loss": 4.8562,
      "mean_token_accuracy": 0.23470979183912277,
      "num_tokens": 163702889.0,
      "step": 78885
    },
    {
      "entropy": 5.68479905128479,
      "epoch": 7.156204644412192,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013538582672672068,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23687610477209092,
      "num_tokens": 163713958.0,
      "step": 78890
    },
    {
      "entropy": 5.626420974731445,
      "epoch": 7.156658200290276,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001353604542478521,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2410464510321617,
      "num_tokens": 163725418.0,
      "step": 78895
    },
    {
      "entropy": 5.7249000549316404,
      "epoch": 7.15711175616836,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013533508465676485,
      "loss": 4.7943,
      "mean_token_accuracy": 0.24052910953760148,
      "num_tokens": 163736185.0,
      "step": 78900
    },
    {
      "entropy": 5.743211174011231,
      "epoch": 7.157565312046444,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001353097179539837,
      "loss": 4.8747,
      "mean_token_accuracy": 0.22826278507709502,
      "num_tokens": 163747884.0,
      "step": 78905
    },
    {
      "entropy": 5.723458099365234,
      "epoch": 7.158018867924528,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013528435414003316,
      "loss": 4.8423,
      "mean_token_accuracy": 0.23356681019067765,
      "num_tokens": 163757703.0,
      "step": 78910
    },
    {
      "entropy": 5.747848796844482,
      "epoch": 7.158472423802612,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013525899321543776,
      "loss": 4.851,
      "mean_token_accuracy": 0.23541519194841384,
      "num_tokens": 163767757.0,
      "step": 78915
    },
    {
      "entropy": 5.700974941253662,
      "epoch": 7.1589259796806965,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013523363518072197,
      "loss": 4.7742,
      "mean_token_accuracy": 0.24454165250062943,
      "num_tokens": 163777792.0,
      "step": 78920
    },
    {
      "entropy": 5.6661182880401615,
      "epoch": 7.159379535558781,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013520828003641026,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23348041176795958,
      "num_tokens": 163788188.0,
      "step": 78925
    },
    {
      "entropy": 5.703259468078613,
      "epoch": 7.159833091436865,
      "grad_norm": 1.375,
      "learning_rate": 0.00013518292778302694,
      "loss": 4.8506,
      "mean_token_accuracy": 0.23202528357505797,
      "num_tokens": 163798935.0,
      "step": 78930
    },
    {
      "entropy": 5.732689762115479,
      "epoch": 7.160286647314949,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001351575784210963,
      "loss": 4.7892,
      "mean_token_accuracy": 0.23559719771146775,
      "num_tokens": 163810415.0,
      "step": 78935
    },
    {
      "entropy": 5.749159860610962,
      "epoch": 7.160740203193034,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001351322319511426,
      "loss": 4.8705,
      "mean_token_accuracy": 0.23065244704484938,
      "num_tokens": 163820960.0,
      "step": 78940
    },
    {
      "entropy": 5.692430019378662,
      "epoch": 7.161193759071118,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013510688837368995,
      "loss": 4.727,
      "mean_token_accuracy": 0.24607384353876113,
      "num_tokens": 163831212.0,
      "step": 78945
    },
    {
      "entropy": 5.664276123046875,
      "epoch": 7.161647314949202,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013508154768926252,
      "loss": 4.7854,
      "mean_token_accuracy": 0.23305121660232545,
      "num_tokens": 163842386.0,
      "step": 78950
    },
    {
      "entropy": 5.75643835067749,
      "epoch": 7.1621008708272855,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013505620989838425,
      "loss": 4.9359,
      "mean_token_accuracy": 0.2310601994395256,
      "num_tokens": 163852228.0,
      "step": 78955
    },
    {
      "entropy": 5.71356430053711,
      "epoch": 7.16255442670537,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001350308750015793,
      "loss": 4.8301,
      "mean_token_accuracy": 0.23157945424318313,
      "num_tokens": 163863322.0,
      "step": 78960
    },
    {
      "entropy": 5.6789562702178955,
      "epoch": 7.163007982583454,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001350055429993715,
      "loss": 4.7316,
      "mean_token_accuracy": 0.24271144568920136,
      "num_tokens": 163874145.0,
      "step": 78965
    },
    {
      "entropy": 5.7319330215454105,
      "epoch": 7.163461538461538,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001349802138922848,
      "loss": 4.8462,
      "mean_token_accuracy": 0.23543092906475066,
      "num_tokens": 163884619.0,
      "step": 78970
    },
    {
      "entropy": 5.705303382873535,
      "epoch": 7.163915094339623,
      "grad_norm": 1.5,
      "learning_rate": 0.00013495488768084293,
      "loss": 4.8341,
      "mean_token_accuracy": 0.2317599833011627,
      "num_tokens": 163895035.0,
      "step": 78975
    },
    {
      "entropy": 5.642584228515625,
      "epoch": 7.164368650217707,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013492956436556965,
      "loss": 4.697,
      "mean_token_accuracy": 0.24685315042734146,
      "num_tokens": 163905156.0,
      "step": 78980
    },
    {
      "entropy": 5.650534963607788,
      "epoch": 7.164822206095791,
      "grad_norm": 1.5,
      "learning_rate": 0.0001349042439469887,
      "loss": 4.8066,
      "mean_token_accuracy": 0.2356936365365982,
      "num_tokens": 163915854.0,
      "step": 78985
    },
    {
      "entropy": 5.649337911605835,
      "epoch": 7.165275761973875,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001348789264256236,
      "loss": 4.7819,
      "mean_token_accuracy": 0.23787357807159423,
      "num_tokens": 163926555.0,
      "step": 78990
    },
    {
      "entropy": 5.68022837638855,
      "epoch": 7.16572931785196,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013485361180199813,
      "loss": 4.808,
      "mean_token_accuracy": 0.2407954916357994,
      "num_tokens": 163936868.0,
      "step": 78995
    },
    {
      "entropy": 5.74219446182251,
      "epoch": 7.166182873730044,
      "grad_norm": 1.375,
      "learning_rate": 0.0001348283000766356,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2210420250892639,
      "num_tokens": 163946860.0,
      "step": 79000
    },
    {
      "entropy": 5.680278253555298,
      "epoch": 7.166636429608127,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001348029912500596,
      "loss": 4.7852,
      "mean_token_accuracy": 0.23885821104049682,
      "num_tokens": 163957039.0,
      "step": 79005
    },
    {
      "entropy": 5.707816123962402,
      "epoch": 7.167089985486212,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013477768532279343,
      "loss": 4.8354,
      "mean_token_accuracy": 0.22776168584823608,
      "num_tokens": 163967352.0,
      "step": 79010
    },
    {
      "entropy": 5.670014381408691,
      "epoch": 7.167543541364296,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001347523822953605,
      "loss": 4.7536,
      "mean_token_accuracy": 0.23629827350378035,
      "num_tokens": 163977680.0,
      "step": 79015
    },
    {
      "entropy": 5.726130723953247,
      "epoch": 7.16799709724238,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013472708216828404,
      "loss": 4.7709,
      "mean_token_accuracy": 0.23841010332107543,
      "num_tokens": 163987267.0,
      "step": 79020
    },
    {
      "entropy": 5.68101863861084,
      "epoch": 7.168450653120464,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013470178494208717,
      "loss": 4.8098,
      "mean_token_accuracy": 0.23135825097560883,
      "num_tokens": 163997702.0,
      "step": 79025
    },
    {
      "entropy": 5.688529968261719,
      "epoch": 7.168904208998549,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013467649061729323,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23202312588691712,
      "num_tokens": 164007946.0,
      "step": 79030
    },
    {
      "entropy": 5.664743375778198,
      "epoch": 7.169357764876633,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001346511991944252,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23611368238925934,
      "num_tokens": 164018697.0,
      "step": 79035
    },
    {
      "entropy": 5.6852884769439695,
      "epoch": 7.169811320754717,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013462591067400615,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2339218571782112,
      "num_tokens": 164029956.0,
      "step": 79040
    },
    {
      "entropy": 5.733752679824829,
      "epoch": 7.170264876632801,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013460062505655913,
      "loss": 4.8179,
      "mean_token_accuracy": 0.23926514983177186,
      "num_tokens": 164039901.0,
      "step": 79045
    },
    {
      "entropy": 5.7172401428222654,
      "epoch": 7.170718432510886,
      "grad_norm": 1.5,
      "learning_rate": 0.00013457534234260677,
      "loss": 4.7892,
      "mean_token_accuracy": 0.23761404007673265,
      "num_tokens": 164050141.0,
      "step": 79050
    },
    {
      "entropy": 5.780180644989014,
      "epoch": 7.17117198838897,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013455006253267223,
      "loss": 4.873,
      "mean_token_accuracy": 0.23118657022714614,
      "num_tokens": 164060460.0,
      "step": 79055
    },
    {
      "entropy": 5.701549482345581,
      "epoch": 7.171625544267053,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013452478562727815,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23336227238178253,
      "num_tokens": 164070695.0,
      "step": 79060
    },
    {
      "entropy": 5.661666631698608,
      "epoch": 7.172079100145138,
      "grad_norm": 1.5,
      "learning_rate": 0.00013449951162694735,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2374382421374321,
      "num_tokens": 164080513.0,
      "step": 79065
    },
    {
      "entropy": 5.68557825088501,
      "epoch": 7.172532656023222,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001344742405322024,
      "loss": 4.832,
      "mean_token_accuracy": 0.23363005518913268,
      "num_tokens": 164090584.0,
      "step": 79070
    },
    {
      "entropy": 5.702570629119873,
      "epoch": 7.172986211901306,
      "grad_norm": 1.515625,
      "learning_rate": 0.000134448972343566,
      "loss": 4.7933,
      "mean_token_accuracy": 0.24515588134527205,
      "num_tokens": 164101374.0,
      "step": 79075
    },
    {
      "entropy": 5.694678401947021,
      "epoch": 7.17343976777939,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013442370706156072,
      "loss": 4.7345,
      "mean_token_accuracy": 0.23585380762815475,
      "num_tokens": 164112068.0,
      "step": 79080
    },
    {
      "entropy": 5.681809282302856,
      "epoch": 7.173893323657475,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013439844468670883,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23046254515647888,
      "num_tokens": 164122681.0,
      "step": 79085
    },
    {
      "entropy": 5.652685880661011,
      "epoch": 7.174346879535559,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013437318521953307,
      "loss": 4.7979,
      "mean_token_accuracy": 0.24561240375041962,
      "num_tokens": 164132937.0,
      "step": 79090
    },
    {
      "entropy": 5.641966104507446,
      "epoch": 7.174800435413643,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013434792866055562,
      "loss": 4.7595,
      "mean_token_accuracy": 0.2351033106446266,
      "num_tokens": 164142947.0,
      "step": 79095
    },
    {
      "entropy": 5.726284027099609,
      "epoch": 7.1752539912917275,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013432267501029888,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2322920724749565,
      "num_tokens": 164152836.0,
      "step": 79100
    },
    {
      "entropy": 5.762383508682251,
      "epoch": 7.175707547169812,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001342974242692851,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22796375304460526,
      "num_tokens": 164163733.0,
      "step": 79105
    },
    {
      "entropy": 5.775234413146973,
      "epoch": 7.176161103047895,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001342721764380364,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23163250386714934,
      "num_tokens": 164174673.0,
      "step": 79110
    },
    {
      "entropy": 5.756480121612549,
      "epoch": 7.176614658925979,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013424693151707496,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22640465050935746,
      "num_tokens": 164185138.0,
      "step": 79115
    },
    {
      "entropy": 5.678897476196289,
      "epoch": 7.177068214804064,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013422168950692276,
      "loss": 4.8041,
      "mean_token_accuracy": 0.23154232800006866,
      "num_tokens": 164194741.0,
      "step": 79120
    },
    {
      "entropy": 5.682958173751831,
      "epoch": 7.177521770682148,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00013419645040810198,
      "loss": 4.7834,
      "mean_token_accuracy": 0.24101862907409669,
      "num_tokens": 164204479.0,
      "step": 79125
    },
    {
      "entropy": 5.710055875778198,
      "epoch": 7.177975326560232,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001341712142211345,
      "loss": 4.9252,
      "mean_token_accuracy": 0.22285580933094024,
      "num_tokens": 164215080.0,
      "step": 79130
    },
    {
      "entropy": 5.6301202297210695,
      "epoch": 7.1784288824383164,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00013414598094654219,
      "loss": 4.7708,
      "mean_token_accuracy": 0.23810305148363115,
      "num_tokens": 164224296.0,
      "step": 79135
    },
    {
      "entropy": 5.698908472061158,
      "epoch": 7.178882438316401,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001341207505848469,
      "loss": 4.7937,
      "mean_token_accuracy": 0.23994258791208267,
      "num_tokens": 164234481.0,
      "step": 79140
    },
    {
      "entropy": 5.726646804809571,
      "epoch": 7.179335994194485,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013409552313657037,
      "loss": 4.8703,
      "mean_token_accuracy": 0.23271180540323258,
      "num_tokens": 164244054.0,
      "step": 79145
    },
    {
      "entropy": 5.751719856262207,
      "epoch": 7.179789550072569,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013407029860223424,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22984601855278014,
      "num_tokens": 164253457.0,
      "step": 79150
    },
    {
      "entropy": 5.6744797706604,
      "epoch": 7.1802431059506535,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013404507698236035,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23122865110635757,
      "num_tokens": 164263404.0,
      "step": 79155
    },
    {
      "entropy": 5.610033130645752,
      "epoch": 7.180696661828737,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001340198582774702,
      "loss": 4.653,
      "mean_token_accuracy": 0.25179181545972823,
      "num_tokens": 164273125.0,
      "step": 79160
    },
    {
      "entropy": 5.725728464126587,
      "epoch": 7.181150217706821,
      "grad_norm": 1.5,
      "learning_rate": 0.00013399464248808527,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22974277138710023,
      "num_tokens": 164283484.0,
      "step": 79165
    },
    {
      "entropy": 5.746385908126831,
      "epoch": 7.181603773584905,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013396942961472707,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2311795398592949,
      "num_tokens": 164293742.0,
      "step": 79170
    },
    {
      "entropy": 5.68889741897583,
      "epoch": 7.18205732946299,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013394421965791704,
      "loss": 4.8015,
      "mean_token_accuracy": 0.23178107291460037,
      "num_tokens": 164304126.0,
      "step": 79175
    },
    {
      "entropy": 5.687674236297608,
      "epoch": 7.182510885341074,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013391901261817647,
      "loss": 4.7957,
      "mean_token_accuracy": 0.23561637848615646,
      "num_tokens": 164314087.0,
      "step": 79180
    },
    {
      "entropy": 5.680413484573364,
      "epoch": 7.182964441219158,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013389380849602656,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23590485751628876,
      "num_tokens": 164324535.0,
      "step": 79185
    },
    {
      "entropy": 5.634315347671508,
      "epoch": 7.1834179970972425,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001338686072919888,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2503853186964989,
      "num_tokens": 164333327.0,
      "step": 79190
    },
    {
      "entropy": 5.708023595809936,
      "epoch": 7.183871552975327,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013384340900658416,
      "loss": 4.7918,
      "mean_token_accuracy": 0.2416976884007454,
      "num_tokens": 164343856.0,
      "step": 79195
    },
    {
      "entropy": 5.687477159500122,
      "epoch": 7.184325108853411,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001338182136403338,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2300134062767029,
      "num_tokens": 164354709.0,
      "step": 79200
    },
    {
      "entropy": 5.721173048019409,
      "epoch": 7.184778664731495,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013379302119375874,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23532824963331223,
      "num_tokens": 164364947.0,
      "step": 79205
    },
    {
      "entropy": 5.735647344589234,
      "epoch": 7.18523222060958,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013376783166737998,
      "loss": 4.8479,
      "mean_token_accuracy": 0.2319689378142357,
      "num_tokens": 164375250.0,
      "step": 79210
    },
    {
      "entropy": 5.721848011016846,
      "epoch": 7.185685776487663,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013374264506171846,
      "loss": 4.8535,
      "mean_token_accuracy": 0.23433109670877456,
      "num_tokens": 164384970.0,
      "step": 79215
    },
    {
      "entropy": 5.717932796478271,
      "epoch": 7.186139332365747,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013371746137729498,
      "loss": 4.8001,
      "mean_token_accuracy": 0.2418528154492378,
      "num_tokens": 164395372.0,
      "step": 79220
    },
    {
      "entropy": 5.7134490489959715,
      "epoch": 7.1865928882438315,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001336922806146305,
      "loss": 4.7606,
      "mean_token_accuracy": 0.2393543541431427,
      "num_tokens": 164405419.0,
      "step": 79225
    },
    {
      "entropy": 5.692757892608642,
      "epoch": 7.187046444121916,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013366710277424562,
      "loss": 4.7499,
      "mean_token_accuracy": 0.25157375931739806,
      "num_tokens": 164415564.0,
      "step": 79230
    },
    {
      "entropy": 5.721685552597046,
      "epoch": 7.1875,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001336419278566611,
      "loss": 4.8773,
      "mean_token_accuracy": 0.23398803174495697,
      "num_tokens": 164426052.0,
      "step": 79235
    },
    {
      "entropy": 5.683283042907715,
      "epoch": 7.187953555878084,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013361675586239752,
      "loss": 4.7465,
      "mean_token_accuracy": 0.24299316108226776,
      "num_tokens": 164436396.0,
      "step": 79240
    },
    {
      "entropy": 5.708822345733642,
      "epoch": 7.1884071117561685,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013359158679197549,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2337556555867195,
      "num_tokens": 164447182.0,
      "step": 79245
    },
    {
      "entropy": 5.633001852035522,
      "epoch": 7.188860667634253,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001335664206459154,
      "loss": 4.7093,
      "mean_token_accuracy": 0.24891120940446854,
      "num_tokens": 164457379.0,
      "step": 79250
    },
    {
      "entropy": 5.701002883911133,
      "epoch": 7.189314223512337,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013354125742473776,
      "loss": 4.8584,
      "mean_token_accuracy": 0.22814934104681014,
      "num_tokens": 164470076.0,
      "step": 79255
    },
    {
      "entropy": 5.642718362808227,
      "epoch": 7.18976777939042,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013351609712896302,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22952521592378616,
      "num_tokens": 164480778.0,
      "step": 79260
    },
    {
      "entropy": 5.766017770767212,
      "epoch": 7.190221335268505,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013349093975911148,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22654128968715667,
      "num_tokens": 164490369.0,
      "step": 79265
    },
    {
      "entropy": 5.678866338729859,
      "epoch": 7.190674891146589,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001334657853157033,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23965320438146592,
      "num_tokens": 164500597.0,
      "step": 79270
    },
    {
      "entropy": 5.710966920852661,
      "epoch": 7.191128447024673,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001334406337992588,
      "loss": 4.8005,
      "mean_token_accuracy": 0.23999689072370528,
      "num_tokens": 164510610.0,
      "step": 79275
    },
    {
      "entropy": 5.696527099609375,
      "epoch": 7.1915820029027575,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013341548521029806,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2420249655842781,
      "num_tokens": 164520215.0,
      "step": 79280
    },
    {
      "entropy": 5.737711715698242,
      "epoch": 7.192035558780842,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00013339033954934117,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2282555565237999,
      "num_tokens": 164530521.0,
      "step": 79285
    },
    {
      "entropy": 5.723300981521606,
      "epoch": 7.192489114658926,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013336519681690805,
      "loss": 4.8239,
      "mean_token_accuracy": 0.23686571568250656,
      "num_tokens": 164540357.0,
      "step": 79290
    },
    {
      "entropy": 5.728692245483399,
      "epoch": 7.19294267053701,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013334005701351882,
      "loss": 4.84,
      "mean_token_accuracy": 0.23361553251743317,
      "num_tokens": 164550696.0,
      "step": 79295
    },
    {
      "entropy": 5.669595956802368,
      "epoch": 7.193396226415095,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013331492013969335,
      "loss": 4.7779,
      "mean_token_accuracy": 0.24014778584241867,
      "num_tokens": 164561299.0,
      "step": 79300
    },
    {
      "entropy": 5.766583204269409,
      "epoch": 7.193849782293179,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001332897861959514,
      "loss": 4.8127,
      "mean_token_accuracy": 0.24205285906791688,
      "num_tokens": 164571352.0,
      "step": 79305
    },
    {
      "entropy": 5.639143371582032,
      "epoch": 7.194303338171263,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001332646551828128,
      "loss": 4.7387,
      "mean_token_accuracy": 0.2368542268872261,
      "num_tokens": 164581756.0,
      "step": 79310
    },
    {
      "entropy": 5.670608711242676,
      "epoch": 7.1947568940493465,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013323952710079723,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2353069454431534,
      "num_tokens": 164592786.0,
      "step": 79315
    },
    {
      "entropy": 5.642485904693603,
      "epoch": 7.195210449927431,
      "grad_norm": 1.625,
      "learning_rate": 0.0001332144019504244,
      "loss": 4.7493,
      "mean_token_accuracy": 0.2398459106683731,
      "num_tokens": 164602585.0,
      "step": 79320
    },
    {
      "entropy": 5.605033254623413,
      "epoch": 7.195664005805515,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013318927973221374,
      "loss": 4.7568,
      "mean_token_accuracy": 0.24592837393283845,
      "num_tokens": 164613022.0,
      "step": 79325
    },
    {
      "entropy": 5.722410535812378,
      "epoch": 7.196117561683599,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013316416044668506,
      "loss": 4.7856,
      "mean_token_accuracy": 0.23757069706916809,
      "num_tokens": 164624017.0,
      "step": 79330
    },
    {
      "entropy": 5.676715660095215,
      "epoch": 7.1965711175616836,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00013313904409435764,
      "loss": 4.7446,
      "mean_token_accuracy": 0.23631299734115602,
      "num_tokens": 164633864.0,
      "step": 79335
    },
    {
      "entropy": 5.724905586242675,
      "epoch": 7.197024673439768,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013311393067575095,
      "loss": 4.9083,
      "mean_token_accuracy": 0.22845513969659806,
      "num_tokens": 164645013.0,
      "step": 79340
    },
    {
      "entropy": 5.740492057800293,
      "epoch": 7.197478229317852,
      "grad_norm": 1.5,
      "learning_rate": 0.00013308882019138433,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22920852899551392,
      "num_tokens": 164655364.0,
      "step": 79345
    },
    {
      "entropy": 5.716395998001099,
      "epoch": 7.197931785195936,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013306371264177705,
      "loss": 4.8067,
      "mean_token_accuracy": 0.23539017885923386,
      "num_tokens": 164665987.0,
      "step": 79350
    },
    {
      "entropy": 5.761898803710937,
      "epoch": 7.198385341074021,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013303860802744833,
      "loss": 4.9553,
      "mean_token_accuracy": 0.22239097207784653,
      "num_tokens": 164677556.0,
      "step": 79355
    },
    {
      "entropy": 5.713170289993286,
      "epoch": 7.198838896952105,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001330135063489174,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2301214024424553,
      "num_tokens": 164687179.0,
      "step": 79360
    },
    {
      "entropy": 5.763459634780884,
      "epoch": 7.199292452830188,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001329884076067034,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2255332887172699,
      "num_tokens": 164697295.0,
      "step": 79365
    },
    {
      "entropy": 5.736675596237182,
      "epoch": 7.1997460087082725,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013296331180132526,
      "loss": 4.8696,
      "mean_token_accuracy": 0.2334085538983345,
      "num_tokens": 164708066.0,
      "step": 79370
    },
    {
      "entropy": 5.7010291576385494,
      "epoch": 7.200199564586357,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013293821893330205,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2453637719154358,
      "num_tokens": 164718578.0,
      "step": 79375
    },
    {
      "entropy": 5.7396961688995365,
      "epoch": 7.200653120464441,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00013291312900315266,
      "loss": 4.8785,
      "mean_token_accuracy": 0.23421580493450164,
      "num_tokens": 164728751.0,
      "step": 79380
    },
    {
      "entropy": 5.700749444961548,
      "epoch": 7.201106676342525,
      "grad_norm": 1.484375,
      "learning_rate": 0.000132888042011396,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23309321701526642,
      "num_tokens": 164739443.0,
      "step": 79385
    },
    {
      "entropy": 5.715386581420899,
      "epoch": 7.20156023222061,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013286295795855076,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23216477483510972,
      "num_tokens": 164749684.0,
      "step": 79390
    },
    {
      "entropy": 5.761003017425537,
      "epoch": 7.202013788098694,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013283787684513582,
      "loss": 4.8823,
      "mean_token_accuracy": 0.2268485128879547,
      "num_tokens": 164761061.0,
      "step": 79395
    },
    {
      "entropy": 5.720073652267456,
      "epoch": 7.202467343976778,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013281279867166984,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22367390990257263,
      "num_tokens": 164771297.0,
      "step": 79400
    },
    {
      "entropy": 5.660160398483276,
      "epoch": 7.202920899854862,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001327877234386714,
      "loss": 4.7427,
      "mean_token_accuracy": 0.2467256158590317,
      "num_tokens": 164781365.0,
      "step": 79405
    },
    {
      "entropy": 5.634328413009643,
      "epoch": 7.203374455732947,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013276265114665913,
      "loss": 4.7539,
      "mean_token_accuracy": 0.23381170332431794,
      "num_tokens": 164791823.0,
      "step": 79410
    },
    {
      "entropy": 5.737472200393677,
      "epoch": 7.20382801161103,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013273758179615142,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23029468208551407,
      "num_tokens": 164803090.0,
      "step": 79415
    },
    {
      "entropy": 5.718024063110351,
      "epoch": 7.204281567489114,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001327125153876668,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2309161901473999,
      "num_tokens": 164812753.0,
      "step": 79420
    },
    {
      "entropy": 5.7328300952911375,
      "epoch": 7.204735123367199,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013268745192172355,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22679451256990432,
      "num_tokens": 164823121.0,
      "step": 79425
    },
    {
      "entropy": 5.630971097946167,
      "epoch": 7.205188679245283,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013266239139884012,
      "loss": 4.7052,
      "mean_token_accuracy": 0.2521443396806717,
      "num_tokens": 164832910.0,
      "step": 79430
    },
    {
      "entropy": 5.696270990371704,
      "epoch": 7.205642235123367,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001326373338195347,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22692478746175765,
      "num_tokens": 164842081.0,
      "step": 79435
    },
    {
      "entropy": 5.704459857940674,
      "epoch": 7.206095791001451,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001326122791843255,
      "loss": 4.8209,
      "mean_token_accuracy": 0.23890772759914397,
      "num_tokens": 164852733.0,
      "step": 79440
    },
    {
      "entropy": 5.619851207733154,
      "epoch": 7.206549346879536,
      "grad_norm": 1.609375,
      "learning_rate": 0.00013258722749373065,
      "loss": 4.741,
      "mean_token_accuracy": 0.24152794629335403,
      "num_tokens": 164862484.0,
      "step": 79445
    },
    {
      "entropy": 5.720030355453491,
      "epoch": 7.20700290275762,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013256217874826827,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23854987621307372,
      "num_tokens": 164872395.0,
      "step": 79450
    },
    {
      "entropy": 5.6490894794464115,
      "epoch": 7.207456458635704,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013253713294845626,
      "loss": 4.6817,
      "mean_token_accuracy": 0.24634997844696044,
      "num_tokens": 164881967.0,
      "step": 79455
    },
    {
      "entropy": 5.673485946655274,
      "epoch": 7.207910014513788,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013251209009481263,
      "loss": 4.7458,
      "mean_token_accuracy": 0.24327891767024995,
      "num_tokens": 164891787.0,
      "step": 79460
    },
    {
      "entropy": 5.642235803604126,
      "epoch": 7.208363570391873,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013248705018785534,
      "loss": 4.7891,
      "mean_token_accuracy": 0.2434471920132637,
      "num_tokens": 164902245.0,
      "step": 79465
    },
    {
      "entropy": 5.798626041412353,
      "epoch": 7.208817126269956,
      "grad_norm": 1.765625,
      "learning_rate": 0.00013246201322810223,
      "loss": 4.919,
      "mean_token_accuracy": 0.22978457510471345,
      "num_tokens": 164912331.0,
      "step": 79470
    },
    {
      "entropy": 5.723922920227051,
      "epoch": 7.20927068214804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013243697921607096,
      "loss": 4.824,
      "mean_token_accuracy": 0.24061060547828675,
      "num_tokens": 164922672.0,
      "step": 79475
    },
    {
      "entropy": 5.707738447189331,
      "epoch": 7.209724238026125,
      "grad_norm": 1.640625,
      "learning_rate": 0.00013241194815227927,
      "loss": 4.848,
      "mean_token_accuracy": 0.23206622749567032,
      "num_tokens": 164933590.0,
      "step": 79480
    },
    {
      "entropy": 5.646531248092652,
      "epoch": 7.210177793904209,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013238692003724476,
      "loss": 4.7327,
      "mean_token_accuracy": 0.23809253573417663,
      "num_tokens": 164944859.0,
      "step": 79485
    },
    {
      "entropy": 5.771115732192993,
      "epoch": 7.210631349782293,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013236189487148516,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22574716061353683,
      "num_tokens": 164955699.0,
      "step": 79490
    },
    {
      "entropy": 5.648886442184448,
      "epoch": 7.211084905660377,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001323368726555179,
      "loss": 4.7293,
      "mean_token_accuracy": 0.24212522953748702,
      "num_tokens": 164966504.0,
      "step": 79495
    },
    {
      "entropy": 5.633896255493164,
      "epoch": 7.211538461538462,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013231185338986051,
      "loss": 4.7541,
      "mean_token_accuracy": 0.24134291857481002,
      "num_tokens": 164977491.0,
      "step": 79500
    },
    {
      "entropy": 5.652782583236695,
      "epoch": 7.211992017416546,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013228683707503032,
      "loss": 4.7003,
      "mean_token_accuracy": 0.2411433205008507,
      "num_tokens": 164988085.0,
      "step": 79505
    },
    {
      "entropy": 5.691669511795044,
      "epoch": 7.21244557329463,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013226182371154472,
      "loss": 4.802,
      "mean_token_accuracy": 0.23614624440670012,
      "num_tokens": 164998432.0,
      "step": 79510
    },
    {
      "entropy": 5.7633140087127686,
      "epoch": 7.2128991291727145,
      "grad_norm": 1.53125,
      "learning_rate": 0.000132236813299921,
      "loss": 4.924,
      "mean_token_accuracy": 0.2275252401828766,
      "num_tokens": 165008775.0,
      "step": 79515
    },
    {
      "entropy": 5.78620080947876,
      "epoch": 7.213352685050798,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013221180584067628,
      "loss": 4.922,
      "mean_token_accuracy": 0.23008792251348495,
      "num_tokens": 165018291.0,
      "step": 79520
    },
    {
      "entropy": 5.755307006835937,
      "epoch": 7.213806240928882,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013218680133432782,
      "loss": 4.8911,
      "mean_token_accuracy": 0.23103784024715424,
      "num_tokens": 165029143.0,
      "step": 79525
    },
    {
      "entropy": 5.729780912399292,
      "epoch": 7.214259796806966,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013216179978139275,
      "loss": 4.8005,
      "mean_token_accuracy": 0.23375299721956252,
      "num_tokens": 165039007.0,
      "step": 79530
    },
    {
      "entropy": 5.767852544784546,
      "epoch": 7.214713352685051,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013213680118238808,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23763471096754074,
      "num_tokens": 165050301.0,
      "step": 79535
    },
    {
      "entropy": 5.608331775665283,
      "epoch": 7.215166908563135,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013211180553783072,
      "loss": 4.7536,
      "mean_token_accuracy": 0.2343786284327507,
      "num_tokens": 165061202.0,
      "step": 79540
    },
    {
      "entropy": 5.613238000869751,
      "epoch": 7.215620464441219,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013208681284823768,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2414446219801903,
      "num_tokens": 165070647.0,
      "step": 79545
    },
    {
      "entropy": 5.691533994674683,
      "epoch": 7.2160740203193035,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013206182311412574,
      "loss": 4.9261,
      "mean_token_accuracy": 0.22604478150606155,
      "num_tokens": 165081165.0,
      "step": 79550
    },
    {
      "entropy": 5.733456039428711,
      "epoch": 7.216527576197388,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013203683633601166,
      "loss": 4.8268,
      "mean_token_accuracy": 0.22918204963207245,
      "num_tokens": 165090290.0,
      "step": 79555
    },
    {
      "entropy": 5.767062664031982,
      "epoch": 7.216981132075472,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013201185251441227,
      "loss": 4.911,
      "mean_token_accuracy": 0.22921855449676515,
      "num_tokens": 165101951.0,
      "step": 79560
    },
    {
      "entropy": 5.714149332046508,
      "epoch": 7.217434687953556,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013198687164984424,
      "loss": 4.8037,
      "mean_token_accuracy": 0.24415821582078934,
      "num_tokens": 165111471.0,
      "step": 79565
    },
    {
      "entropy": 5.691723299026489,
      "epoch": 7.21788824383164,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013196189374282413,
      "loss": 4.6792,
      "mean_token_accuracy": 0.24642592817544937,
      "num_tokens": 165120831.0,
      "step": 79570
    },
    {
      "entropy": 5.759146642684937,
      "epoch": 7.218341799709724,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013193691879386855,
      "loss": 4.8871,
      "mean_token_accuracy": 0.23332500755786895,
      "num_tokens": 165131790.0,
      "step": 79575
    },
    {
      "entropy": 5.649212503433228,
      "epoch": 7.218795355587808,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013191194680349394,
      "loss": 4.7735,
      "mean_token_accuracy": 0.24286335110664367,
      "num_tokens": 165143162.0,
      "step": 79580
    },
    {
      "entropy": 5.739615869522095,
      "epoch": 7.219248911465892,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013188697777221669,
      "loss": 4.8939,
      "mean_token_accuracy": 0.23126439452171327,
      "num_tokens": 165154079.0,
      "step": 79585
    },
    {
      "entropy": 5.692554140090943,
      "epoch": 7.219702467343977,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013186201170055318,
      "loss": 4.8563,
      "mean_token_accuracy": 0.23661043494939804,
      "num_tokens": 165164489.0,
      "step": 79590
    },
    {
      "entropy": 5.699508953094482,
      "epoch": 7.220156023222061,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001318370485890198,
      "loss": 4.81,
      "mean_token_accuracy": 0.2433260515332222,
      "num_tokens": 165175123.0,
      "step": 79595
    },
    {
      "entropy": 5.6830991268157955,
      "epoch": 7.220609579100145,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013181208843813274,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23298896104097366,
      "num_tokens": 165185004.0,
      "step": 79600
    },
    {
      "entropy": 5.707776689529419,
      "epoch": 7.2210631349782295,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013178713124840818,
      "loss": 4.8349,
      "mean_token_accuracy": 0.2310197040438652,
      "num_tokens": 165194382.0,
      "step": 79605
    },
    {
      "entropy": 5.66915717124939,
      "epoch": 7.221516690856314,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001317621770203622,
      "loss": 4.7524,
      "mean_token_accuracy": 0.23747088760137558,
      "num_tokens": 165204741.0,
      "step": 79610
    },
    {
      "entropy": 5.753220844268799,
      "epoch": 7.221970246734398,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013173722575451097,
      "loss": 4.8612,
      "mean_token_accuracy": 0.23261500597000123,
      "num_tokens": 165214626.0,
      "step": 79615
    },
    {
      "entropy": 5.7147400856018065,
      "epoch": 7.222423802612482,
      "grad_norm": 1.5,
      "learning_rate": 0.00013171227745137037,
      "loss": 4.7567,
      "mean_token_accuracy": 0.24036578983068466,
      "num_tokens": 165226452.0,
      "step": 79620
    },
    {
      "entropy": 5.663197183609009,
      "epoch": 7.222877358490566,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00013168733211145633,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23725578635931016,
      "num_tokens": 165235953.0,
      "step": 79625
    },
    {
      "entropy": 5.612479209899902,
      "epoch": 7.22333091436865,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013166238973528487,
      "loss": 4.7264,
      "mean_token_accuracy": 0.2420344114303589,
      "num_tokens": 165245631.0,
      "step": 79630
    },
    {
      "entropy": 5.706305599212646,
      "epoch": 7.223784470246734,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001316374503233717,
      "loss": 4.8256,
      "mean_token_accuracy": 0.23606959730386734,
      "num_tokens": 165256291.0,
      "step": 79635
    },
    {
      "entropy": 5.613236951828003,
      "epoch": 7.2242380261248185,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013161251387623258,
      "loss": 4.6926,
      "mean_token_accuracy": 0.2510713770985603,
      "num_tokens": 165265641.0,
      "step": 79640
    },
    {
      "entropy": 5.738950395584107,
      "epoch": 7.224691582002903,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013158758039438323,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2350037008523941,
      "num_tokens": 165276184.0,
      "step": 79645
    },
    {
      "entropy": 5.715802192687988,
      "epoch": 7.225145137880987,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013156264987833927,
      "loss": 4.916,
      "mean_token_accuracy": 0.22792281955480576,
      "num_tokens": 165286855.0,
      "step": 79650
    },
    {
      "entropy": 5.719462299346924,
      "epoch": 7.225598693759071,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013153772232861615,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22812084555625917,
      "num_tokens": 165297683.0,
      "step": 79655
    },
    {
      "entropy": 5.703419303894043,
      "epoch": 7.2260522496371555,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001315127977457296,
      "loss": 4.8543,
      "mean_token_accuracy": 0.23617464005947114,
      "num_tokens": 165307851.0,
      "step": 79660
    },
    {
      "entropy": 5.643448209762573,
      "epoch": 7.22650580551524,
      "grad_norm": 1.375,
      "learning_rate": 0.00013148787613019496,
      "loss": 4.6959,
      "mean_token_accuracy": 0.24246176034212114,
      "num_tokens": 165319318.0,
      "step": 79665
    },
    {
      "entropy": 5.675776815414428,
      "epoch": 7.226959361393324,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001314629574825276,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2484529048204422,
      "num_tokens": 165329841.0,
      "step": 79670
    },
    {
      "entropy": 5.684766864776611,
      "epoch": 7.2274129172714074,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013143804180324286,
      "loss": 4.7749,
      "mean_token_accuracy": 0.2431527629494667,
      "num_tokens": 165340669.0,
      "step": 79675
    },
    {
      "entropy": 5.679160213470459,
      "epoch": 7.227866473149492,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013141312909285602,
      "loss": 4.809,
      "mean_token_accuracy": 0.23561954796314238,
      "num_tokens": 165351462.0,
      "step": 79680
    },
    {
      "entropy": 5.682598018646241,
      "epoch": 7.228320029027576,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013138821935188222,
      "loss": 4.8179,
      "mean_token_accuracy": 0.23311811089515685,
      "num_tokens": 165361521.0,
      "step": 79685
    },
    {
      "entropy": 5.729411268234253,
      "epoch": 7.22877358490566,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001313633125808366,
      "loss": 4.8712,
      "mean_token_accuracy": 0.22900527268648146,
      "num_tokens": 165370443.0,
      "step": 79690
    },
    {
      "entropy": 5.681548643112182,
      "epoch": 7.2292271407837445,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013133840878023435,
      "loss": 4.7953,
      "mean_token_accuracy": 0.23301055282354355,
      "num_tokens": 165381494.0,
      "step": 79695
    },
    {
      "entropy": 5.6765336990356445,
      "epoch": 7.229680696661829,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013131350795059037,
      "loss": 4.7822,
      "mean_token_accuracy": 0.2279401108622551,
      "num_tokens": 165391353.0,
      "step": 79700
    },
    {
      "entropy": 5.694221067428589,
      "epoch": 7.230134252539913,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013128861009241972,
      "loss": 4.7749,
      "mean_token_accuracy": 0.23811335861682892,
      "num_tokens": 165403560.0,
      "step": 79705
    },
    {
      "entropy": 5.6417571067810055,
      "epoch": 7.230587808417997,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001312637152062372,
      "loss": 4.7701,
      "mean_token_accuracy": 0.2384316697716713,
      "num_tokens": 165414919.0,
      "step": 79710
    },
    {
      "entropy": 5.697670984268188,
      "epoch": 7.231041364296082,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001312388232925577,
      "loss": 4.8458,
      "mean_token_accuracy": 0.23131748884916306,
      "num_tokens": 165425363.0,
      "step": 79715
    },
    {
      "entropy": 5.683913803100586,
      "epoch": 7.231494920174166,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013121393435189594,
      "loss": 4.7927,
      "mean_token_accuracy": 0.24003713428974152,
      "num_tokens": 165436333.0,
      "step": 79720
    },
    {
      "entropy": 5.639953756332398,
      "epoch": 7.231948476052249,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001311890483847666,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2515977382659912,
      "num_tokens": 165445914.0,
      "step": 79725
    },
    {
      "entropy": 5.686485052108765,
      "epoch": 7.2324020319303335,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013116416539168447,
      "loss": 4.7873,
      "mean_token_accuracy": 0.2351452425122261,
      "num_tokens": 165455542.0,
      "step": 79730
    },
    {
      "entropy": 5.656471538543701,
      "epoch": 7.232855587808418,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013113928537316404,
      "loss": 4.6921,
      "mean_token_accuracy": 0.2501168817281723,
      "num_tokens": 165465925.0,
      "step": 79735
    },
    {
      "entropy": 5.675976514816284,
      "epoch": 7.233309143686502,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013111440832971988,
      "loss": 4.8054,
      "mean_token_accuracy": 0.2391568437218666,
      "num_tokens": 165476121.0,
      "step": 79740
    },
    {
      "entropy": 5.6359556198120115,
      "epoch": 7.233762699564586,
      "grad_norm": 1.5,
      "learning_rate": 0.00013108953426186641,
      "loss": 4.7673,
      "mean_token_accuracy": 0.23819286227226258,
      "num_tokens": 165486717.0,
      "step": 79745
    },
    {
      "entropy": 5.687039852142334,
      "epoch": 7.2342162554426706,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000131064663170118,
      "loss": 4.8634,
      "mean_token_accuracy": 0.23974186331033706,
      "num_tokens": 165498408.0,
      "step": 79750
    },
    {
      "entropy": 5.655873775482178,
      "epoch": 7.234669811320755,
      "grad_norm": 1.640625,
      "learning_rate": 0.00013103979505498908,
      "loss": 4.8135,
      "mean_token_accuracy": 0.2391775742173195,
      "num_tokens": 165508840.0,
      "step": 79755
    },
    {
      "entropy": 5.670760202407837,
      "epoch": 7.235123367198839,
      "grad_norm": 1.5,
      "learning_rate": 0.00013101492991699382,
      "loss": 4.7785,
      "mean_token_accuracy": 0.24036636352539062,
      "num_tokens": 165518421.0,
      "step": 79760
    },
    {
      "entropy": 5.681370115280151,
      "epoch": 7.235576923076923,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001309900677566465,
      "loss": 4.7608,
      "mean_token_accuracy": 0.2340863361954689,
      "num_tokens": 165528836.0,
      "step": 79765
    },
    {
      "entropy": 5.690072631835937,
      "epoch": 7.236030478955008,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013096520857446136,
      "loss": 4.8174,
      "mean_token_accuracy": 0.2368500843644142,
      "num_tokens": 165539243.0,
      "step": 79770
    },
    {
      "entropy": 5.672938632965088,
      "epoch": 7.236484034833092,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013094035237095237,
      "loss": 4.7435,
      "mean_token_accuracy": 0.2389140710234642,
      "num_tokens": 165548927.0,
      "step": 79775
    },
    {
      "entropy": 5.692437934875488,
      "epoch": 7.236937590711175,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013091549914663364,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2354038178920746,
      "num_tokens": 165559135.0,
      "step": 79780
    },
    {
      "entropy": 5.649544668197632,
      "epoch": 7.2373911465892595,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013089064890201907,
      "loss": 4.7804,
      "mean_token_accuracy": 0.23286365568637848,
      "num_tokens": 165568443.0,
      "step": 79785
    },
    {
      "entropy": 5.648734092712402,
      "epoch": 7.237844702467344,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001308658016376226,
      "loss": 4.8013,
      "mean_token_accuracy": 0.23579919785261155,
      "num_tokens": 165578499.0,
      "step": 79790
    },
    {
      "entropy": 5.691379117965698,
      "epoch": 7.238298258345428,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013084095735395803,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23724807500839235,
      "num_tokens": 165590034.0,
      "step": 79795
    },
    {
      "entropy": 5.677400541305542,
      "epoch": 7.238751814223512,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013081611605153927,
      "loss": 4.7564,
      "mean_token_accuracy": 0.23463500142097474,
      "num_tokens": 165600248.0,
      "step": 79800
    },
    {
      "entropy": 5.797707891464233,
      "epoch": 7.239205370101597,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013079127773087996,
      "loss": 4.8763,
      "mean_token_accuracy": 0.23342398256063462,
      "num_tokens": 165609911.0,
      "step": 79805
    },
    {
      "entropy": 5.711139106750489,
      "epoch": 7.239658925979681,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001307664423924938,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2244228705763817,
      "num_tokens": 165620765.0,
      "step": 79810
    },
    {
      "entropy": 5.7349237442016605,
      "epoch": 7.240112481857765,
      "grad_norm": 1.5,
      "learning_rate": 0.00013074161003689432,
      "loss": 4.8994,
      "mean_token_accuracy": 0.23024460375308992,
      "num_tokens": 165630982.0,
      "step": 79815
    },
    {
      "entropy": 5.661091327667236,
      "epoch": 7.240566037735849,
      "grad_norm": 1.5,
      "learning_rate": 0.00013071678066459513,
      "loss": 4.7349,
      "mean_token_accuracy": 0.243708136677742,
      "num_tokens": 165641442.0,
      "step": 79820
    },
    {
      "entropy": 5.731976842880249,
      "epoch": 7.241019593613934,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00013069195427610968,
      "loss": 4.8792,
      "mean_token_accuracy": 0.23033960908651352,
      "num_tokens": 165651436.0,
      "step": 79825
    },
    {
      "entropy": 5.738643741607666,
      "epoch": 7.241473149492017,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013066713087195132,
      "loss": 4.8066,
      "mean_token_accuracy": 0.2340743735432625,
      "num_tokens": 165662494.0,
      "step": 79830
    },
    {
      "entropy": 5.759716320037842,
      "epoch": 7.241926705370101,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013064231045263356,
      "loss": 4.8181,
      "mean_token_accuracy": 0.2364825800061226,
      "num_tokens": 165671435.0,
      "step": 79835
    },
    {
      "entropy": 5.815727281570434,
      "epoch": 7.242380261248186,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001306174930186696,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2262369453907013,
      "num_tokens": 165682403.0,
      "step": 79840
    },
    {
      "entropy": 5.755576229095459,
      "epoch": 7.24283381712627,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013059267857057266,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23696912676095963,
      "num_tokens": 165692591.0,
      "step": 79845
    },
    {
      "entropy": 5.740585041046143,
      "epoch": 7.243287373004354,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013056786710885596,
      "loss": 4.941,
      "mean_token_accuracy": 0.22807716131210326,
      "num_tokens": 165703153.0,
      "step": 79850
    },
    {
      "entropy": 5.794582939147949,
      "epoch": 7.243740928882438,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013054305863403255,
      "loss": 4.8551,
      "mean_token_accuracy": 0.23556553423404694,
      "num_tokens": 165714697.0,
      "step": 79855
    },
    {
      "entropy": 5.746880960464478,
      "epoch": 7.244194484760523,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013051825314661542,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2288145124912262,
      "num_tokens": 165726351.0,
      "step": 79860
    },
    {
      "entropy": 5.672469902038574,
      "epoch": 7.244648040638607,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013049345064711774,
      "loss": 4.7617,
      "mean_token_accuracy": 0.24046378582715988,
      "num_tokens": 165737571.0,
      "step": 79865
    },
    {
      "entropy": 5.7325386047363285,
      "epoch": 7.245101596516691,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001304686511360523,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23852331787347794,
      "num_tokens": 165748849.0,
      "step": 79870
    },
    {
      "entropy": 5.747219133377075,
      "epoch": 7.245555152394775,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000130443854613932,
      "loss": 4.9036,
      "mean_token_accuracy": 0.22978721261024476,
      "num_tokens": 165759142.0,
      "step": 79875
    },
    {
      "entropy": 5.680864763259888,
      "epoch": 7.246008708272859,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001304190610812696,
      "loss": 4.726,
      "mean_token_accuracy": 0.2324469193816185,
      "num_tokens": 165769862.0,
      "step": 79880
    },
    {
      "entropy": 5.717576265335083,
      "epoch": 7.246462264150943,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013039427053857783,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2364407390356064,
      "num_tokens": 165779739.0,
      "step": 79885
    },
    {
      "entropy": 5.757901048660278,
      "epoch": 7.246915820029027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013036948298636946,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22873261123895644,
      "num_tokens": 165790006.0,
      "step": 79890
    },
    {
      "entropy": 5.733085823059082,
      "epoch": 7.247369375907112,
      "grad_norm": 1.578125,
      "learning_rate": 0.000130344698425157,
      "loss": 4.8466,
      "mean_token_accuracy": 0.23114182949066162,
      "num_tokens": 165799677.0,
      "step": 79895
    },
    {
      "entropy": 5.669975519180298,
      "epoch": 7.247822931785196,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013031991685545307,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23073423206806182,
      "num_tokens": 165809317.0,
      "step": 79900
    },
    {
      "entropy": 5.684273815155029,
      "epoch": 7.24827648766328,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013029513827777012,
      "loss": 4.8124,
      "mean_token_accuracy": 0.2289959743618965,
      "num_tokens": 165820506.0,
      "step": 79905
    },
    {
      "entropy": 5.7193482398986815,
      "epoch": 7.248730043541364,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013027036269262058,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23068089187145233,
      "num_tokens": 165831787.0,
      "step": 79910
    },
    {
      "entropy": 5.577082538604737,
      "epoch": 7.249183599419449,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001302455901005168,
      "loss": 4.6971,
      "mean_token_accuracy": 0.2460703581571579,
      "num_tokens": 165842216.0,
      "step": 79915
    },
    {
      "entropy": 5.675528621673584,
      "epoch": 7.249637155297533,
      "grad_norm": 1.5,
      "learning_rate": 0.00013022082050197115,
      "loss": 4.7902,
      "mean_token_accuracy": 0.23568252623081207,
      "num_tokens": 165852684.0,
      "step": 79920
    },
    {
      "entropy": 5.671484708786011,
      "epoch": 7.250090711175617,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013019605389749572,
      "loss": 4.8446,
      "mean_token_accuracy": 0.22859311252832412,
      "num_tokens": 165862795.0,
      "step": 79925
    },
    {
      "entropy": 5.669412136077881,
      "epoch": 7.2505442670537015,
      "grad_norm": 1.671875,
      "learning_rate": 0.00013017129028760287,
      "loss": 4.7514,
      "mean_token_accuracy": 0.2393440783023834,
      "num_tokens": 165872414.0,
      "step": 79930
    },
    {
      "entropy": 5.64158616065979,
      "epoch": 7.250997822931785,
      "grad_norm": 1.609375,
      "learning_rate": 0.00013014652967280465,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24554279893636705,
      "num_tokens": 165882909.0,
      "step": 79935
    },
    {
      "entropy": 5.669495296478272,
      "epoch": 7.251451378809869,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013012177205361312,
      "loss": 4.8126,
      "mean_token_accuracy": 0.2346011534333229,
      "num_tokens": 165893292.0,
      "step": 79940
    },
    {
      "entropy": 5.642372417449951,
      "epoch": 7.251904934687953,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013009701743054024,
      "loss": 4.7476,
      "mean_token_accuracy": 0.24023464620113372,
      "num_tokens": 165902897.0,
      "step": 79945
    },
    {
      "entropy": 5.730888557434082,
      "epoch": 7.252358490566038,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013007226580409798,
      "loss": 4.8184,
      "mean_token_accuracy": 0.23267535269260406,
      "num_tokens": 165912486.0,
      "step": 79950
    },
    {
      "entropy": 5.6510992527008055,
      "epoch": 7.252812046444122,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013004751717479818,
      "loss": 4.7872,
      "mean_token_accuracy": 0.2391634002327919,
      "num_tokens": 165923701.0,
      "step": 79955
    },
    {
      "entropy": 5.701243829727173,
      "epoch": 7.253265602322206,
      "grad_norm": 1.6875,
      "learning_rate": 0.00013002277154315262,
      "loss": 4.8064,
      "mean_token_accuracy": 0.24497734904289245,
      "num_tokens": 165933013.0,
      "step": 79960
    },
    {
      "entropy": 5.657495927810669,
      "epoch": 7.2537191582002905,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012999802890967317,
      "loss": 4.7299,
      "mean_token_accuracy": 0.2380698725581169,
      "num_tokens": 165942482.0,
      "step": 79965
    },
    {
      "entropy": 5.710625457763672,
      "epoch": 7.254172714078375,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001299732892748714,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2330002397298813,
      "num_tokens": 165953204.0,
      "step": 79970
    },
    {
      "entropy": 5.648618459701538,
      "epoch": 7.254626269956459,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001299485526392589,
      "loss": 4.7513,
      "mean_token_accuracy": 0.23236252814531327,
      "num_tokens": 165964830.0,
      "step": 79975
    },
    {
      "entropy": 5.699303102493286,
      "epoch": 7.255079825834542,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001299238190033474,
      "loss": 4.819,
      "mean_token_accuracy": 0.23512917757034302,
      "num_tokens": 165974892.0,
      "step": 79980
    },
    {
      "entropy": 5.66225299835205,
      "epoch": 7.255533381712627,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001298990883676483,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2350854754447937,
      "num_tokens": 165984754.0,
      "step": 79985
    },
    {
      "entropy": 5.704316663742065,
      "epoch": 7.255986937590711,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012987436073267288,
      "loss": 4.785,
      "mean_token_accuracy": 0.23651168942451478,
      "num_tokens": 165994404.0,
      "step": 79990
    },
    {
      "entropy": 5.8177143096923825,
      "epoch": 7.256440493468795,
      "grad_norm": 1.5,
      "learning_rate": 0.0001298496360989328,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2292335271835327,
      "num_tokens": 166006481.0,
      "step": 79995
    },
    {
      "entropy": 5.724107313156128,
      "epoch": 7.256894049346879,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001298249144669392,
      "loss": 4.835,
      "mean_token_accuracy": 0.23897299021482468,
      "num_tokens": 166018117.0,
      "step": 80000
    },
    {
      "entropy": 5.678056478500366,
      "epoch": 7.257347605224964,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012980019583720337,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23551883399486542,
      "num_tokens": 166028506.0,
      "step": 80005
    },
    {
      "entropy": 5.725093603134155,
      "epoch": 7.257801161103048,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012977548021023648,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22722053080797194,
      "num_tokens": 166039352.0,
      "step": 80010
    },
    {
      "entropy": 5.705002021789551,
      "epoch": 7.258254716981132,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001297507675865497,
      "loss": 4.843,
      "mean_token_accuracy": 0.23038201928138732,
      "num_tokens": 166049956.0,
      "step": 80015
    },
    {
      "entropy": 5.709950733184814,
      "epoch": 7.2587082728592165,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012972605796665398,
      "loss": 4.7924,
      "mean_token_accuracy": 0.2452584609389305,
      "num_tokens": 166059273.0,
      "step": 80020
    },
    {
      "entropy": 5.694135093688965,
      "epoch": 7.259161828737301,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012970135135106044,
      "loss": 4.797,
      "mean_token_accuracy": 0.23309387564659118,
      "num_tokens": 166069865.0,
      "step": 80025
    },
    {
      "entropy": 5.716419506072998,
      "epoch": 7.259615384615385,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00012967664774028,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22497591525316238,
      "num_tokens": 166080056.0,
      "step": 80030
    },
    {
      "entropy": 5.720234680175781,
      "epoch": 7.260068940493468,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001296519471348235,
      "loss": 4.8037,
      "mean_token_accuracy": 0.24053498059511186,
      "num_tokens": 166090674.0,
      "step": 80035
    },
    {
      "entropy": 5.747909212112427,
      "epoch": 7.260522496371553,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012962724953520177,
      "loss": 4.8452,
      "mean_token_accuracy": 0.23991975486278533,
      "num_tokens": 166101036.0,
      "step": 80040
    },
    {
      "entropy": 5.61997537612915,
      "epoch": 7.260976052249637,
      "grad_norm": 1.625,
      "learning_rate": 0.0001296025549419256,
      "loss": 4.6759,
      "mean_token_accuracy": 0.24376956522464752,
      "num_tokens": 166111229.0,
      "step": 80045
    },
    {
      "entropy": 5.624869298934937,
      "epoch": 7.261429608127721,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012957786335550565,
      "loss": 4.6794,
      "mean_token_accuracy": 0.24185765534639359,
      "num_tokens": 166120724.0,
      "step": 80050
    },
    {
      "entropy": 5.672431325912475,
      "epoch": 7.2618831640058055,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001295531747764525,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23430405259132386,
      "num_tokens": 166131018.0,
      "step": 80055
    },
    {
      "entropy": 5.719108295440674,
      "epoch": 7.26233671988389,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012952848920527672,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2337307944893837,
      "num_tokens": 166142317.0,
      "step": 80060
    },
    {
      "entropy": 5.735792970657348,
      "epoch": 7.262790275761974,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001295038066424889,
      "loss": 4.8503,
      "mean_token_accuracy": 0.23123436272144318,
      "num_tokens": 166153026.0,
      "step": 80065
    },
    {
      "entropy": 5.661243295669555,
      "epoch": 7.263243831640058,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001294791270885995,
      "loss": 4.7547,
      "mean_token_accuracy": 0.24068529307842254,
      "num_tokens": 166162647.0,
      "step": 80070
    },
    {
      "entropy": 5.753008651733398,
      "epoch": 7.2636973875181425,
      "grad_norm": 1.640625,
      "learning_rate": 0.00012945445054411876,
      "loss": 4.7986,
      "mean_token_accuracy": 0.23661804646253587,
      "num_tokens": 166173746.0,
      "step": 80075
    },
    {
      "entropy": 5.788565969467163,
      "epoch": 7.264150943396227,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012942977700955712,
      "loss": 4.9429,
      "mean_token_accuracy": 0.22472820729017257,
      "num_tokens": 166185581.0,
      "step": 80080
    },
    {
      "entropy": 5.678386163711548,
      "epoch": 7.264604499274311,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012940510648542475,
      "loss": 4.8546,
      "mean_token_accuracy": 0.23333043605089188,
      "num_tokens": 166197067.0,
      "step": 80085
    },
    {
      "entropy": 5.780991744995117,
      "epoch": 7.2650580551523944,
      "grad_norm": 1.5,
      "learning_rate": 0.0001293804389722319,
      "loss": 4.886,
      "mean_token_accuracy": 0.22626991271972657,
      "num_tokens": 166207795.0,
      "step": 80090
    },
    {
      "entropy": 5.676224422454834,
      "epoch": 7.265511611030479,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012935577447048858,
      "loss": 4.7692,
      "mean_token_accuracy": 0.24006591141223907,
      "num_tokens": 166217992.0,
      "step": 80095
    },
    {
      "entropy": 5.658362627029419,
      "epoch": 7.265965166908563,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012933111298070509,
      "loss": 4.7784,
      "mean_token_accuracy": 0.244529427587986,
      "num_tokens": 166228698.0,
      "step": 80100
    },
    {
      "entropy": 5.66176404953003,
      "epoch": 7.266418722786647,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012930645450339123,
      "loss": 4.7448,
      "mean_token_accuracy": 0.23563114553689957,
      "num_tokens": 166239421.0,
      "step": 80105
    },
    {
      "entropy": 5.744569540023804,
      "epoch": 7.2668722786647315,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00012928179903905706,
      "loss": 4.8477,
      "mean_token_accuracy": 0.23000955879688262,
      "num_tokens": 166249432.0,
      "step": 80110
    },
    {
      "entropy": 5.659039402008057,
      "epoch": 7.267325834542816,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012925714658821243,
      "loss": 4.734,
      "mean_token_accuracy": 0.24021471440792083,
      "num_tokens": 166259909.0,
      "step": 80115
    },
    {
      "entropy": 5.6980297565460205,
      "epoch": 7.2677793904209,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012923249715136715,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22952256053686143,
      "num_tokens": 166270709.0,
      "step": 80120
    },
    {
      "entropy": 5.659682703018189,
      "epoch": 7.268232946298984,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012920785072903097,
      "loss": 4.6794,
      "mean_token_accuracy": 0.2456708863377571,
      "num_tokens": 166279894.0,
      "step": 80125
    },
    {
      "entropy": 5.676596212387085,
      "epoch": 7.268686502177069,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012918320732171348,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2393201619386673,
      "num_tokens": 166290165.0,
      "step": 80130
    },
    {
      "entropy": 5.685357618331909,
      "epoch": 7.269140058055152,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012915856692992454,
      "loss": 4.832,
      "mean_token_accuracy": 0.23545004576444625,
      "num_tokens": 166300534.0,
      "step": 80135
    },
    {
      "entropy": 5.7265397071838375,
      "epoch": 7.269593613933236,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012913392955417357,
      "loss": 4.8605,
      "mean_token_accuracy": 0.23914008885622023,
      "num_tokens": 166311119.0,
      "step": 80140
    },
    {
      "entropy": 5.663519859313965,
      "epoch": 7.2700471698113205,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001291092951949701,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2568905770778656,
      "num_tokens": 166321281.0,
      "step": 80145
    },
    {
      "entropy": 5.664143705368042,
      "epoch": 7.270500725689405,
      "grad_norm": 1.625,
      "learning_rate": 0.00012908466385282364,
      "loss": 4.8326,
      "mean_token_accuracy": 0.23000639528036118,
      "num_tokens": 166331002.0,
      "step": 80150
    },
    {
      "entropy": 5.720530319213867,
      "epoch": 7.270954281567489,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012906003552824347,
      "loss": 4.8232,
      "mean_token_accuracy": 0.23333876878023146,
      "num_tokens": 166341183.0,
      "step": 80155
    },
    {
      "entropy": 5.774585390090943,
      "epoch": 7.271407837445573,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012903541022173898,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21990306079387664,
      "num_tokens": 166351903.0,
      "step": 80160
    },
    {
      "entropy": 5.776636934280395,
      "epoch": 7.2718613933236576,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001290107879338193,
      "loss": 4.8835,
      "mean_token_accuracy": 0.23364565670490264,
      "num_tokens": 166363608.0,
      "step": 80165
    },
    {
      "entropy": 5.707914543151856,
      "epoch": 7.272314949201742,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001289861686649938,
      "loss": 4.8395,
      "mean_token_accuracy": 0.23762672990560532,
      "num_tokens": 166374739.0,
      "step": 80170
    },
    {
      "entropy": 5.592466354370117,
      "epoch": 7.272768505079826,
      "grad_norm": 1.5,
      "learning_rate": 0.00012896155241577157,
      "loss": 4.6127,
      "mean_token_accuracy": 0.25138787031173704,
      "num_tokens": 166384681.0,
      "step": 80175
    },
    {
      "entropy": 5.654475975036621,
      "epoch": 7.27322206095791,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012893693918666164,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23975809365510942,
      "num_tokens": 166395297.0,
      "step": 80180
    },
    {
      "entropy": 5.74119930267334,
      "epoch": 7.273675616835995,
      "grad_norm": 1.53125,
      "learning_rate": 0.000128912328978173,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2318857192993164,
      "num_tokens": 166406710.0,
      "step": 80185
    },
    {
      "entropy": 5.700164699554444,
      "epoch": 7.274129172714078,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012888772179081465,
      "loss": 4.8253,
      "mean_token_accuracy": 0.2267909675836563,
      "num_tokens": 166416693.0,
      "step": 80190
    },
    {
      "entropy": 5.698992967605591,
      "epoch": 7.274582728592162,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001288631176250954,
      "loss": 4.8667,
      "mean_token_accuracy": 0.23173139840364457,
      "num_tokens": 166427462.0,
      "step": 80195
    },
    {
      "entropy": 5.774973392486572,
      "epoch": 7.2750362844702465,
      "grad_norm": 1.640625,
      "learning_rate": 0.00012883851648152418,
      "loss": 4.8912,
      "mean_token_accuracy": 0.23123562037944795,
      "num_tokens": 166437446.0,
      "step": 80200
    },
    {
      "entropy": 5.68992166519165,
      "epoch": 7.275489840348331,
      "grad_norm": 1.6875,
      "learning_rate": 0.00012881391836060965,
      "loss": 4.737,
      "mean_token_accuracy": 0.24213000535964965,
      "num_tokens": 166447352.0,
      "step": 80205
    },
    {
      "entropy": 5.72064733505249,
      "epoch": 7.275943396226415,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012878932326286052,
      "loss": 4.8411,
      "mean_token_accuracy": 0.23735448271036147,
      "num_tokens": 166458664.0,
      "step": 80210
    },
    {
      "entropy": 5.781466054916382,
      "epoch": 7.276396952104499,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001287647311887855,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22056786268949508,
      "num_tokens": 166469467.0,
      "step": 80215
    },
    {
      "entropy": 5.782115507125854,
      "epoch": 7.276850507982584,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001287401421388931,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23609832525253296,
      "num_tokens": 166480049.0,
      "step": 80220
    },
    {
      "entropy": 5.759831428527832,
      "epoch": 7.277304063860668,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012871555611369183,
      "loss": 4.9021,
      "mean_token_accuracy": 0.23169729262590408,
      "num_tokens": 166489996.0,
      "step": 80225
    },
    {
      "entropy": 5.666635274887085,
      "epoch": 7.277757619738752,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012869097311369004,
      "loss": 4.7436,
      "mean_token_accuracy": 0.24086488783359528,
      "num_tokens": 166501386.0,
      "step": 80230
    },
    {
      "entropy": 5.761150169372558,
      "epoch": 7.278211175616836,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012866639313939633,
      "loss": 4.8757,
      "mean_token_accuracy": 0.23086302280426024,
      "num_tokens": 166511060.0,
      "step": 80235
    },
    {
      "entropy": 5.661197900772095,
      "epoch": 7.27866473149492,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012864181619131885,
      "loss": 4.7839,
      "mean_token_accuracy": 0.24444671720266342,
      "num_tokens": 166520937.0,
      "step": 80240
    },
    {
      "entropy": 5.717969179153442,
      "epoch": 7.279118287373004,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001286172422699659,
      "loss": 4.8885,
      "mean_token_accuracy": 0.22795118391513824,
      "num_tokens": 166531826.0,
      "step": 80245
    },
    {
      "entropy": 5.70911169052124,
      "epoch": 7.279571843251088,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00012859267137584575,
      "loss": 4.8831,
      "mean_token_accuracy": 0.22220389544963837,
      "num_tokens": 166541844.0,
      "step": 80250
    },
    {
      "entropy": 5.692424345016479,
      "epoch": 7.280025399129173,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001285681035094664,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23284814208745958,
      "num_tokens": 166553037.0,
      "step": 80255
    },
    {
      "entropy": 5.728095388412475,
      "epoch": 7.280478955007257,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000128543538671336,
      "loss": 4.8358,
      "mean_token_accuracy": 0.24322753250598908,
      "num_tokens": 166563186.0,
      "step": 80260
    },
    {
      "entropy": 5.690383625030518,
      "epoch": 7.280932510885341,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001285189768619625,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23900243788957595,
      "num_tokens": 166573498.0,
      "step": 80265
    },
    {
      "entropy": 5.731087732315063,
      "epoch": 7.281386066763425,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001284944180818539,
      "loss": 4.8514,
      "mean_token_accuracy": 0.23054523915052413,
      "num_tokens": 166584343.0,
      "step": 80270
    },
    {
      "entropy": 5.699429702758789,
      "epoch": 7.28183962264151,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012846986233151813,
      "loss": 4.794,
      "mean_token_accuracy": 0.23658570498228074,
      "num_tokens": 166595122.0,
      "step": 80275
    },
    {
      "entropy": 5.655314207077026,
      "epoch": 7.282293178519594,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001284453096114629,
      "loss": 4.7718,
      "mean_token_accuracy": 0.2374061167240143,
      "num_tokens": 166605750.0,
      "step": 80280
    },
    {
      "entropy": 5.6863490581512455,
      "epoch": 7.282746734397678,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012842075992219605,
      "loss": 4.8004,
      "mean_token_accuracy": 0.23131698071956636,
      "num_tokens": 166616577.0,
      "step": 80285
    },
    {
      "entropy": 5.727476310729981,
      "epoch": 7.2832002902757615,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012839621326422523,
      "loss": 4.855,
      "mean_token_accuracy": 0.23185361325740814,
      "num_tokens": 166627207.0,
      "step": 80290
    },
    {
      "entropy": 5.814786148071289,
      "epoch": 7.283653846153846,
      "grad_norm": 1.5,
      "learning_rate": 0.00012837166963805807,
      "loss": 4.9434,
      "mean_token_accuracy": 0.22117656469345093,
      "num_tokens": 166637750.0,
      "step": 80295
    },
    {
      "entropy": 5.645334196090698,
      "epoch": 7.28410740203193,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012834712904420206,
      "loss": 4.6987,
      "mean_token_accuracy": 0.2482168197631836,
      "num_tokens": 166648314.0,
      "step": 80300
    },
    {
      "entropy": 5.729985094070434,
      "epoch": 7.284560957910014,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012832259148316488,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22435117810964583,
      "num_tokens": 166659653.0,
      "step": 80305
    },
    {
      "entropy": 5.711084175109863,
      "epoch": 7.285014513788099,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001282980569554539,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2376423165202141,
      "num_tokens": 166670105.0,
      "step": 80310
    },
    {
      "entropy": 5.742158460617065,
      "epoch": 7.285468069666183,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012827352546157647,
      "loss": 4.7792,
      "mean_token_accuracy": 0.24422147423028945,
      "num_tokens": 166679834.0,
      "step": 80315
    },
    {
      "entropy": 5.664431953430176,
      "epoch": 7.285921625544267,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012824899700203995,
      "loss": 4.7468,
      "mean_token_accuracy": 0.23857192546129227,
      "num_tokens": 166690644.0,
      "step": 80320
    },
    {
      "entropy": 5.667792654037475,
      "epoch": 7.286375181422351,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012822447157735152,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2366160273551941,
      "num_tokens": 166701490.0,
      "step": 80325
    },
    {
      "entropy": 5.639398765563965,
      "epoch": 7.286828737300436,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012819994918801845,
      "loss": 4.7335,
      "mean_token_accuracy": 0.23798252642154694,
      "num_tokens": 166711255.0,
      "step": 80330
    },
    {
      "entropy": 5.814866638183593,
      "epoch": 7.28728229317852,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012817542983454782,
      "loss": 4.9014,
      "mean_token_accuracy": 0.22482938915491105,
      "num_tokens": 166720479.0,
      "step": 80335
    },
    {
      "entropy": 5.728391456604004,
      "epoch": 7.287735849056604,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012815091351744677,
      "loss": 4.8206,
      "mean_token_accuracy": 0.23592633455991746,
      "num_tokens": 166730659.0,
      "step": 80340
    },
    {
      "entropy": 5.675257062911987,
      "epoch": 7.288189404934688,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012812640023722222,
      "loss": 4.6906,
      "mean_token_accuracy": 0.2511789470911026,
      "num_tokens": 166741384.0,
      "step": 80345
    },
    {
      "entropy": 5.707175779342651,
      "epoch": 7.288642960812772,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001281018899943811,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2353138655424118,
      "num_tokens": 166752973.0,
      "step": 80350
    },
    {
      "entropy": 5.75395450592041,
      "epoch": 7.289096516690856,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00012807738278943038,
      "loss": 4.8782,
      "mean_token_accuracy": 0.23409016579389572,
      "num_tokens": 166763345.0,
      "step": 80355
    },
    {
      "entropy": 5.701423692703247,
      "epoch": 7.28955007256894,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012805287862287674,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23384274393320084,
      "num_tokens": 166774386.0,
      "step": 80360
    },
    {
      "entropy": 5.680426502227784,
      "epoch": 7.290003628447025,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012802837749522705,
      "loss": 4.8538,
      "mean_token_accuracy": 0.23415073603391648,
      "num_tokens": 166785742.0,
      "step": 80365
    },
    {
      "entropy": 5.698110198974609,
      "epoch": 7.290457184325109,
      "grad_norm": 1.65625,
      "learning_rate": 0.00012800387940698805,
      "loss": 4.8107,
      "mean_token_accuracy": 0.22896721661090852,
      "num_tokens": 166794998.0,
      "step": 80370
    },
    {
      "entropy": 5.6990386009216305,
      "epoch": 7.290910740203193,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012797938435866622,
      "loss": 4.8306,
      "mean_token_accuracy": 0.23999789655208587,
      "num_tokens": 166805236.0,
      "step": 80375
    },
    {
      "entropy": 5.724122095108032,
      "epoch": 7.2913642960812775,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012795489235076818,
      "loss": 4.821,
      "mean_token_accuracy": 0.23497701138257981,
      "num_tokens": 166815172.0,
      "step": 80380
    },
    {
      "entropy": 5.656291437149048,
      "epoch": 7.291817851959362,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012793040338380046,
      "loss": 4.7782,
      "mean_token_accuracy": 0.23463223427534102,
      "num_tokens": 166824957.0,
      "step": 80385
    },
    {
      "entropy": 5.689010286331177,
      "epoch": 7.292271407837446,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001279059174582695,
      "loss": 4.7983,
      "mean_token_accuracy": 0.23771231472492219,
      "num_tokens": 166835769.0,
      "step": 80390
    },
    {
      "entropy": 5.710466098785401,
      "epoch": 7.292724963715529,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012788143457468154,
      "loss": 4.776,
      "mean_token_accuracy": 0.23855147808790206,
      "num_tokens": 166845517.0,
      "step": 80395
    },
    {
      "entropy": 5.698587560653687,
      "epoch": 7.293178519593614,
      "grad_norm": 1.5,
      "learning_rate": 0.0001278569547335431,
      "loss": 4.7614,
      "mean_token_accuracy": 0.243955098092556,
      "num_tokens": 166855276.0,
      "step": 80400
    },
    {
      "entropy": 5.653434181213379,
      "epoch": 7.293632075471698,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001278324779353603,
      "loss": 4.7038,
      "mean_token_accuracy": 0.2516649395227432,
      "num_tokens": 166865852.0,
      "step": 80405
    },
    {
      "entropy": 5.721776103973388,
      "epoch": 7.294085631349782,
      "grad_norm": 1.640625,
      "learning_rate": 0.00012780800418063938,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23028530180454254,
      "num_tokens": 166875484.0,
      "step": 80410
    },
    {
      "entropy": 5.721731376647949,
      "epoch": 7.294539187227866,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012778353346988648,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23703057914972306,
      "num_tokens": 166886963.0,
      "step": 80415
    },
    {
      "entropy": 5.624110317230224,
      "epoch": 7.294992743105951,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012775906580360757,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23750440329313277,
      "num_tokens": 166897998.0,
      "step": 80420
    },
    {
      "entropy": 5.697962856292724,
      "epoch": 7.295446298984035,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001277346011823087,
      "loss": 4.7786,
      "mean_token_accuracy": 0.23649438917636872,
      "num_tokens": 166908129.0,
      "step": 80425
    },
    {
      "entropy": 5.691613388061524,
      "epoch": 7.295899854862119,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012771013960649575,
      "loss": 4.7914,
      "mean_token_accuracy": 0.23383491337299347,
      "num_tokens": 166918048.0,
      "step": 80430
    },
    {
      "entropy": 5.610816574096679,
      "epoch": 7.2963534107402035,
      "grad_norm": 1.375,
      "learning_rate": 0.0001276856810766747,
      "loss": 4.7545,
      "mean_token_accuracy": 0.23359512090682982,
      "num_tokens": 166929274.0,
      "step": 80435
    },
    {
      "entropy": 5.672600412368775,
      "epoch": 7.296806966618288,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012766122559335134,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23339079767465593,
      "num_tokens": 166939819.0,
      "step": 80440
    },
    {
      "entropy": 5.760348463058472,
      "epoch": 7.297260522496371,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012763677315703138,
      "loss": 4.8766,
      "mean_token_accuracy": 0.23532565534114838,
      "num_tokens": 166950418.0,
      "step": 80445
    },
    {
      "entropy": 5.628703355789185,
      "epoch": 7.297714078374455,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012761232376822047,
      "loss": 4.779,
      "mean_token_accuracy": 0.23357848078012466,
      "num_tokens": 166961674.0,
      "step": 80450
    },
    {
      "entropy": 5.6827788829803465,
      "epoch": 7.29816763425254,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012758787742742428,
      "loss": 4.8085,
      "mean_token_accuracy": 0.23649671524763108,
      "num_tokens": 166972177.0,
      "step": 80455
    },
    {
      "entropy": 5.733351325988769,
      "epoch": 7.298621190130624,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012756343413514837,
      "loss": 4.8227,
      "mean_token_accuracy": 0.2362205535173416,
      "num_tokens": 166983212.0,
      "step": 80460
    },
    {
      "entropy": 5.731758785247803,
      "epoch": 7.299074746008708,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001275389938918981,
      "loss": 4.8408,
      "mean_token_accuracy": 0.22960706204175949,
      "num_tokens": 166993553.0,
      "step": 80465
    },
    {
      "entropy": 5.70148458480835,
      "epoch": 7.2995283018867925,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012751455669817913,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2279196262359619,
      "num_tokens": 167004012.0,
      "step": 80470
    },
    {
      "entropy": 5.762228393554688,
      "epoch": 7.299981857764877,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012749012255449665,
      "loss": 4.785,
      "mean_token_accuracy": 0.2440651163458824,
      "num_tokens": 167014101.0,
      "step": 80475
    },
    {
      "entropy": 5.755225658416748,
      "epoch": 7.300435413642961,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001274656914613561,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2290855437517166,
      "num_tokens": 167024651.0,
      "step": 80480
    },
    {
      "entropy": 5.687958812713623,
      "epoch": 7.300888969521045,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001274412634192626,
      "loss": 4.791,
      "mean_token_accuracy": 0.24085622578859328,
      "num_tokens": 167034988.0,
      "step": 80485
    },
    {
      "entropy": 5.687486028671264,
      "epoch": 7.3013425253991295,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012741683842872143,
      "loss": 4.7554,
      "mean_token_accuracy": 0.24264180809259414,
      "num_tokens": 167045413.0,
      "step": 80490
    },
    {
      "entropy": 5.680859994888306,
      "epoch": 7.301796081277214,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001273924164902375,
      "loss": 4.7947,
      "mean_token_accuracy": 0.24310633391141892,
      "num_tokens": 167055848.0,
      "step": 80495
    },
    {
      "entropy": 5.740254688262939,
      "epoch": 7.302249637155297,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001273679976043161,
      "loss": 4.8012,
      "mean_token_accuracy": 0.2433271512389183,
      "num_tokens": 167065803.0,
      "step": 80500
    },
    {
      "entropy": 5.705816030502319,
      "epoch": 7.3027031930333814,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012734358177146214,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23151660114526748,
      "num_tokens": 167075471.0,
      "step": 80505
    },
    {
      "entropy": 5.753402185440064,
      "epoch": 7.303156748911466,
      "grad_norm": 1.640625,
      "learning_rate": 0.00012731916899218058,
      "loss": 4.8592,
      "mean_token_accuracy": 0.23788875341415405,
      "num_tokens": 167085928.0,
      "step": 80510
    },
    {
      "entropy": 5.706054782867431,
      "epoch": 7.30361030478955,
      "grad_norm": 1.5,
      "learning_rate": 0.00012729475926697621,
      "loss": 4.7863,
      "mean_token_accuracy": 0.24524635523557664,
      "num_tokens": 167095915.0,
      "step": 80515
    },
    {
      "entropy": 5.714032173156738,
      "epoch": 7.304063860667634,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012727035259635383,
      "loss": 4.8227,
      "mean_token_accuracy": 0.2349287673830986,
      "num_tokens": 167106194.0,
      "step": 80520
    },
    {
      "entropy": 5.7810282707214355,
      "epoch": 7.3045174165457185,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012724594898081822,
      "loss": 4.9302,
      "mean_token_accuracy": 0.22869990766048431,
      "num_tokens": 167116477.0,
      "step": 80525
    },
    {
      "entropy": 5.625376987457275,
      "epoch": 7.304970972423803,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000127221548420874,
      "loss": 4.6483,
      "mean_token_accuracy": 0.2478260099887848,
      "num_tokens": 167125762.0,
      "step": 80530
    },
    {
      "entropy": 5.713515663146973,
      "epoch": 7.305424528301887,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012719715091702582,
      "loss": 4.8469,
      "mean_token_accuracy": 0.24061265885829924,
      "num_tokens": 167136601.0,
      "step": 80535
    },
    {
      "entropy": 5.670541000366211,
      "epoch": 7.305878084179971,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001271727564697783,
      "loss": 4.8307,
      "mean_token_accuracy": 0.23381952792406083,
      "num_tokens": 167147838.0,
      "step": 80540
    },
    {
      "entropy": 5.70629358291626,
      "epoch": 7.306331640058055,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012714836507963584,
      "loss": 4.8446,
      "mean_token_accuracy": 0.23433381468057632,
      "num_tokens": 167159278.0,
      "step": 80545
    },
    {
      "entropy": 5.734933519363404,
      "epoch": 7.306785195936139,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001271239767471028,
      "loss": 4.8501,
      "mean_token_accuracy": 0.23524681776762008,
      "num_tokens": 167169021.0,
      "step": 80550
    },
    {
      "entropy": 5.751374006271362,
      "epoch": 7.307238751814223,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012709959147268367,
      "loss": 4.8435,
      "mean_token_accuracy": 0.23254607766866683,
      "num_tokens": 167179167.0,
      "step": 80555
    },
    {
      "entropy": 5.724576330184936,
      "epoch": 7.3076923076923075,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012707520925688264,
      "loss": 4.8251,
      "mean_token_accuracy": 0.23715772479772568,
      "num_tokens": 167189316.0,
      "step": 80560
    },
    {
      "entropy": 5.722168731689453,
      "epoch": 7.308145863570392,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001270508301002039,
      "loss": 4.8262,
      "mean_token_accuracy": 0.2361164778470993,
      "num_tokens": 167199521.0,
      "step": 80565
    },
    {
      "entropy": 5.763686800003052,
      "epoch": 7.308599419448476,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001270264540031518,
      "loss": 4.8614,
      "mean_token_accuracy": 0.23351092338562013,
      "num_tokens": 167209635.0,
      "step": 80570
    },
    {
      "entropy": 5.706649971008301,
      "epoch": 7.30905297532656,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012700208096623033,
      "loss": 4.8381,
      "mean_token_accuracy": 0.23450848162174226,
      "num_tokens": 167219835.0,
      "step": 80575
    },
    {
      "entropy": 5.726797866821289,
      "epoch": 7.3095065312046446,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012697771098994353,
      "loss": 4.837,
      "mean_token_accuracy": 0.23234324753284455,
      "num_tokens": 167230692.0,
      "step": 80580
    },
    {
      "entropy": 5.74688835144043,
      "epoch": 7.309960087082729,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012695334407479542,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2258839264512062,
      "num_tokens": 167241211.0,
      "step": 80585
    },
    {
      "entropy": 5.651161575317383,
      "epoch": 7.310413642960813,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012692898022128983,
      "loss": 4.7768,
      "mean_token_accuracy": 0.235834801197052,
      "num_tokens": 167252635.0,
      "step": 80590
    },
    {
      "entropy": 5.7019944190979,
      "epoch": 7.310867198838897,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001269046194299307,
      "loss": 4.7358,
      "mean_token_accuracy": 0.24123466163873672,
      "num_tokens": 167263262.0,
      "step": 80595
    },
    {
      "entropy": 5.703500604629516,
      "epoch": 7.311320754716981,
      "grad_norm": 1.625,
      "learning_rate": 0.00012688026170122168,
      "loss": 4.846,
      "mean_token_accuracy": 0.236411514878273,
      "num_tokens": 167273306.0,
      "step": 80600
    },
    {
      "entropy": 5.736403083801269,
      "epoch": 7.311774310595065,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012685590703566667,
      "loss": 4.8587,
      "mean_token_accuracy": 0.23552679866552353,
      "num_tokens": 167283057.0,
      "step": 80605
    },
    {
      "entropy": 5.622232913970947,
      "epoch": 7.312227866473149,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012683155543376928,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23952912241220475,
      "num_tokens": 167293728.0,
      "step": 80610
    },
    {
      "entropy": 5.713553524017334,
      "epoch": 7.3126814223512335,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012680720689603304,
      "loss": 4.8579,
      "mean_token_accuracy": 0.23622775226831436,
      "num_tokens": 167303122.0,
      "step": 80615
    },
    {
      "entropy": 5.747843885421753,
      "epoch": 7.313134978229318,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001267828614229616,
      "loss": 4.8431,
      "mean_token_accuracy": 0.2351401463150978,
      "num_tokens": 167313346.0,
      "step": 80620
    },
    {
      "entropy": 5.735754156112671,
      "epoch": 7.313588534107402,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001267585190150583,
      "loss": 4.851,
      "mean_token_accuracy": 0.23343707621097565,
      "num_tokens": 167322881.0,
      "step": 80625
    },
    {
      "entropy": 5.640772151947021,
      "epoch": 7.314042089985486,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012673417967282657,
      "loss": 4.7595,
      "mean_token_accuracy": 0.2328380063176155,
      "num_tokens": 167334154.0,
      "step": 80630
    },
    {
      "entropy": 5.727889776229858,
      "epoch": 7.314495645863571,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012670984339676976,
      "loss": 4.8098,
      "mean_token_accuracy": 0.22851155400276185,
      "num_tokens": 167343789.0,
      "step": 80635
    },
    {
      "entropy": 5.657192802429199,
      "epoch": 7.314949201741655,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012668551018739125,
      "loss": 4.7538,
      "mean_token_accuracy": 0.23808301687240602,
      "num_tokens": 167355432.0,
      "step": 80640
    },
    {
      "entropy": 5.646287965774536,
      "epoch": 7.315402757619739,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012666118004519417,
      "loss": 4.7418,
      "mean_token_accuracy": 0.234747277200222,
      "num_tokens": 167365324.0,
      "step": 80645
    },
    {
      "entropy": 5.7184398651123045,
      "epoch": 7.315856313497823,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001266368529706817,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22745178192853927,
      "num_tokens": 167375169.0,
      "step": 80650
    },
    {
      "entropy": 5.679341173171997,
      "epoch": 7.316309869375907,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001266125289643569,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23233786970376968,
      "num_tokens": 167385433.0,
      "step": 80655
    },
    {
      "entropy": 5.78124098777771,
      "epoch": 7.316763425253991,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012658820802672282,
      "loss": 4.8943,
      "mean_token_accuracy": 0.23465390056371688,
      "num_tokens": 167395003.0,
      "step": 80660
    },
    {
      "entropy": 5.695157289505005,
      "epoch": 7.317216981132075,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012656389015828245,
      "loss": 4.8549,
      "mean_token_accuracy": 0.23379410803318024,
      "num_tokens": 167405581.0,
      "step": 80665
    },
    {
      "entropy": 5.681587982177734,
      "epoch": 7.31767053701016,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012653957535953853,
      "loss": 4.7438,
      "mean_token_accuracy": 0.24084278643131257,
      "num_tokens": 167416572.0,
      "step": 80670
    },
    {
      "entropy": 5.67180700302124,
      "epoch": 7.318124092888244,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012651526363099416,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2374172493815422,
      "num_tokens": 167426647.0,
      "step": 80675
    },
    {
      "entropy": 5.6603090286254885,
      "epoch": 7.318577648766328,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012649095497315195,
      "loss": 4.7242,
      "mean_token_accuracy": 0.23760260492563248,
      "num_tokens": 167437457.0,
      "step": 80680
    },
    {
      "entropy": 5.704242277145386,
      "epoch": 7.319031204644412,
      "grad_norm": 1.6875,
      "learning_rate": 0.0001264666493865146,
      "loss": 4.8065,
      "mean_token_accuracy": 0.24488022029399872,
      "num_tokens": 167446925.0,
      "step": 80685
    },
    {
      "entropy": 5.6984764575958256,
      "epoch": 7.319484760522497,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012644234687158483,
      "loss": 4.8105,
      "mean_token_accuracy": 0.24085394144058228,
      "num_tokens": 167456117.0,
      "step": 80690
    },
    {
      "entropy": 5.661560201644898,
      "epoch": 7.319938316400581,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012641804742886516,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23468753695487976,
      "num_tokens": 167466496.0,
      "step": 80695
    },
    {
      "entropy": 5.654313039779663,
      "epoch": 7.320391872278664,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001263937510588581,
      "loss": 4.7424,
      "mean_token_accuracy": 0.23552828878164292,
      "num_tokens": 167476197.0,
      "step": 80700
    },
    {
      "entropy": 5.661574411392212,
      "epoch": 7.3208454281567485,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012636945776206616,
      "loss": 4.7638,
      "mean_token_accuracy": 0.23782849311828613,
      "num_tokens": 167486139.0,
      "step": 80705
    },
    {
      "entropy": 5.683565378189087,
      "epoch": 7.321298984034833,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012634516753899172,
      "loss": 4.7824,
      "mean_token_accuracy": 0.236125910282135,
      "num_tokens": 167495940.0,
      "step": 80710
    },
    {
      "entropy": 5.681276321411133,
      "epoch": 7.321752539912917,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012632088039013715,
      "loss": 4.7589,
      "mean_token_accuracy": 0.24145641177892685,
      "num_tokens": 167506321.0,
      "step": 80715
    },
    {
      "entropy": 5.623831272125244,
      "epoch": 7.322206095791001,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001262965963160046,
      "loss": 4.7269,
      "mean_token_accuracy": 0.24064605832099914,
      "num_tokens": 167515936.0,
      "step": 80720
    },
    {
      "entropy": 5.653921175003052,
      "epoch": 7.322659651669086,
      "grad_norm": 1.625,
      "learning_rate": 0.00012627231531709638,
      "loss": 4.7649,
      "mean_token_accuracy": 0.23956673741340637,
      "num_tokens": 167525146.0,
      "step": 80725
    },
    {
      "entropy": 5.6793529987335205,
      "epoch": 7.32311320754717,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012624803739391455,
      "loss": 4.8058,
      "mean_token_accuracy": 0.23222561180591583,
      "num_tokens": 167534770.0,
      "step": 80730
    },
    {
      "entropy": 5.680341577529907,
      "epoch": 7.323566763425254,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012622376254696114,
      "loss": 4.8824,
      "mean_token_accuracy": 0.23052074909210205,
      "num_tokens": 167545810.0,
      "step": 80735
    },
    {
      "entropy": 5.766911602020263,
      "epoch": 7.324020319303338,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001261994907767383,
      "loss": 4.9084,
      "mean_token_accuracy": 0.22543840557336808,
      "num_tokens": 167556704.0,
      "step": 80740
    },
    {
      "entropy": 5.69844388961792,
      "epoch": 7.324473875181423,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012617522208374792,
      "loss": 4.8317,
      "mean_token_accuracy": 0.22984735369682313,
      "num_tokens": 167567071.0,
      "step": 80745
    },
    {
      "entropy": 5.695381784439087,
      "epoch": 7.324927431059507,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012615095646849195,
      "loss": 4.8273,
      "mean_token_accuracy": 0.23178349137306214,
      "num_tokens": 167577904.0,
      "step": 80750
    },
    {
      "entropy": 5.722815322875976,
      "epoch": 7.32538098693759,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012612669393147205,
      "loss": 4.6973,
      "mean_token_accuracy": 0.24778030216693878,
      "num_tokens": 167586925.0,
      "step": 80755
    },
    {
      "entropy": 5.644959545135498,
      "epoch": 7.325834542815675,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001261024344731901,
      "loss": 4.7231,
      "mean_token_accuracy": 0.24167320132255554,
      "num_tokens": 167597167.0,
      "step": 80760
    },
    {
      "entropy": 5.686674356460571,
      "epoch": 7.326288098693759,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012607817809414762,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22857246696949005,
      "num_tokens": 167608217.0,
      "step": 80765
    },
    {
      "entropy": 5.596011972427368,
      "epoch": 7.326741654571843,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012605392479484646,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24157433807849885,
      "num_tokens": 167619033.0,
      "step": 80770
    },
    {
      "entropy": 5.6949131965637205,
      "epoch": 7.327195210449927,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012602967457578807,
      "loss": 4.8156,
      "mean_token_accuracy": 0.23699064254760743,
      "num_tokens": 167629293.0,
      "step": 80775
    },
    {
      "entropy": 5.665574073791504,
      "epoch": 7.327648766328012,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000126005427437474,
      "loss": 4.7483,
      "mean_token_accuracy": 0.23795510530471803,
      "num_tokens": 167639515.0,
      "step": 80780
    },
    {
      "entropy": 5.720299196243286,
      "epoch": 7.328102322206096,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012598118338040564,
      "loss": 4.7809,
      "mean_token_accuracy": 0.24705965220928192,
      "num_tokens": 167651364.0,
      "step": 80785
    },
    {
      "entropy": 5.705311489105225,
      "epoch": 7.32855587808418,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012595694240508437,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23982430398464202,
      "num_tokens": 167662406.0,
      "step": 80790
    },
    {
      "entropy": 5.794840240478516,
      "epoch": 7.3290094339622645,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001259327045120115,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22947699427604676,
      "num_tokens": 167673430.0,
      "step": 80795
    },
    {
      "entropy": 5.690776777267456,
      "epoch": 7.329462989840349,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00012590846970168828,
      "loss": 4.8024,
      "mean_token_accuracy": 0.23994238376617433,
      "num_tokens": 167683608.0,
      "step": 80800
    },
    {
      "entropy": 5.686628961563111,
      "epoch": 7.329916545718432,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012588423797461597,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23502685278654098,
      "num_tokens": 167694160.0,
      "step": 80805
    },
    {
      "entropy": 5.691329002380371,
      "epoch": 7.330370101596516,
      "grad_norm": 1.5,
      "learning_rate": 0.00012586000933129557,
      "loss": 4.7431,
      "mean_token_accuracy": 0.23779174387454988,
      "num_tokens": 167703583.0,
      "step": 80810
    },
    {
      "entropy": 5.702123785018921,
      "epoch": 7.330823657474601,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001258357837722282,
      "loss": 4.7615,
      "mean_token_accuracy": 0.24127877801656722,
      "num_tokens": 167714622.0,
      "step": 80815
    },
    {
      "entropy": 5.7088782787323,
      "epoch": 7.331277213352685,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001258115612979148,
      "loss": 4.8171,
      "mean_token_accuracy": 0.24035807996988295,
      "num_tokens": 167724649.0,
      "step": 80820
    },
    {
      "entropy": 5.648400974273682,
      "epoch": 7.331730769230769,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001257873419088564,
      "loss": 4.7543,
      "mean_token_accuracy": 0.23912548571825026,
      "num_tokens": 167734675.0,
      "step": 80825
    },
    {
      "entropy": 5.686031627655029,
      "epoch": 7.332184325108853,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001257631256055537,
      "loss": 4.8297,
      "mean_token_accuracy": 0.233046992123127,
      "num_tokens": 167745558.0,
      "step": 80830
    },
    {
      "entropy": 5.669600009918213,
      "epoch": 7.332637880986938,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012573891238850762,
      "loss": 4.7812,
      "mean_token_accuracy": 0.2304044932126999,
      "num_tokens": 167755582.0,
      "step": 80835
    },
    {
      "entropy": 5.6780433177948,
      "epoch": 7.333091436865022,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001257147022582189,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22660826593637468,
      "num_tokens": 167766631.0,
      "step": 80840
    },
    {
      "entropy": 5.648380088806152,
      "epoch": 7.333544992743106,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012569049521518822,
      "loss": 4.7958,
      "mean_token_accuracy": 0.2281759887933731,
      "num_tokens": 167777041.0,
      "step": 80845
    },
    {
      "entropy": 5.708081245422363,
      "epoch": 7.3339985486211905,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001256662912599161,
      "loss": 4.8539,
      "mean_token_accuracy": 0.23272907435894014,
      "num_tokens": 167786977.0,
      "step": 80850
    },
    {
      "entropy": 5.6605278015136715,
      "epoch": 7.334452104499274,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012564209039290312,
      "loss": 4.7042,
      "mean_token_accuracy": 0.23998372107744217,
      "num_tokens": 167798492.0,
      "step": 80855
    },
    {
      "entropy": 5.686869192123413,
      "epoch": 7.334905660377358,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001256178926146498,
      "loss": 4.8167,
      "mean_token_accuracy": 0.23495445996522904,
      "num_tokens": 167809559.0,
      "step": 80860
    },
    {
      "entropy": 5.618239450454712,
      "epoch": 7.335359216255442,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001255936979256564,
      "loss": 4.7205,
      "mean_token_accuracy": 0.2449051946401596,
      "num_tokens": 167818795.0,
      "step": 80865
    },
    {
      "entropy": 5.627130174636841,
      "epoch": 7.335812772133527,
      "grad_norm": 1.375,
      "learning_rate": 0.00012556950632642357,
      "loss": 4.7096,
      "mean_token_accuracy": 0.24400675147771836,
      "num_tokens": 167829860.0,
      "step": 80870
    },
    {
      "entropy": 5.682979249954224,
      "epoch": 7.336266328011611,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012554531781745136,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23817257732152938,
      "num_tokens": 167841174.0,
      "step": 80875
    },
    {
      "entropy": 5.651905155181884,
      "epoch": 7.336719883889695,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012552113239924006,
      "loss": 4.7411,
      "mean_token_accuracy": 0.24221429526805877,
      "num_tokens": 167851750.0,
      "step": 80880
    },
    {
      "entropy": 5.74464225769043,
      "epoch": 7.3371734397677795,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012549695007228984,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22670807540416718,
      "num_tokens": 167861697.0,
      "step": 80885
    },
    {
      "entropy": 5.686903381347657,
      "epoch": 7.337626995645864,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012547277083710078,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2350178375840187,
      "num_tokens": 167872653.0,
      "step": 80890
    },
    {
      "entropy": 5.724564886093139,
      "epoch": 7.338080551523948,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012544859469417296,
      "loss": 4.863,
      "mean_token_accuracy": 0.23435367196798323,
      "num_tokens": 167882862.0,
      "step": 80895
    },
    {
      "entropy": 5.725908899307251,
      "epoch": 7.338534107402032,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012542442164400626,
      "loss": 4.8043,
      "mean_token_accuracy": 0.2336816668510437,
      "num_tokens": 167893003.0,
      "step": 80900
    },
    {
      "entropy": 5.6231646060943605,
      "epoch": 7.3389876632801165,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012540025168710064,
      "loss": 4.6932,
      "mean_token_accuracy": 0.24326329827308654,
      "num_tokens": 167903029.0,
      "step": 80905
    },
    {
      "entropy": 5.632634782791138,
      "epoch": 7.3394412191582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012537608482395595,
      "loss": 4.7188,
      "mean_token_accuracy": 0.2470138415694237,
      "num_tokens": 167913107.0,
      "step": 80910
    },
    {
      "entropy": 5.711182975769043,
      "epoch": 7.339894775036284,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000125351921055072,
      "loss": 4.8113,
      "mean_token_accuracy": 0.23873778879642488,
      "num_tokens": 167922886.0,
      "step": 80915
    },
    {
      "entropy": 5.675186347961426,
      "epoch": 7.3403483309143684,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012532776038094845,
      "loss": 4.7883,
      "mean_token_accuracy": 0.23654431849718094,
      "num_tokens": 167933149.0,
      "step": 80920
    },
    {
      "entropy": 5.747275352478027,
      "epoch": 7.340801886792453,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012530360280208492,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22459416091442108,
      "num_tokens": 167943608.0,
      "step": 80925
    },
    {
      "entropy": 5.710912752151489,
      "epoch": 7.341255442670537,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012527944831898106,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23788411021232606,
      "num_tokens": 167954424.0,
      "step": 80930
    },
    {
      "entropy": 5.687103700637818,
      "epoch": 7.341708998548621,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012525529693213627,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23271915018558503,
      "num_tokens": 167965199.0,
      "step": 80935
    },
    {
      "entropy": 5.642490100860596,
      "epoch": 7.3421625544267055,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012523114864205017,
      "loss": 4.7532,
      "mean_token_accuracy": 0.24464649856090545,
      "num_tokens": 167974644.0,
      "step": 80940
    },
    {
      "entropy": 5.631093215942383,
      "epoch": 7.34261611030479,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001252070034492221,
      "loss": 4.7098,
      "mean_token_accuracy": 0.24282160848379136,
      "num_tokens": 167985263.0,
      "step": 80945
    },
    {
      "entropy": 5.645437955856323,
      "epoch": 7.343069666182874,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012518286135415136,
      "loss": 4.7195,
      "mean_token_accuracy": 0.24734914302825928,
      "num_tokens": 167996414.0,
      "step": 80950
    },
    {
      "entropy": 5.643903017044067,
      "epoch": 7.343523222060958,
      "grad_norm": 1.5,
      "learning_rate": 0.00012515872235733725,
      "loss": 4.7418,
      "mean_token_accuracy": 0.24859345704317093,
      "num_tokens": 168007218.0,
      "step": 80955
    },
    {
      "entropy": 5.65173306465149,
      "epoch": 7.343976777939042,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012513458645927892,
      "loss": 4.8087,
      "mean_token_accuracy": 0.23665431886911392,
      "num_tokens": 168017238.0,
      "step": 80960
    },
    {
      "entropy": 5.726324129104614,
      "epoch": 7.344430333817126,
      "grad_norm": 1.5,
      "learning_rate": 0.00012511045366047553,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22558677345514297,
      "num_tokens": 168027762.0,
      "step": 80965
    },
    {
      "entropy": 5.69609112739563,
      "epoch": 7.34488388969521,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001250863239614261,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23654776215553283,
      "num_tokens": 168038825.0,
      "step": 80970
    },
    {
      "entropy": 5.6562049865722654,
      "epoch": 7.3453374455732945,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012506219736262974,
      "loss": 4.7618,
      "mean_token_accuracy": 0.24025674313306808,
      "num_tokens": 168050205.0,
      "step": 80975
    },
    {
      "entropy": 5.631498861312866,
      "epoch": 7.345791001451379,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012503807386458538,
      "loss": 4.7385,
      "mean_token_accuracy": 0.24203567951917648,
      "num_tokens": 168061171.0,
      "step": 80980
    },
    {
      "entropy": 5.714876937866211,
      "epoch": 7.346244557329463,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012501395346779185,
      "loss": 4.7932,
      "mean_token_accuracy": 0.2262965112924576,
      "num_tokens": 168070887.0,
      "step": 80985
    },
    {
      "entropy": 5.6817230701446535,
      "epoch": 7.346698113207547,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012498983617274798,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23724264651536942,
      "num_tokens": 168081949.0,
      "step": 80990
    },
    {
      "entropy": 5.688386058807373,
      "epoch": 7.3471516690856316,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012496572197995251,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2364712044596672,
      "num_tokens": 168092675.0,
      "step": 80995
    },
    {
      "entropy": 5.7780803680419925,
      "epoch": 7.347605224963716,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001249416108899041,
      "loss": 4.9267,
      "mean_token_accuracy": 0.2221642717719078,
      "num_tokens": 168102777.0,
      "step": 81000
    },
    {
      "epoch": 7.347605224963716,
      "eval_entropy": 5.413833413296695,
      "eval_loss": 4.9188313484191895,
      "eval_mean_token_accuracy": 0.23778447433126187,
      "eval_num_tokens": 168102777.0,
      "eval_runtime": 20.7106,
      "eval_samples_per_second": 1707.341,
      "eval_steps_per_second": 213.418,
      "step": 81000
    },
    {
      "entropy": 5.723496007919311,
      "epoch": 7.3480587808418,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012491750290310146,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23453736305236816,
      "num_tokens": 168113242.0,
      "step": 81005
    },
    {
      "entropy": 5.7141234397888185,
      "epoch": 7.3485123367198835,
      "grad_norm": 1.5,
      "learning_rate": 0.00012489339802004314,
      "loss": 4.7645,
      "mean_token_accuracy": 0.235772705078125,
      "num_tokens": 168123631.0,
      "step": 81010
    },
    {
      "entropy": 5.6361274242401125,
      "epoch": 7.348965892597968,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001248692962412276,
      "loss": 4.7698,
      "mean_token_accuracy": 0.23719546645879747,
      "num_tokens": 168134261.0,
      "step": 81015
    },
    {
      "entropy": 5.649791812896728,
      "epoch": 7.349419448476052,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012484519756715323,
      "loss": 4.7694,
      "mean_token_accuracy": 0.2487760603427887,
      "num_tokens": 168144701.0,
      "step": 81020
    },
    {
      "entropy": 5.637010812759399,
      "epoch": 7.349873004354136,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012482110199831846,
      "loss": 4.7359,
      "mean_token_accuracy": 0.24220097810029984,
      "num_tokens": 168155286.0,
      "step": 81025
    },
    {
      "entropy": 5.687467241287232,
      "epoch": 7.3503265602322205,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012479700953522158,
      "loss": 4.7733,
      "mean_token_accuracy": 0.2391968548297882,
      "num_tokens": 168165967.0,
      "step": 81030
    },
    {
      "entropy": 5.638638925552368,
      "epoch": 7.350780116110305,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012477292017836072,
      "loss": 4.7085,
      "mean_token_accuracy": 0.24182909429073335,
      "num_tokens": 168175839.0,
      "step": 81035
    },
    {
      "entropy": 5.685779047012329,
      "epoch": 7.351233671988389,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012474883392823427,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23238018453121184,
      "num_tokens": 168186565.0,
      "step": 81040
    },
    {
      "entropy": 5.651682329177857,
      "epoch": 7.351687227866473,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012472475078534017,
      "loss": 4.7389,
      "mean_token_accuracy": 0.2532279446721077,
      "num_tokens": 168196425.0,
      "step": 81045
    },
    {
      "entropy": 5.7207599639892575,
      "epoch": 7.352140783744558,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001247006707501766,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23891181498765945,
      "num_tokens": 168205436.0,
      "step": 81050
    },
    {
      "entropy": 5.711791706085205,
      "epoch": 7.352594339622642,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012467659382324136,
      "loss": 4.8122,
      "mean_token_accuracy": 0.23741666078567505,
      "num_tokens": 168215467.0,
      "step": 81055
    },
    {
      "entropy": 5.636174440383911,
      "epoch": 7.353047895500726,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012465252000503252,
      "loss": 4.7398,
      "mean_token_accuracy": 0.2426847293972969,
      "num_tokens": 168225423.0,
      "step": 81060
    },
    {
      "entropy": 5.718926239013672,
      "epoch": 7.3535014513788095,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012462844929604788,
      "loss": 4.799,
      "mean_token_accuracy": 0.2370886579155922,
      "num_tokens": 168235606.0,
      "step": 81065
    },
    {
      "entropy": 5.658138370513916,
      "epoch": 7.353955007256894,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012460438169678516,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24236348271369934,
      "num_tokens": 168245703.0,
      "step": 81070
    },
    {
      "entropy": 5.73297610282898,
      "epoch": 7.354408563134978,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001245803172077422,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22629911452531815,
      "num_tokens": 168256409.0,
      "step": 81075
    },
    {
      "entropy": 5.651335144042969,
      "epoch": 7.354862119013062,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012455625582941663,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2442840814590454,
      "num_tokens": 168267395.0,
      "step": 81080
    },
    {
      "entropy": 5.637912178039551,
      "epoch": 7.355315674891147,
      "grad_norm": 1.5,
      "learning_rate": 0.000124532197562306,
      "loss": 4.7722,
      "mean_token_accuracy": 0.23232924193143845,
      "num_tokens": 168277522.0,
      "step": 81085
    },
    {
      "entropy": 5.617005681991577,
      "epoch": 7.355769230769231,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001245081424069079,
      "loss": 4.7029,
      "mean_token_accuracy": 0.2472250744700432,
      "num_tokens": 168288275.0,
      "step": 81090
    },
    {
      "entropy": 5.6828210830688475,
      "epoch": 7.356222786647315,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012448409036371977,
      "loss": 4.817,
      "mean_token_accuracy": 0.23590667843818663,
      "num_tokens": 168299096.0,
      "step": 81095
    },
    {
      "entropy": 5.64278826713562,
      "epoch": 7.356676342525399,
      "grad_norm": 1.375,
      "learning_rate": 0.00012446004143323899,
      "loss": 4.7841,
      "mean_token_accuracy": 0.23885442316532135,
      "num_tokens": 168309638.0,
      "step": 81100
    },
    {
      "entropy": 5.7244439125061035,
      "epoch": 7.357129898403484,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012443599561596283,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22163786888122558,
      "num_tokens": 168319277.0,
      "step": 81105
    },
    {
      "entropy": 5.647344589233398,
      "epoch": 7.357583454281568,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012441195291238875,
      "loss": 4.7458,
      "mean_token_accuracy": 0.24620737880468369,
      "num_tokens": 168329096.0,
      "step": 81110
    },
    {
      "entropy": 5.741960668563843,
      "epoch": 7.358037010159651,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012438791332301384,
      "loss": 4.8332,
      "mean_token_accuracy": 0.24001302272081376,
      "num_tokens": 168338981.0,
      "step": 81115
    },
    {
      "entropy": 5.7166626930236815,
      "epoch": 7.3584905660377355,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012436387684833526,
      "loss": 4.8105,
      "mean_token_accuracy": 0.2392224609851837,
      "num_tokens": 168350184.0,
      "step": 81120
    },
    {
      "entropy": 5.647202444076538,
      "epoch": 7.35894412191582,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012433984348885009,
      "loss": 4.6974,
      "mean_token_accuracy": 0.24252939671278,
      "num_tokens": 168360665.0,
      "step": 81125
    },
    {
      "entropy": 5.652266931533814,
      "epoch": 7.359397677793904,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012431581324505537,
      "loss": 4.8143,
      "mean_token_accuracy": 0.22904779762029648,
      "num_tokens": 168371379.0,
      "step": 81130
    },
    {
      "entropy": 5.700074815750122,
      "epoch": 7.359851233671988,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012429178611744803,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23175902664661407,
      "num_tokens": 168382056.0,
      "step": 81135
    },
    {
      "entropy": 5.69833812713623,
      "epoch": 7.360304789550073,
      "grad_norm": 1.5,
      "learning_rate": 0.00012426776210652487,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23529503345489503,
      "num_tokens": 168391424.0,
      "step": 81140
    },
    {
      "entropy": 5.686516571044922,
      "epoch": 7.360758345428157,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001242437412127829,
      "loss": 4.7795,
      "mean_token_accuracy": 0.239681276679039,
      "num_tokens": 168401818.0,
      "step": 81145
    },
    {
      "entropy": 5.737078237533569,
      "epoch": 7.361211901306241,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012421972343671872,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23851344138383865,
      "num_tokens": 168410867.0,
      "step": 81150
    },
    {
      "entropy": 5.654119825363159,
      "epoch": 7.361665457184325,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00012419570877882915,
      "loss": 4.8061,
      "mean_token_accuracy": 0.23847198635339736,
      "num_tokens": 168420570.0,
      "step": 81155
    },
    {
      "entropy": 5.617822074890137,
      "epoch": 7.36211901306241,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001241716972396107,
      "loss": 4.6758,
      "mean_token_accuracy": 0.24284279048442842,
      "num_tokens": 168430888.0,
      "step": 81160
    },
    {
      "entropy": 5.667983150482177,
      "epoch": 7.362572568940493,
      "grad_norm": 1.6875,
      "learning_rate": 0.00012414768881956,
      "loss": 4.7729,
      "mean_token_accuracy": 0.23998259007930756,
      "num_tokens": 168440932.0,
      "step": 81165
    },
    {
      "entropy": 5.655382061004639,
      "epoch": 7.363026124818577,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012412368351917357,
      "loss": 4.739,
      "mean_token_accuracy": 0.24163853079080583,
      "num_tokens": 168451567.0,
      "step": 81170
    },
    {
      "entropy": 5.718571043014526,
      "epoch": 7.363479680696662,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001240996813389478,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23902571946382523,
      "num_tokens": 168461922.0,
      "step": 81175
    },
    {
      "entropy": 5.671838569641113,
      "epoch": 7.363933236574746,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012407568227937904,
      "loss": 4.793,
      "mean_token_accuracy": 0.23984011709690095,
      "num_tokens": 168471885.0,
      "step": 81180
    },
    {
      "entropy": 5.710716295242309,
      "epoch": 7.36438679245283,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012405168634096366,
      "loss": 4.7961,
      "mean_token_accuracy": 0.2350583180785179,
      "num_tokens": 168481770.0,
      "step": 81185
    },
    {
      "entropy": 5.691390180587769,
      "epoch": 7.364840348330914,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012402769352419787,
      "loss": 4.7502,
      "mean_token_accuracy": 0.2406336784362793,
      "num_tokens": 168491590.0,
      "step": 81190
    },
    {
      "entropy": 5.79954423904419,
      "epoch": 7.365293904208999,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012400370382957784,
      "loss": 4.9443,
      "mean_token_accuracy": 0.22671602964401244,
      "num_tokens": 168503819.0,
      "step": 81195
    },
    {
      "entropy": 5.777622365951538,
      "epoch": 7.365747460087083,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012397971725759964,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21724750995635986,
      "num_tokens": 168514531.0,
      "step": 81200
    },
    {
      "entropy": 5.720914745330811,
      "epoch": 7.366201015965167,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012395573380875942,
      "loss": 4.8078,
      "mean_token_accuracy": 0.24003355354070663,
      "num_tokens": 168523940.0,
      "step": 81205
    },
    {
      "entropy": 5.6789360523223875,
      "epoch": 7.3666545718432515,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012393175348355314,
      "loss": 4.75,
      "mean_token_accuracy": 0.24255448579788208,
      "num_tokens": 168532901.0,
      "step": 81210
    },
    {
      "entropy": 5.706609201431275,
      "epoch": 7.367108127721336,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012390777628247667,
      "loss": 4.8698,
      "mean_token_accuracy": 0.23419414460659027,
      "num_tokens": 168543813.0,
      "step": 81215
    },
    {
      "entropy": 5.643485069274902,
      "epoch": 7.367561683599419,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012388380220602592,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2337145522236824,
      "num_tokens": 168555224.0,
      "step": 81220
    },
    {
      "entropy": 5.692464017868042,
      "epoch": 7.368015239477503,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012385983125469662,
      "loss": 4.7852,
      "mean_token_accuracy": 0.23622420579195022,
      "num_tokens": 168565046.0,
      "step": 81225
    },
    {
      "entropy": 5.6798272132873535,
      "epoch": 7.368468795355588,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012383586342898454,
      "loss": 4.8368,
      "mean_token_accuracy": 0.2378340795636177,
      "num_tokens": 168576633.0,
      "step": 81230
    },
    {
      "entropy": 5.667960739135742,
      "epoch": 7.368922351233672,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012381189872938525,
      "loss": 4.7306,
      "mean_token_accuracy": 0.24198979139328003,
      "num_tokens": 168586785.0,
      "step": 81235
    },
    {
      "entropy": 5.649985504150391,
      "epoch": 7.369375907111756,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001237879371563945,
      "loss": 4.8661,
      "mean_token_accuracy": 0.23147812634706497,
      "num_tokens": 168597619.0,
      "step": 81240
    },
    {
      "entropy": 5.6721577644348145,
      "epoch": 7.36982946298984,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012376397871050773,
      "loss": 4.8179,
      "mean_token_accuracy": 0.23690919429063798,
      "num_tokens": 168607141.0,
      "step": 81245
    },
    {
      "entropy": 5.66845293045044,
      "epoch": 7.370283018867925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001237400233922204,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2368020623922348,
      "num_tokens": 168617526.0,
      "step": 81250
    },
    {
      "entropy": 5.768713092803955,
      "epoch": 7.370736574746009,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012371607120202798,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2299398437142372,
      "num_tokens": 168629373.0,
      "step": 81255
    },
    {
      "entropy": 5.705772018432617,
      "epoch": 7.371190130624093,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012369212214042574,
      "loss": 4.8091,
      "mean_token_accuracy": 0.23023316264152527,
      "num_tokens": 168640634.0,
      "step": 81260
    },
    {
      "entropy": 5.737506008148193,
      "epoch": 7.371643686502177,
      "grad_norm": 1.6875,
      "learning_rate": 0.00012366817620790898,
      "loss": 4.8304,
      "mean_token_accuracy": 0.23069190829992295,
      "num_tokens": 168649767.0,
      "step": 81265
    },
    {
      "entropy": 5.649160623550415,
      "epoch": 7.372097242380261,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001236442334049728,
      "loss": 4.7685,
      "mean_token_accuracy": 0.23788317590951918,
      "num_tokens": 168660229.0,
      "step": 81270
    },
    {
      "entropy": 5.604401063919068,
      "epoch": 7.372550798258345,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012362029373211254,
      "loss": 4.6845,
      "mean_token_accuracy": 0.2521725565195084,
      "num_tokens": 168669588.0,
      "step": 81275
    },
    {
      "entropy": 5.753988933563233,
      "epoch": 7.373004354136429,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012359635718982317,
      "loss": 4.8707,
      "mean_token_accuracy": 0.23003207445144652,
      "num_tokens": 168679488.0,
      "step": 81280
    },
    {
      "entropy": 5.725850248336792,
      "epoch": 7.373457910014514,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001235724237785997,
      "loss": 4.8622,
      "mean_token_accuracy": 0.24083968251943588,
      "num_tokens": 168689619.0,
      "step": 81285
    },
    {
      "entropy": 5.763457870483398,
      "epoch": 7.373911465892598,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012354849349893713,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22393730729818345,
      "num_tokens": 168701818.0,
      "step": 81290
    },
    {
      "entropy": 5.632351207733154,
      "epoch": 7.374365021770682,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001235245663513303,
      "loss": 4.7053,
      "mean_token_accuracy": 0.25177421420812607,
      "num_tokens": 168711842.0,
      "step": 81295
    },
    {
      "entropy": 5.737472629547119,
      "epoch": 7.3748185776487665,
      "grad_norm": 1.5,
      "learning_rate": 0.00012350064233627395,
      "loss": 4.8891,
      "mean_token_accuracy": 0.2269837275147438,
      "num_tokens": 168721919.0,
      "step": 81300
    },
    {
      "entropy": 5.7060943126678465,
      "epoch": 7.375272133526851,
      "grad_norm": 1.515625,
      "learning_rate": 0.000123476721454263,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23341753333806992,
      "num_tokens": 168732077.0,
      "step": 81305
    },
    {
      "entropy": 5.731432962417602,
      "epoch": 7.375725689404935,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00012345280370579207,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23594329953193666,
      "num_tokens": 168742171.0,
      "step": 81310
    },
    {
      "entropy": 5.76609845161438,
      "epoch": 7.376179245283019,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012342888909135577,
      "loss": 4.9344,
      "mean_token_accuracy": 0.22323248237371446,
      "num_tokens": 168752529.0,
      "step": 81315
    },
    {
      "entropy": 5.799232339859008,
      "epoch": 7.376632801161103,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0001234049776114487,
      "loss": 4.9125,
      "mean_token_accuracy": 0.23862666189670562,
      "num_tokens": 168762847.0,
      "step": 81320
    },
    {
      "entropy": 5.735550355911255,
      "epoch": 7.377086357039187,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001233810692665653,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22803432792425155,
      "num_tokens": 168773107.0,
      "step": 81325
    },
    {
      "entropy": 5.63005428314209,
      "epoch": 7.377539912917271,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001233571640572,
      "loss": 4.6867,
      "mean_token_accuracy": 0.24535962492227553,
      "num_tokens": 168783364.0,
      "step": 81330
    },
    {
      "entropy": 5.665647315979004,
      "epoch": 7.3779934687953554,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012333326198384715,
      "loss": 4.7791,
      "mean_token_accuracy": 0.23043323308229446,
      "num_tokens": 168793412.0,
      "step": 81335
    },
    {
      "entropy": 5.680971384048462,
      "epoch": 7.37844702467344,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012330936304700115,
      "loss": 4.8171,
      "mean_token_accuracy": 0.23007944077253342,
      "num_tokens": 168803480.0,
      "step": 81340
    },
    {
      "entropy": 5.762677526473999,
      "epoch": 7.378900580551524,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001232854672471562,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22479903250932692,
      "num_tokens": 168814764.0,
      "step": 81345
    },
    {
      "entropy": 5.7548846244812015,
      "epoch": 7.379354136429608,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012326157458480644,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22856409400701522,
      "num_tokens": 168825925.0,
      "step": 81350
    },
    {
      "entropy": 5.752668809890747,
      "epoch": 7.3798076923076925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012323768506044595,
      "loss": 4.8423,
      "mean_token_accuracy": 0.23108072876930236,
      "num_tokens": 168837783.0,
      "step": 81355
    },
    {
      "entropy": 5.728699207305908,
      "epoch": 7.380261248185777,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012321379867456884,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23514918535947799,
      "num_tokens": 168848600.0,
      "step": 81360
    },
    {
      "entropy": 5.702346563339233,
      "epoch": 7.380714804063861,
      "grad_norm": 1.5,
      "learning_rate": 0.00012318991542766903,
      "loss": 4.874,
      "mean_token_accuracy": 0.22502328604459762,
      "num_tokens": 168859946.0,
      "step": 81365
    },
    {
      "entropy": 5.779552841186524,
      "epoch": 7.381168359941945,
      "grad_norm": 1.765625,
      "learning_rate": 0.0001231660353202404,
      "loss": 4.9972,
      "mean_token_accuracy": 0.22577069699764252,
      "num_tokens": 168870665.0,
      "step": 81370
    },
    {
      "entropy": 5.623283958435058,
      "epoch": 7.381621915820029,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012314215835277693,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23984850198030472,
      "num_tokens": 168881240.0,
      "step": 81375
    },
    {
      "entropy": 5.752943897247315,
      "epoch": 7.382075471698113,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012311828452577232,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2376105159521103,
      "num_tokens": 168892205.0,
      "step": 81380
    },
    {
      "entropy": 5.66025390625,
      "epoch": 7.382529027576197,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012309441383972028,
      "loss": 4.7356,
      "mean_token_accuracy": 0.24317531436681747,
      "num_tokens": 168902035.0,
      "step": 81385
    },
    {
      "entropy": 5.6759134292602536,
      "epoch": 7.3829825834542815,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012307054629511448,
      "loss": 4.7273,
      "mean_token_accuracy": 0.24124190360307693,
      "num_tokens": 168912537.0,
      "step": 81390
    },
    {
      "entropy": 5.713928365707398,
      "epoch": 7.383436139332366,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012304668189244849,
      "loss": 4.776,
      "mean_token_accuracy": 0.23019180148839952,
      "num_tokens": 168922975.0,
      "step": 81395
    },
    {
      "entropy": 5.697358846664429,
      "epoch": 7.38388969521045,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012302282063221586,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23887412548065184,
      "num_tokens": 168932658.0,
      "step": 81400
    },
    {
      "entropy": 5.671048927307129,
      "epoch": 7.384343251088534,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012299896251490993,
      "loss": 4.7087,
      "mean_token_accuracy": 0.24286148250102996,
      "num_tokens": 168942680.0,
      "step": 81405
    },
    {
      "entropy": 5.697747707366943,
      "epoch": 7.3847968069666186,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00012297510754102426,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22668686211109162,
      "num_tokens": 168951799.0,
      "step": 81410
    },
    {
      "entropy": 5.754725551605224,
      "epoch": 7.385250362844703,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012295125571105215,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2253000721335411,
      "num_tokens": 168961954.0,
      "step": 81415
    },
    {
      "entropy": 5.594652080535889,
      "epoch": 7.385703918722786,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012292740702548677,
      "loss": 4.7043,
      "mean_token_accuracy": 0.24979418367147446,
      "num_tokens": 168971546.0,
      "step": 81420
    },
    {
      "entropy": 5.647523736953735,
      "epoch": 7.3861574746008705,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001229035614848214,
      "loss": 4.7514,
      "mean_token_accuracy": 0.24436900466680528,
      "num_tokens": 168982124.0,
      "step": 81425
    },
    {
      "entropy": 5.649971771240234,
      "epoch": 7.386611030478955,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001228797190895491,
      "loss": 4.7134,
      "mean_token_accuracy": 0.2478869691491127,
      "num_tokens": 168992979.0,
      "step": 81430
    },
    {
      "entropy": 5.6642309665679935,
      "epoch": 7.387064586357039,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000122855879840163,
      "loss": 4.7495,
      "mean_token_accuracy": 0.2412780687212944,
      "num_tokens": 169003505.0,
      "step": 81435
    },
    {
      "entropy": 5.713238859176636,
      "epoch": 7.387518142235123,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012283204373715603,
      "loss": 4.7865,
      "mean_token_accuracy": 0.2328994631767273,
      "num_tokens": 169014209.0,
      "step": 81440
    },
    {
      "entropy": 5.656588554382324,
      "epoch": 7.3879716981132075,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001228082107810212,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2476410076022148,
      "num_tokens": 169024034.0,
      "step": 81445
    },
    {
      "entropy": 5.710478496551514,
      "epoch": 7.388425253991292,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00012278438097225135,
      "loss": 4.7433,
      "mean_token_accuracy": 0.24171810895204543,
      "num_tokens": 169034677.0,
      "step": 81450
    },
    {
      "entropy": 5.687245798110962,
      "epoch": 7.388878809869376,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012276055431133934,
      "loss": 4.7535,
      "mean_token_accuracy": 0.24508186280727387,
      "num_tokens": 169044667.0,
      "step": 81455
    },
    {
      "entropy": 5.711621618270874,
      "epoch": 7.38933236574746,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001227367307987778,
      "loss": 4.8752,
      "mean_token_accuracy": 0.23636381477117538,
      "num_tokens": 169054451.0,
      "step": 81460
    },
    {
      "entropy": 5.738630533218384,
      "epoch": 7.389785921625545,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001227129104350595,
      "loss": 4.8382,
      "mean_token_accuracy": 0.2341847911477089,
      "num_tokens": 169064753.0,
      "step": 81465
    },
    {
      "entropy": 5.751779270172119,
      "epoch": 7.390239477503629,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012268909322067704,
      "loss": 4.8467,
      "mean_token_accuracy": 0.2368702158331871,
      "num_tokens": 169075463.0,
      "step": 81470
    },
    {
      "entropy": 5.6980925559997555,
      "epoch": 7.390693033381712,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012266527915612285,
      "loss": 4.8298,
      "mean_token_accuracy": 0.2388140857219696,
      "num_tokens": 169084378.0,
      "step": 81475
    },
    {
      "entropy": 5.613417339324951,
      "epoch": 7.3911465892597965,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001226414682418896,
      "loss": 4.7856,
      "mean_token_accuracy": 0.23629797399044036,
      "num_tokens": 169094890.0,
      "step": 81480
    },
    {
      "entropy": 5.7465372562408445,
      "epoch": 7.391600145137881,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001226176604784696,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22274529337882995,
      "num_tokens": 169105648.0,
      "step": 81485
    },
    {
      "entropy": 5.689800167083741,
      "epoch": 7.392053701015965,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012259385586635516,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23402854204177856,
      "num_tokens": 169115831.0,
      "step": 81490
    },
    {
      "entropy": 5.686041593551636,
      "epoch": 7.392507256894049,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001225700544060387,
      "loss": 4.7528,
      "mean_token_accuracy": 0.24324069172143936,
      "num_tokens": 169125567.0,
      "step": 81495
    },
    {
      "entropy": 5.6978926181793215,
      "epoch": 7.392960812772134,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012254625609801233,
      "loss": 4.8009,
      "mean_token_accuracy": 0.2380684956908226,
      "num_tokens": 169137239.0,
      "step": 81500
    },
    {
      "entropy": 5.626588153839111,
      "epoch": 7.393414368650218,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012252246094276816,
      "loss": 4.7197,
      "mean_token_accuracy": 0.246528160572052,
      "num_tokens": 169147223.0,
      "step": 81505
    },
    {
      "entropy": 5.663716983795166,
      "epoch": 7.393867924528302,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001224986689407984,
      "loss": 4.7682,
      "mean_token_accuracy": 0.2355859249830246,
      "num_tokens": 169156993.0,
      "step": 81510
    },
    {
      "entropy": 5.7580643653869625,
      "epoch": 7.394321480406386,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00012247488009259504,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2347245529294014,
      "num_tokens": 169166872.0,
      "step": 81515
    },
    {
      "entropy": 5.6508008480072025,
      "epoch": 7.394775036284471,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012245109439865007,
      "loss": 4.8434,
      "mean_token_accuracy": 0.2351182743906975,
      "num_tokens": 169176967.0,
      "step": 81520
    },
    {
      "entropy": 5.682506895065307,
      "epoch": 7.395228592162554,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001224273118594553,
      "loss": 4.8191,
      "mean_token_accuracy": 0.2333638623356819,
      "num_tokens": 169187107.0,
      "step": 81525
    },
    {
      "entropy": 5.697214508056641,
      "epoch": 7.395682148040638,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012240353247550261,
      "loss": 4.7636,
      "mean_token_accuracy": 0.24076257199048995,
      "num_tokens": 169197490.0,
      "step": 81530
    },
    {
      "entropy": 5.622847986221314,
      "epoch": 7.3961357039187225,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012237975624728375,
      "loss": 4.7008,
      "mean_token_accuracy": 0.24428580552339554,
      "num_tokens": 169208037.0,
      "step": 81535
    },
    {
      "entropy": 5.69072790145874,
      "epoch": 7.396589259796807,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012235598317529036,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23071001321077347,
      "num_tokens": 169217200.0,
      "step": 81540
    },
    {
      "entropy": 5.738064908981324,
      "epoch": 7.397042815674891,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001223322132600142,
      "loss": 4.8044,
      "mean_token_accuracy": 0.22943001091480256,
      "num_tokens": 169227660.0,
      "step": 81545
    },
    {
      "entropy": 5.723073053359985,
      "epoch": 7.397496371552975,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012230844650194678,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22978173047304154,
      "num_tokens": 169238430.0,
      "step": 81550
    },
    {
      "entropy": 5.671173858642578,
      "epoch": 7.39794992743106,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001222846829015796,
      "loss": 4.8101,
      "mean_token_accuracy": 0.23083504736423494,
      "num_tokens": 169248044.0,
      "step": 81555
    },
    {
      "entropy": 5.721630048751831,
      "epoch": 7.398403483309144,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012226092245940407,
      "loss": 4.8674,
      "mean_token_accuracy": 0.23637733310461045,
      "num_tokens": 169257939.0,
      "step": 81560
    },
    {
      "entropy": 5.691953992843628,
      "epoch": 7.398857039187228,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001222371651759116,
      "loss": 4.8094,
      "mean_token_accuracy": 0.22957893460988998,
      "num_tokens": 169268386.0,
      "step": 81565
    },
    {
      "entropy": 5.719910383224487,
      "epoch": 7.399310595065312,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001222134110515935,
      "loss": 4.9051,
      "mean_token_accuracy": 0.23638111650943755,
      "num_tokens": 169278815.0,
      "step": 81570
    },
    {
      "entropy": 5.678946447372437,
      "epoch": 7.399764150943396,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012218966008694092,
      "loss": 4.728,
      "mean_token_accuracy": 0.2399652346968651,
      "num_tokens": 169289203.0,
      "step": 81575
    },
    {
      "entropy": 5.671596336364746,
      "epoch": 7.40021770682148,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00012216591228244518,
      "loss": 4.7762,
      "mean_token_accuracy": 0.24165356755256653,
      "num_tokens": 169299045.0,
      "step": 81580
    },
    {
      "entropy": 5.696849250793457,
      "epoch": 7.400671262699564,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012214216763859728,
      "loss": 4.8525,
      "mean_token_accuracy": 0.2310642868280411,
      "num_tokens": 169309400.0,
      "step": 81585
    },
    {
      "entropy": 5.735404825210571,
      "epoch": 7.401124818577649,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012211842615588835,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23358568400144578,
      "num_tokens": 169319641.0,
      "step": 81590
    },
    {
      "entropy": 5.720919513702393,
      "epoch": 7.401578374455733,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012209468783480932,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23880073577165603,
      "num_tokens": 169330027.0,
      "step": 81595
    },
    {
      "entropy": 5.639342403411865,
      "epoch": 7.402031930333817,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001220709526758511,
      "loss": 4.7445,
      "mean_token_accuracy": 0.23728809952735902,
      "num_tokens": 169339493.0,
      "step": 81600
    },
    {
      "entropy": 5.66121997833252,
      "epoch": 7.402485486211901,
      "grad_norm": 1.796875,
      "learning_rate": 0.00012204722067950456,
      "loss": 4.8072,
      "mean_token_accuracy": 0.23726499378681182,
      "num_tokens": 169350565.0,
      "step": 81605
    },
    {
      "entropy": 5.649004554748535,
      "epoch": 7.402939042089986,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00012202349184626046,
      "loss": 4.7621,
      "mean_token_accuracy": 0.24366765469312668,
      "num_tokens": 169360325.0,
      "step": 81610
    },
    {
      "entropy": 5.6914341926574705,
      "epoch": 7.40339259796807,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012199976617660954,
      "loss": 4.7997,
      "mean_token_accuracy": 0.23515740633010865,
      "num_tokens": 169370910.0,
      "step": 81615
    },
    {
      "entropy": 5.7027122497558596,
      "epoch": 7.403846153846154,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012197604367104245,
      "loss": 4.8499,
      "mean_token_accuracy": 0.23220687955617905,
      "num_tokens": 169381496.0,
      "step": 81620
    },
    {
      "entropy": 5.707665300369262,
      "epoch": 7.4042997097242385,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012195232433004977,
      "loss": 4.8265,
      "mean_token_accuracy": 0.23507474064826966,
      "num_tokens": 169391477.0,
      "step": 81625
    },
    {
      "entropy": 5.725276041030884,
      "epoch": 7.404753265602322,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012192860815412202,
      "loss": 4.8063,
      "mean_token_accuracy": 0.23531776070594787,
      "num_tokens": 169402588.0,
      "step": 81630
    },
    {
      "entropy": 5.648385000228882,
      "epoch": 7.405206821480406,
      "grad_norm": 1.375,
      "learning_rate": 0.0001219048951437496,
      "loss": 4.8099,
      "mean_token_accuracy": 0.22977572828531265,
      "num_tokens": 169414087.0,
      "step": 81635
    },
    {
      "entropy": 5.678192186355591,
      "epoch": 7.40566037735849,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012188118529942302,
      "loss": 4.7659,
      "mean_token_accuracy": 0.24209673553705216,
      "num_tokens": 169424541.0,
      "step": 81640
    },
    {
      "entropy": 5.70583872795105,
      "epoch": 7.406113933236575,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012185747862163253,
      "loss": 4.7735,
      "mean_token_accuracy": 0.23275058418512345,
      "num_tokens": 169434973.0,
      "step": 81645
    },
    {
      "entropy": 5.718557929992675,
      "epoch": 7.406567489114659,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012183377511086844,
      "loss": 4.7892,
      "mean_token_accuracy": 0.2354325994849205,
      "num_tokens": 169445049.0,
      "step": 81650
    },
    {
      "entropy": 5.7290346145629885,
      "epoch": 7.407021044992743,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012181007476762089,
      "loss": 4.8161,
      "mean_token_accuracy": 0.2391987130045891,
      "num_tokens": 169454437.0,
      "step": 81655
    },
    {
      "entropy": 5.648699522018433,
      "epoch": 7.407474600870827,
      "grad_norm": 1.6875,
      "learning_rate": 0.00012178637759238005,
      "loss": 4.7311,
      "mean_token_accuracy": 0.2409927099943161,
      "num_tokens": 169463411.0,
      "step": 81660
    },
    {
      "entropy": 5.699513483047485,
      "epoch": 7.407928156748912,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012176268358563595,
      "loss": 4.8022,
      "mean_token_accuracy": 0.2334605410695076,
      "num_tokens": 169473400.0,
      "step": 81665
    },
    {
      "entropy": 5.707895231246948,
      "epoch": 7.408381712626996,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012173899274787855,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23492905348539353,
      "num_tokens": 169483435.0,
      "step": 81670
    },
    {
      "entropy": 5.668398904800415,
      "epoch": 7.40883526850508,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012171530507959788,
      "loss": 4.7685,
      "mean_token_accuracy": 0.24057261496782303,
      "num_tokens": 169493598.0,
      "step": 81675
    },
    {
      "entropy": 5.635529375076294,
      "epoch": 7.409288824383164,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012169162058128379,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23355320543050767,
      "num_tokens": 169504553.0,
      "step": 81680
    },
    {
      "entropy": 5.675246334075927,
      "epoch": 7.409742380261248,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012166793925342605,
      "loss": 4.7847,
      "mean_token_accuracy": 0.24261100888252257,
      "num_tokens": 169514429.0,
      "step": 81685
    },
    {
      "entropy": 5.663343143463135,
      "epoch": 7.410195936139332,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012164426109651437,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23655266761779786,
      "num_tokens": 169526135.0,
      "step": 81690
    },
    {
      "entropy": 5.722339820861817,
      "epoch": 7.410649492017416,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012162058611103849,
      "loss": 4.7633,
      "mean_token_accuracy": 0.24689801037311554,
      "num_tokens": 169535673.0,
      "step": 81695
    },
    {
      "entropy": 5.696670722961426,
      "epoch": 7.411103047895501,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012159691429748795,
      "loss": 4.7347,
      "mean_token_accuracy": 0.2497272178530693,
      "num_tokens": 169545126.0,
      "step": 81700
    },
    {
      "entropy": 5.6616006851196286,
      "epoch": 7.411556603773585,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012157324565635226,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24212064296007157,
      "num_tokens": 169556115.0,
      "step": 81705
    },
    {
      "entropy": 5.639784908294677,
      "epoch": 7.412010159651669,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012154958018812099,
      "loss": 4.7763,
      "mean_token_accuracy": 0.23940693885087966,
      "num_tokens": 169565642.0,
      "step": 81710
    },
    {
      "entropy": 5.611612224578858,
      "epoch": 7.4124637155297535,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012152591789328354,
      "loss": 4.7034,
      "mean_token_accuracy": 0.24298497438430786,
      "num_tokens": 169574942.0,
      "step": 81715
    },
    {
      "entropy": 5.697449254989624,
      "epoch": 7.412917271407838,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001215022587723292,
      "loss": 4.7982,
      "mean_token_accuracy": 0.23454180210828782,
      "num_tokens": 169584994.0,
      "step": 81720
    },
    {
      "entropy": 5.743783283233642,
      "epoch": 7.413370827285922,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012147860282574726,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22931042462587356,
      "num_tokens": 169595641.0,
      "step": 81725
    },
    {
      "entropy": 5.643787956237793,
      "epoch": 7.413824383164005,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012145495005402694,
      "loss": 4.7304,
      "mean_token_accuracy": 0.23726305067539216,
      "num_tokens": 169606826.0,
      "step": 81730
    },
    {
      "entropy": 5.691174936294556,
      "epoch": 7.41427793904209,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012143130045765737,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23688192963600158,
      "num_tokens": 169617563.0,
      "step": 81735
    },
    {
      "entropy": 5.639278554916382,
      "epoch": 7.414731494920174,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012140765403712758,
      "loss": 4.7464,
      "mean_token_accuracy": 0.23159608393907546,
      "num_tokens": 169626958.0,
      "step": 81740
    },
    {
      "entropy": 5.624629878997803,
      "epoch": 7.415185050798258,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012138401079292673,
      "loss": 4.7041,
      "mean_token_accuracy": 0.24747368395328523,
      "num_tokens": 169637886.0,
      "step": 81745
    },
    {
      "entropy": 5.704370784759521,
      "epoch": 7.4156386066763424,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012136037072554366,
      "loss": 4.8299,
      "mean_token_accuracy": 0.23352744728326796,
      "num_tokens": 169647772.0,
      "step": 81750
    },
    {
      "entropy": 5.61042046546936,
      "epoch": 7.416092162554427,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012133673383546728,
      "loss": 4.7338,
      "mean_token_accuracy": 0.2474440023303032,
      "num_tokens": 169658354.0,
      "step": 81755
    },
    {
      "entropy": 5.595696496963501,
      "epoch": 7.416545718432511,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012131310012318643,
      "loss": 4.708,
      "mean_token_accuracy": 0.25012904554605486,
      "num_tokens": 169668572.0,
      "step": 81760
    },
    {
      "entropy": 5.639694881439209,
      "epoch": 7.416999274310595,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001212894695891898,
      "loss": 4.7748,
      "mean_token_accuracy": 0.24095564782619477,
      "num_tokens": 169679411.0,
      "step": 81765
    },
    {
      "entropy": 5.706206941604615,
      "epoch": 7.4174528301886795,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012126584223396612,
      "loss": 4.8322,
      "mean_token_accuracy": 0.235731840133667,
      "num_tokens": 169688655.0,
      "step": 81770
    },
    {
      "entropy": 5.671988201141358,
      "epoch": 7.417906386066764,
      "grad_norm": 1.375,
      "learning_rate": 0.00012124221805800393,
      "loss": 4.7755,
      "mean_token_accuracy": 0.22989991009235383,
      "num_tokens": 169700164.0,
      "step": 81775
    },
    {
      "entropy": 5.735798597335815,
      "epoch": 7.418359941944848,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012121859706179194,
      "loss": 4.8507,
      "mean_token_accuracy": 0.235653118789196,
      "num_tokens": 169710456.0,
      "step": 81780
    },
    {
      "entropy": 5.678575134277343,
      "epoch": 7.418813497822931,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012119497924581851,
      "loss": 4.777,
      "mean_token_accuracy": 0.241900035738945,
      "num_tokens": 169719689.0,
      "step": 81785
    },
    {
      "entropy": 5.693374967575073,
      "epoch": 7.419267053701016,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012117136461057213,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23414182662963867,
      "num_tokens": 169729591.0,
      "step": 81790
    },
    {
      "entropy": 5.69946255683899,
      "epoch": 7.4197206095791,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012114775315654115,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23841458708047866,
      "num_tokens": 169739252.0,
      "step": 81795
    },
    {
      "entropy": 5.781071329116822,
      "epoch": 7.420174165457184,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001211241448842138,
      "loss": 4.9522,
      "mean_token_accuracy": 0.21924320012331008,
      "num_tokens": 169749746.0,
      "step": 81800
    },
    {
      "entropy": 5.678777980804443,
      "epoch": 7.4206277213352685,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012110053979407828,
      "loss": 4.7269,
      "mean_token_accuracy": 0.2430211454629898,
      "num_tokens": 169759881.0,
      "step": 81805
    },
    {
      "entropy": 5.678756284713745,
      "epoch": 7.421081277213353,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001210769378866229,
      "loss": 4.7816,
      "mean_token_accuracy": 0.24253743290901184,
      "num_tokens": 169769881.0,
      "step": 81810
    },
    {
      "entropy": 5.659210443496704,
      "epoch": 7.421534833091437,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012105333916233569,
      "loss": 4.7634,
      "mean_token_accuracy": 0.23254372626543046,
      "num_tokens": 169780759.0,
      "step": 81815
    },
    {
      "entropy": 5.733157205581665,
      "epoch": 7.421988388969521,
      "grad_norm": 1.5,
      "learning_rate": 0.00012102974362170461,
      "loss": 4.8439,
      "mean_token_accuracy": 0.23638661354780197,
      "num_tokens": 169790747.0,
      "step": 81820
    },
    {
      "entropy": 5.6423609256744385,
      "epoch": 7.4224419448476056,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001210061512652177,
      "loss": 4.7983,
      "mean_token_accuracy": 0.23472455143928528,
      "num_tokens": 169801592.0,
      "step": 81825
    },
    {
      "entropy": 5.630546998977661,
      "epoch": 7.422895500725689,
      "grad_norm": 1.671875,
      "learning_rate": 0.00012098256209336281,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2425663113594055,
      "num_tokens": 169810866.0,
      "step": 81830
    },
    {
      "entropy": 5.669723987579346,
      "epoch": 7.423349056603773,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001209589761066278,
      "loss": 4.762,
      "mean_token_accuracy": 0.23732836544513702,
      "num_tokens": 169821317.0,
      "step": 81835
    },
    {
      "entropy": 5.67207441329956,
      "epoch": 7.4238026124818575,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012093539330550034,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23708195239305496,
      "num_tokens": 169831119.0,
      "step": 81840
    },
    {
      "entropy": 5.721826839447021,
      "epoch": 7.424256168359942,
      "grad_norm": 1.625,
      "learning_rate": 0.00012091181369046826,
      "loss": 4.8017,
      "mean_token_accuracy": 0.24435980170965194,
      "num_tokens": 169840738.0,
      "step": 81845
    },
    {
      "entropy": 5.718198823928833,
      "epoch": 7.424709724238026,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012088823726201917,
      "loss": 4.8613,
      "mean_token_accuracy": 0.23288204669952392,
      "num_tokens": 169851095.0,
      "step": 81850
    },
    {
      "entropy": 5.768139600753784,
      "epoch": 7.42516328011611,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012086466402064057,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22984125912189485,
      "num_tokens": 169861551.0,
      "step": 81855
    },
    {
      "entropy": 5.648501443862915,
      "epoch": 7.4256168359941945,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012084109396682002,
      "loss": 4.778,
      "mean_token_accuracy": 0.23965219110250474,
      "num_tokens": 169871868.0,
      "step": 81860
    },
    {
      "entropy": 5.6156844139099125,
      "epoch": 7.426070391872279,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012081752710104493,
      "loss": 4.7048,
      "mean_token_accuracy": 0.25038735568523407,
      "num_tokens": 169881535.0,
      "step": 81865
    },
    {
      "entropy": 5.685941123962403,
      "epoch": 7.426523947750363,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012079396342380267,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22340679317712783,
      "num_tokens": 169892216.0,
      "step": 81870
    },
    {
      "entropy": 5.75584921836853,
      "epoch": 7.426977503628447,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012077040293558046,
      "loss": 4.8959,
      "mean_token_accuracy": 0.22871956080198289,
      "num_tokens": 169902961.0,
      "step": 81875
    },
    {
      "entropy": 5.6899701118469235,
      "epoch": 7.427431059506532,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001207468456368657,
      "loss": 4.7574,
      "mean_token_accuracy": 0.2358209416270256,
      "num_tokens": 169913406.0,
      "step": 81880
    },
    {
      "entropy": 5.716890954971314,
      "epoch": 7.427884615384615,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001207232915281455,
      "loss": 4.8181,
      "mean_token_accuracy": 0.237673057615757,
      "num_tokens": 169923859.0,
      "step": 81885
    },
    {
      "entropy": 5.706766414642334,
      "epoch": 7.428338171262699,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012069974060990692,
      "loss": 4.7507,
      "mean_token_accuracy": 0.2331315353512764,
      "num_tokens": 169933435.0,
      "step": 81890
    },
    {
      "entropy": 5.711969041824341,
      "epoch": 7.4287917271407835,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012067619288263702,
      "loss": 4.7863,
      "mean_token_accuracy": 0.24251692444086076,
      "num_tokens": 169942631.0,
      "step": 81895
    },
    {
      "entropy": 5.658561944961548,
      "epoch": 7.429245283018868,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012065264834682279,
      "loss": 4.7744,
      "mean_token_accuracy": 0.24193817675113677,
      "num_tokens": 169952699.0,
      "step": 81900
    },
    {
      "entropy": 5.649479007720947,
      "epoch": 7.429698838896952,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012062910700295112,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2375403195619583,
      "num_tokens": 169962210.0,
      "step": 81905
    },
    {
      "entropy": 5.655820608139038,
      "epoch": 7.430152394775036,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012060556885150879,
      "loss": 4.6417,
      "mean_token_accuracy": 0.24866674542427064,
      "num_tokens": 169972062.0,
      "step": 81910
    },
    {
      "entropy": 5.617575597763062,
      "epoch": 7.430605950653121,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012058203389298272,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2371774360537529,
      "num_tokens": 169982023.0,
      "step": 81915
    },
    {
      "entropy": 5.683704423904419,
      "epoch": 7.431059506531205,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012055850212785954,
      "loss": 4.8392,
      "mean_token_accuracy": 0.23351691663265228,
      "num_tokens": 169992217.0,
      "step": 81920
    },
    {
      "entropy": 5.744466400146484,
      "epoch": 7.431513062409289,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012053497355662588,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22975030094385146,
      "num_tokens": 170002887.0,
      "step": 81925
    },
    {
      "entropy": 5.651812505722046,
      "epoch": 7.431966618287373,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012051144817976831,
      "loss": 4.7315,
      "mean_token_accuracy": 0.23835998326539992,
      "num_tokens": 170012947.0,
      "step": 81930
    },
    {
      "entropy": 5.667786359786987,
      "epoch": 7.432420174165458,
      "grad_norm": 1.5,
      "learning_rate": 0.00012048792599777339,
      "loss": 4.7934,
      "mean_token_accuracy": 0.2328647032380104,
      "num_tokens": 170022748.0,
      "step": 81935
    },
    {
      "entropy": 5.73884539604187,
      "epoch": 7.432873730043541,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012046440701112752,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22594269663095473,
      "num_tokens": 170033933.0,
      "step": 81940
    },
    {
      "entropy": 5.701661443710327,
      "epoch": 7.433327285921625,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012044089122031703,
      "loss": 4.7942,
      "mean_token_accuracy": 0.23493710905313492,
      "num_tokens": 170044662.0,
      "step": 81945
    },
    {
      "entropy": 5.696150302886963,
      "epoch": 7.4337808417997095,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012041737862582836,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22179128974676132,
      "num_tokens": 170056607.0,
      "step": 81950
    },
    {
      "entropy": 5.705140113830566,
      "epoch": 7.434234397677794,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001203938692281477,
      "loss": 4.8436,
      "mean_token_accuracy": 0.22922383248806,
      "num_tokens": 170065829.0,
      "step": 81955
    },
    {
      "entropy": 5.717015647888184,
      "epoch": 7.434687953555878,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001203703630277612,
      "loss": 4.8669,
      "mean_token_accuracy": 0.23210162669420242,
      "num_tokens": 170076710.0,
      "step": 81960
    },
    {
      "entropy": 5.7242179870605465,
      "epoch": 7.435141509433962,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012034686002515504,
      "loss": 4.8442,
      "mean_token_accuracy": 0.23385944068431855,
      "num_tokens": 170086113.0,
      "step": 81965
    },
    {
      "entropy": 5.660913705825806,
      "epoch": 7.435595065312047,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001203233602208152,
      "loss": 4.7379,
      "mean_token_accuracy": 0.2398462250828743,
      "num_tokens": 170096655.0,
      "step": 81970
    },
    {
      "entropy": 5.741151523590088,
      "epoch": 7.436048621190131,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012029986361522769,
      "loss": 4.8812,
      "mean_token_accuracy": 0.23609177470207215,
      "num_tokens": 170106450.0,
      "step": 81975
    },
    {
      "entropy": 5.655555963516235,
      "epoch": 7.436502177068215,
      "grad_norm": 1.703125,
      "learning_rate": 0.00012027637020887834,
      "loss": 4.759,
      "mean_token_accuracy": 0.23218743354082108,
      "num_tokens": 170118171.0,
      "step": 81980
    },
    {
      "entropy": 5.670216274261475,
      "epoch": 7.4369557329462985,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012025288000225318,
      "loss": 4.789,
      "mean_token_accuracy": 0.24277092516422272,
      "num_tokens": 170128551.0,
      "step": 81985
    },
    {
      "entropy": 5.695877599716186,
      "epoch": 7.437409288824383,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012022939299583786,
      "loss": 4.7942,
      "mean_token_accuracy": 0.23245633244514466,
      "num_tokens": 170138806.0,
      "step": 81990
    },
    {
      "entropy": 5.675938606262207,
      "epoch": 7.437862844702467,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012020590919011818,
      "loss": 4.8208,
      "mean_token_accuracy": 0.2274032488465309,
      "num_tokens": 170149116.0,
      "step": 81995
    },
    {
      "entropy": 5.768739414215088,
      "epoch": 7.438316400580551,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001201824285855797,
      "loss": 4.8504,
      "mean_token_accuracy": 0.23101221770048141,
      "num_tokens": 170159290.0,
      "step": 82000
    },
    {
      "entropy": 5.771636724472046,
      "epoch": 7.438769956458636,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012015895118270807,
      "loss": 4.934,
      "mean_token_accuracy": 0.22797711342573165,
      "num_tokens": 170170682.0,
      "step": 82005
    },
    {
      "entropy": 5.7337242603302006,
      "epoch": 7.43922351233672,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012013547698198871,
      "loss": 4.8606,
      "mean_token_accuracy": 0.231476029753685,
      "num_tokens": 170182231.0,
      "step": 82010
    },
    {
      "entropy": 5.638474369049073,
      "epoch": 7.439677068214804,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012011200598390722,
      "loss": 4.7831,
      "mean_token_accuracy": 0.23846717774868012,
      "num_tokens": 170191932.0,
      "step": 82015
    },
    {
      "entropy": 5.720299148559571,
      "epoch": 7.440130624092888,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012008853818894892,
      "loss": 4.8029,
      "mean_token_accuracy": 0.22829538732767105,
      "num_tokens": 170202289.0,
      "step": 82020
    },
    {
      "entropy": 5.670910358428955,
      "epoch": 7.440584179970973,
      "grad_norm": 1.5,
      "learning_rate": 0.00012006507359759912,
      "loss": 4.7412,
      "mean_token_accuracy": 0.2395218774676323,
      "num_tokens": 170212127.0,
      "step": 82025
    },
    {
      "entropy": 5.798388242721558,
      "epoch": 7.441037735849057,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012004161221034317,
      "loss": 4.9227,
      "mean_token_accuracy": 0.23034819811582566,
      "num_tokens": 170223113.0,
      "step": 82030
    },
    {
      "entropy": 5.739327764511108,
      "epoch": 7.441491291727141,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012001815402766607,
      "loss": 4.8744,
      "mean_token_accuracy": 0.23561985939741134,
      "num_tokens": 170233446.0,
      "step": 82035
    },
    {
      "entropy": 5.676852512359619,
      "epoch": 7.441944847605225,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000119994699050053,
      "loss": 4.7425,
      "mean_token_accuracy": 0.24358588308095933,
      "num_tokens": 170243573.0,
      "step": 82040
    },
    {
      "entropy": 5.689122915267944,
      "epoch": 7.442398403483309,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011997124727798911,
      "loss": 4.7742,
      "mean_token_accuracy": 0.23944949358701706,
      "num_tokens": 170253974.0,
      "step": 82045
    },
    {
      "entropy": 5.74721360206604,
      "epoch": 7.442851959361393,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001199477987119593,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2280567020177841,
      "num_tokens": 170264178.0,
      "step": 82050
    },
    {
      "entropy": 5.667952632904052,
      "epoch": 7.443305515239477,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011992435335244855,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2356905683875084,
      "num_tokens": 170274810.0,
      "step": 82055
    },
    {
      "entropy": 5.669204521179199,
      "epoch": 7.443759071117562,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001199009111999417,
      "loss": 4.7951,
      "mean_token_accuracy": 0.2340357705950737,
      "num_tokens": 170284839.0,
      "step": 82060
    },
    {
      "entropy": 5.666364431381226,
      "epoch": 7.444212626995646,
      "grad_norm": 1.5,
      "learning_rate": 0.00011987747225492354,
      "loss": 4.7728,
      "mean_token_accuracy": 0.23561263233423232,
      "num_tokens": 170295893.0,
      "step": 82065
    },
    {
      "entropy": 5.6509768009185795,
      "epoch": 7.44466618287373,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011985403651787876,
      "loss": 4.7947,
      "mean_token_accuracy": 0.23079959899187089,
      "num_tokens": 170306411.0,
      "step": 82070
    },
    {
      "entropy": 5.712678384780884,
      "epoch": 7.445119738751814,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011983060398929198,
      "loss": 4.8002,
      "mean_token_accuracy": 0.24504948258399964,
      "num_tokens": 170316995.0,
      "step": 82075
    },
    {
      "entropy": 5.735393810272217,
      "epoch": 7.445573294629899,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00011980717466964791,
      "loss": 4.83,
      "mean_token_accuracy": 0.23321364670991898,
      "num_tokens": 170326092.0,
      "step": 82080
    },
    {
      "entropy": 5.682204818725586,
      "epoch": 7.446026850507983,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011978374855943102,
      "loss": 4.7911,
      "mean_token_accuracy": 0.22644547373056412,
      "num_tokens": 170336213.0,
      "step": 82085
    },
    {
      "entropy": 5.6828125,
      "epoch": 7.446480406386067,
      "grad_norm": 1.625,
      "learning_rate": 0.00011976032565912577,
      "loss": 4.7867,
      "mean_token_accuracy": 0.23514269590377807,
      "num_tokens": 170346635.0,
      "step": 82090
    },
    {
      "entropy": 5.670386171340942,
      "epoch": 7.446933962264151,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011973690596921655,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23955421149730682,
      "num_tokens": 170356704.0,
      "step": 82095
    },
    {
      "entropy": 5.696506690979004,
      "epoch": 7.447387518142235,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011971348949018769,
      "loss": 4.7871,
      "mean_token_accuracy": 0.23889367282390594,
      "num_tokens": 170368107.0,
      "step": 82100
    },
    {
      "entropy": 5.706119155883789,
      "epoch": 7.447841074020319,
      "grad_norm": 1.734375,
      "learning_rate": 0.00011969007622252342,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23483374416828157,
      "num_tokens": 170378286.0,
      "step": 82105
    },
    {
      "entropy": 5.641286325454712,
      "epoch": 7.448294629898403,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011966666616670789,
      "loss": 4.7459,
      "mean_token_accuracy": 0.2433878153562546,
      "num_tokens": 170388765.0,
      "step": 82110
    },
    {
      "entropy": 5.582980632781982,
      "epoch": 7.448748185776488,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011964325932322533,
      "loss": 4.7363,
      "mean_token_accuracy": 0.23619820326566696,
      "num_tokens": 170398180.0,
      "step": 82115
    },
    {
      "entropy": 5.705201196670532,
      "epoch": 7.449201741654572,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001196198556925598,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23139962702989578,
      "num_tokens": 170408984.0,
      "step": 82120
    },
    {
      "entropy": 5.761727285385132,
      "epoch": 7.449655297532656,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011959645527519524,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2239574208855629,
      "num_tokens": 170418891.0,
      "step": 82125
    },
    {
      "entropy": 5.723606204986572,
      "epoch": 7.4501088534107405,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011957305807161555,
      "loss": 4.857,
      "mean_token_accuracy": 0.23257341980934143,
      "num_tokens": 170428544.0,
      "step": 82130
    },
    {
      "entropy": 5.721006059646607,
      "epoch": 7.450562409288825,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011954966408230468,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23592900335788727,
      "num_tokens": 170438905.0,
      "step": 82135
    },
    {
      "entropy": 5.647617959976197,
      "epoch": 7.451015965166908,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011952627330774626,
      "loss": 4.7441,
      "mean_token_accuracy": 0.23763959109783173,
      "num_tokens": 170449100.0,
      "step": 82140
    },
    {
      "entropy": 5.685591459274292,
      "epoch": 7.451469521044992,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001195028857484242,
      "loss": 4.7873,
      "mean_token_accuracy": 0.24084055721759795,
      "num_tokens": 170459464.0,
      "step": 82145
    },
    {
      "entropy": 5.6593505382537845,
      "epoch": 7.451923076923077,
      "grad_norm": 1.5,
      "learning_rate": 0.00011947950140482211,
      "loss": 4.779,
      "mean_token_accuracy": 0.23609941750764846,
      "num_tokens": 170469419.0,
      "step": 82150
    },
    {
      "entropy": 5.70257339477539,
      "epoch": 7.452376632801161,
      "grad_norm": 1.640625,
      "learning_rate": 0.00011945612027742355,
      "loss": 4.8171,
      "mean_token_accuracy": 0.23577926456928253,
      "num_tokens": 170479630.0,
      "step": 82155
    },
    {
      "entropy": 5.6146985530853275,
      "epoch": 7.452830188679245,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011943274236671207,
      "loss": 4.6984,
      "mean_token_accuracy": 0.24211020469665528,
      "num_tokens": 170490835.0,
      "step": 82160
    },
    {
      "entropy": 5.607745885848999,
      "epoch": 7.4532837445573294,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011940936767317109,
      "loss": 4.69,
      "mean_token_accuracy": 0.2514627411961555,
      "num_tokens": 170502166.0,
      "step": 82165
    },
    {
      "entropy": 5.763886404037476,
      "epoch": 7.453737300435414,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011938599619728406,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2325771152973175,
      "num_tokens": 170511955.0,
      "step": 82170
    },
    {
      "entropy": 5.675064420700073,
      "epoch": 7.454190856313498,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011936262793953422,
      "loss": 4.7447,
      "mean_token_accuracy": 0.2386399656534195,
      "num_tokens": 170522384.0,
      "step": 82175
    },
    {
      "entropy": 5.6078773021698,
      "epoch": 7.454644412191582,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011933926290040495,
      "loss": 4.6344,
      "mean_token_accuracy": 0.2504678308963776,
      "num_tokens": 170532440.0,
      "step": 82180
    },
    {
      "entropy": 5.651521968841553,
      "epoch": 7.4550979680696665,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001193159010803794,
      "loss": 4.7528,
      "mean_token_accuracy": 0.23234965801239013,
      "num_tokens": 170542791.0,
      "step": 82185
    },
    {
      "entropy": 5.72489743232727,
      "epoch": 7.455551523947751,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011929254247994068,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2374740794301033,
      "num_tokens": 170553979.0,
      "step": 82190
    },
    {
      "entropy": 5.649919319152832,
      "epoch": 7.456005079825834,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011926918709957188,
      "loss": 4.8234,
      "mean_token_accuracy": 0.2354747846722603,
      "num_tokens": 170564982.0,
      "step": 82195
    },
    {
      "entropy": 5.649632167816162,
      "epoch": 7.456458635703918,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011924583493975594,
      "loss": 4.7313,
      "mean_token_accuracy": 0.23948212713003159,
      "num_tokens": 170575194.0,
      "step": 82200
    },
    {
      "entropy": 5.6439732074737545,
      "epoch": 7.456912191582003,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011922248600097585,
      "loss": 4.782,
      "mean_token_accuracy": 0.2400775283575058,
      "num_tokens": 170585454.0,
      "step": 82205
    },
    {
      "entropy": 5.733838415145874,
      "epoch": 7.457365747460087,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011919914028371438,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2334489643573761,
      "num_tokens": 170596298.0,
      "step": 82210
    },
    {
      "entropy": 5.6646030902862545,
      "epoch": 7.457819303338171,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011917579778845445,
      "loss": 4.7796,
      "mean_token_accuracy": 0.23307473808526993,
      "num_tokens": 170605546.0,
      "step": 82215
    },
    {
      "entropy": 5.656720495223999,
      "epoch": 7.4582728592162555,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011915245851567873,
      "loss": 4.7299,
      "mean_token_accuracy": 0.23849676549434662,
      "num_tokens": 170615672.0,
      "step": 82220
    },
    {
      "entropy": 5.660548257827759,
      "epoch": 7.45872641509434,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011912912246586989,
      "loss": 4.7494,
      "mean_token_accuracy": 0.23827740103006362,
      "num_tokens": 170625149.0,
      "step": 82225
    },
    {
      "entropy": 5.700597858428955,
      "epoch": 7.459179970972424,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011910578963951052,
      "loss": 4.8206,
      "mean_token_accuracy": 0.23222015500068666,
      "num_tokens": 170635608.0,
      "step": 82230
    },
    {
      "entropy": 5.632126426696777,
      "epoch": 7.459633526850508,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011908246003708314,
      "loss": 4.7193,
      "mean_token_accuracy": 0.24446091502904893,
      "num_tokens": 170645701.0,
      "step": 82235
    },
    {
      "entropy": 5.672017812728882,
      "epoch": 7.4600870827285926,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011905913365907023,
      "loss": 4.8252,
      "mean_token_accuracy": 0.23472318053245544,
      "num_tokens": 170656243.0,
      "step": 82240
    },
    {
      "entropy": 5.668290519714356,
      "epoch": 7.460540638606676,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011903581050595407,
      "loss": 4.7671,
      "mean_token_accuracy": 0.24076107740402222,
      "num_tokens": 170666435.0,
      "step": 82245
    },
    {
      "entropy": 5.67756028175354,
      "epoch": 7.46099419448476,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011901249057821721,
      "loss": 4.7727,
      "mean_token_accuracy": 0.2380557283759117,
      "num_tokens": 170676886.0,
      "step": 82250
    },
    {
      "entropy": 5.692611598968506,
      "epoch": 7.4614477503628445,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011898917387634176,
      "loss": 4.7871,
      "mean_token_accuracy": 0.2360011264681816,
      "num_tokens": 170687252.0,
      "step": 82255
    },
    {
      "entropy": 5.656498146057129,
      "epoch": 7.461901306240929,
      "grad_norm": 1.625,
      "learning_rate": 0.00011896586040080998,
      "loss": 4.6922,
      "mean_token_accuracy": 0.24128201156854628,
      "num_tokens": 170697407.0,
      "step": 82260
    },
    {
      "entropy": 5.638920021057129,
      "epoch": 7.462354862119013,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011894255015210395,
      "loss": 4.769,
      "mean_token_accuracy": 0.2387396439909935,
      "num_tokens": 170707457.0,
      "step": 82265
    },
    {
      "entropy": 5.714733982086182,
      "epoch": 7.462808417997097,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00011891924313070573,
      "loss": 4.8987,
      "mean_token_accuracy": 0.23683644384145736,
      "num_tokens": 170717743.0,
      "step": 82270
    },
    {
      "entropy": 5.748761081695557,
      "epoch": 7.4632619738751815,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011889593933709737,
      "loss": 4.8756,
      "mean_token_accuracy": 0.23553256690502167,
      "num_tokens": 170727438.0,
      "step": 82275
    },
    {
      "entropy": 5.66643271446228,
      "epoch": 7.463715529753266,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001188726387717607,
      "loss": 4.7555,
      "mean_token_accuracy": 0.2459407702088356,
      "num_tokens": 170738166.0,
      "step": 82280
    },
    {
      "entropy": 5.72619309425354,
      "epoch": 7.46416908563135,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001188493414351777,
      "loss": 4.871,
      "mean_token_accuracy": 0.2320113107562065,
      "num_tokens": 170748551.0,
      "step": 82285
    },
    {
      "entropy": 5.695669555664063,
      "epoch": 7.464622641509434,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001188260473278301,
      "loss": 4.8196,
      "mean_token_accuracy": 0.23379953652620317,
      "num_tokens": 170759969.0,
      "step": 82290
    },
    {
      "entropy": 5.714319658279419,
      "epoch": 7.465076197387518,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011880275645019966,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2300044685602188,
      "num_tokens": 170771003.0,
      "step": 82295
    },
    {
      "entropy": 5.679366636276245,
      "epoch": 7.465529753265602,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000118779468802768,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24131496399641036,
      "num_tokens": 170780258.0,
      "step": 82300
    },
    {
      "entropy": 5.71941990852356,
      "epoch": 7.465983309143686,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011875618438601676,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22941233366727828,
      "num_tokens": 170790846.0,
      "step": 82305
    },
    {
      "entropy": 5.6187139511108395,
      "epoch": 7.4664368650217705,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011873290320042735,
      "loss": 4.6818,
      "mean_token_accuracy": 0.24442839622497559,
      "num_tokens": 170801651.0,
      "step": 82310
    },
    {
      "entropy": 5.70516357421875,
      "epoch": 7.466890420899855,
      "grad_norm": 1.5,
      "learning_rate": 0.00011870962524648141,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23608005344867705,
      "num_tokens": 170812113.0,
      "step": 82315
    },
    {
      "entropy": 5.650691270828247,
      "epoch": 7.467343976777939,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011868635052466025,
      "loss": 4.7976,
      "mean_token_accuracy": 0.23605429381132126,
      "num_tokens": 170823093.0,
      "step": 82320
    },
    {
      "entropy": 5.647961521148682,
      "epoch": 7.467797532656023,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011866307903544515,
      "loss": 4.7935,
      "mean_token_accuracy": 0.24173090904951094,
      "num_tokens": 170832691.0,
      "step": 82325
    },
    {
      "entropy": 5.600640058517456,
      "epoch": 7.468251088534108,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011863981077931748,
      "loss": 4.7322,
      "mean_token_accuracy": 0.24672621935606004,
      "num_tokens": 170842481.0,
      "step": 82330
    },
    {
      "entropy": 5.726885747909546,
      "epoch": 7.468704644412192,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011861654575675831,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22949544042348863,
      "num_tokens": 170852975.0,
      "step": 82335
    },
    {
      "entropy": 5.720180988311768,
      "epoch": 7.469158200290276,
      "grad_norm": 1.703125,
      "learning_rate": 0.00011859328396824888,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2328108459711075,
      "num_tokens": 170864166.0,
      "step": 82340
    },
    {
      "entropy": 5.710850572586059,
      "epoch": 7.46961175616836,
      "grad_norm": 1.5,
      "learning_rate": 0.0001185700254142701,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2286963313817978,
      "num_tokens": 170874376.0,
      "step": 82345
    },
    {
      "entropy": 5.635263681411743,
      "epoch": 7.470065312046444,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011854677009530309,
      "loss": 4.7402,
      "mean_token_accuracy": 0.23818876594305038,
      "num_tokens": 170884956.0,
      "step": 82350
    },
    {
      "entropy": 5.7250084400177,
      "epoch": 7.470518867924528,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011852351801182882,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23614016622304917,
      "num_tokens": 170895940.0,
      "step": 82355
    },
    {
      "entropy": 5.717491483688354,
      "epoch": 7.470972423802612,
      "grad_norm": 1.515625,
      "learning_rate": 0.000118500269164328,
      "loss": 4.8356,
      "mean_token_accuracy": 0.23697734773159027,
      "num_tokens": 170906869.0,
      "step": 82360
    },
    {
      "entropy": 5.6924537181854244,
      "epoch": 7.4714259796806965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011847702355328154,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23249533474445344,
      "num_tokens": 170917757.0,
      "step": 82365
    },
    {
      "entropy": 5.653545141220093,
      "epoch": 7.471879535558781,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001184537811791701,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23294393718242645,
      "num_tokens": 170927773.0,
      "step": 82370
    },
    {
      "entropy": 5.696524667739868,
      "epoch": 7.472333091436865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011843054204247437,
      "loss": 4.839,
      "mean_token_accuracy": 0.23542846441268922,
      "num_tokens": 170939051.0,
      "step": 82375
    },
    {
      "entropy": 5.711836624145508,
      "epoch": 7.472786647314949,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011840730614367486,
      "loss": 4.791,
      "mean_token_accuracy": 0.24107400476932525,
      "num_tokens": 170949791.0,
      "step": 82380
    },
    {
      "entropy": 5.7587298393249515,
      "epoch": 7.473240203193034,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011838407348325222,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2273171067237854,
      "num_tokens": 170959657.0,
      "step": 82385
    },
    {
      "entropy": 5.666163921356201,
      "epoch": 7.473693759071118,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011836084406168685,
      "loss": 4.7612,
      "mean_token_accuracy": 0.23900034576654433,
      "num_tokens": 170969709.0,
      "step": 82390
    },
    {
      "entropy": 5.694407844543457,
      "epoch": 7.474147314949202,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011833761787945915,
      "loss": 4.8452,
      "mean_token_accuracy": 0.22730033546686174,
      "num_tokens": 170980849.0,
      "step": 82395
    },
    {
      "entropy": 5.73897271156311,
      "epoch": 7.4746008708272855,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011831439493704945,
      "loss": 4.85,
      "mean_token_accuracy": 0.23810755014419555,
      "num_tokens": 170991055.0,
      "step": 82400
    },
    {
      "entropy": 5.757911825180054,
      "epoch": 7.47505442670537,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011829117523493795,
      "loss": 4.9063,
      "mean_token_accuracy": 0.23309336751699447,
      "num_tokens": 171001967.0,
      "step": 82405
    },
    {
      "entropy": 5.745773267745972,
      "epoch": 7.475507982583454,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011826795877360491,
      "loss": 4.8693,
      "mean_token_accuracy": 0.2329438090324402,
      "num_tokens": 171011802.0,
      "step": 82410
    },
    {
      "entropy": 5.748556995391846,
      "epoch": 7.475961538461538,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011824474555353033,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2296000599861145,
      "num_tokens": 171023498.0,
      "step": 82415
    },
    {
      "entropy": 5.707155084609985,
      "epoch": 7.476415094339623,
      "grad_norm": 1.5,
      "learning_rate": 0.00011822153557519447,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23702149838209152,
      "num_tokens": 171033044.0,
      "step": 82420
    },
    {
      "entropy": 5.73694167137146,
      "epoch": 7.476868650217707,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011819832883907717,
      "loss": 4.791,
      "mean_token_accuracy": 0.23517031818628312,
      "num_tokens": 171043909.0,
      "step": 82425
    },
    {
      "entropy": 5.707014656066894,
      "epoch": 7.477322206095791,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001181751253456584,
      "loss": 4.7465,
      "mean_token_accuracy": 0.23830821067094804,
      "num_tokens": 171054298.0,
      "step": 82430
    },
    {
      "entropy": 5.7504034519195555,
      "epoch": 7.477775761973875,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000118151925095418,
      "loss": 4.8642,
      "mean_token_accuracy": 0.23310166746377944,
      "num_tokens": 171063754.0,
      "step": 82435
    },
    {
      "entropy": 5.606510400772095,
      "epoch": 7.47822931785196,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011812872808883577,
      "loss": 4.651,
      "mean_token_accuracy": 0.2471102550625801,
      "num_tokens": 171074725.0,
      "step": 82440
    },
    {
      "entropy": 5.680151700973511,
      "epoch": 7.478682873730044,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011810553432639138,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23631786406040192,
      "num_tokens": 171085139.0,
      "step": 82445
    },
    {
      "entropy": 5.652435159683227,
      "epoch": 7.479136429608127,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001180823438085645,
      "loss": 4.745,
      "mean_token_accuracy": 0.24600565135478974,
      "num_tokens": 171096275.0,
      "step": 82450
    },
    {
      "entropy": 5.647460031509399,
      "epoch": 7.479589985486212,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011805915653583485,
      "loss": 4.7315,
      "mean_token_accuracy": 0.24837521314620972,
      "num_tokens": 171107061.0,
      "step": 82455
    },
    {
      "entropy": 5.690073823928833,
      "epoch": 7.480043541364296,
      "grad_norm": 1.5,
      "learning_rate": 0.00011803597250868176,
      "loss": 4.8616,
      "mean_token_accuracy": 0.23060386180877684,
      "num_tokens": 171117913.0,
      "step": 82460
    },
    {
      "entropy": 5.62269172668457,
      "epoch": 7.48049709724238,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011801279172758479,
      "loss": 4.7384,
      "mean_token_accuracy": 0.2456885412335396,
      "num_tokens": 171128410.0,
      "step": 82465
    },
    {
      "entropy": 5.700246858596802,
      "epoch": 7.480950653120464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011798961419302326,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2358858272433281,
      "num_tokens": 171139838.0,
      "step": 82470
    },
    {
      "entropy": 5.6877968311309814,
      "epoch": 7.481404208998549,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011796643990547647,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23691661208868026,
      "num_tokens": 171151155.0,
      "step": 82475
    },
    {
      "entropy": 5.661692810058594,
      "epoch": 7.481857764876633,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011794326886542377,
      "loss": 4.8117,
      "mean_token_accuracy": 0.23564085811376573,
      "num_tokens": 171162448.0,
      "step": 82480
    },
    {
      "entropy": 5.704004621505737,
      "epoch": 7.482311320754717,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011792010107334432,
      "loss": 4.8357,
      "mean_token_accuracy": 0.2267945572733879,
      "num_tokens": 171172587.0,
      "step": 82485
    },
    {
      "entropy": 5.683737754821777,
      "epoch": 7.482764876632801,
      "grad_norm": 1.5,
      "learning_rate": 0.00011789693652971719,
      "loss": 4.799,
      "mean_token_accuracy": 0.2334206685423851,
      "num_tokens": 171182904.0,
      "step": 82490
    },
    {
      "entropy": 5.711511898040771,
      "epoch": 7.483218432510886,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011787377523502143,
      "loss": 4.8446,
      "mean_token_accuracy": 0.22781636863946914,
      "num_tokens": 171193432.0,
      "step": 82495
    },
    {
      "entropy": 5.794775199890137,
      "epoch": 7.48367198838897,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011785061718973608,
      "loss": 4.9061,
      "mean_token_accuracy": 0.23495739549398423,
      "num_tokens": 171203037.0,
      "step": 82500
    },
    {
      "entropy": 5.698377370834351,
      "epoch": 7.484125544267053,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011782746239433998,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23448212146759034,
      "num_tokens": 171212665.0,
      "step": 82505
    },
    {
      "entropy": 5.674925374984741,
      "epoch": 7.484579100145138,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011780431084931195,
      "loss": 4.8157,
      "mean_token_accuracy": 0.23146655261516572,
      "num_tokens": 171223222.0,
      "step": 82510
    },
    {
      "entropy": 5.770322465896607,
      "epoch": 7.485032656023222,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011778116255513091,
      "loss": 4.923,
      "mean_token_accuracy": 0.2173786774277687,
      "num_tokens": 171233894.0,
      "step": 82515
    },
    {
      "entropy": 5.726304197311402,
      "epoch": 7.485486211901306,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011775801751227549,
      "loss": 4.854,
      "mean_token_accuracy": 0.23120349943637847,
      "num_tokens": 171245144.0,
      "step": 82520
    },
    {
      "entropy": 5.630112886428833,
      "epoch": 7.48593976777939,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001177348757212243,
      "loss": 4.752,
      "mean_token_accuracy": 0.2492365285754204,
      "num_tokens": 171254743.0,
      "step": 82525
    },
    {
      "entropy": 5.71283597946167,
      "epoch": 7.486393323657475,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011771173718245601,
      "loss": 4.8024,
      "mean_token_accuracy": 0.2299635663628578,
      "num_tokens": 171265042.0,
      "step": 82530
    },
    {
      "entropy": 5.686265659332276,
      "epoch": 7.486846879535559,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011768860189644907,
      "loss": 4.7443,
      "mean_token_accuracy": 0.24482636600732804,
      "num_tokens": 171275095.0,
      "step": 82535
    },
    {
      "entropy": 5.642235040664673,
      "epoch": 7.487300435413643,
      "grad_norm": 1.5,
      "learning_rate": 0.00011766546986368189,
      "loss": 4.7773,
      "mean_token_accuracy": 0.23832616209983826,
      "num_tokens": 171286786.0,
      "step": 82540
    },
    {
      "entropy": 5.65437798500061,
      "epoch": 7.4877539912917275,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011764234108463284,
      "loss": 4.7379,
      "mean_token_accuracy": 0.23889190107584,
      "num_tokens": 171297102.0,
      "step": 82545
    },
    {
      "entropy": 5.685315370559692,
      "epoch": 7.488207547169811,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011761921555978035,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22858558893203734,
      "num_tokens": 171307466.0,
      "step": 82550
    },
    {
      "entropy": 5.701445484161377,
      "epoch": 7.488661103047895,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001175960932896026,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23973275870084762,
      "num_tokens": 171319204.0,
      "step": 82555
    },
    {
      "entropy": 5.754260444641114,
      "epoch": 7.489114658925979,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00011757297427457774,
      "loss": 4.861,
      "mean_token_accuracy": 0.23172779232263566,
      "num_tokens": 171329149.0,
      "step": 82560
    },
    {
      "entropy": 5.6954244613647464,
      "epoch": 7.489568214804064,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011754985851518388,
      "loss": 4.7738,
      "mean_token_accuracy": 0.24125525802373887,
      "num_tokens": 171339265.0,
      "step": 82565
    },
    {
      "entropy": 5.662315797805786,
      "epoch": 7.490021770682148,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011752674601189906,
      "loss": 4.7524,
      "mean_token_accuracy": 0.24108128994703293,
      "num_tokens": 171349325.0,
      "step": 82570
    },
    {
      "entropy": 5.722105169296265,
      "epoch": 7.490475326560232,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011750363676520128,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22704992443323135,
      "num_tokens": 171360236.0,
      "step": 82575
    },
    {
      "entropy": 5.71274094581604,
      "epoch": 7.4909288824383164,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011748053077556836,
      "loss": 4.8326,
      "mean_token_accuracy": 0.22893455177545546,
      "num_tokens": 171371793.0,
      "step": 82580
    },
    {
      "entropy": 5.734305095672608,
      "epoch": 7.491382438316401,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011745742804347826,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22662105411291122,
      "num_tokens": 171382042.0,
      "step": 82585
    },
    {
      "entropy": 5.6716978549957275,
      "epoch": 7.491835994194485,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011743432856940871,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23404400944709777,
      "num_tokens": 171391685.0,
      "step": 82590
    },
    {
      "entropy": 5.714800691604614,
      "epoch": 7.492289550072569,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011741123235383741,
      "loss": 4.8072,
      "mean_token_accuracy": 0.2351088836789131,
      "num_tokens": 171401372.0,
      "step": 82595
    },
    {
      "entropy": 5.686866712570191,
      "epoch": 7.4927431059506535,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00011738813939724195,
      "loss": 4.7785,
      "mean_token_accuracy": 0.23754211664199829,
      "num_tokens": 171412638.0,
      "step": 82600
    },
    {
      "entropy": 5.661996126174927,
      "epoch": 7.493196661828737,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011736504970009994,
      "loss": 4.7507,
      "mean_token_accuracy": 0.23842352330684663,
      "num_tokens": 171422145.0,
      "step": 82605
    },
    {
      "entropy": 5.7124875545501705,
      "epoch": 7.493650217706821,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011734196326288887,
      "loss": 4.8075,
      "mean_token_accuracy": 0.22955403625965118,
      "num_tokens": 171432825.0,
      "step": 82610
    },
    {
      "entropy": 5.719291019439697,
      "epoch": 7.494103773584905,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011731888008608612,
      "loss": 4.8333,
      "mean_token_accuracy": 0.23119871765375138,
      "num_tokens": 171442579.0,
      "step": 82615
    },
    {
      "entropy": 5.67109260559082,
      "epoch": 7.49455732946299,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011729580017016917,
      "loss": 4.7624,
      "mean_token_accuracy": 0.24629207104444503,
      "num_tokens": 171452899.0,
      "step": 82620
    },
    {
      "entropy": 5.683236837387085,
      "epoch": 7.495010885341074,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011727272351561527,
      "loss": 4.7489,
      "mean_token_accuracy": 0.24504191130399705,
      "num_tokens": 171462382.0,
      "step": 82625
    },
    {
      "entropy": 5.718303251266479,
      "epoch": 7.495464441219158,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011724965012290163,
      "loss": 4.8557,
      "mean_token_accuracy": 0.24014696776866912,
      "num_tokens": 171472192.0,
      "step": 82630
    },
    {
      "entropy": 5.668425703048706,
      "epoch": 7.4959179970972425,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011722657999250544,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2386636808514595,
      "num_tokens": 171482662.0,
      "step": 82635
    },
    {
      "entropy": 5.666841650009156,
      "epoch": 7.496371552975327,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011720351312490377,
      "loss": 4.8592,
      "mean_token_accuracy": 0.23107832074165344,
      "num_tokens": 171492486.0,
      "step": 82640
    },
    {
      "entropy": 5.736710262298584,
      "epoch": 7.496825108853411,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011718044952057358,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2400682345032692,
      "num_tokens": 171501962.0,
      "step": 82645
    },
    {
      "entropy": 5.677612257003784,
      "epoch": 7.497278664731495,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000117157389179992,
      "loss": 4.7462,
      "mean_token_accuracy": 0.23605752140283584,
      "num_tokens": 171511572.0,
      "step": 82650
    },
    {
      "entropy": 5.724182271957398,
      "epoch": 7.49773222060958,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011713433210363583,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2289244756102562,
      "num_tokens": 171521468.0,
      "step": 82655
    },
    {
      "entropy": 5.651597881317139,
      "epoch": 7.498185776487663,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001171112782919819,
      "loss": 4.7484,
      "mean_token_accuracy": 0.2435622215270996,
      "num_tokens": 171531212.0,
      "step": 82660
    },
    {
      "entropy": 5.673616600036621,
      "epoch": 7.498639332365747,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011708822774550697,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2345760852098465,
      "num_tokens": 171541694.0,
      "step": 82665
    },
    {
      "entropy": 5.704024934768677,
      "epoch": 7.4990928882438315,
      "grad_norm": 1.625,
      "learning_rate": 0.00011706518046468774,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22791072130203247,
      "num_tokens": 171552287.0,
      "step": 82670
    },
    {
      "entropy": 5.61456036567688,
      "epoch": 7.499546444121916,
      "grad_norm": 1.6875,
      "learning_rate": 0.0001170421364500008,
      "loss": 4.6461,
      "mean_token_accuracy": 0.2508475616574287,
      "num_tokens": 171562134.0,
      "step": 82675
    },
    {
      "entropy": 5.696462154388428,
      "epoch": 7.5,
      "grad_norm": 1.5,
      "learning_rate": 0.0001170190957019227,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23282849937677383,
      "num_tokens": 171573995.0,
      "step": 82680
    },
    {
      "entropy": 5.752775192260742,
      "epoch": 7.500453555878084,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011699605822092998,
      "loss": 4.8036,
      "mean_token_accuracy": 0.23951748609542847,
      "num_tokens": 171584136.0,
      "step": 82685
    },
    {
      "entropy": 5.676340961456299,
      "epoch": 7.5009071117561685,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011697302400749908,
      "loss": 4.7783,
      "mean_token_accuracy": 0.24016357809305192,
      "num_tokens": 171593123.0,
      "step": 82690
    },
    {
      "entropy": 5.736552715301514,
      "epoch": 7.501360667634253,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001169499930621063,
      "loss": 4.8623,
      "mean_token_accuracy": 0.23401568084955215,
      "num_tokens": 171604460.0,
      "step": 82695
    },
    {
      "entropy": 5.677992057800293,
      "epoch": 7.501814223512337,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00011692696538522795,
      "loss": 4.8315,
      "mean_token_accuracy": 0.23825973719358445,
      "num_tokens": 171614687.0,
      "step": 82700
    },
    {
      "entropy": 5.712870407104492,
      "epoch": 7.50226777939042,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011690394097734022,
      "loss": 4.805,
      "mean_token_accuracy": 0.23744404464960098,
      "num_tokens": 171625206.0,
      "step": 82705
    },
    {
      "entropy": 5.723920488357544,
      "epoch": 7.502721335268505,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011688091983891929,
      "loss": 4.8936,
      "mean_token_accuracy": 0.23181127309799193,
      "num_tokens": 171635323.0,
      "step": 82710
    },
    {
      "entropy": 5.678724050521851,
      "epoch": 7.503174891146589,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011685790197044114,
      "loss": 4.7958,
      "mean_token_accuracy": 0.23783492147922516,
      "num_tokens": 171645237.0,
      "step": 82715
    },
    {
      "entropy": 5.667114639282227,
      "epoch": 7.503628447024673,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011683488737238197,
      "loss": 4.7761,
      "mean_token_accuracy": 0.22791056782007219,
      "num_tokens": 171656052.0,
      "step": 82720
    },
    {
      "entropy": 5.724498224258423,
      "epoch": 7.5040820029027575,
      "grad_norm": 1.71875,
      "learning_rate": 0.00011681187604521761,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2352926641702652,
      "num_tokens": 171666780.0,
      "step": 82725
    },
    {
      "entropy": 5.703934240341186,
      "epoch": 7.504535558780842,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00011678886798942398,
      "loss": 4.7894,
      "mean_token_accuracy": 0.24519841223955155,
      "num_tokens": 171676607.0,
      "step": 82730
    },
    {
      "entropy": 5.641823148727417,
      "epoch": 7.504989114658926,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001167658632054769,
      "loss": 4.7727,
      "mean_token_accuracy": 0.2426133334636688,
      "num_tokens": 171686870.0,
      "step": 82735
    },
    {
      "entropy": 5.637161350250244,
      "epoch": 7.50544267053701,
      "grad_norm": 1.5,
      "learning_rate": 0.00011674286169385207,
      "loss": 4.7307,
      "mean_token_accuracy": 0.2363320618867874,
      "num_tokens": 171697009.0,
      "step": 82740
    },
    {
      "entropy": 5.704566526412964,
      "epoch": 7.505896226415095,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011671986345502518,
      "loss": 4.8297,
      "mean_token_accuracy": 0.24017961025238038,
      "num_tokens": 171707839.0,
      "step": 82745
    },
    {
      "entropy": 5.686008739471435,
      "epoch": 7.506349782293179,
      "grad_norm": 1.5,
      "learning_rate": 0.00011669686848947178,
      "loss": 4.7424,
      "mean_token_accuracy": 0.2417706474661827,
      "num_tokens": 171717651.0,
      "step": 82750
    },
    {
      "entropy": 5.652339744567871,
      "epoch": 7.506803338171263,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011667387679766756,
      "loss": 4.7939,
      "mean_token_accuracy": 0.23298774063587188,
      "num_tokens": 171728049.0,
      "step": 82755
    },
    {
      "entropy": 5.698211288452148,
      "epoch": 7.5072568940493465,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011665088838008792,
      "loss": 4.7889,
      "mean_token_accuracy": 0.24276248812675477,
      "num_tokens": 171739065.0,
      "step": 82760
    },
    {
      "entropy": 5.674100112915039,
      "epoch": 7.507710449927431,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011662790323720825,
      "loss": 4.7893,
      "mean_token_accuracy": 0.24291263669729232,
      "num_tokens": 171749372.0,
      "step": 82765
    },
    {
      "entropy": 5.734217405319214,
      "epoch": 7.508164005805515,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011660492136950388,
      "loss": 4.8298,
      "mean_token_accuracy": 0.22980074286460878,
      "num_tokens": 171759208.0,
      "step": 82770
    },
    {
      "entropy": 5.69556188583374,
      "epoch": 7.508617561683599,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011658194277745013,
      "loss": 4.7712,
      "mean_token_accuracy": 0.23998657912015914,
      "num_tokens": 171768781.0,
      "step": 82775
    },
    {
      "entropy": 5.716873741149902,
      "epoch": 7.5090711175616836,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011655896746152215,
      "loss": 4.8663,
      "mean_token_accuracy": 0.24004873782396316,
      "num_tokens": 171779230.0,
      "step": 82780
    },
    {
      "entropy": 5.700922155380249,
      "epoch": 7.509524673439768,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011653599542219506,
      "loss": 4.7842,
      "mean_token_accuracy": 0.23820294439792633,
      "num_tokens": 171788827.0,
      "step": 82785
    },
    {
      "entropy": 5.699626207351685,
      "epoch": 7.509978229317852,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011651302665994403,
      "loss": 4.8275,
      "mean_token_accuracy": 0.2292943477630615,
      "num_tokens": 171799273.0,
      "step": 82790
    },
    {
      "entropy": 5.66239447593689,
      "epoch": 7.510431785195936,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011649006117524397,
      "loss": 4.7682,
      "mean_token_accuracy": 0.24395185112953185,
      "num_tokens": 171809413.0,
      "step": 82795
    },
    {
      "entropy": 5.694342517852784,
      "epoch": 7.510885341074021,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011646709896856985,
      "loss": 4.7844,
      "mean_token_accuracy": 0.23408839106559753,
      "num_tokens": 171820423.0,
      "step": 82800
    },
    {
      "entropy": 5.673445558547973,
      "epoch": 7.511338896952105,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011644414004039653,
      "loss": 4.8226,
      "mean_token_accuracy": 0.23521362692117692,
      "num_tokens": 171830267.0,
      "step": 82805
    },
    {
      "entropy": 5.680794715881348,
      "epoch": 7.511792452830189,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001164211843911988,
      "loss": 4.7489,
      "mean_token_accuracy": 0.24643397331237793,
      "num_tokens": 171839598.0,
      "step": 82810
    },
    {
      "entropy": 5.679993534088135,
      "epoch": 7.5122460087082725,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011639823202145137,
      "loss": 4.8362,
      "mean_token_accuracy": 0.23720603734254836,
      "num_tokens": 171850297.0,
      "step": 82815
    },
    {
      "entropy": 5.681590795516968,
      "epoch": 7.512699564586357,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011637528293162887,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22651907801628113,
      "num_tokens": 171860670.0,
      "step": 82820
    },
    {
      "entropy": 5.701354837417602,
      "epoch": 7.513153120464441,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011635233712220599,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23047226369380952,
      "num_tokens": 171870056.0,
      "step": 82825
    },
    {
      "entropy": 5.6688408851623535,
      "epoch": 7.513606676342525,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001163293945936572,
      "loss": 4.7119,
      "mean_token_accuracy": 0.24888333231210708,
      "num_tokens": 171879439.0,
      "step": 82830
    },
    {
      "entropy": 5.670438480377197,
      "epoch": 7.51406023222061,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011630645534645701,
      "loss": 4.7652,
      "mean_token_accuracy": 0.2393222853541374,
      "num_tokens": 171889760.0,
      "step": 82835
    },
    {
      "entropy": 5.72156171798706,
      "epoch": 7.514513788098694,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011628351938107973,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23670365363359452,
      "num_tokens": 171900131.0,
      "step": 82840
    },
    {
      "entropy": 5.733314466476441,
      "epoch": 7.514967343976778,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011626058669799971,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23275552839040756,
      "num_tokens": 171910615.0,
      "step": 82845
    },
    {
      "entropy": 5.709409999847412,
      "epoch": 7.515420899854862,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011623765729769116,
      "loss": 4.8103,
      "mean_token_accuracy": 0.23835280239582063,
      "num_tokens": 171922430.0,
      "step": 82850
    },
    {
      "entropy": 5.750828790664673,
      "epoch": 7.515874455732947,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011621473118062838,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2357093349099159,
      "num_tokens": 171933390.0,
      "step": 82855
    },
    {
      "entropy": 5.687496137619019,
      "epoch": 7.51632801161103,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011619180834728539,
      "loss": 4.8204,
      "mean_token_accuracy": 0.23568494617938995,
      "num_tokens": 171943857.0,
      "step": 82860
    },
    {
      "entropy": 5.696945714950561,
      "epoch": 7.516781567489114,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011616888879813629,
      "loss": 4.8859,
      "mean_token_accuracy": 0.23057149797677995,
      "num_tokens": 171954393.0,
      "step": 82865
    },
    {
      "entropy": 5.692142057418823,
      "epoch": 7.517235123367199,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011614597253365507,
      "loss": 4.7652,
      "mean_token_accuracy": 0.2359791174530983,
      "num_tokens": 171963508.0,
      "step": 82870
    },
    {
      "entropy": 5.668784093856812,
      "epoch": 7.517688679245283,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001161230595543156,
      "loss": 4.7572,
      "mean_token_accuracy": 0.239236544072628,
      "num_tokens": 171973092.0,
      "step": 82875
    },
    {
      "entropy": 5.757874965667725,
      "epoch": 7.518142235123367,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011610014986059175,
      "loss": 4.8891,
      "mean_token_accuracy": 0.23196842521429062,
      "num_tokens": 171983573.0,
      "step": 82880
    },
    {
      "entropy": 5.63690037727356,
      "epoch": 7.518595791001451,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00011607724345295728,
      "loss": 4.683,
      "mean_token_accuracy": 0.2440018981695175,
      "num_tokens": 171993680.0,
      "step": 82885
    },
    {
      "entropy": 5.683474397659301,
      "epoch": 7.519049346879536,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011605434033188593,
      "loss": 4.7728,
      "mean_token_accuracy": 0.2479969933629036,
      "num_tokens": 172003313.0,
      "step": 82890
    },
    {
      "entropy": 5.64459261894226,
      "epoch": 7.51950290275762,
      "grad_norm": 1.375,
      "learning_rate": 0.00011603144049785133,
      "loss": 4.78,
      "mean_token_accuracy": 0.23514410406351088,
      "num_tokens": 172013347.0,
      "step": 82895
    },
    {
      "entropy": 5.629031705856323,
      "epoch": 7.519956458635704,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011600854395132703,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2350270166993141,
      "num_tokens": 172024647.0,
      "step": 82900
    },
    {
      "entropy": 5.6628580570220945,
      "epoch": 7.520410014513788,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011598565069278657,
      "loss": 4.7979,
      "mean_token_accuracy": 0.23688989132642746,
      "num_tokens": 172035266.0,
      "step": 82905
    },
    {
      "entropy": 5.657684230804444,
      "epoch": 7.520863570391873,
      "grad_norm": 1.640625,
      "learning_rate": 0.00011596276072270338,
      "loss": 4.7722,
      "mean_token_accuracy": 0.23648664355278015,
      "num_tokens": 172046243.0,
      "step": 82910
    },
    {
      "entropy": 5.67142276763916,
      "epoch": 7.521317126269956,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011593987404155075,
      "loss": 4.8035,
      "mean_token_accuracy": 0.23187480866909027,
      "num_tokens": 172056607.0,
      "step": 82915
    },
    {
      "entropy": 5.735681200027466,
      "epoch": 7.52177068214804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011591699064980213,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2248491495847702,
      "num_tokens": 172066960.0,
      "step": 82920
    },
    {
      "entropy": 5.768231153488159,
      "epoch": 7.522224238026125,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011589411054793067,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23264780789613723,
      "num_tokens": 172076607.0,
      "step": 82925
    },
    {
      "entropy": 5.733028411865234,
      "epoch": 7.522677793904209,
      "grad_norm": 1.5,
      "learning_rate": 0.00011587123373640957,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22495394796133042,
      "num_tokens": 172087623.0,
      "step": 82930
    },
    {
      "entropy": 5.688844203948975,
      "epoch": 7.523131349782293,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011584836021571187,
      "loss": 4.739,
      "mean_token_accuracy": 0.2475421890616417,
      "num_tokens": 172098269.0,
      "step": 82935
    },
    {
      "entropy": 5.675072193145752,
      "epoch": 7.523584905660377,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011582548998631066,
      "loss": 4.7652,
      "mean_token_accuracy": 0.23788215219974518,
      "num_tokens": 172109030.0,
      "step": 82940
    },
    {
      "entropy": 5.717048168182373,
      "epoch": 7.524038461538462,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011580262304867884,
      "loss": 4.8062,
      "mean_token_accuracy": 0.2314256966114044,
      "num_tokens": 172118794.0,
      "step": 82945
    },
    {
      "entropy": 5.681716728210449,
      "epoch": 7.524492017416546,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011577975940328936,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23543429076671601,
      "num_tokens": 172128909.0,
      "step": 82950
    },
    {
      "entropy": 5.7108320713043215,
      "epoch": 7.52494557329463,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011575689905061506,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23619056195020677,
      "num_tokens": 172139039.0,
      "step": 82955
    },
    {
      "entropy": 5.710452461242676,
      "epoch": 7.525399129172714,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011573404199112866,
      "loss": 4.747,
      "mean_token_accuracy": 0.24145395010709764,
      "num_tokens": 172148877.0,
      "step": 82960
    },
    {
      "entropy": 5.7061803340911865,
      "epoch": 7.525852685050799,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011571118822530286,
      "loss": 4.7257,
      "mean_token_accuracy": 0.2410976305603981,
      "num_tokens": 172159177.0,
      "step": 82965
    },
    {
      "entropy": 5.698020076751709,
      "epoch": 7.526306240928882,
      "grad_norm": 1.5,
      "learning_rate": 0.00011568833775361028,
      "loss": 4.736,
      "mean_token_accuracy": 0.24919187128543854,
      "num_tokens": 172169345.0,
      "step": 82970
    },
    {
      "entropy": 5.7038818359375,
      "epoch": 7.526759796806966,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011566549057652347,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2261480763554573,
      "num_tokens": 172179651.0,
      "step": 82975
    },
    {
      "entropy": 5.708933925628662,
      "epoch": 7.527213352685051,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011564264669451485,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2363829419016838,
      "num_tokens": 172190808.0,
      "step": 82980
    },
    {
      "entropy": 5.725775718688965,
      "epoch": 7.527666908563135,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00011561980610805699,
      "loss": 4.8307,
      "mean_token_accuracy": 0.2335733726620674,
      "num_tokens": 172201196.0,
      "step": 82985
    },
    {
      "entropy": 5.665519666671753,
      "epoch": 7.528120464441219,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011559696881762214,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23734779804944992,
      "num_tokens": 172212176.0,
      "step": 82990
    },
    {
      "entropy": 5.724420022964478,
      "epoch": 7.5285740203193035,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011557413482368257,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23317391574382781,
      "num_tokens": 172222386.0,
      "step": 82995
    },
    {
      "entropy": 5.776415061950684,
      "epoch": 7.529027576197388,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011555130412671056,
      "loss": 4.8614,
      "mean_token_accuracy": 0.23039446920156478,
      "num_tokens": 172232991.0,
      "step": 83000
    },
    {
      "entropy": 5.675754976272583,
      "epoch": 7.529481132075472,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001155284767271782,
      "loss": 4.7528,
      "mean_token_accuracy": 0.24227285087108613,
      "num_tokens": 172243401.0,
      "step": 83005
    },
    {
      "entropy": 5.718887042999268,
      "epoch": 7.529934687953556,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00011550565262555759,
      "loss": 4.7791,
      "mean_token_accuracy": 0.24260727912187577,
      "num_tokens": 172252880.0,
      "step": 83010
    },
    {
      "entropy": 5.686279296875,
      "epoch": 7.53038824383164,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011548283182232064,
      "loss": 4.8315,
      "mean_token_accuracy": 0.23461531847715378,
      "num_tokens": 172263294.0,
      "step": 83015
    },
    {
      "entropy": 5.662045478820801,
      "epoch": 7.530841799709724,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011546001431793945,
      "loss": 4.8108,
      "mean_token_accuracy": 0.23196005672216416,
      "num_tokens": 172273597.0,
      "step": 83020
    },
    {
      "entropy": 5.656449460983277,
      "epoch": 7.531295355587808,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001154372001128858,
      "loss": 4.6868,
      "mean_token_accuracy": 0.23735601902008058,
      "num_tokens": 172284269.0,
      "step": 83025
    },
    {
      "entropy": 5.6345744132995605,
      "epoch": 7.531748911465892,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011541438920763154,
      "loss": 4.7776,
      "mean_token_accuracy": 0.238577701151371,
      "num_tokens": 172294633.0,
      "step": 83030
    },
    {
      "entropy": 5.673733472824097,
      "epoch": 7.532202467343977,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011539158160264839,
      "loss": 4.7781,
      "mean_token_accuracy": 0.24205999970436096,
      "num_tokens": 172304640.0,
      "step": 83035
    },
    {
      "entropy": 5.645149374008179,
      "epoch": 7.532656023222061,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011536877729840797,
      "loss": 4.7735,
      "mean_token_accuracy": 0.23376611471176148,
      "num_tokens": 172314914.0,
      "step": 83040
    },
    {
      "entropy": 5.697492551803589,
      "epoch": 7.533109579100145,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011534597629538194,
      "loss": 4.8008,
      "mean_token_accuracy": 0.2341560497879982,
      "num_tokens": 172325063.0,
      "step": 83045
    },
    {
      "entropy": 5.654654550552368,
      "epoch": 7.5335631349782295,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011532317859404172,
      "loss": 4.7371,
      "mean_token_accuracy": 0.2351870909333229,
      "num_tokens": 172335144.0,
      "step": 83050
    },
    {
      "entropy": 5.755621099472046,
      "epoch": 7.534016690856314,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011530038419485895,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2241345837712288,
      "num_tokens": 172344895.0,
      "step": 83055
    },
    {
      "entropy": 5.725408506393433,
      "epoch": 7.534470246734398,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00011527759309830491,
      "loss": 4.87,
      "mean_token_accuracy": 0.22960888892412185,
      "num_tokens": 172355600.0,
      "step": 83060
    },
    {
      "entropy": 5.6734518051147464,
      "epoch": 7.534923802612482,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011525480530485094,
      "loss": 4.7694,
      "mean_token_accuracy": 0.2347310721874237,
      "num_tokens": 172366295.0,
      "step": 83065
    },
    {
      "entropy": 5.677165222167969,
      "epoch": 7.535377358490566,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011523202081496834,
      "loss": 4.7786,
      "mean_token_accuracy": 0.24151491820812226,
      "num_tokens": 172375164.0,
      "step": 83070
    },
    {
      "entropy": 5.6552635669708256,
      "epoch": 7.53583091436865,
      "grad_norm": 1.5,
      "learning_rate": 0.00011520923962912824,
      "loss": 4.7355,
      "mean_token_accuracy": 0.243499955534935,
      "num_tokens": 172385487.0,
      "step": 83075
    },
    {
      "entropy": 5.712437486648559,
      "epoch": 7.536284470246734,
      "grad_norm": 1.5,
      "learning_rate": 0.00011518646174780178,
      "loss": 4.8478,
      "mean_token_accuracy": 0.23083169758319855,
      "num_tokens": 172396021.0,
      "step": 83080
    },
    {
      "entropy": 5.689449214935303,
      "epoch": 7.5367380261248185,
      "grad_norm": 1.6875,
      "learning_rate": 0.00011516368717145996,
      "loss": 4.7581,
      "mean_token_accuracy": 0.24398435205221175,
      "num_tokens": 172404989.0,
      "step": 83085
    },
    {
      "entropy": 5.609430599212646,
      "epoch": 7.537191582002903,
      "grad_norm": 1.5,
      "learning_rate": 0.00011514091590057387,
      "loss": 4.7255,
      "mean_token_accuracy": 0.24550487250089645,
      "num_tokens": 172415561.0,
      "step": 83090
    },
    {
      "entropy": 5.667363929748535,
      "epoch": 7.537645137880987,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011511814793561438,
      "loss": 4.8335,
      "mean_token_accuracy": 0.22794902622699736,
      "num_tokens": 172425890.0,
      "step": 83095
    },
    {
      "entropy": 5.714703989028931,
      "epoch": 7.538098693759071,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011509538327705236,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23491137176752092,
      "num_tokens": 172435964.0,
      "step": 83100
    },
    {
      "entropy": 5.701128625869751,
      "epoch": 7.5385522496371555,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011507262192535854,
      "loss": 4.8083,
      "mean_token_accuracy": 0.22947175651788712,
      "num_tokens": 172446028.0,
      "step": 83105
    },
    {
      "entropy": 5.651557302474975,
      "epoch": 7.53900580551524,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011504986388100365,
      "loss": 4.7843,
      "mean_token_accuracy": 0.2399594247341156,
      "num_tokens": 172456923.0,
      "step": 83110
    },
    {
      "entropy": 5.674364185333252,
      "epoch": 7.539459361393323,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011502710914445836,
      "loss": 4.7762,
      "mean_token_accuracy": 0.24744750708341598,
      "num_tokens": 172466751.0,
      "step": 83115
    },
    {
      "entropy": 5.675386524200439,
      "epoch": 7.539912917271408,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011500435771619312,
      "loss": 4.7828,
      "mean_token_accuracy": 0.23737753182649612,
      "num_tokens": 172477273.0,
      "step": 83120
    },
    {
      "entropy": 5.690413522720337,
      "epoch": 7.540366473149492,
      "grad_norm": 1.703125,
      "learning_rate": 0.00011498160959667861,
      "loss": 4.825,
      "mean_token_accuracy": 0.23084429055452346,
      "num_tokens": 172486802.0,
      "step": 83125
    },
    {
      "entropy": 5.6890229225158695,
      "epoch": 7.540820029027576,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011495886478638521,
      "loss": 4.877,
      "mean_token_accuracy": 0.2260482519865036,
      "num_tokens": 172498140.0,
      "step": 83130
    },
    {
      "entropy": 5.670880508422852,
      "epoch": 7.54127358490566,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011493612328578323,
      "loss": 4.802,
      "mean_token_accuracy": 0.23918316811323165,
      "num_tokens": 172508019.0,
      "step": 83135
    },
    {
      "entropy": 5.769548749923706,
      "epoch": 7.5417271407837445,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011491338509534305,
      "loss": 4.9489,
      "mean_token_accuracy": 0.23148145377635956,
      "num_tokens": 172517378.0,
      "step": 83140
    },
    {
      "entropy": 5.713857746124267,
      "epoch": 7.542180696661829,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011489065021553484,
      "loss": 4.8173,
      "mean_token_accuracy": 0.2308536395430565,
      "num_tokens": 172527776.0,
      "step": 83145
    },
    {
      "entropy": 5.66889123916626,
      "epoch": 7.542634252539913,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011486791864682872,
      "loss": 4.7551,
      "mean_token_accuracy": 0.23973950743675232,
      "num_tokens": 172537889.0,
      "step": 83150
    },
    {
      "entropy": 5.601785707473755,
      "epoch": 7.543087808417997,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011484519038969491,
      "loss": 4.6804,
      "mean_token_accuracy": 0.24318053871393203,
      "num_tokens": 172547993.0,
      "step": 83155
    },
    {
      "entropy": 5.642045545578003,
      "epoch": 7.543541364296082,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001148224654446034,
      "loss": 4.749,
      "mean_token_accuracy": 0.2358465775847435,
      "num_tokens": 172559092.0,
      "step": 83160
    },
    {
      "entropy": 5.672921752929687,
      "epoch": 7.543994920174166,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011479974381202411,
      "loss": 4.7482,
      "mean_token_accuracy": 0.24270201325416565,
      "num_tokens": 172570001.0,
      "step": 83165
    },
    {
      "entropy": 5.662831258773804,
      "epoch": 7.544448476052249,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011477702549242694,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2394967719912529,
      "num_tokens": 172579572.0,
      "step": 83170
    },
    {
      "entropy": 5.683291864395142,
      "epoch": 7.5449020319303335,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001147543104862817,
      "loss": 4.8281,
      "mean_token_accuracy": 0.23868699073791505,
      "num_tokens": 172589389.0,
      "step": 83175
    },
    {
      "entropy": 5.678320693969726,
      "epoch": 7.545355587808418,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011473159879405814,
      "loss": 4.8087,
      "mean_token_accuracy": 0.24054549336433412,
      "num_tokens": 172598780.0,
      "step": 83180
    },
    {
      "entropy": 5.670759105682373,
      "epoch": 7.545809143686502,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011470889041622594,
      "loss": 4.7913,
      "mean_token_accuracy": 0.23697266727685928,
      "num_tokens": 172610840.0,
      "step": 83185
    },
    {
      "entropy": 5.7239762306213375,
      "epoch": 7.546262699564586,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011468618535325475,
      "loss": 4.839,
      "mean_token_accuracy": 0.23003097623586655,
      "num_tokens": 172620478.0,
      "step": 83190
    },
    {
      "entropy": 5.717563676834106,
      "epoch": 7.5467162554426706,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011466348360561414,
      "loss": 4.8284,
      "mean_token_accuracy": 0.23527166098356248,
      "num_tokens": 172630574.0,
      "step": 83195
    },
    {
      "entropy": 5.675572061538697,
      "epoch": 7.547169811320755,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011464078517377353,
      "loss": 4.8237,
      "mean_token_accuracy": 0.23425098657608032,
      "num_tokens": 172641045.0,
      "step": 83200
    },
    {
      "entropy": 5.6755352973937985,
      "epoch": 7.547623367198839,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011461809005820237,
      "loss": 4.7658,
      "mean_token_accuracy": 0.24352861046791077,
      "num_tokens": 172650510.0,
      "step": 83205
    },
    {
      "entropy": 5.677806806564331,
      "epoch": 7.548076923076923,
      "grad_norm": 1.5,
      "learning_rate": 0.00011459539825936996,
      "loss": 4.8113,
      "mean_token_accuracy": 0.23322986364364623,
      "num_tokens": 172660509.0,
      "step": 83210
    },
    {
      "entropy": 5.6801230907440186,
      "epoch": 7.548530478955008,
      "grad_norm": 1.5,
      "learning_rate": 0.00011457270977774557,
      "loss": 4.7798,
      "mean_token_accuracy": 0.23402228206396103,
      "num_tokens": 172670821.0,
      "step": 83215
    },
    {
      "entropy": 5.704159307479858,
      "epoch": 7.548984034833092,
      "grad_norm": 1.625,
      "learning_rate": 0.00011455002461379838,
      "loss": 4.8486,
      "mean_token_accuracy": 0.23988111317157745,
      "num_tokens": 172682043.0,
      "step": 83220
    },
    {
      "entropy": 5.644045114517212,
      "epoch": 7.549437590711175,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011452734276799763,
      "loss": 4.7373,
      "mean_token_accuracy": 0.23454324156045914,
      "num_tokens": 172692468.0,
      "step": 83225
    },
    {
      "entropy": 5.755618381500244,
      "epoch": 7.5498911465892595,
      "grad_norm": 1.671875,
      "learning_rate": 0.00011450466424081232,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2334960773587227,
      "num_tokens": 172702531.0,
      "step": 83230
    },
    {
      "entropy": 5.683874607086182,
      "epoch": 7.550344702467344,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011448198903271149,
      "loss": 4.7899,
      "mean_token_accuracy": 0.24525615572929382,
      "num_tokens": 172713276.0,
      "step": 83235
    },
    {
      "entropy": 5.695186185836792,
      "epoch": 7.550798258345428,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011445931714416398,
      "loss": 4.8325,
      "mean_token_accuracy": 0.2364784523844719,
      "num_tokens": 172723719.0,
      "step": 83240
    },
    {
      "entropy": 5.679843091964722,
      "epoch": 7.551251814223512,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00011443664857563872,
      "loss": 4.7285,
      "mean_token_accuracy": 0.24632879942655564,
      "num_tokens": 172732708.0,
      "step": 83245
    },
    {
      "entropy": 5.711436176300049,
      "epoch": 7.551705370101597,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001144139833276045,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22912688553333282,
      "num_tokens": 172742398.0,
      "step": 83250
    },
    {
      "entropy": 5.684877538681031,
      "epoch": 7.552158925979681,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011439132140052992,
      "loss": 4.8293,
      "mean_token_accuracy": 0.23520802855491638,
      "num_tokens": 172752838.0,
      "step": 83255
    },
    {
      "entropy": 5.693961477279663,
      "epoch": 7.552612481857765,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011436866279488386,
      "loss": 4.7951,
      "mean_token_accuracy": 0.2388746753334999,
      "num_tokens": 172762997.0,
      "step": 83260
    },
    {
      "entropy": 5.671948528289795,
      "epoch": 7.553066037735849,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011434600751113475,
      "loss": 4.7724,
      "mean_token_accuracy": 0.24266267269849778,
      "num_tokens": 172773168.0,
      "step": 83265
    },
    {
      "entropy": 5.672827816009521,
      "epoch": 7.553519593613933,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011432335554975115,
      "loss": 4.7596,
      "mean_token_accuracy": 0.23185529112815856,
      "num_tokens": 172784697.0,
      "step": 83270
    },
    {
      "entropy": 5.680236339569092,
      "epoch": 7.553973149492017,
      "grad_norm": 1.5,
      "learning_rate": 0.0001143007069112015,
      "loss": 4.7552,
      "mean_token_accuracy": 0.24111237972974778,
      "num_tokens": 172794704.0,
      "step": 83275
    },
    {
      "entropy": 5.6884101867675785,
      "epoch": 7.554426705370101,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001142780615959542,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23674551844596864,
      "num_tokens": 172804501.0,
      "step": 83280
    },
    {
      "entropy": 5.669151306152344,
      "epoch": 7.554880261248186,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011425541960447753,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23892078995704652,
      "num_tokens": 172815147.0,
      "step": 83285
    },
    {
      "entropy": 5.707792949676514,
      "epoch": 7.55533381712627,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011423278093723969,
      "loss": 4.8644,
      "mean_token_accuracy": 0.23081612288951875,
      "num_tokens": 172824215.0,
      "step": 83290
    },
    {
      "entropy": 5.6473894119262695,
      "epoch": 7.555787373004354,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011421014559470897,
      "loss": 4.7287,
      "mean_token_accuracy": 0.2534963026642799,
      "num_tokens": 172833993.0,
      "step": 83295
    },
    {
      "entropy": 5.70927848815918,
      "epoch": 7.556240928882438,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011418751357735342,
      "loss": 4.7943,
      "mean_token_accuracy": 0.2411731392145157,
      "num_tokens": 172844362.0,
      "step": 83300
    },
    {
      "entropy": 5.670987224578857,
      "epoch": 7.556694484760523,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011416488488564105,
      "loss": 4.7519,
      "mean_token_accuracy": 0.24576616138219834,
      "num_tokens": 172854083.0,
      "step": 83305
    },
    {
      "entropy": 5.70896430015564,
      "epoch": 7.557148040638607,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011414225952003994,
      "loss": 4.8411,
      "mean_token_accuracy": 0.2347422108054161,
      "num_tokens": 172864452.0,
      "step": 83310
    },
    {
      "entropy": 5.664637422561645,
      "epoch": 7.557601596516691,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011411963748101772,
      "loss": 4.7476,
      "mean_token_accuracy": 0.2437704861164093,
      "num_tokens": 172874040.0,
      "step": 83315
    },
    {
      "entropy": 5.68825044631958,
      "epoch": 7.558055152394775,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001140970187690425,
      "loss": 4.7669,
      "mean_token_accuracy": 0.24411387294530867,
      "num_tokens": 172883623.0,
      "step": 83320
    },
    {
      "entropy": 5.6686177253723145,
      "epoch": 7.558508708272859,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011407440338458191,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23620298653841018,
      "num_tokens": 172893884.0,
      "step": 83325
    },
    {
      "entropy": 5.64952826499939,
      "epoch": 7.558962264150943,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011405179132810367,
      "loss": 4.7866,
      "mean_token_accuracy": 0.2376190334558487,
      "num_tokens": 172904008.0,
      "step": 83330
    },
    {
      "entropy": 5.652129650115967,
      "epoch": 7.559415820029027,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011402918260007542,
      "loss": 4.7692,
      "mean_token_accuracy": 0.23607661277055741,
      "num_tokens": 172914917.0,
      "step": 83335
    },
    {
      "entropy": 5.695153141021729,
      "epoch": 7.559869375907112,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011400657720096466,
      "loss": 4.7954,
      "mean_token_accuracy": 0.23471648544073104,
      "num_tokens": 172924944.0,
      "step": 83340
    },
    {
      "entropy": 5.6938032627105715,
      "epoch": 7.560322931785196,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011398397513123894,
      "loss": 4.8144,
      "mean_token_accuracy": 0.23569868355989457,
      "num_tokens": 172934153.0,
      "step": 83345
    },
    {
      "entropy": 5.702463388442993,
      "epoch": 7.56077648766328,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011396137639136557,
      "loss": 4.8122,
      "mean_token_accuracy": 0.2306219071149826,
      "num_tokens": 172944899.0,
      "step": 83350
    },
    {
      "entropy": 5.745291423797608,
      "epoch": 7.561230043541364,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011393878098181204,
      "loss": 4.8573,
      "mean_token_accuracy": 0.2336154192686081,
      "num_tokens": 172955420.0,
      "step": 83355
    },
    {
      "entropy": 5.688716220855713,
      "epoch": 7.561683599419449,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011391618890304555,
      "loss": 4.795,
      "mean_token_accuracy": 0.2388277143239975,
      "num_tokens": 172965312.0,
      "step": 83360
    },
    {
      "entropy": 5.6401204586029055,
      "epoch": 7.562137155297533,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011389360015553333,
      "loss": 4.7023,
      "mean_token_accuracy": 0.24611851423978806,
      "num_tokens": 172975326.0,
      "step": 83365
    },
    {
      "entropy": 5.69426155090332,
      "epoch": 7.562590711175617,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011387101473974253,
      "loss": 4.7727,
      "mean_token_accuracy": 0.24241530150175095,
      "num_tokens": 172985794.0,
      "step": 83370
    },
    {
      "entropy": 5.684609985351562,
      "epoch": 7.5630442670537015,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011384843265614019,
      "loss": 4.8422,
      "mean_token_accuracy": 0.23200657665729524,
      "num_tokens": 172996361.0,
      "step": 83375
    },
    {
      "entropy": 5.7509050369262695,
      "epoch": 7.563497822931785,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011382585390519333,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23188749998807906,
      "num_tokens": 173006894.0,
      "step": 83380
    },
    {
      "entropy": 5.673059272766113,
      "epoch": 7.563951378809869,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011380327848736882,
      "loss": 4.777,
      "mean_token_accuracy": 0.23623957633972167,
      "num_tokens": 173018230.0,
      "step": 83385
    },
    {
      "entropy": 5.721418237686157,
      "epoch": 7.564404934687953,
      "grad_norm": 1.5,
      "learning_rate": 0.00011378070640313366,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2321941837668419,
      "num_tokens": 173028809.0,
      "step": 83390
    },
    {
      "entropy": 5.637228059768677,
      "epoch": 7.564858490566038,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011375813765295456,
      "loss": 4.7155,
      "mean_token_accuracy": 0.24771819412708282,
      "num_tokens": 173038940.0,
      "step": 83395
    },
    {
      "entropy": 5.687128400802612,
      "epoch": 7.565312046444122,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011373557223729827,
      "loss": 4.7591,
      "mean_token_accuracy": 0.24001488834619522,
      "num_tokens": 173049414.0,
      "step": 83400
    },
    {
      "entropy": 5.6379842281341555,
      "epoch": 7.565765602322206,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011371301015663145,
      "loss": 4.7227,
      "mean_token_accuracy": 0.24501180797815322,
      "num_tokens": 173059123.0,
      "step": 83405
    },
    {
      "entropy": 5.677895355224609,
      "epoch": 7.5662191582002905,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011369045141142067,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23469818532466888,
      "num_tokens": 173069423.0,
      "step": 83410
    },
    {
      "entropy": 5.710721349716186,
      "epoch": 7.566672714078375,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011366789600213248,
      "loss": 4.7804,
      "mean_token_accuracy": 0.2335665389895439,
      "num_tokens": 173079290.0,
      "step": 83415
    },
    {
      "entropy": 5.698455095291138,
      "epoch": 7.567126269956459,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011364534392923322,
      "loss": 4.7439,
      "mean_token_accuracy": 0.2380705401301384,
      "num_tokens": 173088907.0,
      "step": 83420
    },
    {
      "entropy": 5.672198677062989,
      "epoch": 7.567579825834542,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011362279519318944,
      "loss": 4.794,
      "mean_token_accuracy": 0.22940537184476853,
      "num_tokens": 173100239.0,
      "step": 83425
    },
    {
      "entropy": 5.686069011688232,
      "epoch": 7.568033381712627,
      "grad_norm": 1.5,
      "learning_rate": 0.00011360024979446739,
      "loss": 4.7505,
      "mean_token_accuracy": 0.23494194746017455,
      "num_tokens": 173110861.0,
      "step": 83430
    },
    {
      "entropy": 5.624919176101685,
      "epoch": 7.568486937590711,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001135777077335333,
      "loss": 4.6907,
      "mean_token_accuracy": 0.24401615262031556,
      "num_tokens": 173121369.0,
      "step": 83435
    },
    {
      "entropy": 5.642443418502808,
      "epoch": 7.568940493468795,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011355516901085335,
      "loss": 4.7336,
      "mean_token_accuracy": 0.24472754299640656,
      "num_tokens": 173131794.0,
      "step": 83440
    },
    {
      "entropy": 5.697180891036988,
      "epoch": 7.569394049346879,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011353263362689363,
      "loss": 4.8091,
      "mean_token_accuracy": 0.23430525958538057,
      "num_tokens": 173142628.0,
      "step": 83445
    },
    {
      "entropy": 5.7154566764831545,
      "epoch": 7.569847605224964,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011351010158212015,
      "loss": 4.8782,
      "mean_token_accuracy": 0.23634139001369475,
      "num_tokens": 173152845.0,
      "step": 83450
    },
    {
      "entropy": 5.68208794593811,
      "epoch": 7.570301161103048,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011348757287699896,
      "loss": 4.7893,
      "mean_token_accuracy": 0.2382381245493889,
      "num_tokens": 173163754.0,
      "step": 83455
    },
    {
      "entropy": 5.652309226989746,
      "epoch": 7.570754716981132,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011346504751199595,
      "loss": 4.8583,
      "mean_token_accuracy": 0.22949514538049698,
      "num_tokens": 173175615.0,
      "step": 83460
    },
    {
      "entropy": 5.606964111328125,
      "epoch": 7.5712082728592165,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011344252548757692,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2504023268818855,
      "num_tokens": 173185295.0,
      "step": 83465
    },
    {
      "entropy": 5.643199920654297,
      "epoch": 7.571661828737301,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011342000680420761,
      "loss": 4.774,
      "mean_token_accuracy": 0.238978511095047,
      "num_tokens": 173196183.0,
      "step": 83470
    },
    {
      "entropy": 5.726722955703735,
      "epoch": 7.572115384615385,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011339749146235376,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22494578808546067,
      "num_tokens": 173206211.0,
      "step": 83475
    },
    {
      "entropy": 5.74275050163269,
      "epoch": 7.572568940493468,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011337497946248096,
      "loss": 4.8482,
      "mean_token_accuracy": 0.23932321965694428,
      "num_tokens": 173216639.0,
      "step": 83480
    },
    {
      "entropy": 5.715053939819336,
      "epoch": 7.573022496371553,
      "grad_norm": 1.5,
      "learning_rate": 0.00011335247080505469,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22879806458950042,
      "num_tokens": 173226721.0,
      "step": 83485
    },
    {
      "entropy": 5.73384428024292,
      "epoch": 7.573476052249637,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00011332996549054058,
      "loss": 4.847,
      "mean_token_accuracy": 0.2322703093290329,
      "num_tokens": 173237083.0,
      "step": 83490
    },
    {
      "entropy": 5.682819461822509,
      "epoch": 7.573929608127721,
      "grad_norm": 1.34375,
      "learning_rate": 0.000113307463519404,
      "loss": 4.7448,
      "mean_token_accuracy": 0.24593787491321564,
      "num_tokens": 173247473.0,
      "step": 83495
    },
    {
      "entropy": 5.696189069747925,
      "epoch": 7.5743831640058055,
      "grad_norm": 1.5,
      "learning_rate": 0.0001132849648921103,
      "loss": 4.7912,
      "mean_token_accuracy": 0.23092774003744126,
      "num_tokens": 173257908.0,
      "step": 83500
    },
    {
      "entropy": 5.681456613540649,
      "epoch": 7.57483671988389,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00011326246960912469,
      "loss": 4.8122,
      "mean_token_accuracy": 0.2353632241487503,
      "num_tokens": 173268287.0,
      "step": 83505
    },
    {
      "entropy": 5.680482864379883,
      "epoch": 7.575290275761974,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011323997767091245,
      "loss": 4.8206,
      "mean_token_accuracy": 0.23046404272317886,
      "num_tokens": 173278828.0,
      "step": 83510
    },
    {
      "entropy": 5.726144027709961,
      "epoch": 7.575743831640058,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001132174890779387,
      "loss": 4.8202,
      "mean_token_accuracy": 0.23099009394645692,
      "num_tokens": 173290306.0,
      "step": 83515
    },
    {
      "entropy": 5.648488330841064,
      "epoch": 7.5761973875181425,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001131950038306684,
      "loss": 4.7607,
      "mean_token_accuracy": 0.24162009358406067,
      "num_tokens": 173300831.0,
      "step": 83520
    },
    {
      "entropy": 5.703284645080567,
      "epoch": 7.576650943396227,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011317252192956675,
      "loss": 4.776,
      "mean_token_accuracy": 0.23483503013849258,
      "num_tokens": 173311521.0,
      "step": 83525
    },
    {
      "entropy": 5.882217502593994,
      "epoch": 7.577104499274311,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011315004337509858,
      "loss": 5.0562,
      "mean_token_accuracy": 0.21440766304731368,
      "num_tokens": 173321785.0,
      "step": 83530
    },
    {
      "entropy": 5.743297433853149,
      "epoch": 7.5775580551523944,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011312756816772876,
      "loss": 4.815,
      "mean_token_accuracy": 0.23726718872785568,
      "num_tokens": 173332230.0,
      "step": 83535
    },
    {
      "entropy": 5.723842716217041,
      "epoch": 7.578011611030479,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001131050963079221,
      "loss": 4.8737,
      "mean_token_accuracy": 0.23132576495409013,
      "num_tokens": 173343160.0,
      "step": 83540
    },
    {
      "entropy": 5.738211822509766,
      "epoch": 7.578465166908563,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011308262779614324,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2349212110042572,
      "num_tokens": 173353153.0,
      "step": 83545
    },
    {
      "entropy": 5.7591899871826175,
      "epoch": 7.578918722786647,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011306016263285698,
      "loss": 4.8558,
      "mean_token_accuracy": 0.2399126410484314,
      "num_tokens": 173363033.0,
      "step": 83550
    },
    {
      "entropy": 5.741361951828003,
      "epoch": 7.5793722786647315,
      "grad_norm": 1.375,
      "learning_rate": 0.0001130377008185277,
      "loss": 4.8465,
      "mean_token_accuracy": 0.2293159395456314,
      "num_tokens": 173374108.0,
      "step": 83555
    },
    {
      "entropy": 5.707822990417481,
      "epoch": 7.579825834542816,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011301524235362011,
      "loss": 4.8698,
      "mean_token_accuracy": 0.231499320268631,
      "num_tokens": 173383505.0,
      "step": 83560
    },
    {
      "entropy": 5.6770689487457275,
      "epoch": 7.5802793904209,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001129927872385986,
      "loss": 4.8065,
      "mean_token_accuracy": 0.2393037661910057,
      "num_tokens": 173394963.0,
      "step": 83565
    },
    {
      "entropy": 5.7055457592010494,
      "epoch": 7.580732946298984,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011297033547392752,
      "loss": 4.8311,
      "mean_token_accuracy": 0.23130760788917543,
      "num_tokens": 173404951.0,
      "step": 83570
    },
    {
      "entropy": 5.671956157684326,
      "epoch": 7.581186502177069,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011294788706007125,
      "loss": 4.7755,
      "mean_token_accuracy": 0.2342960312962532,
      "num_tokens": 173415198.0,
      "step": 83575
    },
    {
      "entropy": 5.6873396873474125,
      "epoch": 7.581640058055152,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001129254419974939,
      "loss": 4.7808,
      "mean_token_accuracy": 0.24221838861703873,
      "num_tokens": 173425456.0,
      "step": 83580
    },
    {
      "entropy": 5.698033475875855,
      "epoch": 7.582093613933236,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011290300028665975,
      "loss": 4.8144,
      "mean_token_accuracy": 0.23813683688640594,
      "num_tokens": 173435582.0,
      "step": 83585
    },
    {
      "entropy": 5.699863862991333,
      "epoch": 7.5825471698113205,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011288056192803278,
      "loss": 4.6982,
      "mean_token_accuracy": 0.24149938821792602,
      "num_tokens": 173445000.0,
      "step": 83590
    },
    {
      "entropy": 5.712002468109131,
      "epoch": 7.583000725689405,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011285812692207719,
      "loss": 4.8098,
      "mean_token_accuracy": 0.24558672606945037,
      "num_tokens": 173455562.0,
      "step": 83595
    },
    {
      "entropy": 5.65025897026062,
      "epoch": 7.583454281567489,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011283569526925683,
      "loss": 4.8083,
      "mean_token_accuracy": 0.24239256083965302,
      "num_tokens": 173465331.0,
      "step": 83600
    },
    {
      "entropy": 5.661040639877319,
      "epoch": 7.583907837445573,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011281326697003562,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2446054220199585,
      "num_tokens": 173475436.0,
      "step": 83605
    },
    {
      "entropy": 5.704891347885132,
      "epoch": 7.5843613933236576,
      "grad_norm": 1.5,
      "learning_rate": 0.00011279084202487738,
      "loss": 4.7725,
      "mean_token_accuracy": 0.23937655985355377,
      "num_tokens": 173486082.0,
      "step": 83610
    },
    {
      "entropy": 5.681444597244263,
      "epoch": 7.584814949201742,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00011276842043424586,
      "loss": 4.8511,
      "mean_token_accuracy": 0.23779212534427643,
      "num_tokens": 173497010.0,
      "step": 83615
    },
    {
      "entropy": 5.706721782684326,
      "epoch": 7.585268505079826,
      "grad_norm": 1.5,
      "learning_rate": 0.00011274600219860475,
      "loss": 4.8085,
      "mean_token_accuracy": 0.23464771956205369,
      "num_tokens": 173506843.0,
      "step": 83620
    },
    {
      "entropy": 5.682535314559937,
      "epoch": 7.58572206095791,
      "grad_norm": 1.375,
      "learning_rate": 0.0001127235873184176,
      "loss": 4.7706,
      "mean_token_accuracy": 0.24330561310052873,
      "num_tokens": 173516804.0,
      "step": 83625
    },
    {
      "entropy": 5.748626708984375,
      "epoch": 7.586175616835995,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00011270117579414807,
      "loss": 4.831,
      "mean_token_accuracy": 0.2389477238059044,
      "num_tokens": 173527255.0,
      "step": 83630
    },
    {
      "entropy": 5.788983535766602,
      "epoch": 7.586629172714078,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011267876762625962,
      "loss": 4.877,
      "mean_token_accuracy": 0.2275131955742836,
      "num_tokens": 173537121.0,
      "step": 83635
    },
    {
      "entropy": 5.715200805664063,
      "epoch": 7.587082728592162,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011265636281521557,
      "loss": 4.7786,
      "mean_token_accuracy": 0.23443466126918794,
      "num_tokens": 173546889.0,
      "step": 83640
    },
    {
      "entropy": 5.662035512924194,
      "epoch": 7.5875362844702465,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011263396136147932,
      "loss": 4.822,
      "mean_token_accuracy": 0.22768602520227432,
      "num_tokens": 173557423.0,
      "step": 83645
    },
    {
      "entropy": 5.7077384948730465,
      "epoch": 7.587989840348331,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011261156326551414,
      "loss": 4.8927,
      "mean_token_accuracy": 0.2352728769183159,
      "num_tokens": 173566977.0,
      "step": 83650
    },
    {
      "entropy": 5.693960428237915,
      "epoch": 7.588443396226415,
      "grad_norm": 1.65625,
      "learning_rate": 0.00011258916852778315,
      "loss": 4.7643,
      "mean_token_accuracy": 0.23895489871501924,
      "num_tokens": 173576277.0,
      "step": 83655
    },
    {
      "entropy": 5.700793647766114,
      "epoch": 7.588896952104499,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011256677714874959,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2356919452548027,
      "num_tokens": 173587546.0,
      "step": 83660
    },
    {
      "entropy": 5.674040269851685,
      "epoch": 7.589350507982584,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011254438912887649,
      "loss": 4.8157,
      "mean_token_accuracy": 0.2352731093764305,
      "num_tokens": 173597606.0,
      "step": 83665
    },
    {
      "entropy": 5.673433065414429,
      "epoch": 7.589804063860668,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011252200446862683,
      "loss": 4.7869,
      "mean_token_accuracy": 0.24005222767591478,
      "num_tokens": 173607815.0,
      "step": 83670
    },
    {
      "entropy": 5.711996364593506,
      "epoch": 7.590257619738752,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001124996231684635,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2384713962674141,
      "num_tokens": 173618800.0,
      "step": 83675
    },
    {
      "entropy": 5.645258474349975,
      "epoch": 7.5907111756168355,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001124772452288494,
      "loss": 4.6958,
      "mean_token_accuracy": 0.24559382647275924,
      "num_tokens": 173628802.0,
      "step": 83680
    },
    {
      "entropy": 5.691353511810303,
      "epoch": 7.591164731494921,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011245487065024723,
      "loss": 4.8434,
      "mean_token_accuracy": 0.2350208804011345,
      "num_tokens": 173638587.0,
      "step": 83685
    },
    {
      "entropy": 5.691893100738525,
      "epoch": 7.591618287373004,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011243249943311972,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23318164199590682,
      "num_tokens": 173648634.0,
      "step": 83690
    },
    {
      "entropy": 5.666100883483887,
      "epoch": 7.592071843251088,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011241013157792963,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2420487329363823,
      "num_tokens": 173660225.0,
      "step": 83695
    },
    {
      "entropy": 5.644812154769897,
      "epoch": 7.592525399129173,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011238776708513942,
      "loss": 4.7586,
      "mean_token_accuracy": 0.2398919031023979,
      "num_tokens": 173670365.0,
      "step": 83700
    },
    {
      "entropy": 5.635463285446167,
      "epoch": 7.592978955007257,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011236540595521167,
      "loss": 4.7317,
      "mean_token_accuracy": 0.23644431233406066,
      "num_tokens": 173680848.0,
      "step": 83705
    },
    {
      "entropy": 5.661846351623535,
      "epoch": 7.593432510885341,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011234304818860875,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2379024177789688,
      "num_tokens": 173691202.0,
      "step": 83710
    },
    {
      "entropy": 5.7108112335205075,
      "epoch": 7.593886066763425,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011232069378579301,
      "loss": 4.7833,
      "mean_token_accuracy": 0.2411651372909546,
      "num_tokens": 173701046.0,
      "step": 83715
    },
    {
      "entropy": 5.667090034484863,
      "epoch": 7.59433962264151,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011229834274722683,
      "loss": 4.7,
      "mean_token_accuracy": 0.24778658747673035,
      "num_tokens": 173711995.0,
      "step": 83720
    },
    {
      "entropy": 5.717611265182495,
      "epoch": 7.594793178519594,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011227599507337228,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22775918990373611,
      "num_tokens": 173723830.0,
      "step": 83725
    },
    {
      "entropy": 5.686549186706543,
      "epoch": 7.595246734397678,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001122536507646917,
      "loss": 4.7941,
      "mean_token_accuracy": 0.2409241646528244,
      "num_tokens": 173733632.0,
      "step": 83730
    },
    {
      "entropy": 5.68307638168335,
      "epoch": 7.5957002902757615,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011223130982164715,
      "loss": 4.791,
      "mean_token_accuracy": 0.24179677218198775,
      "num_tokens": 173742875.0,
      "step": 83735
    },
    {
      "entropy": 5.664987564086914,
      "epoch": 7.596153846153846,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011220897224470054,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23805470317602156,
      "num_tokens": 173753263.0,
      "step": 83740
    },
    {
      "entropy": 5.678294897079468,
      "epoch": 7.59660740203193,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001121866380343139,
      "loss": 4.7843,
      "mean_token_accuracy": 0.24293871223926544,
      "num_tokens": 173763217.0,
      "step": 83745
    },
    {
      "entropy": 5.5931093215942385,
      "epoch": 7.597060957910014,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011216430719094894,
      "loss": 4.688,
      "mean_token_accuracy": 0.2495583787560463,
      "num_tokens": 173774441.0,
      "step": 83750
    },
    {
      "entropy": 5.7241205215454105,
      "epoch": 7.597514513788099,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011214197971506771,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22966910749673844,
      "num_tokens": 173785483.0,
      "step": 83755
    },
    {
      "entropy": 5.624571228027344,
      "epoch": 7.597968069666183,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011211965560713182,
      "loss": 4.7618,
      "mean_token_accuracy": 0.23231335282325744,
      "num_tokens": 173796325.0,
      "step": 83760
    },
    {
      "entropy": 5.713738965988159,
      "epoch": 7.598421625544267,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011209733486760296,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23573267757892608,
      "num_tokens": 173806190.0,
      "step": 83765
    },
    {
      "entropy": 5.763010263442993,
      "epoch": 7.598875181422351,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011207501749694274,
      "loss": 4.9299,
      "mean_token_accuracy": 0.22512554228305817,
      "num_tokens": 173816245.0,
      "step": 83770
    },
    {
      "entropy": 5.6413827419281,
      "epoch": 7.599328737300436,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011205270349561269,
      "loss": 4.7322,
      "mean_token_accuracy": 0.24368294179439545,
      "num_tokens": 173826007.0,
      "step": 83775
    },
    {
      "entropy": 5.7080302238464355,
      "epoch": 7.59978229317852,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011203039286407422,
      "loss": 4.8303,
      "mean_token_accuracy": 0.22792435586452484,
      "num_tokens": 173836789.0,
      "step": 83780
    },
    {
      "entropy": 5.659118413925171,
      "epoch": 7.600235849056604,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011200808560278872,
      "loss": 4.7378,
      "mean_token_accuracy": 0.2507086515426636,
      "num_tokens": 173847232.0,
      "step": 83785
    },
    {
      "entropy": 5.688721704483032,
      "epoch": 7.600689404934688,
      "grad_norm": 1.375,
      "learning_rate": 0.0001119857817122176,
      "loss": 4.7958,
      "mean_token_accuracy": 0.23816125243902206,
      "num_tokens": 173857167.0,
      "step": 83790
    },
    {
      "entropy": 5.68041877746582,
      "epoch": 7.601142960812772,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011196348119282204,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2300206497311592,
      "num_tokens": 173866203.0,
      "step": 83795
    },
    {
      "entropy": 5.640534114837647,
      "epoch": 7.601596516690856,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011194118404506327,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2444537475705147,
      "num_tokens": 173877948.0,
      "step": 83800
    },
    {
      "entropy": 5.63453106880188,
      "epoch": 7.60205007256894,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011191889026940234,
      "loss": 4.7325,
      "mean_token_accuracy": 0.24428839683532716,
      "num_tokens": 173888270.0,
      "step": 83805
    },
    {
      "entropy": 5.78384838104248,
      "epoch": 7.602503628447025,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011189659986630032,
      "loss": 4.917,
      "mean_token_accuracy": 0.22283659726381302,
      "num_tokens": 173899310.0,
      "step": 83810
    },
    {
      "entropy": 5.730911016464233,
      "epoch": 7.602957184325109,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011187431283621819,
      "loss": 4.8889,
      "mean_token_accuracy": 0.2290959507226944,
      "num_tokens": 173909680.0,
      "step": 83815
    },
    {
      "entropy": 5.7374927520751955,
      "epoch": 7.603410740203193,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011185202917961676,
      "loss": 4.8966,
      "mean_token_accuracy": 0.22380589842796325,
      "num_tokens": 173920330.0,
      "step": 83820
    },
    {
      "entropy": 5.725018262863159,
      "epoch": 7.6038642960812775,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011182974889695702,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22990585416555404,
      "num_tokens": 173931541.0,
      "step": 83825
    },
    {
      "entropy": 5.657117080688477,
      "epoch": 7.604317851959362,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011180747198869964,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2275623068213463,
      "num_tokens": 173941939.0,
      "step": 83830
    },
    {
      "entropy": 5.6380524158477785,
      "epoch": 7.604771407837445,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011178519845530533,
      "loss": 4.6627,
      "mean_token_accuracy": 0.24702646285295488,
      "num_tokens": 173951558.0,
      "step": 83835
    },
    {
      "entropy": 5.668971395492553,
      "epoch": 7.60522496371553,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011176292829723473,
      "loss": 4.7883,
      "mean_token_accuracy": 0.23659078627824784,
      "num_tokens": 173962579.0,
      "step": 83840
    },
    {
      "entropy": 5.656063365936279,
      "epoch": 7.605678519593614,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011174066151494835,
      "loss": 4.7377,
      "mean_token_accuracy": 0.2422298714518547,
      "num_tokens": 173973326.0,
      "step": 83845
    },
    {
      "entropy": 5.7047051906585695,
      "epoch": 7.606132075471698,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011171839810890668,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2269509255886078,
      "num_tokens": 173983363.0,
      "step": 83850
    },
    {
      "entropy": 5.7009382247924805,
      "epoch": 7.606585631349782,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011169613807957009,
      "loss": 4.8172,
      "mean_token_accuracy": 0.24290809780359268,
      "num_tokens": 173994612.0,
      "step": 83855
    },
    {
      "entropy": 5.697957611083984,
      "epoch": 7.607039187227866,
      "grad_norm": 1.625,
      "learning_rate": 0.00011167388142739906,
      "loss": 4.7734,
      "mean_token_accuracy": 0.23940186500549315,
      "num_tokens": 174004158.0,
      "step": 83860
    },
    {
      "entropy": 5.715658569335938,
      "epoch": 7.607492743105951,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011165162815285378,
      "loss": 4.7894,
      "mean_token_accuracy": 0.24095505475997925,
      "num_tokens": 174013993.0,
      "step": 83865
    },
    {
      "entropy": 5.681912660598755,
      "epoch": 7.607946298984035,
      "grad_norm": 1.5,
      "learning_rate": 0.00011162937825639447,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22987115383148193,
      "num_tokens": 174023902.0,
      "step": 83870
    },
    {
      "entropy": 5.661837911605835,
      "epoch": 7.608399854862119,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001116071317384812,
      "loss": 4.7707,
      "mean_token_accuracy": 0.24123817980289458,
      "num_tokens": 174034831.0,
      "step": 83875
    },
    {
      "entropy": 5.740160608291626,
      "epoch": 7.6088534107402035,
      "grad_norm": 1.5,
      "learning_rate": 0.00011158488859957414,
      "loss": 4.863,
      "mean_token_accuracy": 0.22716025412082672,
      "num_tokens": 174045056.0,
      "step": 83880
    },
    {
      "entropy": 5.71494722366333,
      "epoch": 7.609306966618288,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011156264884013318,
      "loss": 4.855,
      "mean_token_accuracy": 0.23105367869138718,
      "num_tokens": 174055335.0,
      "step": 83885
    },
    {
      "entropy": 5.6987542629241945,
      "epoch": 7.609760522496371,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011154041246061822,
      "loss": 4.7954,
      "mean_token_accuracy": 0.23274642527103423,
      "num_tokens": 174065156.0,
      "step": 83890
    },
    {
      "entropy": 5.728152751922607,
      "epoch": 7.610214078374455,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011151817946148929,
      "loss": 4.8895,
      "mean_token_accuracy": 0.23102173805236817,
      "num_tokens": 174075457.0,
      "step": 83895
    },
    {
      "entropy": 5.74791989326477,
      "epoch": 7.61066763425254,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011149594984320606,
      "loss": 4.7975,
      "mean_token_accuracy": 0.2437887966632843,
      "num_tokens": 174086705.0,
      "step": 83900
    },
    {
      "entropy": 5.743889427185058,
      "epoch": 7.611121190130624,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011147372360622824,
      "loss": 4.874,
      "mean_token_accuracy": 0.23743629306554795,
      "num_tokens": 174097264.0,
      "step": 83905
    },
    {
      "entropy": 5.721566295623779,
      "epoch": 7.611574746008708,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011145150075101548,
      "loss": 4.8763,
      "mean_token_accuracy": 0.2335299700498581,
      "num_tokens": 174108094.0,
      "step": 83910
    },
    {
      "entropy": 5.63741626739502,
      "epoch": 7.6120283018867925,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011142928127802737,
      "loss": 4.738,
      "mean_token_accuracy": 0.24036391079425812,
      "num_tokens": 174118452.0,
      "step": 83915
    },
    {
      "entropy": 5.719272470474243,
      "epoch": 7.612481857764877,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00011140706518772344,
      "loss": 4.8585,
      "mean_token_accuracy": 0.24117291867733,
      "num_tokens": 174128932.0,
      "step": 83920
    },
    {
      "entropy": 5.715784358978271,
      "epoch": 7.612935413642961,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011138485248056299,
      "loss": 4.8196,
      "mean_token_accuracy": 0.22892747074365616,
      "num_tokens": 174138861.0,
      "step": 83925
    },
    {
      "entropy": 5.7307531356811525,
      "epoch": 7.613388969521045,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011136264315700557,
      "loss": 4.8426,
      "mean_token_accuracy": 0.23059200793504714,
      "num_tokens": 174149157.0,
      "step": 83930
    },
    {
      "entropy": 5.687130117416382,
      "epoch": 7.6138425253991295,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011134043721751037,
      "loss": 4.8142,
      "mean_token_accuracy": 0.23312708139419555,
      "num_tokens": 174160302.0,
      "step": 83935
    },
    {
      "entropy": 5.719525718688965,
      "epoch": 7.614296081277214,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011131823466253662,
      "loss": 4.7757,
      "mean_token_accuracy": 0.2396411955356598,
      "num_tokens": 174169933.0,
      "step": 83940
    },
    {
      "entropy": 5.641562795639038,
      "epoch": 7.614749637155297,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011129603549254353,
      "loss": 4.7225,
      "mean_token_accuracy": 0.23648441582918167,
      "num_tokens": 174180979.0,
      "step": 83945
    },
    {
      "entropy": 5.678743314743042,
      "epoch": 7.6152031930333814,
      "grad_norm": 1.765625,
      "learning_rate": 0.0001112738397079901,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2384340062737465,
      "num_tokens": 174190705.0,
      "step": 83950
    },
    {
      "entropy": 5.708654499053955,
      "epoch": 7.615656748911466,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011125164730933532,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23838768005371094,
      "num_tokens": 174201845.0,
      "step": 83955
    },
    {
      "entropy": 5.716143465042114,
      "epoch": 7.61611030478955,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011122945829703831,
      "loss": 4.825,
      "mean_token_accuracy": 0.23697344213724136,
      "num_tokens": 174212224.0,
      "step": 83960
    },
    {
      "entropy": 5.705365324020386,
      "epoch": 7.616563860667634,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011120727267155781,
      "loss": 4.764,
      "mean_token_accuracy": 0.23622593134641648,
      "num_tokens": 174223364.0,
      "step": 83965
    },
    {
      "entropy": 5.804047250747681,
      "epoch": 7.6170174165457185,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011118509043335267,
      "loss": 4.9593,
      "mean_token_accuracy": 0.22302047461271285,
      "num_tokens": 174234671.0,
      "step": 83970
    },
    {
      "entropy": 5.726512002944946,
      "epoch": 7.617470972423803,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011116291158288159,
      "loss": 4.8688,
      "mean_token_accuracy": 0.2355018749833107,
      "num_tokens": 174245278.0,
      "step": 83975
    },
    {
      "entropy": 5.7361198425292965,
      "epoch": 7.617924528301887,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011114073612060324,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2309002846479416,
      "num_tokens": 174254592.0,
      "step": 83980
    },
    {
      "entropy": 5.660242033004761,
      "epoch": 7.618378084179971,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00011111856404697623,
      "loss": 4.7505,
      "mean_token_accuracy": 0.23851866126060486,
      "num_tokens": 174264473.0,
      "step": 83985
    },
    {
      "entropy": 5.694737911224365,
      "epoch": 7.618831640058055,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00011109639536245902,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23568824231624602,
      "num_tokens": 174274960.0,
      "step": 83990
    },
    {
      "entropy": 5.728572225570678,
      "epoch": 7.619285195936139,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011107423006751018,
      "loss": 4.8289,
      "mean_token_accuracy": 0.2437902122735977,
      "num_tokens": 174286175.0,
      "step": 83995
    },
    {
      "entropy": 5.719342851638794,
      "epoch": 7.619738751814223,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011105206816258803,
      "loss": 4.7653,
      "mean_token_accuracy": 0.2411285623908043,
      "num_tokens": 174296199.0,
      "step": 84000
    },
    {
      "epoch": 7.619738751814223,
      "eval_entropy": 5.395252584548018,
      "eval_loss": 4.916695594787598,
      "eval_mean_token_accuracy": 0.23800491928649703,
      "eval_num_tokens": 174296199.0,
      "eval_runtime": 20.7225,
      "eval_samples_per_second": 1706.361,
      "eval_steps_per_second": 213.295,
      "step": 84000
    },
    {
      "entropy": 5.716356897354126,
      "epoch": 7.6201923076923075,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001110299096481509,
      "loss": 4.8257,
      "mean_token_accuracy": 0.23265095204114913,
      "num_tokens": 174306144.0,
      "step": 84005
    },
    {
      "entropy": 5.619827604293823,
      "epoch": 7.620645863570392,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011100775452465703,
      "loss": 4.7715,
      "mean_token_accuracy": 0.24110664278268815,
      "num_tokens": 174317452.0,
      "step": 84010
    },
    {
      "entropy": 5.698990392684936,
      "epoch": 7.621099419448476,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011098560279256456,
      "loss": 4.836,
      "mean_token_accuracy": 0.23400939255952835,
      "num_tokens": 174328214.0,
      "step": 84015
    },
    {
      "entropy": 5.678552293777466,
      "epoch": 7.62155297532656,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011096345445233164,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23424726873636245,
      "num_tokens": 174337932.0,
      "step": 84020
    },
    {
      "entropy": 5.702295398712158,
      "epoch": 7.6220065312046446,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001109413095044162,
      "loss": 4.7711,
      "mean_token_accuracy": 0.2410706862807274,
      "num_tokens": 174347851.0,
      "step": 84025
    },
    {
      "entropy": 5.727661752700806,
      "epoch": 7.622460087082729,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011091916794927638,
      "loss": 4.8695,
      "mean_token_accuracy": 0.229923315346241,
      "num_tokens": 174359197.0,
      "step": 84030
    },
    {
      "entropy": 5.682359457015991,
      "epoch": 7.622913642960813,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011089702978736998,
      "loss": 4.7535,
      "mean_token_accuracy": 0.23463456332683563,
      "num_tokens": 174370046.0,
      "step": 84035
    },
    {
      "entropy": 5.6895476341247555,
      "epoch": 7.623367198838897,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011087489501915479,
      "loss": 4.7677,
      "mean_token_accuracy": 0.24518459737300874,
      "num_tokens": 174379478.0,
      "step": 84040
    },
    {
      "entropy": 5.750435733795166,
      "epoch": 7.623820754716981,
      "grad_norm": 1.5,
      "learning_rate": 0.00011085276364508861,
      "loss": 4.9319,
      "mean_token_accuracy": 0.22658315896987916,
      "num_tokens": 174390398.0,
      "step": 84045
    },
    {
      "entropy": 5.629548645019531,
      "epoch": 7.624274310595065,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011083063566562912,
      "loss": 4.7271,
      "mean_token_accuracy": 0.24264998584985734,
      "num_tokens": 174401253.0,
      "step": 84050
    },
    {
      "entropy": 5.675577020645141,
      "epoch": 7.624727866473149,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011080851108123392,
      "loss": 4.7207,
      "mean_token_accuracy": 0.23718957155942916,
      "num_tokens": 174411738.0,
      "step": 84055
    },
    {
      "entropy": 5.693033933639526,
      "epoch": 7.6251814223512335,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011078638989236048,
      "loss": 4.8413,
      "mean_token_accuracy": 0.24327694177627562,
      "num_tokens": 174422214.0,
      "step": 84060
    },
    {
      "entropy": 5.713444614410401,
      "epoch": 7.625634978229318,
      "grad_norm": 1.625,
      "learning_rate": 0.00011076427209946638,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23435532450675964,
      "num_tokens": 174431547.0,
      "step": 84065
    },
    {
      "entropy": 5.7115990161895756,
      "epoch": 7.626088534107402,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011074215770300904,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2345958858728409,
      "num_tokens": 174441303.0,
      "step": 84070
    },
    {
      "entropy": 5.685244941711426,
      "epoch": 7.626542089985486,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011072004670344571,
      "loss": 4.8312,
      "mean_token_accuracy": 0.23625317215919495,
      "num_tokens": 174451816.0,
      "step": 84075
    },
    {
      "entropy": 5.65763292312622,
      "epoch": 7.626995645863571,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011069793910123365,
      "loss": 4.7987,
      "mean_token_accuracy": 0.23101042658090593,
      "num_tokens": 174461436.0,
      "step": 84080
    },
    {
      "entropy": 5.637791633605957,
      "epoch": 7.627449201741655,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011067583489683012,
      "loss": 4.7297,
      "mean_token_accuracy": 0.24713925868272782,
      "num_tokens": 174471906.0,
      "step": 84085
    },
    {
      "entropy": 5.7329143524169925,
      "epoch": 7.627902757619739,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011065373409069218,
      "loss": 4.8626,
      "mean_token_accuracy": 0.22759314775466918,
      "num_tokens": 174482798.0,
      "step": 84090
    },
    {
      "entropy": 5.751374387741089,
      "epoch": 7.628356313497823,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011063163668327684,
      "loss": 4.8573,
      "mean_token_accuracy": 0.22840225845575332,
      "num_tokens": 174493210.0,
      "step": 84095
    },
    {
      "entropy": 5.704670190811157,
      "epoch": 7.628809869375907,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001106095426750412,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22741565704345704,
      "num_tokens": 174503930.0,
      "step": 84100
    },
    {
      "entropy": 5.659931659698486,
      "epoch": 7.629263425253991,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001105874520664421,
      "loss": 4.7793,
      "mean_token_accuracy": 0.23354572653770447,
      "num_tokens": 174513337.0,
      "step": 84105
    },
    {
      "entropy": 5.669081830978394,
      "epoch": 7.629716981132075,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011056536485793642,
      "loss": 4.7521,
      "mean_token_accuracy": 0.24070166796445847,
      "num_tokens": 174523838.0,
      "step": 84110
    },
    {
      "entropy": 5.711907577514649,
      "epoch": 7.63017053701016,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011054328104998087,
      "loss": 4.8291,
      "mean_token_accuracy": 0.23827932626008988,
      "num_tokens": 174534319.0,
      "step": 84115
    },
    {
      "entropy": 5.605140733718872,
      "epoch": 7.630624092888244,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011052120064303217,
      "loss": 4.6618,
      "mean_token_accuracy": 0.2512822911143303,
      "num_tokens": 174545278.0,
      "step": 84120
    },
    {
      "entropy": 5.6320531368255615,
      "epoch": 7.631077648766328,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011049912363754699,
      "loss": 4.7873,
      "mean_token_accuracy": 0.2399717941880226,
      "num_tokens": 174555861.0,
      "step": 84125
    },
    {
      "entropy": 5.691706132888794,
      "epoch": 7.631531204644412,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011047705003398175,
      "loss": 4.8087,
      "mean_token_accuracy": 0.22785147726535798,
      "num_tokens": 174567241.0,
      "step": 84130
    },
    {
      "entropy": 5.570748376846313,
      "epoch": 7.631984760522497,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011045497983279312,
      "loss": 4.6811,
      "mean_token_accuracy": 0.24366651475429535,
      "num_tokens": 174577622.0,
      "step": 84135
    },
    {
      "entropy": 5.750010251998901,
      "epoch": 7.632438316400581,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011043291303443745,
      "loss": 4.9076,
      "mean_token_accuracy": 0.230992628633976,
      "num_tokens": 174588623.0,
      "step": 84140
    },
    {
      "entropy": 5.6856237888336185,
      "epoch": 7.632891872278664,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011041084963937106,
      "loss": 4.835,
      "mean_token_accuracy": 0.23836689293384553,
      "num_tokens": 174600054.0,
      "step": 84145
    },
    {
      "entropy": 5.718937826156616,
      "epoch": 7.6333454281567485,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011038878964805026,
      "loss": 4.8217,
      "mean_token_accuracy": 0.2393082708120346,
      "num_tokens": 174611141.0,
      "step": 84150
    },
    {
      "entropy": 5.691985273361206,
      "epoch": 7.633798984034833,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011036673306093123,
      "loss": 4.7618,
      "mean_token_accuracy": 0.23945507407188416,
      "num_tokens": 174621661.0,
      "step": 84155
    },
    {
      "entropy": 5.750765371322632,
      "epoch": 7.634252539912917,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00011034467987847002,
      "loss": 4.8894,
      "mean_token_accuracy": 0.23243104368448259,
      "num_tokens": 174632213.0,
      "step": 84160
    },
    {
      "entropy": 5.60904369354248,
      "epoch": 7.634706095791001,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00011032263010112295,
      "loss": 4.7311,
      "mean_token_accuracy": 0.24832914471626283,
      "num_tokens": 174643104.0,
      "step": 84165
    },
    {
      "entropy": 5.667706775665283,
      "epoch": 7.635159651669086,
      "grad_norm": 1.5,
      "learning_rate": 0.00011030058372934578,
      "loss": 4.7498,
      "mean_token_accuracy": 0.24222686290740966,
      "num_tokens": 174653837.0,
      "step": 84170
    },
    {
      "entropy": 5.7071085453033445,
      "epoch": 7.63561320754717,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011027854076359452,
      "loss": 4.7675,
      "mean_token_accuracy": 0.24243251830339432,
      "num_tokens": 174663631.0,
      "step": 84175
    },
    {
      "entropy": 5.6442484855651855,
      "epoch": 7.636066763425254,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00011025650120432501,
      "loss": 4.7495,
      "mean_token_accuracy": 0.23939675837755203,
      "num_tokens": 174674341.0,
      "step": 84180
    },
    {
      "entropy": 5.641862630844116,
      "epoch": 7.636520319303338,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011023446505199302,
      "loss": 4.7747,
      "mean_token_accuracy": 0.24040648490190505,
      "num_tokens": 174684659.0,
      "step": 84185
    },
    {
      "entropy": 5.6702690601348875,
      "epoch": 7.636973875181423,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011021243230705424,
      "loss": 4.7709,
      "mean_token_accuracy": 0.23261283189058304,
      "num_tokens": 174695448.0,
      "step": 84190
    },
    {
      "entropy": 5.7607893466949465,
      "epoch": 7.637427431059507,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011019040296996438,
      "loss": 4.9112,
      "mean_token_accuracy": 0.23219645470380784,
      "num_tokens": 174705689.0,
      "step": 84195
    },
    {
      "entropy": 5.69315185546875,
      "epoch": 7.63788098693759,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011016837704117902,
      "loss": 4.7566,
      "mean_token_accuracy": 0.2375543624162674,
      "num_tokens": 174716262.0,
      "step": 84200
    },
    {
      "entropy": 5.703664064407349,
      "epoch": 7.638334542815675,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011014635452115363,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23468830585479736,
      "num_tokens": 174726611.0,
      "step": 84205
    },
    {
      "entropy": 5.7106242179870605,
      "epoch": 7.638788098693759,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011012433541034364,
      "loss": 4.8345,
      "mean_token_accuracy": 0.23024053871631622,
      "num_tokens": 174736651.0,
      "step": 84210
    },
    {
      "entropy": 5.675400686264038,
      "epoch": 7.639241654571843,
      "grad_norm": 1.5,
      "learning_rate": 0.00011010231970920437,
      "loss": 4.7688,
      "mean_token_accuracy": 0.24421483278274536,
      "num_tokens": 174747301.0,
      "step": 84215
    },
    {
      "entropy": 5.676346635818481,
      "epoch": 7.639695210449927,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011008030741819117,
      "loss": 4.7702,
      "mean_token_accuracy": 0.2410260334610939,
      "num_tokens": 174757483.0,
      "step": 84220
    },
    {
      "entropy": 5.718687200546265,
      "epoch": 7.640148766328012,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00011005829853775918,
      "loss": 4.7964,
      "mean_token_accuracy": 0.2377586767077446,
      "num_tokens": 174767339.0,
      "step": 84225
    },
    {
      "entropy": 5.68795256614685,
      "epoch": 7.640602322206096,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011003629306836369,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2336046054959297,
      "num_tokens": 174777470.0,
      "step": 84230
    },
    {
      "entropy": 5.697873878479004,
      "epoch": 7.64105587808418,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011001429101045966,
      "loss": 4.8041,
      "mean_token_accuracy": 0.23794040828943253,
      "num_tokens": 174788022.0,
      "step": 84235
    },
    {
      "entropy": 5.662185430526733,
      "epoch": 7.6415094339622645,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010999229236450214,
      "loss": 4.7462,
      "mean_token_accuracy": 0.24300743639469147,
      "num_tokens": 174799027.0,
      "step": 84240
    },
    {
      "entropy": 5.664727401733399,
      "epoch": 7.641962989840349,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010997029713094609,
      "loss": 4.8183,
      "mean_token_accuracy": 0.23352073282003402,
      "num_tokens": 174809208.0,
      "step": 84245
    },
    {
      "entropy": 5.644902133941651,
      "epoch": 7.642416545718433,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010994830531024632,
      "loss": 4.6841,
      "mean_token_accuracy": 0.24417064487934112,
      "num_tokens": 174818735.0,
      "step": 84250
    },
    {
      "entropy": 5.685443258285522,
      "epoch": 7.642870101596516,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010992631690285762,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23981162309646606,
      "num_tokens": 174828138.0,
      "step": 84255
    },
    {
      "entropy": 5.676176977157593,
      "epoch": 7.643323657474601,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010990433190923481,
      "loss": 4.8627,
      "mean_token_accuracy": 0.23329456895589828,
      "num_tokens": 174838857.0,
      "step": 84260
    },
    {
      "entropy": 5.633644390106201,
      "epoch": 7.643777213352685,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010988235032983249,
      "loss": 4.7835,
      "mean_token_accuracy": 0.24406892508268357,
      "num_tokens": 174848870.0,
      "step": 84265
    },
    {
      "entropy": 5.752386093139648,
      "epoch": 7.644230769230769,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001098603721651052,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23568666875362396,
      "num_tokens": 174858786.0,
      "step": 84270
    },
    {
      "entropy": 5.651988744735718,
      "epoch": 7.644684325108853,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010983839741550755,
      "loss": 4.7616,
      "mean_token_accuracy": 0.2459567368030548,
      "num_tokens": 174869262.0,
      "step": 84275
    },
    {
      "entropy": 5.698151636123657,
      "epoch": 7.645137880986938,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010981642608149389,
      "loss": 4.8508,
      "mean_token_accuracy": 0.23041881024837493,
      "num_tokens": 174878961.0,
      "step": 84280
    },
    {
      "entropy": 5.63091287612915,
      "epoch": 7.645591436865022,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010979445816351865,
      "loss": 4.7234,
      "mean_token_accuracy": 0.23808145076036452,
      "num_tokens": 174889966.0,
      "step": 84285
    },
    {
      "entropy": 5.770605039596558,
      "epoch": 7.646044992743106,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010977249366203606,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2342644065618515,
      "num_tokens": 174900060.0,
      "step": 84290
    },
    {
      "entropy": 5.64465298652649,
      "epoch": 7.6464985486211905,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010975053257750048,
      "loss": 4.7191,
      "mean_token_accuracy": 0.2435617282986641,
      "num_tokens": 174909998.0,
      "step": 84295
    },
    {
      "entropy": 5.670325708389282,
      "epoch": 7.646952104499274,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010972857491036598,
      "loss": 4.8075,
      "mean_token_accuracy": 0.2299002692103386,
      "num_tokens": 174920421.0,
      "step": 84300
    },
    {
      "entropy": 5.642437267303467,
      "epoch": 7.647405660377358,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010970662066108666,
      "loss": 4.7088,
      "mean_token_accuracy": 0.2452804610133171,
      "num_tokens": 174931417.0,
      "step": 84305
    },
    {
      "entropy": 5.67553825378418,
      "epoch": 7.647859216255442,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010968466983011655,
      "loss": 4.7494,
      "mean_token_accuracy": 0.23738245964050292,
      "num_tokens": 174941991.0,
      "step": 84310
    },
    {
      "entropy": 5.666906213760376,
      "epoch": 7.648312772133527,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010966272241790959,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2358412966132164,
      "num_tokens": 174952677.0,
      "step": 84315
    },
    {
      "entropy": 5.666139602661133,
      "epoch": 7.648766328011611,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010964077842491968,
      "loss": 4.7735,
      "mean_token_accuracy": 0.24274397045373916,
      "num_tokens": 174963701.0,
      "step": 84320
    },
    {
      "entropy": 5.695900392532349,
      "epoch": 7.649219883889695,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010961883785160054,
      "loss": 4.8873,
      "mean_token_accuracy": 0.23083521872758866,
      "num_tokens": 174975086.0,
      "step": 84325
    },
    {
      "entropy": 5.664107418060302,
      "epoch": 7.6496734397677795,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010959690069840605,
      "loss": 4.7397,
      "mean_token_accuracy": 0.24333584010601045,
      "num_tokens": 174985604.0,
      "step": 84330
    },
    {
      "entropy": 5.650161218643189,
      "epoch": 7.650126995645864,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010957496696578978,
      "loss": 4.7593,
      "mean_token_accuracy": 0.24117717295885086,
      "num_tokens": 174995774.0,
      "step": 84335
    },
    {
      "entropy": 5.783895063400268,
      "epoch": 7.650580551523948,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00010955303665420535,
      "loss": 4.9005,
      "mean_token_accuracy": 0.23025432080030442,
      "num_tokens": 175006059.0,
      "step": 84340
    },
    {
      "entropy": 5.683767318725586,
      "epoch": 7.651034107402032,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010953110976410632,
      "loss": 4.8195,
      "mean_token_accuracy": 0.24037139415740966,
      "num_tokens": 175016297.0,
      "step": 84345
    },
    {
      "entropy": 5.670510292053223,
      "epoch": 7.6514876632801165,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010950918629594607,
      "loss": 4.7278,
      "mean_token_accuracy": 0.23736998438835144,
      "num_tokens": 175026571.0,
      "step": 84350
    },
    {
      "entropy": 5.6121964931488035,
      "epoch": 7.6519412191582,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010948726625017804,
      "loss": 4.7295,
      "mean_token_accuracy": 0.24114463478326797,
      "num_tokens": 175036569.0,
      "step": 84355
    },
    {
      "entropy": 5.646055126190186,
      "epoch": 7.652394775036284,
      "grad_norm": 1.5,
      "learning_rate": 0.00010946534962725543,
      "loss": 4.7655,
      "mean_token_accuracy": 0.24008501917123795,
      "num_tokens": 175047190.0,
      "step": 84360
    },
    {
      "entropy": 5.64576096534729,
      "epoch": 7.6528483309143684,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00010944343642763168,
      "loss": 4.7773,
      "mean_token_accuracy": 0.24177418649196625,
      "num_tokens": 175056413.0,
      "step": 84365
    },
    {
      "entropy": 5.6406783103942875,
      "epoch": 7.653301886792453,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010942152665175984,
      "loss": 4.6791,
      "mean_token_accuracy": 0.24701838791370392,
      "num_tokens": 175066613.0,
      "step": 84370
    },
    {
      "entropy": 5.64328899383545,
      "epoch": 7.653755442670537,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010939962030009299,
      "loss": 4.7728,
      "mean_token_accuracy": 0.24655575156211854,
      "num_tokens": 175075985.0,
      "step": 84375
    },
    {
      "entropy": 5.6312110900878904,
      "epoch": 7.654208998548621,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010937771737308421,
      "loss": 4.7134,
      "mean_token_accuracy": 0.24653242230415345,
      "num_tokens": 175086407.0,
      "step": 84380
    },
    {
      "entropy": 5.6621095657348635,
      "epoch": 7.6546625544267055,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010935581787118644,
      "loss": 4.7723,
      "mean_token_accuracy": 0.2391234666109085,
      "num_tokens": 175097638.0,
      "step": 84385
    },
    {
      "entropy": 5.666798639297485,
      "epoch": 7.65511611030479,
      "grad_norm": 1.625,
      "learning_rate": 0.00010933392179485257,
      "loss": 4.7656,
      "mean_token_accuracy": 0.2543972462415695,
      "num_tokens": 175107962.0,
      "step": 84390
    },
    {
      "entropy": 5.714268350601197,
      "epoch": 7.655569666182874,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010931202914453536,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22758493721485137,
      "num_tokens": 175118931.0,
      "step": 84395
    },
    {
      "entropy": 5.672896862030029,
      "epoch": 7.656023222060957,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010929013992068767,
      "loss": 4.715,
      "mean_token_accuracy": 0.2389421820640564,
      "num_tokens": 175127994.0,
      "step": 84400
    },
    {
      "entropy": 5.662539339065551,
      "epoch": 7.656476777939043,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001092682541237621,
      "loss": 4.757,
      "mean_token_accuracy": 0.23394109308719635,
      "num_tokens": 175137353.0,
      "step": 84405
    },
    {
      "entropy": 5.742432641983032,
      "epoch": 7.656930333817126,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001092463717542113,
      "loss": 4.8703,
      "mean_token_accuracy": 0.23407489508390428,
      "num_tokens": 175148380.0,
      "step": 84410
    },
    {
      "entropy": 5.663908004760742,
      "epoch": 7.65738388969521,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010922449281248773,
      "loss": 4.8188,
      "mean_token_accuracy": 0.23685349822044371,
      "num_tokens": 175158815.0,
      "step": 84415
    },
    {
      "entropy": 5.757571697235107,
      "epoch": 7.6578374455732945,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010920261729904391,
      "loss": 4.9229,
      "mean_token_accuracy": 0.22234316766262055,
      "num_tokens": 175168492.0,
      "step": 84420
    },
    {
      "entropy": 5.731603765487671,
      "epoch": 7.658291001451379,
      "grad_norm": 1.5,
      "learning_rate": 0.00010918074521433221,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22994445264339447,
      "num_tokens": 175178616.0,
      "step": 84425
    },
    {
      "entropy": 5.718447256088257,
      "epoch": 7.658744557329463,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010915887655880491,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22968159168958663,
      "num_tokens": 175188045.0,
      "step": 84430
    },
    {
      "entropy": 5.697771072387695,
      "epoch": 7.659198113207547,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010913701133291434,
      "loss": 4.7586,
      "mean_token_accuracy": 0.24218652099370958,
      "num_tokens": 175198471.0,
      "step": 84435
    },
    {
      "entropy": 5.706306076049804,
      "epoch": 7.6596516690856316,
      "grad_norm": 1.5,
      "learning_rate": 0.00010911514953711269,
      "loss": 4.8227,
      "mean_token_accuracy": 0.23768647760152817,
      "num_tokens": 175208494.0,
      "step": 84440
    },
    {
      "entropy": 5.747741603851319,
      "epoch": 7.660105224963716,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010909329117185198,
      "loss": 4.7867,
      "mean_token_accuracy": 0.2398211434483528,
      "num_tokens": 175218170.0,
      "step": 84445
    },
    {
      "entropy": 5.66434416770935,
      "epoch": 7.6605587808418,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010907143623758428,
      "loss": 4.7074,
      "mean_token_accuracy": 0.25354094952344897,
      "num_tokens": 175228208.0,
      "step": 84450
    },
    {
      "entropy": 5.696089267730713,
      "epoch": 7.6610123367198835,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010904958473476159,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2297880932688713,
      "num_tokens": 175239188.0,
      "step": 84455
    },
    {
      "entropy": 5.693049335479737,
      "epoch": 7.661465892597968,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001090277366638357,
      "loss": 4.798,
      "mean_token_accuracy": 0.23125956058502198,
      "num_tokens": 175249244.0,
      "step": 84460
    },
    {
      "entropy": 5.737937355041504,
      "epoch": 7.661919448476052,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010900589202525855,
      "loss": 4.917,
      "mean_token_accuracy": 0.23161399066448213,
      "num_tokens": 175259674.0,
      "step": 84465
    },
    {
      "entropy": 5.707055377960205,
      "epoch": 7.662373004354136,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001089840508194819,
      "loss": 4.7965,
      "mean_token_accuracy": 0.22958102971315383,
      "num_tokens": 175270508.0,
      "step": 84470
    },
    {
      "entropy": 5.656643056869507,
      "epoch": 7.6628265602322205,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010896221304695736,
      "loss": 4.7219,
      "mean_token_accuracy": 0.23851592391729354,
      "num_tokens": 175280437.0,
      "step": 84475
    },
    {
      "entropy": 5.657727146148682,
      "epoch": 7.663280116110305,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010894037870813657,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2391742929816246,
      "num_tokens": 175290779.0,
      "step": 84480
    },
    {
      "entropy": 5.644654083251953,
      "epoch": 7.663733671988389,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010891854780347106,
      "loss": 4.7581,
      "mean_token_accuracy": 0.24241146743297576,
      "num_tokens": 175300855.0,
      "step": 84485
    },
    {
      "entropy": 5.634581899642944,
      "epoch": 7.664187227866473,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001088967203334123,
      "loss": 4.7434,
      "mean_token_accuracy": 0.24260884821414946,
      "num_tokens": 175310464.0,
      "step": 84490
    },
    {
      "entropy": 5.65996527671814,
      "epoch": 7.664640783744558,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010887489629841166,
      "loss": 4.7498,
      "mean_token_accuracy": 0.2464767038822174,
      "num_tokens": 175320903.0,
      "step": 84495
    },
    {
      "entropy": 5.7757820129394535,
      "epoch": 7.665094339622642,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010885307569892053,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22783162593841552,
      "num_tokens": 175330948.0,
      "step": 84500
    },
    {
      "entropy": 5.727266025543213,
      "epoch": 7.665547895500726,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010883125853539012,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2360101118683815,
      "num_tokens": 175341906.0,
      "step": 84505
    },
    {
      "entropy": 5.688679647445679,
      "epoch": 7.6660014513788095,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010880944480827169,
      "loss": 4.789,
      "mean_token_accuracy": 0.23994005024433135,
      "num_tokens": 175352146.0,
      "step": 84510
    },
    {
      "entropy": 5.75362000465393,
      "epoch": 7.666455007256894,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010878763451801621,
      "loss": 4.867,
      "mean_token_accuracy": 0.23396522104740142,
      "num_tokens": 175361916.0,
      "step": 84515
    },
    {
      "entropy": 5.64525408744812,
      "epoch": 7.666908563134978,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010876582766507487,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2437494844198227,
      "num_tokens": 175371635.0,
      "step": 84520
    },
    {
      "entropy": 5.700886678695679,
      "epoch": 7.667362119013062,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010874402424989854,
      "loss": 4.788,
      "mean_token_accuracy": 0.24228933900594712,
      "num_tokens": 175382566.0,
      "step": 84525
    },
    {
      "entropy": 5.657766246795655,
      "epoch": 7.667815674891147,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001087222242729381,
      "loss": 4.7323,
      "mean_token_accuracy": 0.24158019870519637,
      "num_tokens": 175392659.0,
      "step": 84530
    },
    {
      "entropy": 5.626302576065063,
      "epoch": 7.668269230769231,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001087004277346445,
      "loss": 4.7481,
      "mean_token_accuracy": 0.24388831555843354,
      "num_tokens": 175403813.0,
      "step": 84535
    },
    {
      "entropy": 5.691702938079834,
      "epoch": 7.668722786647315,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010867863463546842,
      "loss": 4.817,
      "mean_token_accuracy": 0.23388817757368088,
      "num_tokens": 175413894.0,
      "step": 84540
    },
    {
      "entropy": 5.707453870773316,
      "epoch": 7.669176342525399,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010865684497586059,
      "loss": 4.793,
      "mean_token_accuracy": 0.23590748608112336,
      "num_tokens": 175424705.0,
      "step": 84545
    },
    {
      "entropy": 5.697444820404053,
      "epoch": 7.669629898403484,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010863505875627158,
      "loss": 4.8603,
      "mean_token_accuracy": 0.2356006160378456,
      "num_tokens": 175435780.0,
      "step": 84550
    },
    {
      "entropy": 5.68165135383606,
      "epoch": 7.670083454281567,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010861327597715195,
      "loss": 4.8227,
      "mean_token_accuracy": 0.23808947205543518,
      "num_tokens": 175445027.0,
      "step": 84555
    },
    {
      "entropy": 5.743770694732666,
      "epoch": 7.670537010159651,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010859149663895216,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23892529904842377,
      "num_tokens": 175454808.0,
      "step": 84560
    },
    {
      "entropy": 5.689342498779297,
      "epoch": 7.6709905660377355,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001085697207421226,
      "loss": 4.7987,
      "mean_token_accuracy": 0.23145609647035598,
      "num_tokens": 175465608.0,
      "step": 84565
    },
    {
      "entropy": 5.7051599502563475,
      "epoch": 7.67144412191582,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010854794828711364,
      "loss": 4.7894,
      "mean_token_accuracy": 0.23487520962953568,
      "num_tokens": 175476227.0,
      "step": 84570
    },
    {
      "entropy": 5.6758626937866214,
      "epoch": 7.671897677793904,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010852617927437555,
      "loss": 4.7557,
      "mean_token_accuracy": 0.24391781091690062,
      "num_tokens": 175486187.0,
      "step": 84575
    },
    {
      "entropy": 5.636924982070923,
      "epoch": 7.672351233671988,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010850441370435851,
      "loss": 4.8274,
      "mean_token_accuracy": 0.2301945224404335,
      "num_tokens": 175496262.0,
      "step": 84580
    },
    {
      "entropy": 5.704106950759888,
      "epoch": 7.672804789550073,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001084826515775126,
      "loss": 4.8879,
      "mean_token_accuracy": 0.23502283841371535,
      "num_tokens": 175508499.0,
      "step": 84585
    },
    {
      "entropy": 5.688471460342408,
      "epoch": 7.673258345428157,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001084608928942879,
      "loss": 4.8363,
      "mean_token_accuracy": 0.23695410639047623,
      "num_tokens": 175518640.0,
      "step": 84590
    },
    {
      "entropy": 5.681609058380127,
      "epoch": 7.673711901306241,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010843913765513439,
      "loss": 4.7962,
      "mean_token_accuracy": 0.24094754159450532,
      "num_tokens": 175528547.0,
      "step": 84595
    },
    {
      "entropy": 5.676098299026489,
      "epoch": 7.674165457184325,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010841738586050192,
      "loss": 4.7384,
      "mean_token_accuracy": 0.23355382233858107,
      "num_tokens": 175539081.0,
      "step": 84600
    },
    {
      "entropy": 5.723686647415161,
      "epoch": 7.67461901306241,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010839563751084039,
      "loss": 4.8363,
      "mean_token_accuracy": 0.23156394213438034,
      "num_tokens": 175548273.0,
      "step": 84605
    },
    {
      "entropy": 5.651850605010987,
      "epoch": 7.675072568940493,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001083738926065995,
      "loss": 4.7236,
      "mean_token_accuracy": 0.24107781499624253,
      "num_tokens": 175558612.0,
      "step": 84610
    },
    {
      "entropy": 5.649007272720337,
      "epoch": 7.675526124818577,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010835215114822902,
      "loss": 4.7527,
      "mean_token_accuracy": 0.23077468127012252,
      "num_tokens": 175569438.0,
      "step": 84615
    },
    {
      "entropy": 5.550062036514282,
      "epoch": 7.675979680696662,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001083304131361785,
      "loss": 4.6888,
      "mean_token_accuracy": 0.2402544856071472,
      "num_tokens": 175579646.0,
      "step": 84620
    },
    {
      "entropy": 5.672089910507202,
      "epoch": 7.676433236574746,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00010830867857089742,
      "loss": 4.7647,
      "mean_token_accuracy": 0.23705214262008667,
      "num_tokens": 175590864.0,
      "step": 84625
    },
    {
      "entropy": 5.6443249702453615,
      "epoch": 7.67688679245283,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010828694745283544,
      "loss": 4.771,
      "mean_token_accuracy": 0.24346132427453995,
      "num_tokens": 175601233.0,
      "step": 84630
    },
    {
      "entropy": 5.697715520858765,
      "epoch": 7.677340348330914,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010826521978244186,
      "loss": 4.7915,
      "mean_token_accuracy": 0.2443598836660385,
      "num_tokens": 175610800.0,
      "step": 84635
    },
    {
      "entropy": 5.724925518035889,
      "epoch": 7.677793904208999,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010824349556016602,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23317067921161652,
      "num_tokens": 175620386.0,
      "step": 84640
    },
    {
      "entropy": 5.685139179229736,
      "epoch": 7.678247460087083,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001082217747864572,
      "loss": 4.7745,
      "mean_token_accuracy": 0.23624382317066192,
      "num_tokens": 175630054.0,
      "step": 84645
    },
    {
      "entropy": 5.652391338348389,
      "epoch": 7.678701015965167,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010820005746176457,
      "loss": 4.776,
      "mean_token_accuracy": 0.24294458627700805,
      "num_tokens": 175640291.0,
      "step": 84650
    },
    {
      "entropy": 5.686648654937744,
      "epoch": 7.6791545718432515,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010817834358653725,
      "loss": 4.7808,
      "mean_token_accuracy": 0.24433837234973907,
      "num_tokens": 175650403.0,
      "step": 84655
    },
    {
      "entropy": 5.698321580886841,
      "epoch": 7.679608127721336,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010815663316122423,
      "loss": 4.8609,
      "mean_token_accuracy": 0.23496873527765275,
      "num_tokens": 175661945.0,
      "step": 84660
    },
    {
      "entropy": 5.668729496002197,
      "epoch": 7.680061683599419,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010813492618627463,
      "loss": 4.7685,
      "mean_token_accuracy": 0.23636545538902282,
      "num_tokens": 175672806.0,
      "step": 84665
    },
    {
      "entropy": 5.741792154312134,
      "epoch": 7.680515239477503,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010811322266213728,
      "loss": 4.8508,
      "mean_token_accuracy": 0.23820461332798004,
      "num_tokens": 175683032.0,
      "step": 84670
    },
    {
      "entropy": 5.643901014328003,
      "epoch": 7.680968795355588,
      "grad_norm": 1.5625,
      "learning_rate": 0.000108091522589261,
      "loss": 4.7315,
      "mean_token_accuracy": 0.2416547045111656,
      "num_tokens": 175693147.0,
      "step": 84675
    },
    {
      "entropy": 5.678544616699218,
      "epoch": 7.681422351233672,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010806982596809459,
      "loss": 4.7974,
      "mean_token_accuracy": 0.2478821262717247,
      "num_tokens": 175703783.0,
      "step": 84680
    },
    {
      "entropy": 5.655767774581909,
      "epoch": 7.681875907111756,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010804813279908668,
      "loss": 4.7855,
      "mean_token_accuracy": 0.23910187631845475,
      "num_tokens": 175714733.0,
      "step": 84685
    },
    {
      "entropy": 5.706062507629395,
      "epoch": 7.68232946298984,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010802644308268594,
      "loss": 4.7563,
      "mean_token_accuracy": 0.24500386714935302,
      "num_tokens": 175724224.0,
      "step": 84690
    },
    {
      "entropy": 5.677576494216919,
      "epoch": 7.682783018867925,
      "grad_norm": 1.78125,
      "learning_rate": 0.00010800475681934086,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2381932720541954,
      "num_tokens": 175733320.0,
      "step": 84695
    },
    {
      "entropy": 5.723559951782226,
      "epoch": 7.683236574746009,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010798307400950003,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22073474973440171,
      "num_tokens": 175743805.0,
      "step": 84700
    },
    {
      "entropy": 5.642357683181762,
      "epoch": 7.683690130624093,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010796139465361177,
      "loss": 4.7352,
      "mean_token_accuracy": 0.2493190959095955,
      "num_tokens": 175753579.0,
      "step": 84705
    },
    {
      "entropy": 5.710572195053101,
      "epoch": 7.684143686502177,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010793971875212444,
      "loss": 4.7558,
      "mean_token_accuracy": 0.2392566680908203,
      "num_tokens": 175762535.0,
      "step": 84710
    },
    {
      "entropy": 5.669097757339477,
      "epoch": 7.684597242380261,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010791804630548632,
      "loss": 4.8397,
      "mean_token_accuracy": 0.22807037383317946,
      "num_tokens": 175773139.0,
      "step": 84715
    },
    {
      "entropy": 5.674318265914917,
      "epoch": 7.685050798258345,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010789637731414556,
      "loss": 4.7594,
      "mean_token_accuracy": 0.24798588305711747,
      "num_tokens": 175783285.0,
      "step": 84720
    },
    {
      "entropy": 5.776847314834595,
      "epoch": 7.685504354136429,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010787471177855028,
      "loss": 4.9102,
      "mean_token_accuracy": 0.23121562004089355,
      "num_tokens": 175793683.0,
      "step": 84725
    },
    {
      "entropy": 5.671490955352783,
      "epoch": 7.685957910014514,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010785304969914851,
      "loss": 4.7279,
      "mean_token_accuracy": 0.24427422136068344,
      "num_tokens": 175804412.0,
      "step": 84730
    },
    {
      "entropy": 5.661520099639892,
      "epoch": 7.686411465892598,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010783139107638835,
      "loss": 4.7213,
      "mean_token_accuracy": 0.24109646081924438,
      "num_tokens": 175813654.0,
      "step": 84735
    },
    {
      "entropy": 5.69855751991272,
      "epoch": 7.686865021770682,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010780973591071759,
      "loss": 4.8556,
      "mean_token_accuracy": 0.2324543073773384,
      "num_tokens": 175824523.0,
      "step": 84740
    },
    {
      "entropy": 5.694000482559204,
      "epoch": 7.6873185776487665,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010778808420258409,
      "loss": 4.8196,
      "mean_token_accuracy": 0.23649154007434844,
      "num_tokens": 175833971.0,
      "step": 84745
    },
    {
      "entropy": 5.669891691207885,
      "epoch": 7.687772133526851,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010776643595243565,
      "loss": 4.7693,
      "mean_token_accuracy": 0.2444642722606659,
      "num_tokens": 175845201.0,
      "step": 84750
    },
    {
      "entropy": 5.665609884262085,
      "epoch": 7.688225689404935,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010774479116071992,
      "loss": 4.7499,
      "mean_token_accuracy": 0.24438760280609131,
      "num_tokens": 175855557.0,
      "step": 84755
    },
    {
      "entropy": 5.636026382446289,
      "epoch": 7.688679245283019,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010772314982788453,
      "loss": 4.7267,
      "mean_token_accuracy": 0.24740840196609498,
      "num_tokens": 175866155.0,
      "step": 84760
    },
    {
      "entropy": 5.6466669082641605,
      "epoch": 7.689132801161103,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010770151195437697,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23775675296783447,
      "num_tokens": 175876731.0,
      "step": 84765
    },
    {
      "entropy": 5.734694719314575,
      "epoch": 7.689586357039187,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010767987754064482,
      "loss": 4.9152,
      "mean_token_accuracy": 0.22412612438201904,
      "num_tokens": 175887702.0,
      "step": 84770
    },
    {
      "entropy": 5.694736051559448,
      "epoch": 7.690039912917271,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010765824658713544,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23280410766601561,
      "num_tokens": 175898007.0,
      "step": 84775
    },
    {
      "entropy": 5.685818243026733,
      "epoch": 7.6904934687953554,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001076366190942962,
      "loss": 4.8072,
      "mean_token_accuracy": 0.2339922934770584,
      "num_tokens": 175908559.0,
      "step": 84780
    },
    {
      "entropy": 5.695411634445191,
      "epoch": 7.69094702467344,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010761499506257428,
      "loss": 4.8713,
      "mean_token_accuracy": 0.22952475249767304,
      "num_tokens": 175918148.0,
      "step": 84785
    },
    {
      "entropy": 5.703229904174805,
      "epoch": 7.691400580551524,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010759337449241696,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22995987683534622,
      "num_tokens": 175929424.0,
      "step": 84790
    },
    {
      "entropy": 5.725583457946778,
      "epoch": 7.691854136429608,
      "grad_norm": 1.5,
      "learning_rate": 0.00010757175738427122,
      "loss": 4.865,
      "mean_token_accuracy": 0.23501451909542084,
      "num_tokens": 175940062.0,
      "step": 84795
    },
    {
      "entropy": 5.799626398086548,
      "epoch": 7.6923076923076925,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010755014373858429,
      "loss": 4.8647,
      "mean_token_accuracy": 0.2312662795186043,
      "num_tokens": 175950800.0,
      "step": 84800
    },
    {
      "entropy": 5.703770017623901,
      "epoch": 7.692761248185777,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010752853355580306,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23295449763536452,
      "num_tokens": 175961821.0,
      "step": 84805
    },
    {
      "entropy": 5.7118665218353275,
      "epoch": 7.693214804063861,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010750692683637442,
      "loss": 4.7992,
      "mean_token_accuracy": 0.24035113751888276,
      "num_tokens": 175972734.0,
      "step": 84810
    },
    {
      "entropy": 5.626008701324463,
      "epoch": 7.693668359941945,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001074853235807452,
      "loss": 4.72,
      "mean_token_accuracy": 0.24448170214891435,
      "num_tokens": 175982894.0,
      "step": 84815
    },
    {
      "entropy": 5.689994430541992,
      "epoch": 7.694121915820029,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001074637237893622,
      "loss": 4.7668,
      "mean_token_accuracy": 0.2398929476737976,
      "num_tokens": 175993250.0,
      "step": 84820
    },
    {
      "entropy": 5.723377084732055,
      "epoch": 7.694575471698113,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001074421274626721,
      "loss": 4.8686,
      "mean_token_accuracy": 0.23268771320581436,
      "num_tokens": 176003266.0,
      "step": 84825
    },
    {
      "entropy": 5.727598714828491,
      "epoch": 7.695029027576197,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010742053460112139,
      "loss": 4.7776,
      "mean_token_accuracy": 0.24197258949279785,
      "num_tokens": 176013440.0,
      "step": 84830
    },
    {
      "entropy": 5.679803991317749,
      "epoch": 7.6954825834542815,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010739894520515684,
      "loss": 4.803,
      "mean_token_accuracy": 0.2343592882156372,
      "num_tokens": 176023383.0,
      "step": 84835
    },
    {
      "entropy": 5.731428623199463,
      "epoch": 7.695936139332366,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010737735927522476,
      "loss": 4.8166,
      "mean_token_accuracy": 0.2350016087293625,
      "num_tokens": 176033512.0,
      "step": 84840
    },
    {
      "entropy": 5.684135770797729,
      "epoch": 7.69638969521045,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010735577681177166,
      "loss": 4.8246,
      "mean_token_accuracy": 0.22937773615121843,
      "num_tokens": 176043321.0,
      "step": 84845
    },
    {
      "entropy": 5.666730833053589,
      "epoch": 7.696843251088534,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001073341978152438,
      "loss": 4.7779,
      "mean_token_accuracy": 0.2388827070593834,
      "num_tokens": 176053878.0,
      "step": 84850
    },
    {
      "entropy": 5.747332906723022,
      "epoch": 7.6972968069666186,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010731262228608742,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22345561981201173,
      "num_tokens": 176064440.0,
      "step": 84855
    },
    {
      "entropy": 5.71242241859436,
      "epoch": 7.697750362844703,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010729105022474878,
      "loss": 4.7696,
      "mean_token_accuracy": 0.23986511379480363,
      "num_tokens": 176074507.0,
      "step": 84860
    },
    {
      "entropy": 5.6567240238189695,
      "epoch": 7.698203918722786,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010726948163167391,
      "loss": 4.7742,
      "mean_token_accuracy": 0.23667015582323075,
      "num_tokens": 176085141.0,
      "step": 84865
    },
    {
      "entropy": 5.691426706314087,
      "epoch": 7.6986574746008705,
      "grad_norm": 1.65625,
      "learning_rate": 0.00010724791650730894,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23601479530334474,
      "num_tokens": 176095615.0,
      "step": 84870
    },
    {
      "entropy": 5.666571521759034,
      "epoch": 7.699111030478955,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001072263548520998,
      "loss": 4.7746,
      "mean_token_accuracy": 0.2428348422050476,
      "num_tokens": 176105684.0,
      "step": 84875
    },
    {
      "entropy": 5.703186750411987,
      "epoch": 7.699564586357039,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001072047966664924,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22172591984272003,
      "num_tokens": 176117143.0,
      "step": 84880
    },
    {
      "entropy": 5.679004144668579,
      "epoch": 7.700018142235123,
      "grad_norm": 1.625,
      "learning_rate": 0.0001071832419509326,
      "loss": 4.773,
      "mean_token_accuracy": 0.23964154720306396,
      "num_tokens": 176127599.0,
      "step": 84885
    },
    {
      "entropy": 5.6983719825744625,
      "epoch": 7.7004716981132075,
      "grad_norm": 1.375,
      "learning_rate": 0.00010716169070586608,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2283088117837906,
      "num_tokens": 176138772.0,
      "step": 84890
    },
    {
      "entropy": 5.649540519714355,
      "epoch": 7.700925253991292,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00010714014293173857,
      "loss": 4.7615,
      "mean_token_accuracy": 0.24315565675497056,
      "num_tokens": 176148389.0,
      "step": 84895
    },
    {
      "entropy": 5.706053113937378,
      "epoch": 7.701378809869376,
      "grad_norm": 1.6875,
      "learning_rate": 0.00010711859862899562,
      "loss": 4.7655,
      "mean_token_accuracy": 0.24183543175458908,
      "num_tokens": 176158368.0,
      "step": 84900
    },
    {
      "entropy": 5.712331056594849,
      "epoch": 7.70183236574746,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001070970577980829,
      "loss": 4.801,
      "mean_token_accuracy": 0.2332739219069481,
      "num_tokens": 176168277.0,
      "step": 84905
    },
    {
      "entropy": 5.71025652885437,
      "epoch": 7.702285921625545,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001070755204394458,
      "loss": 4.7852,
      "mean_token_accuracy": 0.24188446551561354,
      "num_tokens": 176179167.0,
      "step": 84910
    },
    {
      "entropy": 5.791369867324829,
      "epoch": 7.702739477503629,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010705398655352974,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22751090973615645,
      "num_tokens": 176190536.0,
      "step": 84915
    },
    {
      "entropy": 5.6414146900177,
      "epoch": 7.703193033381712,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010703245614078003,
      "loss": 4.6959,
      "mean_token_accuracy": 0.24044378995895385,
      "num_tokens": 176200319.0,
      "step": 84920
    },
    {
      "entropy": 5.667777872085571,
      "epoch": 7.7036465892597965,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010701092920164195,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2398851126432419,
      "num_tokens": 176211018.0,
      "step": 84925
    },
    {
      "entropy": 5.653370094299317,
      "epoch": 7.704100145137881,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010698940573656066,
      "loss": 4.7722,
      "mean_token_accuracy": 0.24018557220697404,
      "num_tokens": 176221990.0,
      "step": 84930
    },
    {
      "entropy": 5.687936544418335,
      "epoch": 7.704553701015965,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001069678857459812,
      "loss": 4.798,
      "mean_token_accuracy": 0.23653199672698974,
      "num_tokens": 176232706.0,
      "step": 84935
    },
    {
      "entropy": 5.75677261352539,
      "epoch": 7.705007256894049,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010694636923034874,
      "loss": 4.879,
      "mean_token_accuracy": 0.2372664123773575,
      "num_tokens": 176244149.0,
      "step": 84940
    },
    {
      "entropy": 5.693623971939087,
      "epoch": 7.705460812772134,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001069248561901082,
      "loss": 4.8284,
      "mean_token_accuracy": 0.23294290006160737,
      "num_tokens": 176254431.0,
      "step": 84945
    },
    {
      "entropy": 5.726635122299195,
      "epoch": 7.705914368650218,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010690334662570445,
      "loss": 4.8269,
      "mean_token_accuracy": 0.23033232390880584,
      "num_tokens": 176265772.0,
      "step": 84950
    },
    {
      "entropy": 5.704879522323608,
      "epoch": 7.706367924528302,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010688184053758232,
      "loss": 4.7686,
      "mean_token_accuracy": 0.2372826889157295,
      "num_tokens": 176275737.0,
      "step": 84955
    },
    {
      "entropy": 5.724411582946777,
      "epoch": 7.706821480406386,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010686033792618659,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2344055414199829,
      "num_tokens": 176287718.0,
      "step": 84960
    },
    {
      "entropy": 5.70329532623291,
      "epoch": 7.70727503628447,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010683883879196194,
      "loss": 4.807,
      "mean_token_accuracy": 0.24320995360612868,
      "num_tokens": 176297859.0,
      "step": 84965
    },
    {
      "entropy": 5.735558080673218,
      "epoch": 7.707728592162555,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010681734313535285,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23698141872882844,
      "num_tokens": 176309377.0,
      "step": 84970
    },
    {
      "entropy": 5.690344429016113,
      "epoch": 7.708182148040638,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010679585095680403,
      "loss": 4.818,
      "mean_token_accuracy": 0.22807858288288116,
      "num_tokens": 176319659.0,
      "step": 84975
    },
    {
      "entropy": 5.673630285263061,
      "epoch": 7.7086357039187225,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001067743622567599,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23695580065250396,
      "num_tokens": 176330662.0,
      "step": 84980
    },
    {
      "entropy": 5.689888525009155,
      "epoch": 7.709089259796807,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010675287703566482,
      "loss": 4.8277,
      "mean_token_accuracy": 0.24070339947938918,
      "num_tokens": 176341047.0,
      "step": 84985
    },
    {
      "entropy": 5.741056299209594,
      "epoch": 7.709542815674891,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001067313952939631,
      "loss": 4.8219,
      "mean_token_accuracy": 0.24181724935770035,
      "num_tokens": 176350697.0,
      "step": 84990
    },
    {
      "entropy": 5.676215887069702,
      "epoch": 7.709996371552975,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010670991703209902,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23513637632131576,
      "num_tokens": 176361633.0,
      "step": 84995
    },
    {
      "entropy": 5.692743444442749,
      "epoch": 7.71044992743106,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010668844225051669,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2395537480711937,
      "num_tokens": 176372344.0,
      "step": 85000
    },
    {
      "entropy": 5.636342334747314,
      "epoch": 7.710903483309144,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001066669709496603,
      "loss": 4.7818,
      "mean_token_accuracy": 0.23677716553211212,
      "num_tokens": 176383296.0,
      "step": 85005
    },
    {
      "entropy": 5.749105215072632,
      "epoch": 7.711357039187228,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010664550312997387,
      "loss": 4.8566,
      "mean_token_accuracy": 0.23887959271669387,
      "num_tokens": 176394785.0,
      "step": 85010
    },
    {
      "entropy": 5.716761350631714,
      "epoch": 7.711810595065312,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001066240387919014,
      "loss": 4.8081,
      "mean_token_accuracy": 0.22367448806762696,
      "num_tokens": 176404413.0,
      "step": 85015
    },
    {
      "entropy": 5.673671674728394,
      "epoch": 7.712264150943396,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010660257793588664,
      "loss": 4.6943,
      "mean_token_accuracy": 0.24627304524183274,
      "num_tokens": 176414657.0,
      "step": 85020
    },
    {
      "entropy": 5.698151159286499,
      "epoch": 7.71271770682148,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010658112056237348,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23390621542930604,
      "num_tokens": 176425088.0,
      "step": 85025
    },
    {
      "entropy": 5.696037340164184,
      "epoch": 7.713171262699564,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001065596666718056,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23131799101829528,
      "num_tokens": 176434525.0,
      "step": 85030
    },
    {
      "entropy": 5.674731683731079,
      "epoch": 7.713624818577649,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010653821626462683,
      "loss": 4.8041,
      "mean_token_accuracy": 0.23803952634334563,
      "num_tokens": 176444541.0,
      "step": 85035
    },
    {
      "entropy": 5.605048131942749,
      "epoch": 7.714078374455733,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010651676934128063,
      "loss": 4.6664,
      "mean_token_accuracy": 0.2441364660859108,
      "num_tokens": 176454997.0,
      "step": 85040
    },
    {
      "entropy": 5.681187868118286,
      "epoch": 7.714531930333817,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010649532590221062,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23435551673173904,
      "num_tokens": 176465232.0,
      "step": 85045
    },
    {
      "entropy": 5.739282321929932,
      "epoch": 7.714985486211901,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001064738859478602,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23123730719089508,
      "num_tokens": 176475889.0,
      "step": 85050
    },
    {
      "entropy": 5.7108726978302,
      "epoch": 7.715439042089986,
      "grad_norm": 1.671875,
      "learning_rate": 0.00010645244947867275,
      "loss": 4.7817,
      "mean_token_accuracy": 0.23619965761899947,
      "num_tokens": 176485709.0,
      "step": 85055
    },
    {
      "entropy": 5.721382331848145,
      "epoch": 7.71589259796807,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010643101649509159,
      "loss": 4.8747,
      "mean_token_accuracy": 0.22423248291015624,
      "num_tokens": 176495722.0,
      "step": 85060
    },
    {
      "entropy": 5.707239818572998,
      "epoch": 7.716346153846154,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001064095869975599,
      "loss": 4.799,
      "mean_token_accuracy": 0.24110052287578582,
      "num_tokens": 176506543.0,
      "step": 85065
    },
    {
      "entropy": 5.715398168563842,
      "epoch": 7.7167997097242385,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000106388160986521,
      "loss": 4.8644,
      "mean_token_accuracy": 0.22584544867277145,
      "num_tokens": 176516475.0,
      "step": 85070
    },
    {
      "entropy": 5.642709112167358,
      "epoch": 7.717253265602322,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010636673846241787,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2426581025123596,
      "num_tokens": 176526036.0,
      "step": 85075
    },
    {
      "entropy": 5.667078304290771,
      "epoch": 7.717706821480406,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010634531942569358,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23865239322185516,
      "num_tokens": 176536726.0,
      "step": 85080
    },
    {
      "entropy": 5.652373743057251,
      "epoch": 7.71816037735849,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010632390387679109,
      "loss": 4.8017,
      "mean_token_accuracy": 0.23602332025766373,
      "num_tokens": 176547695.0,
      "step": 85085
    },
    {
      "entropy": 5.715457487106323,
      "epoch": 7.718613933236575,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001063024918161532,
      "loss": 4.7456,
      "mean_token_accuracy": 0.23954505175352098,
      "num_tokens": 176557235.0,
      "step": 85090
    },
    {
      "entropy": 5.684514141082763,
      "epoch": 7.719067489114659,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010628108324422272,
      "loss": 4.7745,
      "mean_token_accuracy": 0.23974183052778245,
      "num_tokens": 176567827.0,
      "step": 85095
    },
    {
      "entropy": 5.656181526184082,
      "epoch": 7.719521044992743,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010625967816144253,
      "loss": 4.7382,
      "mean_token_accuracy": 0.2396529033780098,
      "num_tokens": 176578276.0,
      "step": 85100
    },
    {
      "entropy": 5.65914888381958,
      "epoch": 7.719974600870827,
      "grad_norm": 1.828125,
      "learning_rate": 0.00010623827656825516,
      "loss": 4.7565,
      "mean_token_accuracy": 0.24595843851566315,
      "num_tokens": 176587558.0,
      "step": 85105
    },
    {
      "entropy": 5.741402816772461,
      "epoch": 7.720428156748912,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010621687846510322,
      "loss": 4.8916,
      "mean_token_accuracy": 0.23866128474473952,
      "num_tokens": 176598750.0,
      "step": 85110
    },
    {
      "entropy": 5.685157299041748,
      "epoch": 7.720881712626996,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010619548385242927,
      "loss": 4.7435,
      "mean_token_accuracy": 0.2442406862974167,
      "num_tokens": 176607920.0,
      "step": 85115
    },
    {
      "entropy": 5.603433132171631,
      "epoch": 7.721335268505079,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010617409273067575,
      "loss": 4.7851,
      "mean_token_accuracy": 0.24096280187368393,
      "num_tokens": 176617337.0,
      "step": 85120
    },
    {
      "entropy": 5.6333201885223385,
      "epoch": 7.7217888243831645,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010615270510028499,
      "loss": 4.7999,
      "mean_token_accuracy": 0.23606844395399093,
      "num_tokens": 176627546.0,
      "step": 85125
    },
    {
      "entropy": 5.639571762084961,
      "epoch": 7.722242380261248,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010613132096169925,
      "loss": 4.7146,
      "mean_token_accuracy": 0.24074763655662537,
      "num_tokens": 176637587.0,
      "step": 85130
    },
    {
      "entropy": 5.787112188339234,
      "epoch": 7.722695936139332,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001061099403153609,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2313857853412628,
      "num_tokens": 176648106.0,
      "step": 85135
    },
    {
      "entropy": 5.6435709476470945,
      "epoch": 7.723149492017416,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000106088563161712,
      "loss": 4.7249,
      "mean_token_accuracy": 0.23838075548410415,
      "num_tokens": 176658868.0,
      "step": 85140
    },
    {
      "entropy": 5.673800468444824,
      "epoch": 7.723603047895501,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001060671895011947,
      "loss": 4.7791,
      "mean_token_accuracy": 0.23818932473659515,
      "num_tokens": 176669307.0,
      "step": 85145
    },
    {
      "entropy": 5.661696243286133,
      "epoch": 7.724056603773585,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010604581933425094,
      "loss": 4.7491,
      "mean_token_accuracy": 0.23782203048467637,
      "num_tokens": 176679762.0,
      "step": 85150
    },
    {
      "entropy": 5.676333045959472,
      "epoch": 7.724510159651669,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001060244526613227,
      "loss": 4.7939,
      "mean_token_accuracy": 0.23233700394630433,
      "num_tokens": 176689403.0,
      "step": 85155
    },
    {
      "entropy": 5.672534656524658,
      "epoch": 7.7249637155297535,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010600308948285184,
      "loss": 4.8133,
      "mean_token_accuracy": 0.232011678814888,
      "num_tokens": 176699253.0,
      "step": 85160
    },
    {
      "entropy": 5.60011739730835,
      "epoch": 7.725417271407838,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010598172979928009,
      "loss": 4.7315,
      "mean_token_accuracy": 0.23864664286375045,
      "num_tokens": 176710358.0,
      "step": 85165
    },
    {
      "entropy": 5.673085927963257,
      "epoch": 7.725870827285922,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010596037361104928,
      "loss": 4.8479,
      "mean_token_accuracy": 0.23706871569156646,
      "num_tokens": 176720483.0,
      "step": 85170
    },
    {
      "entropy": 5.656337738037109,
      "epoch": 7.726324383164005,
      "grad_norm": 1.5,
      "learning_rate": 0.00010593902091860106,
      "loss": 4.8084,
      "mean_token_accuracy": 0.23900435119867325,
      "num_tokens": 176731039.0,
      "step": 85175
    },
    {
      "entropy": 5.734070682525635,
      "epoch": 7.72677793904209,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010591767172237695,
      "loss": 4.8767,
      "mean_token_accuracy": 0.23030120879411697,
      "num_tokens": 176740649.0,
      "step": 85180
    },
    {
      "entropy": 5.7078265190124515,
      "epoch": 7.727231494920174,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010589632602281846,
      "loss": 4.8114,
      "mean_token_accuracy": 0.23194426000118257,
      "num_tokens": 176751816.0,
      "step": 85185
    },
    {
      "entropy": 5.702360820770264,
      "epoch": 7.727685050798258,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010587498382036706,
      "loss": 4.8591,
      "mean_token_accuracy": 0.2296292781829834,
      "num_tokens": 176761487.0,
      "step": 85190
    },
    {
      "entropy": 5.697126960754394,
      "epoch": 7.7281386066763424,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010585364511546408,
      "loss": 4.7674,
      "mean_token_accuracy": 0.23846134841442107,
      "num_tokens": 176770572.0,
      "step": 85195
    },
    {
      "entropy": 5.676544237136841,
      "epoch": 7.728592162554427,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010583230990855077,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23758988678455353,
      "num_tokens": 176780902.0,
      "step": 85200
    },
    {
      "entropy": 5.741615581512451,
      "epoch": 7.729045718432511,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010581097820006846,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22743198424577712,
      "num_tokens": 176791886.0,
      "step": 85205
    },
    {
      "entropy": 5.720429468154907,
      "epoch": 7.729499274310595,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010578964999045823,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2335293784737587,
      "num_tokens": 176801747.0,
      "step": 85210
    },
    {
      "entropy": 5.649673795700073,
      "epoch": 7.7299528301886795,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010576832528016117,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23265302777290345,
      "num_tokens": 176812526.0,
      "step": 85215
    },
    {
      "entropy": 5.63809232711792,
      "epoch": 7.730406386066764,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010574700406961827,
      "loss": 4.7667,
      "mean_token_accuracy": 0.24026407152414322,
      "num_tokens": 176822820.0,
      "step": 85220
    },
    {
      "entropy": 5.688310956954956,
      "epoch": 7.730859941944848,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010572568635927047,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23106064945459365,
      "num_tokens": 176834119.0,
      "step": 85225
    },
    {
      "entropy": 5.701862049102783,
      "epoch": 7.731313497822931,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010570437214955858,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2358265072107315,
      "num_tokens": 176844590.0,
      "step": 85230
    },
    {
      "entropy": 5.709168815612793,
      "epoch": 7.731767053701016,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010568306144092336,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23704487532377244,
      "num_tokens": 176854540.0,
      "step": 85235
    },
    {
      "entropy": 5.6478582382202145,
      "epoch": 7.7322206095791,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010566175423380564,
      "loss": 4.7471,
      "mean_token_accuracy": 0.2438343346118927,
      "num_tokens": 176864879.0,
      "step": 85240
    },
    {
      "entropy": 5.680026960372925,
      "epoch": 7.732674165457184,
      "grad_norm": 1.625,
      "learning_rate": 0.00010564045052864599,
      "loss": 4.8261,
      "mean_token_accuracy": 0.22864626199007035,
      "num_tokens": 176874861.0,
      "step": 85245
    },
    {
      "entropy": 5.76047534942627,
      "epoch": 7.7331277213352685,
      "grad_norm": 1.46875,
      "learning_rate": 0.000105619150325885,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2296413689851761,
      "num_tokens": 176885200.0,
      "step": 85250
    },
    {
      "entropy": 5.739051437377929,
      "epoch": 7.733581277213353,
      "grad_norm": 1.5,
      "learning_rate": 0.00010559785362596314,
      "loss": 4.9354,
      "mean_token_accuracy": 0.2339744746685028,
      "num_tokens": 176895706.0,
      "step": 85255
    },
    {
      "entropy": 5.602507257461548,
      "epoch": 7.734034833091437,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010557656042932085,
      "loss": 4.6813,
      "mean_token_accuracy": 0.2491481438279152,
      "num_tokens": 176906009.0,
      "step": 85260
    },
    {
      "entropy": 5.726395893096924,
      "epoch": 7.734488388969521,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010555527073639842,
      "loss": 4.8918,
      "mean_token_accuracy": 0.22916273176670074,
      "num_tokens": 176918252.0,
      "step": 85265
    },
    {
      "entropy": 5.621360969543457,
      "epoch": 7.7349419448476056,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010553398454763613,
      "loss": 4.723,
      "mean_token_accuracy": 0.24279148131608963,
      "num_tokens": 176928458.0,
      "step": 85270
    },
    {
      "entropy": 5.731056547164917,
      "epoch": 7.735395500725689,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010551270186347431,
      "loss": 4.8372,
      "mean_token_accuracy": 0.23056381642818452,
      "num_tokens": 176937929.0,
      "step": 85275
    },
    {
      "entropy": 5.723809003829956,
      "epoch": 7.735849056603773,
      "grad_norm": 1.5,
      "learning_rate": 0.00010549142268435297,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23943334966897964,
      "num_tokens": 176947955.0,
      "step": 85280
    },
    {
      "entropy": 5.643531084060669,
      "epoch": 7.7363026124818575,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010547014701071223,
      "loss": 4.7589,
      "mean_token_accuracy": 0.2385540187358856,
      "num_tokens": 176959019.0,
      "step": 85285
    },
    {
      "entropy": 5.627083492279053,
      "epoch": 7.736756168359942,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010544887484299203,
      "loss": 4.7663,
      "mean_token_accuracy": 0.23721059113740922,
      "num_tokens": 176970145.0,
      "step": 85290
    },
    {
      "entropy": 5.635013294219971,
      "epoch": 7.737209724238026,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010542760618163233,
      "loss": 4.7276,
      "mean_token_accuracy": 0.2437549889087677,
      "num_tokens": 176980489.0,
      "step": 85295
    },
    {
      "entropy": 5.716020298004151,
      "epoch": 7.73766328011611,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010540634102707286,
      "loss": 4.8628,
      "mean_token_accuracy": 0.23241778761148452,
      "num_tokens": 176991708.0,
      "step": 85300
    },
    {
      "entropy": 5.704592704772949,
      "epoch": 7.7381168359941945,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010538507937975355,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22977657616138458,
      "num_tokens": 177002617.0,
      "step": 85305
    },
    {
      "entropy": 5.7428971290588375,
      "epoch": 7.738570391872279,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010536382124011404,
      "loss": 4.9,
      "mean_token_accuracy": 0.22464606314897537,
      "num_tokens": 177013567.0,
      "step": 85310
    },
    {
      "entropy": 5.600673961639404,
      "epoch": 7.739023947750363,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001053425666085939,
      "loss": 4.6752,
      "mean_token_accuracy": 0.24919829666614532,
      "num_tokens": 177024126.0,
      "step": 85315
    },
    {
      "entropy": 5.691568040847779,
      "epoch": 7.739477503628447,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010532131548563273,
      "loss": 4.8321,
      "mean_token_accuracy": 0.22613049298524857,
      "num_tokens": 177033759.0,
      "step": 85320
    },
    {
      "entropy": 5.733746528625488,
      "epoch": 7.739931059506532,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010530006787167002,
      "loss": 4.802,
      "mean_token_accuracy": 0.23784958720207214,
      "num_tokens": 177044007.0,
      "step": 85325
    },
    {
      "entropy": 5.732452392578125,
      "epoch": 7.740384615384615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001052788237671451,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22766190022230148,
      "num_tokens": 177056054.0,
      "step": 85330
    },
    {
      "entropy": 5.731256675720215,
      "epoch": 7.740838171262699,
      "grad_norm": 1.609375,
      "learning_rate": 0.00010525758317249734,
      "loss": 4.8129,
      "mean_token_accuracy": 0.23965531587600708,
      "num_tokens": 177065085.0,
      "step": 85335
    },
    {
      "entropy": 5.845173406600952,
      "epoch": 7.7412917271407835,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010523634608816607,
      "loss": 5.0304,
      "mean_token_accuracy": 0.21676595360040665,
      "num_tokens": 177075942.0,
      "step": 85340
    },
    {
      "entropy": 5.788197994232178,
      "epoch": 7.741745283018868,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001052151125145904,
      "loss": 4.9124,
      "mean_token_accuracy": 0.22885061353445052,
      "num_tokens": 177084975.0,
      "step": 85345
    },
    {
      "entropy": 5.696478271484375,
      "epoch": 7.742198838896952,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010519388245220948,
      "loss": 4.7898,
      "mean_token_accuracy": 0.2366633802652359,
      "num_tokens": 177095973.0,
      "step": 85350
    },
    {
      "entropy": 5.69132513999939,
      "epoch": 7.742652394775036,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010517265590146232,
      "loss": 4.7542,
      "mean_token_accuracy": 0.24091623723506927,
      "num_tokens": 177107236.0,
      "step": 85355
    },
    {
      "entropy": 5.738664007186889,
      "epoch": 7.743105950653121,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010515143286278797,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22899530977010726,
      "num_tokens": 177117038.0,
      "step": 85360
    },
    {
      "entropy": 5.738428115844727,
      "epoch": 7.743559506531205,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010513021333662523,
      "loss": 4.8768,
      "mean_token_accuracy": 0.23088705837726592,
      "num_tokens": 177127640.0,
      "step": 85365
    },
    {
      "entropy": 5.653015851974487,
      "epoch": 7.744013062409289,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001051089973234129,
      "loss": 4.7816,
      "mean_token_accuracy": 0.23623951524496078,
      "num_tokens": 177138238.0,
      "step": 85370
    },
    {
      "entropy": 5.5821229934692385,
      "epoch": 7.744466618287373,
      "grad_norm": 1.375,
      "learning_rate": 0.00010508778482358988,
      "loss": 4.7004,
      "mean_token_accuracy": 0.24862547367811202,
      "num_tokens": 177149660.0,
      "step": 85375
    },
    {
      "entropy": 5.6360620021820065,
      "epoch": 7.744920174165458,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010506657583759477,
      "loss": 4.785,
      "mean_token_accuracy": 0.22838643789291382,
      "num_tokens": 177160376.0,
      "step": 85380
    },
    {
      "entropy": 5.641961765289307,
      "epoch": 7.745373730043541,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010504537036586616,
      "loss": 4.7681,
      "mean_token_accuracy": 0.23653019964694977,
      "num_tokens": 177170418.0,
      "step": 85385
    },
    {
      "entropy": 5.587708616256714,
      "epoch": 7.745827285921625,
      "grad_norm": 1.609375,
      "learning_rate": 0.00010502416840884262,
      "loss": 4.6699,
      "mean_token_accuracy": 0.24346197247505189,
      "num_tokens": 177180521.0,
      "step": 85390
    },
    {
      "entropy": 5.627298641204834,
      "epoch": 7.7462808417997095,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010500296996696257,
      "loss": 4.7127,
      "mean_token_accuracy": 0.24792238920927048,
      "num_tokens": 177191039.0,
      "step": 85395
    },
    {
      "entropy": 5.726149606704712,
      "epoch": 7.746734397677794,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010498177504066444,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23526682555675507,
      "num_tokens": 177201065.0,
      "step": 85400
    },
    {
      "entropy": 5.651192998886108,
      "epoch": 7.747187953555878,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010496058363038644,
      "loss": 4.7797,
      "mean_token_accuracy": 0.24601693749427794,
      "num_tokens": 177211946.0,
      "step": 85405
    },
    {
      "entropy": 5.673475456237793,
      "epoch": 7.747641509433962,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010493939573656699,
      "loss": 4.8846,
      "mean_token_accuracy": 0.23411154597997666,
      "num_tokens": 177222589.0,
      "step": 85410
    },
    {
      "entropy": 5.757125425338745,
      "epoch": 7.748095065312047,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010491821135964416,
      "loss": 4.8884,
      "mean_token_accuracy": 0.21959104090929032,
      "num_tokens": 177233639.0,
      "step": 85415
    },
    {
      "entropy": 5.653888797760009,
      "epoch": 7.748548621190131,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010489703050005606,
      "loss": 4.8021,
      "mean_token_accuracy": 0.23348983526229858,
      "num_tokens": 177245029.0,
      "step": 85420
    },
    {
      "entropy": 5.662559986114502,
      "epoch": 7.749002177068215,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010487585315824075,
      "loss": 4.7632,
      "mean_token_accuracy": 0.24590455442667009,
      "num_tokens": 177256020.0,
      "step": 85425
    },
    {
      "entropy": 5.694324922561646,
      "epoch": 7.7494557329462985,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010485467933463611,
      "loss": 4.8026,
      "mean_token_accuracy": 0.24239596873521804,
      "num_tokens": 177267261.0,
      "step": 85430
    },
    {
      "entropy": 5.735349798202515,
      "epoch": 7.749909288824383,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010483350902968006,
      "loss": 4.8972,
      "mean_token_accuracy": 0.22787030786275864,
      "num_tokens": 177277963.0,
      "step": 85435
    },
    {
      "entropy": 5.69495530128479,
      "epoch": 7.750362844702467,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010481234224381037,
      "loss": 4.8493,
      "mean_token_accuracy": 0.23109967708587648,
      "num_tokens": 177289730.0,
      "step": 85440
    },
    {
      "entropy": 5.6773511409759525,
      "epoch": 7.750816400580551,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010479117897746492,
      "loss": 4.7748,
      "mean_token_accuracy": 0.24533764272928238,
      "num_tokens": 177300917.0,
      "step": 85445
    },
    {
      "entropy": 5.691782093048095,
      "epoch": 7.751269956458636,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001047700192310812,
      "loss": 4.7688,
      "mean_token_accuracy": 0.2419234871864319,
      "num_tokens": 177310469.0,
      "step": 85450
    },
    {
      "entropy": 5.710037660598755,
      "epoch": 7.75172351233672,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010474886300509686,
      "loss": 4.916,
      "mean_token_accuracy": 0.23219090551137925,
      "num_tokens": 177320244.0,
      "step": 85455
    },
    {
      "entropy": 5.687626886367798,
      "epoch": 7.752177068214804,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010472771029994943,
      "loss": 4.8991,
      "mean_token_accuracy": 0.2254070058465004,
      "num_tokens": 177330987.0,
      "step": 85460
    },
    {
      "entropy": 5.684991073608399,
      "epoch": 7.752630624092888,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010470656111607629,
      "loss": 4.8417,
      "mean_token_accuracy": 0.23819301128387452,
      "num_tokens": 177341877.0,
      "step": 85465
    },
    {
      "entropy": 5.704334688186646,
      "epoch": 7.753084179970973,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010468541545391489,
      "loss": 4.8242,
      "mean_token_accuracy": 0.23172320872545243,
      "num_tokens": 177352144.0,
      "step": 85470
    },
    {
      "entropy": 5.671601152420044,
      "epoch": 7.753537735849057,
      "grad_norm": 1.375,
      "learning_rate": 0.00010466427331390252,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2462722733616829,
      "num_tokens": 177362559.0,
      "step": 85475
    },
    {
      "entropy": 5.768140649795532,
      "epoch": 7.753991291727141,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010464313469647638,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2256608709692955,
      "num_tokens": 177372780.0,
      "step": 85480
    },
    {
      "entropy": 5.728974676132202,
      "epoch": 7.754444847605225,
      "grad_norm": 1.5,
      "learning_rate": 0.00010462199960207361,
      "loss": 4.8477,
      "mean_token_accuracy": 0.23191864639520646,
      "num_tokens": 177382562.0,
      "step": 85485
    },
    {
      "entropy": 5.7578184604644775,
      "epoch": 7.754898403483309,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010460086803113133,
      "loss": 4.8708,
      "mean_token_accuracy": 0.23370311707258223,
      "num_tokens": 177393605.0,
      "step": 85490
    },
    {
      "entropy": 5.760343933105469,
      "epoch": 7.755351959361393,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010457973998408654,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23367548137903213,
      "num_tokens": 177403739.0,
      "step": 85495
    },
    {
      "entropy": 5.7160036087036135,
      "epoch": 7.755805515239477,
      "grad_norm": 1.5,
      "learning_rate": 0.00010455861546137605,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23268719017505646,
      "num_tokens": 177414441.0,
      "step": 85500
    },
    {
      "entropy": 5.6457459926605225,
      "epoch": 7.756259071117562,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010453749446343692,
      "loss": 4.7875,
      "mean_token_accuracy": 0.23767043352127076,
      "num_tokens": 177424366.0,
      "step": 85505
    },
    {
      "entropy": 5.611693429946899,
      "epoch": 7.756712626995646,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010451637699070582,
      "loss": 4.6914,
      "mean_token_accuracy": 0.25015769302845003,
      "num_tokens": 177434516.0,
      "step": 85510
    },
    {
      "entropy": 5.688226747512817,
      "epoch": 7.75716618287373,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001044952630436195,
      "loss": 4.7854,
      "mean_token_accuracy": 0.24235329180955886,
      "num_tokens": 177444628.0,
      "step": 85515
    },
    {
      "entropy": 5.681096172332763,
      "epoch": 7.757619738751814,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010447415262261458,
      "loss": 4.8023,
      "mean_token_accuracy": 0.24312182664871215,
      "num_tokens": 177456009.0,
      "step": 85520
    },
    {
      "entropy": 5.756582307815552,
      "epoch": 7.758073294629899,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010445304572812766,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23564021289348602,
      "num_tokens": 177465770.0,
      "step": 85525
    },
    {
      "entropy": 5.7060833930969235,
      "epoch": 7.758526850507983,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001044319423605952,
      "loss": 4.8465,
      "mean_token_accuracy": 0.23458099961280823,
      "num_tokens": 177475992.0,
      "step": 85530
    },
    {
      "entropy": 5.688169384002686,
      "epoch": 7.758980406386067,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00010441084252045358,
      "loss": 4.8535,
      "mean_token_accuracy": 0.2339552164077759,
      "num_tokens": 177485964.0,
      "step": 85535
    },
    {
      "entropy": 5.7006690979003904,
      "epoch": 7.759433962264151,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010438974620813926,
      "loss": 4.8607,
      "mean_token_accuracy": 0.23202950060367583,
      "num_tokens": 177496755.0,
      "step": 85540
    },
    {
      "entropy": 5.670661926269531,
      "epoch": 7.759887518142235,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010436865342408848,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22725465893745422,
      "num_tokens": 177506195.0,
      "step": 85545
    },
    {
      "entropy": 5.708387994766236,
      "epoch": 7.760341074020319,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010434756416873744,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22669960409402848,
      "num_tokens": 177516660.0,
      "step": 85550
    },
    {
      "entropy": 5.705891132354736,
      "epoch": 7.760794629898403,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010432647844252222,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23113973587751388,
      "num_tokens": 177526893.0,
      "step": 85555
    },
    {
      "entropy": 5.756480312347412,
      "epoch": 7.761248185776488,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010430539624587893,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23793764263391495,
      "num_tokens": 177537083.0,
      "step": 85560
    },
    {
      "entropy": 5.734438753128051,
      "epoch": 7.761701741654572,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010428431757924356,
      "loss": 4.8212,
      "mean_token_accuracy": 0.2337159425020218,
      "num_tokens": 177547829.0,
      "step": 85565
    },
    {
      "entropy": 5.729514741897583,
      "epoch": 7.762155297532656,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010426324244305194,
      "loss": 4.8503,
      "mean_token_accuracy": 0.24180425554513932,
      "num_tokens": 177558681.0,
      "step": 85570
    },
    {
      "entropy": 5.678193521499634,
      "epoch": 7.7626088534107405,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010424217083774004,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23856198340654372,
      "num_tokens": 177568840.0,
      "step": 85575
    },
    {
      "entropy": 5.734144067764282,
      "epoch": 7.763062409288825,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010422110276374356,
      "loss": 4.9125,
      "mean_token_accuracy": 0.23219259232282638,
      "num_tokens": 177579665.0,
      "step": 85580
    },
    {
      "entropy": 5.6918964862823485,
      "epoch": 7.763515965166908,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001042000382214982,
      "loss": 4.8099,
      "mean_token_accuracy": 0.23963652700185775,
      "num_tokens": 177591081.0,
      "step": 85585
    },
    {
      "entropy": 5.687898111343384,
      "epoch": 7.763969521044992,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010417897721143957,
      "loss": 4.8233,
      "mean_token_accuracy": 0.23346817195415498,
      "num_tokens": 177601071.0,
      "step": 85590
    },
    {
      "entropy": 5.694834280014038,
      "epoch": 7.764423076923077,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001041579197340032,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23862191885709763,
      "num_tokens": 177611735.0,
      "step": 85595
    },
    {
      "entropy": 5.714218378067017,
      "epoch": 7.764876632801161,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010413686578962453,
      "loss": 4.8248,
      "mean_token_accuracy": 0.24128559231758118,
      "num_tokens": 177622070.0,
      "step": 85600
    },
    {
      "entropy": 5.778224992752075,
      "epoch": 7.765330188679245,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010411581537873911,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22932581305503846,
      "num_tokens": 177632648.0,
      "step": 85605
    },
    {
      "entropy": 5.705072641372681,
      "epoch": 7.7657837445573294,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010409476850178215,
      "loss": 4.7914,
      "mean_token_accuracy": 0.23859187811613083,
      "num_tokens": 177642346.0,
      "step": 85610
    },
    {
      "entropy": 5.649119186401367,
      "epoch": 7.766237300435414,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010407372515918891,
      "loss": 4.6998,
      "mean_token_accuracy": 0.24857074171304702,
      "num_tokens": 177652520.0,
      "step": 85615
    },
    {
      "entropy": 5.748463249206543,
      "epoch": 7.766690856313498,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010405268535139465,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22567216157913209,
      "num_tokens": 177662428.0,
      "step": 85620
    },
    {
      "entropy": 5.729219532012939,
      "epoch": 7.767144412191582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010403164907883439,
      "loss": 4.8065,
      "mean_token_accuracy": 0.2386387348175049,
      "num_tokens": 177673430.0,
      "step": 85625
    },
    {
      "entropy": 5.770590353012085,
      "epoch": 7.7675979680696665,
      "grad_norm": 1.609375,
      "learning_rate": 0.00010401061634194317,
      "loss": 4.892,
      "mean_token_accuracy": 0.22602829337120056,
      "num_tokens": 177684590.0,
      "step": 85630
    },
    {
      "entropy": 5.702650737762451,
      "epoch": 7.768051523947751,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010398958714115595,
      "loss": 4.8136,
      "mean_token_accuracy": 0.24190887957811355,
      "num_tokens": 177695755.0,
      "step": 85635
    },
    {
      "entropy": 5.69180154800415,
      "epoch": 7.768505079825834,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001039685614769077,
      "loss": 4.7326,
      "mean_token_accuracy": 0.23745538294315338,
      "num_tokens": 177706831.0,
      "step": 85640
    },
    {
      "entropy": 5.66959433555603,
      "epoch": 7.768958635703918,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010394753934963316,
      "loss": 4.8034,
      "mean_token_accuracy": 0.2354728564620018,
      "num_tokens": 177718344.0,
      "step": 85645
    },
    {
      "entropy": 5.639937877655029,
      "epoch": 7.769412191582003,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010392652075976713,
      "loss": 4.7415,
      "mean_token_accuracy": 0.23707587867975236,
      "num_tokens": 177729791.0,
      "step": 85650
    },
    {
      "entropy": 5.626002502441406,
      "epoch": 7.769865747460087,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010390550570774421,
      "loss": 4.6962,
      "mean_token_accuracy": 0.24244726747274398,
      "num_tokens": 177740497.0,
      "step": 85655
    },
    {
      "entropy": 5.674116420745849,
      "epoch": 7.770319303338171,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00010388449419399906,
      "loss": 4.7774,
      "mean_token_accuracy": 0.23733894824981688,
      "num_tokens": 177751245.0,
      "step": 85660
    },
    {
      "entropy": 5.657079887390137,
      "epoch": 7.7707728592162555,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010386348621896618,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24173085391521454,
      "num_tokens": 177761175.0,
      "step": 85665
    },
    {
      "entropy": 5.5869011878967285,
      "epoch": 7.77122641509434,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010384248178307993,
      "loss": 4.724,
      "mean_token_accuracy": 0.24698318243026735,
      "num_tokens": 177771345.0,
      "step": 85670
    },
    {
      "entropy": 5.712271976470947,
      "epoch": 7.771679970972424,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010382148088677483,
      "loss": 4.8843,
      "mean_token_accuracy": 0.2221337839961052,
      "num_tokens": 177781672.0,
      "step": 85675
    },
    {
      "entropy": 5.697598314285278,
      "epoch": 7.772133526850508,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010380048353048514,
      "loss": 4.7366,
      "mean_token_accuracy": 0.23394632935523987,
      "num_tokens": 177790805.0,
      "step": 85680
    },
    {
      "entropy": 5.703327417373657,
      "epoch": 7.772587082728592,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001037794897146451,
      "loss": 4.8344,
      "mean_token_accuracy": 0.23163216412067414,
      "num_tokens": 177801285.0,
      "step": 85685
    },
    {
      "entropy": 5.752950286865234,
      "epoch": 7.773040638606677,
      "grad_norm": 1.5,
      "learning_rate": 0.00010375849943968879,
      "loss": 4.8144,
      "mean_token_accuracy": 0.24157941341400146,
      "num_tokens": 177811159.0,
      "step": 85690
    },
    {
      "entropy": 5.636379766464233,
      "epoch": 7.77349419448476,
      "grad_norm": 1.5,
      "learning_rate": 0.00010373751270605038,
      "loss": 4.8059,
      "mean_token_accuracy": 0.24551165997982025,
      "num_tokens": 177822437.0,
      "step": 85695
    },
    {
      "entropy": 5.658304929733276,
      "epoch": 7.7739477503628445,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010371652951416386,
      "loss": 4.8057,
      "mean_token_accuracy": 0.23342348486185074,
      "num_tokens": 177832487.0,
      "step": 85700
    },
    {
      "entropy": 5.705261135101319,
      "epoch": 7.774401306240929,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010369554986446309,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23616620600223542,
      "num_tokens": 177842521.0,
      "step": 85705
    },
    {
      "entropy": 5.6603827476501465,
      "epoch": 7.774854862119013,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010367457375738207,
      "loss": 4.7593,
      "mean_token_accuracy": 0.24589310586452484,
      "num_tokens": 177854162.0,
      "step": 85710
    },
    {
      "entropy": 5.747497177124023,
      "epoch": 7.775308417997097,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010365360119335452,
      "loss": 4.889,
      "mean_token_accuracy": 0.22456276267766953,
      "num_tokens": 177864128.0,
      "step": 85715
    },
    {
      "entropy": 5.7081585884094235,
      "epoch": 7.7757619738751815,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010363263217281412,
      "loss": 4.7732,
      "mean_token_accuracy": 0.24235527962446213,
      "num_tokens": 177875509.0,
      "step": 85720
    },
    {
      "entropy": 5.657920169830322,
      "epoch": 7.776215529753266,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00010361166669619462,
      "loss": 4.712,
      "mean_token_accuracy": 0.24744511246681214,
      "num_tokens": 177885103.0,
      "step": 85725
    },
    {
      "entropy": 5.723322772979737,
      "epoch": 7.77666908563135,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010359070476392951,
      "loss": 4.8614,
      "mean_token_accuracy": 0.23511679768562316,
      "num_tokens": 177894966.0,
      "step": 85730
    },
    {
      "entropy": 5.69950761795044,
      "epoch": 7.777122641509434,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010356974637645231,
      "loss": 4.798,
      "mean_token_accuracy": 0.2340112328529358,
      "num_tokens": 177904448.0,
      "step": 85735
    },
    {
      "entropy": 5.657710361480713,
      "epoch": 7.777576197387518,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010354879153419636,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23943931609392166,
      "num_tokens": 177915380.0,
      "step": 85740
    },
    {
      "entropy": 5.71664776802063,
      "epoch": 7.778029753265602,
      "grad_norm": 1.703125,
      "learning_rate": 0.00010352784023759515,
      "loss": 4.8381,
      "mean_token_accuracy": 0.23520132750272751,
      "num_tokens": 177925133.0,
      "step": 85745
    },
    {
      "entropy": 5.710671854019165,
      "epoch": 7.778483309143686,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001035068924870819,
      "loss": 4.8413,
      "mean_token_accuracy": 0.24051326811313628,
      "num_tokens": 177935888.0,
      "step": 85750
    },
    {
      "entropy": 5.730961894989013,
      "epoch": 7.7789368650217705,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010348594828308983,
      "loss": 4.9076,
      "mean_token_accuracy": 0.22837493121623992,
      "num_tokens": 177946639.0,
      "step": 85755
    },
    {
      "entropy": 5.6886577129364015,
      "epoch": 7.779390420899855,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010346500762605207,
      "loss": 4.7077,
      "mean_token_accuracy": 0.24626678824424744,
      "num_tokens": 177956705.0,
      "step": 85760
    },
    {
      "entropy": 5.635003995895386,
      "epoch": 7.779843976777939,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010344407051640162,
      "loss": 4.7119,
      "mean_token_accuracy": 0.24451713263988495,
      "num_tokens": 177966764.0,
      "step": 85765
    },
    {
      "entropy": 5.679439306259155,
      "epoch": 7.780297532656023,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010342313695457156,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23644677102565764,
      "num_tokens": 177976519.0,
      "step": 85770
    },
    {
      "entropy": 5.716339254379273,
      "epoch": 7.780751088534108,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010340220694099464,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22000498473644256,
      "num_tokens": 177985540.0,
      "step": 85775
    },
    {
      "entropy": 5.64702353477478,
      "epoch": 7.781204644412192,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010338128047610388,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2374136656522751,
      "num_tokens": 177995785.0,
      "step": 85780
    },
    {
      "entropy": 5.714316654205322,
      "epoch": 7.781658200290276,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000103360357560332,
      "loss": 4.8324,
      "mean_token_accuracy": 0.23004477322101594,
      "num_tokens": 178006767.0,
      "step": 85785
    },
    {
      "entropy": 5.663460731506348,
      "epoch": 7.78211175616836,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010333943819411166,
      "loss": 4.7851,
      "mean_token_accuracy": 0.2404533550143242,
      "num_tokens": 178017288.0,
      "step": 85790
    },
    {
      "entropy": 5.63821849822998,
      "epoch": 7.782565312046444,
      "grad_norm": 1.5,
      "learning_rate": 0.00010331852237787548,
      "loss": 4.7385,
      "mean_token_accuracy": 0.24220570176839828,
      "num_tokens": 178027720.0,
      "step": 85795
    },
    {
      "entropy": 5.768410062789917,
      "epoch": 7.783018867924528,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000103297610112056,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22969971150159835,
      "num_tokens": 178039299.0,
      "step": 85800
    },
    {
      "entropy": 5.66511607170105,
      "epoch": 7.783472423802612,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010327670139708564,
      "loss": 4.7906,
      "mean_token_accuracy": 0.24051644206047057,
      "num_tokens": 178049452.0,
      "step": 85805
    },
    {
      "entropy": 5.762681150436402,
      "epoch": 7.7839259796806965,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010325579623339691,
      "loss": 4.8647,
      "mean_token_accuracy": 0.23605238348245622,
      "num_tokens": 178059626.0,
      "step": 85810
    },
    {
      "entropy": 5.6831952095031735,
      "epoch": 7.784379535558781,
      "grad_norm": 1.609375,
      "learning_rate": 0.00010323489462142208,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23050255179405213,
      "num_tokens": 178071046.0,
      "step": 85815
    },
    {
      "entropy": 5.711097526550293,
      "epoch": 7.784833091436865,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010321399656159341,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23808197528123856,
      "num_tokens": 178082213.0,
      "step": 85820
    },
    {
      "entropy": 5.671383476257324,
      "epoch": 7.785286647314949,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010319310205434307,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2419242188334465,
      "num_tokens": 178093920.0,
      "step": 85825
    },
    {
      "entropy": 5.706444263458252,
      "epoch": 7.785740203193034,
      "grad_norm": 1.6875,
      "learning_rate": 0.00010317221110010314,
      "loss": 4.856,
      "mean_token_accuracy": 0.23516449183225632,
      "num_tokens": 178103476.0,
      "step": 85830
    },
    {
      "entropy": 5.677838945388794,
      "epoch": 7.786193759071118,
      "grad_norm": 1.65625,
      "learning_rate": 0.00010315132369930568,
      "loss": 4.8,
      "mean_token_accuracy": 0.23358982801437378,
      "num_tokens": 178113780.0,
      "step": 85835
    },
    {
      "entropy": 5.7026876449584964,
      "epoch": 7.786647314949201,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001031304398523826,
      "loss": 4.8131,
      "mean_token_accuracy": 0.2362490028142929,
      "num_tokens": 178125206.0,
      "step": 85840
    },
    {
      "entropy": 5.732986974716186,
      "epoch": 7.787100870827286,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010310955955976586,
      "loss": 4.7839,
      "mean_token_accuracy": 0.2423274651169777,
      "num_tokens": 178135487.0,
      "step": 85845
    },
    {
      "entropy": 5.732972574234009,
      "epoch": 7.78755442670537,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00010308868282188725,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2339868500828743,
      "num_tokens": 178146673.0,
      "step": 85850
    },
    {
      "entropy": 5.7525458335876465,
      "epoch": 7.788007982583454,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010306780963917848,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2355559676885605,
      "num_tokens": 178157916.0,
      "step": 85855
    },
    {
      "entropy": 5.807825231552124,
      "epoch": 7.788461538461538,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010304694001207118,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21875371634960175,
      "num_tokens": 178168930.0,
      "step": 85860
    },
    {
      "entropy": 5.764727592468262,
      "epoch": 7.788915094339623,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010302607394099703,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22686728090047836,
      "num_tokens": 178178770.0,
      "step": 85865
    },
    {
      "entropy": 5.769053888320923,
      "epoch": 7.789368650217707,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010300521142638747,
      "loss": 4.8778,
      "mean_token_accuracy": 0.23704951107501984,
      "num_tokens": 178188706.0,
      "step": 85870
    },
    {
      "entropy": 5.723030471801758,
      "epoch": 7.789822206095791,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010298435246867399,
      "loss": 4.831,
      "mean_token_accuracy": 0.2307754635810852,
      "num_tokens": 178199789.0,
      "step": 85875
    },
    {
      "entropy": 5.689100360870361,
      "epoch": 7.790275761973875,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010296349706828792,
      "loss": 4.8411,
      "mean_token_accuracy": 0.23039606362581252,
      "num_tokens": 178210060.0,
      "step": 85880
    },
    {
      "entropy": 5.6724049091339115,
      "epoch": 7.79072931785196,
      "grad_norm": 1.5,
      "learning_rate": 0.00010294264522566057,
      "loss": 4.7765,
      "mean_token_accuracy": 0.2379220247268677,
      "num_tokens": 178220575.0,
      "step": 85885
    },
    {
      "entropy": 5.724989891052246,
      "epoch": 7.791182873730044,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010292179694122316,
      "loss": 4.9021,
      "mean_token_accuracy": 0.2312943458557129,
      "num_tokens": 178231331.0,
      "step": 85890
    },
    {
      "entropy": 5.645988035202026,
      "epoch": 7.791636429608127,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010290095221540685,
      "loss": 4.7413,
      "mean_token_accuracy": 0.23974779695272447,
      "num_tokens": 178240809.0,
      "step": 85895
    },
    {
      "entropy": 5.687858438491821,
      "epoch": 7.792089985486212,
      "grad_norm": 1.5,
      "learning_rate": 0.00010288011104864262,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23079836815595628,
      "num_tokens": 178252053.0,
      "step": 85900
    },
    {
      "entropy": 5.691532135009766,
      "epoch": 7.792543541364296,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010285927344136165,
      "loss": 4.87,
      "mean_token_accuracy": 0.2327770173549652,
      "num_tokens": 178262966.0,
      "step": 85905
    },
    {
      "entropy": 5.654448223114014,
      "epoch": 7.79299709724238,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010283843939399476,
      "loss": 4.7665,
      "mean_token_accuracy": 0.23842001855373382,
      "num_tokens": 178272932.0,
      "step": 85910
    },
    {
      "entropy": 5.630277872085571,
      "epoch": 7.793450653120464,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010281760890697278,
      "loss": 4.7215,
      "mean_token_accuracy": 0.24856132715940477,
      "num_tokens": 178282613.0,
      "step": 85915
    },
    {
      "entropy": 5.737717056274414,
      "epoch": 7.793904208998549,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010279678198072659,
      "loss": 4.8438,
      "mean_token_accuracy": 0.23504901081323623,
      "num_tokens": 178293494.0,
      "step": 85920
    },
    {
      "entropy": 5.74928617477417,
      "epoch": 7.794357764876633,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010277595861568678,
      "loss": 4.8984,
      "mean_token_accuracy": 0.23097391426563263,
      "num_tokens": 178304463.0,
      "step": 85925
    },
    {
      "entropy": 5.619329977035522,
      "epoch": 7.794811320754717,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010275513881228406,
      "loss": 4.71,
      "mean_token_accuracy": 0.24248361140489577,
      "num_tokens": 178315252.0,
      "step": 85930
    },
    {
      "entropy": 5.7599067211151125,
      "epoch": 7.795264876632801,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010273432257094888,
      "loss": 4.8849,
      "mean_token_accuracy": 0.23365606218576432,
      "num_tokens": 178326061.0,
      "step": 85935
    },
    {
      "entropy": 5.711139917373657,
      "epoch": 7.795718432510886,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010271350989211187,
      "loss": 4.7593,
      "mean_token_accuracy": 0.23473141491413116,
      "num_tokens": 178336233.0,
      "step": 85940
    },
    {
      "entropy": 5.629272890090943,
      "epoch": 7.79617198838897,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001026927007762034,
      "loss": 4.7196,
      "mean_token_accuracy": 0.24845019727945328,
      "num_tokens": 178346558.0,
      "step": 85945
    },
    {
      "entropy": 5.741444540023804,
      "epoch": 7.796625544267053,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00010267189522365377,
      "loss": 4.8174,
      "mean_token_accuracy": 0.23131094127893448,
      "num_tokens": 178355301.0,
      "step": 85950
    },
    {
      "entropy": 5.718047761917115,
      "epoch": 7.797079100145138,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010265109323489327,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2267212152481079,
      "num_tokens": 178364958.0,
      "step": 85955
    },
    {
      "entropy": 5.709626483917236,
      "epoch": 7.797532656023222,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010263029481035207,
      "loss": 4.8027,
      "mean_token_accuracy": 0.2398900032043457,
      "num_tokens": 178374514.0,
      "step": 85960
    },
    {
      "entropy": 5.708483076095581,
      "epoch": 7.797986211901306,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010260949995046032,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23241982758045196,
      "num_tokens": 178384860.0,
      "step": 85965
    },
    {
      "entropy": 5.733905029296875,
      "epoch": 7.79843976777939,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010258870865564796,
      "loss": 4.8501,
      "mean_token_accuracy": 0.22958568036556243,
      "num_tokens": 178394947.0,
      "step": 85970
    },
    {
      "entropy": 5.730617713928223,
      "epoch": 7.798893323657475,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00010256792092634512,
      "loss": 4.804,
      "mean_token_accuracy": 0.23675638288259507,
      "num_tokens": 178405108.0,
      "step": 85975
    },
    {
      "entropy": 5.640832567214966,
      "epoch": 7.799346879535559,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001025471367629816,
      "loss": 4.7277,
      "mean_token_accuracy": 0.2410445138812065,
      "num_tokens": 178415441.0,
      "step": 85980
    },
    {
      "entropy": 5.702263975143433,
      "epoch": 7.799800435413643,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010252635616598723,
      "loss": 4.774,
      "mean_token_accuracy": 0.23362423330545426,
      "num_tokens": 178424649.0,
      "step": 85985
    },
    {
      "entropy": 5.728178548812866,
      "epoch": 7.8002539912917275,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010250557913579181,
      "loss": 4.8352,
      "mean_token_accuracy": 0.2320543110370636,
      "num_tokens": 178435628.0,
      "step": 85990
    },
    {
      "entropy": 5.6826544284820555,
      "epoch": 7.800707547169811,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010248480567282495,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2310528948903084,
      "num_tokens": 178445730.0,
      "step": 85995
    },
    {
      "entropy": 5.785158395767212,
      "epoch": 7.801161103047895,
      "grad_norm": 1.5,
      "learning_rate": 0.00010246403577751623,
      "loss": 4.962,
      "mean_token_accuracy": 0.22243584096431732,
      "num_tokens": 178455516.0,
      "step": 86000
    },
    {
      "entropy": 5.725975561141968,
      "epoch": 7.801614658925979,
      "grad_norm": 1.625,
      "learning_rate": 0.0001024432694502952,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22843396961688994,
      "num_tokens": 178466131.0,
      "step": 86005
    },
    {
      "entropy": 5.743707656860352,
      "epoch": 7.802068214804064,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010242250669159137,
      "loss": 4.8819,
      "mean_token_accuracy": 0.23293963372707366,
      "num_tokens": 178476837.0,
      "step": 86010
    },
    {
      "entropy": 5.6551100730896,
      "epoch": 7.802521770682148,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001024017475018341,
      "loss": 4.7774,
      "mean_token_accuracy": 0.24316612035036086,
      "num_tokens": 178487730.0,
      "step": 86015
    },
    {
      "entropy": 5.69727258682251,
      "epoch": 7.802975326560232,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010238099188145266,
      "loss": 4.7398,
      "mean_token_accuracy": 0.2441292941570282,
      "num_tokens": 178498476.0,
      "step": 86020
    },
    {
      "entropy": 5.672516107559204,
      "epoch": 7.8034288824383164,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010236023983087631,
      "loss": 4.798,
      "mean_token_accuracy": 0.24199607819318772,
      "num_tokens": 178509323.0,
      "step": 86025
    },
    {
      "entropy": 5.7187196731567385,
      "epoch": 7.803882438316401,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010233949135053416,
      "loss": 4.802,
      "mean_token_accuracy": 0.2401932582259178,
      "num_tokens": 178520271.0,
      "step": 86030
    },
    {
      "entropy": 5.658328819274902,
      "epoch": 7.804335994194485,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010231874644085537,
      "loss": 4.7626,
      "mean_token_accuracy": 0.24428032636642455,
      "num_tokens": 178531229.0,
      "step": 86035
    },
    {
      "entropy": 5.6900896549224855,
      "epoch": 7.804789550072569,
      "grad_norm": 1.5,
      "learning_rate": 0.00010229800510226883,
      "loss": 4.7161,
      "mean_token_accuracy": 0.242729651927948,
      "num_tokens": 178541132.0,
      "step": 86040
    },
    {
      "entropy": 5.69422402381897,
      "epoch": 7.8052431059506535,
      "grad_norm": 1.5,
      "learning_rate": 0.00010227726733520361,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23920318037271499,
      "num_tokens": 178552089.0,
      "step": 86045
    },
    {
      "entropy": 5.710457229614258,
      "epoch": 7.805696661828737,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010225653314008852,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23951925486326217,
      "num_tokens": 178562198.0,
      "step": 86050
    },
    {
      "entropy": 5.6377252578735355,
      "epoch": 7.806150217706821,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00010223580251735235,
      "loss": 4.7186,
      "mean_token_accuracy": 0.25125600695610045,
      "num_tokens": 178573245.0,
      "step": 86055
    },
    {
      "entropy": 5.67108564376831,
      "epoch": 7.806603773584905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010221507546742381,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22680799216032027,
      "num_tokens": 178583636.0,
      "step": 86060
    },
    {
      "entropy": 5.653881072998047,
      "epoch": 7.80705732946299,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010219435199073152,
      "loss": 4.784,
      "mean_token_accuracy": 0.23303370475769042,
      "num_tokens": 178594586.0,
      "step": 86065
    },
    {
      "entropy": 5.663761472702026,
      "epoch": 7.807510885341074,
      "grad_norm": 1.5,
      "learning_rate": 0.00010217363208770412,
      "loss": 4.7549,
      "mean_token_accuracy": 0.2408065140247345,
      "num_tokens": 178605315.0,
      "step": 86070
    },
    {
      "entropy": 5.696752738952637,
      "epoch": 7.807964441219158,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010215291575876992,
      "loss": 4.8281,
      "mean_token_accuracy": 0.23420400172472,
      "num_tokens": 178615497.0,
      "step": 86075
    },
    {
      "entropy": 5.718396377563477,
      "epoch": 7.8084179970972425,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010213220300435755,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23014033436775208,
      "num_tokens": 178625376.0,
      "step": 86080
    },
    {
      "entropy": 5.616784620285034,
      "epoch": 7.808871552975327,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001021114938248953,
      "loss": 4.806,
      "mean_token_accuracy": 0.23536321818828582,
      "num_tokens": 178637589.0,
      "step": 86085
    },
    {
      "entropy": 5.731557559967041,
      "epoch": 7.809325108853411,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001020907882208114,
      "loss": 4.875,
      "mean_token_accuracy": 0.23482240438461305,
      "num_tokens": 178649432.0,
      "step": 86090
    },
    {
      "entropy": 5.702654075622559,
      "epoch": 7.809778664731495,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010207008619253406,
      "loss": 4.7959,
      "mean_token_accuracy": 0.24470877051353454,
      "num_tokens": 178659063.0,
      "step": 86095
    },
    {
      "entropy": 5.742581748962403,
      "epoch": 7.81023222060958,
      "grad_norm": 1.625,
      "learning_rate": 0.0001020493877404914,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22521349638700486,
      "num_tokens": 178669761.0,
      "step": 86100
    },
    {
      "entropy": 5.662538194656372,
      "epoch": 7.810685776487663,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010202869286511143,
      "loss": 4.7725,
      "mean_token_accuracy": 0.2394942596554756,
      "num_tokens": 178680144.0,
      "step": 86105
    },
    {
      "entropy": 5.720706987380981,
      "epoch": 7.811139332365747,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010200800156682222,
      "loss": 4.8366,
      "mean_token_accuracy": 0.2353087529540062,
      "num_tokens": 178690582.0,
      "step": 86110
    },
    {
      "entropy": 5.738883543014526,
      "epoch": 7.8115928882438315,
      "grad_norm": 1.703125,
      "learning_rate": 0.00010198731384605163,
      "loss": 4.8767,
      "mean_token_accuracy": 0.2302405521273613,
      "num_tokens": 178700200.0,
      "step": 86115
    },
    {
      "entropy": 5.7072797298431395,
      "epoch": 7.812046444121916,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010196662970322749,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22845925837755204,
      "num_tokens": 178710216.0,
      "step": 86120
    },
    {
      "entropy": 5.756412935256958,
      "epoch": 7.8125,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001019459491387775,
      "loss": 4.8639,
      "mean_token_accuracy": 0.22948063313961028,
      "num_tokens": 178720133.0,
      "step": 86125
    },
    {
      "entropy": 5.667602491378784,
      "epoch": 7.812953555878084,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010192527215312942,
      "loss": 4.7447,
      "mean_token_accuracy": 0.23655675649642943,
      "num_tokens": 178729947.0,
      "step": 86130
    },
    {
      "entropy": 5.679594421386719,
      "epoch": 7.8134071117561685,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010190459874671079,
      "loss": 4.808,
      "mean_token_accuracy": 0.2453747883439064,
      "num_tokens": 178739950.0,
      "step": 86135
    },
    {
      "entropy": 5.688198852539062,
      "epoch": 7.813860667634253,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001018839289199491,
      "loss": 4.759,
      "mean_token_accuracy": 0.2491427555680275,
      "num_tokens": 178749333.0,
      "step": 86140
    },
    {
      "entropy": 5.682958507537842,
      "epoch": 7.814314223512337,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010186326267327198,
      "loss": 4.7519,
      "mean_token_accuracy": 0.243946373462677,
      "num_tokens": 178760659.0,
      "step": 86145
    },
    {
      "entropy": 5.657117557525635,
      "epoch": 7.81476777939042,
      "grad_norm": 1.5,
      "learning_rate": 0.00010184260000710668,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23208971023559571,
      "num_tokens": 178771187.0,
      "step": 86150
    },
    {
      "entropy": 5.723793649673462,
      "epoch": 7.815221335268505,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010182194092188052,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2293289750814438,
      "num_tokens": 178780609.0,
      "step": 86155
    },
    {
      "entropy": 5.606597375869751,
      "epoch": 7.815674891146589,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010180128541802078,
      "loss": 4.7407,
      "mean_token_accuracy": 0.23878353983163833,
      "num_tokens": 178791529.0,
      "step": 86160
    },
    {
      "entropy": 5.684730005264282,
      "epoch": 7.816128447024673,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010178063349595459,
      "loss": 4.7965,
      "mean_token_accuracy": 0.2362723395228386,
      "num_tokens": 178802159.0,
      "step": 86165
    },
    {
      "entropy": 5.730052375793457,
      "epoch": 7.8165820029027575,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010175998515610905,
      "loss": 4.814,
      "mean_token_accuracy": 0.24270088523626326,
      "num_tokens": 178812135.0,
      "step": 86170
    },
    {
      "entropy": 5.681087064743042,
      "epoch": 7.817035558780842,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010173934039891108,
      "loss": 4.7756,
      "mean_token_accuracy": 0.23725817054510118,
      "num_tokens": 178823413.0,
      "step": 86175
    },
    {
      "entropy": 5.667197322845459,
      "epoch": 7.817489114658926,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001017186992247878,
      "loss": 4.7814,
      "mean_token_accuracy": 0.23518887609243394,
      "num_tokens": 178832716.0,
      "step": 86180
    },
    {
      "entropy": 5.638728189468384,
      "epoch": 7.81794267053701,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010169806163416593,
      "loss": 4.762,
      "mean_token_accuracy": 0.23659597784280778,
      "num_tokens": 178844584.0,
      "step": 86185
    },
    {
      "entropy": 5.714158582687378,
      "epoch": 7.818396226415095,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010167742762747234,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23811686635017396,
      "num_tokens": 178853866.0,
      "step": 86190
    },
    {
      "entropy": 5.720226097106933,
      "epoch": 7.818849782293179,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010165679720513368,
      "loss": 4.8932,
      "mean_token_accuracy": 0.23529156744480134,
      "num_tokens": 178864340.0,
      "step": 86195
    },
    {
      "entropy": 5.720113325119018,
      "epoch": 7.819303338171263,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010163617036757668,
      "loss": 4.8613,
      "mean_token_accuracy": 0.23011814504861833,
      "num_tokens": 178873890.0,
      "step": 86200
    },
    {
      "entropy": 5.743004369735718,
      "epoch": 7.8197568940493465,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010161554711522782,
      "loss": 4.8716,
      "mean_token_accuracy": 0.23582170456647872,
      "num_tokens": 178883761.0,
      "step": 86205
    },
    {
      "entropy": 5.724093580245972,
      "epoch": 7.820210449927431,
      "grad_norm": 1.734375,
      "learning_rate": 0.00010159492744851357,
      "loss": 4.8425,
      "mean_token_accuracy": 0.23831141889095306,
      "num_tokens": 178892879.0,
      "step": 86210
    },
    {
      "entropy": 5.719255352020264,
      "epoch": 7.820664005805515,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010157431136786046,
      "loss": 4.8237,
      "mean_token_accuracy": 0.23381189703941346,
      "num_tokens": 178903511.0,
      "step": 86215
    },
    {
      "entropy": 5.631912899017334,
      "epoch": 7.821117561683599,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001015536988736948,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23858722746372224,
      "num_tokens": 178913179.0,
      "step": 86220
    },
    {
      "entropy": 5.6542699337005615,
      "epoch": 7.8215711175616836,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010153308996644286,
      "loss": 4.7654,
      "mean_token_accuracy": 0.2440857008099556,
      "num_tokens": 178922946.0,
      "step": 86225
    },
    {
      "entropy": 5.701928567886353,
      "epoch": 7.822024673439768,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010151248464653079,
      "loss": 4.8456,
      "mean_token_accuracy": 0.23106411695480347,
      "num_tokens": 178933527.0,
      "step": 86230
    },
    {
      "entropy": 5.7197963237762455,
      "epoch": 7.822478229317852,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010149188291438479,
      "loss": 4.8538,
      "mean_token_accuracy": 0.23216979652643205,
      "num_tokens": 178943650.0,
      "step": 86235
    },
    {
      "entropy": 5.688753938674926,
      "epoch": 7.822931785195936,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010147128477043083,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2375294879078865,
      "num_tokens": 178954208.0,
      "step": 86240
    },
    {
      "entropy": 5.722055816650391,
      "epoch": 7.823385341074021,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001014506902150949,
      "loss": 4.8139,
      "mean_token_accuracy": 0.242082841694355,
      "num_tokens": 178964352.0,
      "step": 86245
    },
    {
      "entropy": 5.718559741973877,
      "epoch": 7.823838896952105,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00010143009924880295,
      "loss": 4.8227,
      "mean_token_accuracy": 0.24026339799165725,
      "num_tokens": 178974218.0,
      "step": 86250
    },
    {
      "entropy": 5.6428765773773195,
      "epoch": 7.824292452830189,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001014095118719808,
      "loss": 4.7695,
      "mean_token_accuracy": 0.23885969519615174,
      "num_tokens": 178985288.0,
      "step": 86255
    },
    {
      "entropy": 5.71118016242981,
      "epoch": 7.8247460087082725,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010138892808505413,
      "loss": 4.8421,
      "mean_token_accuracy": 0.23163340985774994,
      "num_tokens": 178995458.0,
      "step": 86260
    },
    {
      "entropy": 5.719559669494629,
      "epoch": 7.825199564586357,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010136834788844873,
      "loss": 4.8434,
      "mean_token_accuracy": 0.23904865980148315,
      "num_tokens": 179006592.0,
      "step": 86265
    },
    {
      "entropy": 5.7151713371276855,
      "epoch": 7.825653120464441,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001013477712825901,
      "loss": 4.83,
      "mean_token_accuracy": 0.233533576130867,
      "num_tokens": 179015877.0,
      "step": 86270
    },
    {
      "entropy": 5.663133430480957,
      "epoch": 7.826106676342525,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010132719826790385,
      "loss": 4.7544,
      "mean_token_accuracy": 0.23984033912420272,
      "num_tokens": 179025776.0,
      "step": 86275
    },
    {
      "entropy": 5.709198093414306,
      "epoch": 7.82656023222061,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010130662884481531,
      "loss": 4.8476,
      "mean_token_accuracy": 0.23995332717895507,
      "num_tokens": 179036204.0,
      "step": 86280
    },
    {
      "entropy": 5.6702831268310545,
      "epoch": 7.827013788098694,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010128606301375001,
      "loss": 4.8035,
      "mean_token_accuracy": 0.2453102722764015,
      "num_tokens": 179046743.0,
      "step": 86285
    },
    {
      "entropy": 5.636453247070312,
      "epoch": 7.827467343976778,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010126550077513322,
      "loss": 4.778,
      "mean_token_accuracy": 0.23973690420389177,
      "num_tokens": 179057671.0,
      "step": 86290
    },
    {
      "entropy": 5.655806255340576,
      "epoch": 7.827920899854862,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010124494212939018,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23149520307779312,
      "num_tokens": 179066807.0,
      "step": 86295
    },
    {
      "entropy": 5.671860599517823,
      "epoch": 7.828374455732947,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010122438707694595,
      "loss": 4.7811,
      "mean_token_accuracy": 0.23212903887033462,
      "num_tokens": 179076820.0,
      "step": 86300
    },
    {
      "entropy": 5.630180597305298,
      "epoch": 7.82882801161103,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010120383561822563,
      "loss": 4.7267,
      "mean_token_accuracy": 0.24701191186904908,
      "num_tokens": 179086781.0,
      "step": 86305
    },
    {
      "entropy": 5.724128150939942,
      "epoch": 7.829281567489114,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010118328775365434,
      "loss": 4.9597,
      "mean_token_accuracy": 0.23499705046415328,
      "num_tokens": 179097055.0,
      "step": 86310
    },
    {
      "entropy": 5.6570515632629395,
      "epoch": 7.829735123367199,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010116274348365694,
      "loss": 4.7671,
      "mean_token_accuracy": 0.23524074703454972,
      "num_tokens": 179107691.0,
      "step": 86315
    },
    {
      "entropy": 5.6919771194458,
      "epoch": 7.830188679245283,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010114220280865831,
      "loss": 4.8555,
      "mean_token_accuracy": 0.23105887919664383,
      "num_tokens": 179117917.0,
      "step": 86320
    },
    {
      "entropy": 5.759331274032593,
      "epoch": 7.830642235123367,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010112166572908324,
      "loss": 4.8685,
      "mean_token_accuracy": 0.2363256722688675,
      "num_tokens": 179128308.0,
      "step": 86325
    },
    {
      "entropy": 5.698671579360962,
      "epoch": 7.831095791001451,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010110113224535644,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23412371575832366,
      "num_tokens": 179139038.0,
      "step": 86330
    },
    {
      "entropy": 5.652106046676636,
      "epoch": 7.831549346879536,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001010806023579025,
      "loss": 4.7727,
      "mean_token_accuracy": 0.23806941211223603,
      "num_tokens": 179149440.0,
      "step": 86335
    },
    {
      "entropy": 5.80371470451355,
      "epoch": 7.83200290275762,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00010106007606714602,
      "loss": 4.9388,
      "mean_token_accuracy": 0.22728834748268129,
      "num_tokens": 179160794.0,
      "step": 86340
    },
    {
      "entropy": 5.664030313491821,
      "epoch": 7.832456458635704,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001010395533735115,
      "loss": 4.7295,
      "mean_token_accuracy": 0.2351341053843498,
      "num_tokens": 179171763.0,
      "step": 86345
    },
    {
      "entropy": 5.690585994720459,
      "epoch": 7.832910014513788,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010101903427742336,
      "loss": 4.8328,
      "mean_token_accuracy": 0.23345271199941636,
      "num_tokens": 179182170.0,
      "step": 86350
    },
    {
      "entropy": 5.698733186721801,
      "epoch": 7.833363570391873,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00010099851877930594,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23639626950025558,
      "num_tokens": 179192092.0,
      "step": 86355
    },
    {
      "entropy": 5.6814781665802006,
      "epoch": 7.833817126269956,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001009780068795835,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23678032606840133,
      "num_tokens": 179202368.0,
      "step": 86360
    },
    {
      "entropy": 5.647631454467773,
      "epoch": 7.83427068214804,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001009574985786802,
      "loss": 4.8062,
      "mean_token_accuracy": 0.2385762467980385,
      "num_tokens": 179213092.0,
      "step": 86365
    },
    {
      "entropy": 5.673862791061401,
      "epoch": 7.834724238026125,
      "grad_norm": 1.84375,
      "learning_rate": 0.00010093699387702019,
      "loss": 4.7352,
      "mean_token_accuracy": 0.24091971963644027,
      "num_tokens": 179223430.0,
      "step": 86370
    },
    {
      "entropy": 5.735047435760498,
      "epoch": 7.835177793904209,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00010091649277502744,
      "loss": 4.8738,
      "mean_token_accuracy": 0.22536443471908568,
      "num_tokens": 179233576.0,
      "step": 86375
    },
    {
      "entropy": 5.727531766891479,
      "epoch": 7.835631349782293,
      "grad_norm": 1.65625,
      "learning_rate": 0.00010089599527312607,
      "loss": 4.8127,
      "mean_token_accuracy": 0.24291347712278366,
      "num_tokens": 179242849.0,
      "step": 86380
    },
    {
      "entropy": 5.694306421279907,
      "epoch": 7.836084905660377,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010087550137173985,
      "loss": 4.7556,
      "mean_token_accuracy": 0.23726035207509993,
      "num_tokens": 179253104.0,
      "step": 86385
    },
    {
      "entropy": 5.722910404205322,
      "epoch": 7.836538461538462,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010085501107129264,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23750170320272446,
      "num_tokens": 179264842.0,
      "step": 86390
    },
    {
      "entropy": 5.673921346664429,
      "epoch": 7.836992017416546,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010083452437220818,
      "loss": 4.7682,
      "mean_token_accuracy": 0.2378324955701828,
      "num_tokens": 179275270.0,
      "step": 86395
    },
    {
      "entropy": 5.778890895843506,
      "epoch": 7.83744557329463,
      "grad_norm": 1.671875,
      "learning_rate": 0.00010081404127491013,
      "loss": 4.915,
      "mean_token_accuracy": 0.2234516531229019,
      "num_tokens": 179285602.0,
      "step": 86400
    },
    {
      "entropy": 5.618832015991211,
      "epoch": 7.837899129172714,
      "grad_norm": 1.65625,
      "learning_rate": 0.00010079356177982208,
      "loss": 4.737,
      "mean_token_accuracy": 0.24127089232206345,
      "num_tokens": 179295778.0,
      "step": 86405
    },
    {
      "entropy": 5.618314123153686,
      "epoch": 7.838352685050799,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010077308588736755,
      "loss": 4.7351,
      "mean_token_accuracy": 0.24436119943857193,
      "num_tokens": 179306824.0,
      "step": 86410
    },
    {
      "entropy": 5.651763391494751,
      "epoch": 7.838806240928882,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010075261359797007,
      "loss": 4.8194,
      "mean_token_accuracy": 0.23072778731584548,
      "num_tokens": 179317818.0,
      "step": 86415
    },
    {
      "entropy": 5.6758404731750485,
      "epoch": 7.839259796806966,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010073214491205291,
      "loss": 4.8424,
      "mean_token_accuracy": 0.2360488221049309,
      "num_tokens": 179327939.0,
      "step": 86420
    },
    {
      "entropy": 5.699243783950806,
      "epoch": 7.839713352685051,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001007116798300394,
      "loss": 4.7912,
      "mean_token_accuracy": 0.23792139887809755,
      "num_tokens": 179338173.0,
      "step": 86425
    },
    {
      "entropy": 5.710375833511352,
      "epoch": 7.840166908563135,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010069121835235279,
      "loss": 4.8392,
      "mean_token_accuracy": 0.24055782556533814,
      "num_tokens": 179349219.0,
      "step": 86430
    },
    {
      "entropy": 5.704298400878907,
      "epoch": 7.840620464441219,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010067076047941619,
      "loss": 4.7832,
      "mean_token_accuracy": 0.2422276809811592,
      "num_tokens": 179359136.0,
      "step": 86435
    },
    {
      "entropy": 5.694571018218994,
      "epoch": 7.8410740203193035,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010065030621165263,
      "loss": 4.8355,
      "mean_token_accuracy": 0.23942452371120454,
      "num_tokens": 179370016.0,
      "step": 86440
    },
    {
      "entropy": 5.710232973098755,
      "epoch": 7.841527576197388,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001006298555494852,
      "loss": 4.8469,
      "mean_token_accuracy": 0.23596456050872802,
      "num_tokens": 179381361.0,
      "step": 86445
    },
    {
      "entropy": 5.708321189880371,
      "epoch": 7.841981132075472,
      "grad_norm": 1.625,
      "learning_rate": 0.00010060940849333682,
      "loss": 4.8083,
      "mean_token_accuracy": 0.2406468451023102,
      "num_tokens": 179391623.0,
      "step": 86450
    },
    {
      "entropy": 5.707652854919433,
      "epoch": 7.842434687953556,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010058896504363033,
      "loss": 4.9037,
      "mean_token_accuracy": 0.23392214328050615,
      "num_tokens": 179402885.0,
      "step": 86455
    },
    {
      "entropy": 5.74942626953125,
      "epoch": 7.84288824383164,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010056852520078843,
      "loss": 4.8469,
      "mean_token_accuracy": 0.23376302570104598,
      "num_tokens": 179413373.0,
      "step": 86460
    },
    {
      "entropy": 5.670090913772583,
      "epoch": 7.843341799709724,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010054808896523393,
      "loss": 4.7443,
      "mean_token_accuracy": 0.24494732022285462,
      "num_tokens": 179424323.0,
      "step": 86465
    },
    {
      "entropy": 5.640061330795288,
      "epoch": 7.843795355587808,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001005276563373894,
      "loss": 4.7014,
      "mean_token_accuracy": 0.2452957287430763,
      "num_tokens": 179434218.0,
      "step": 86470
    },
    {
      "entropy": 5.681114196777344,
      "epoch": 7.844248911465892,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010050722731767731,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23694107234477996,
      "num_tokens": 179444634.0,
      "step": 86475
    },
    {
      "entropy": 5.731177520751953,
      "epoch": 7.844702467343977,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010048680190652033,
      "loss": 4.808,
      "mean_token_accuracy": 0.2342948004603386,
      "num_tokens": 179453750.0,
      "step": 86480
    },
    {
      "entropy": 5.641708040237427,
      "epoch": 7.845156023222061,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010046638010434071,
      "loss": 4.8014,
      "mean_token_accuracy": 0.23531881123781204,
      "num_tokens": 179463591.0,
      "step": 86485
    },
    {
      "entropy": 5.703209733963012,
      "epoch": 7.845609579100145,
      "grad_norm": 1.5,
      "learning_rate": 0.00010044596191156085,
      "loss": 4.8035,
      "mean_token_accuracy": 0.23855825513601303,
      "num_tokens": 179474433.0,
      "step": 86490
    },
    {
      "entropy": 5.657539987564087,
      "epoch": 7.8460631349782295,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010042554732860299,
      "loss": 4.7835,
      "mean_token_accuracy": 0.23092390298843385,
      "num_tokens": 179484464.0,
      "step": 86495
    },
    {
      "entropy": 5.7364644527435305,
      "epoch": 7.846516690856314,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010040513635588929,
      "loss": 4.8714,
      "mean_token_accuracy": 0.23807385116815566,
      "num_tokens": 179494809.0,
      "step": 86500
    },
    {
      "entropy": 5.696652460098266,
      "epoch": 7.846970246734398,
      "grad_norm": 1.5,
      "learning_rate": 0.00010038472899384183,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23288713842630387,
      "num_tokens": 179504423.0,
      "step": 86505
    },
    {
      "entropy": 5.7624493598937985,
      "epoch": 7.847423802612482,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010036432524288264,
      "loss": 4.8736,
      "mean_token_accuracy": 0.22644934952259063,
      "num_tokens": 179515158.0,
      "step": 86510
    },
    {
      "entropy": 5.627625465393066,
      "epoch": 7.847877358490566,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001003439251034338,
      "loss": 4.684,
      "mean_token_accuracy": 0.24280933886766434,
      "num_tokens": 179526100.0,
      "step": 86515
    },
    {
      "entropy": 5.73229022026062,
      "epoch": 7.84833091436865,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010032352857591704,
      "loss": 4.8718,
      "mean_token_accuracy": 0.22528907060623168,
      "num_tokens": 179536919.0,
      "step": 86520
    },
    {
      "entropy": 5.686948013305664,
      "epoch": 7.848784470246734,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010030313566075424,
      "loss": 4.7443,
      "mean_token_accuracy": 0.24693213403224945,
      "num_tokens": 179546782.0,
      "step": 86525
    },
    {
      "entropy": 5.658621883392334,
      "epoch": 7.8492380261248185,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010028274635836712,
      "loss": 4.7714,
      "mean_token_accuracy": 0.24492305964231492,
      "num_tokens": 179557813.0,
      "step": 86530
    },
    {
      "entropy": 5.714878702163697,
      "epoch": 7.849691582002903,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010026236066917733,
      "loss": 4.8217,
      "mean_token_accuracy": 0.2296708956360817,
      "num_tokens": 179568965.0,
      "step": 86535
    },
    {
      "entropy": 5.707384729385376,
      "epoch": 7.850145137880987,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010024197859360642,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22908064723014832,
      "num_tokens": 179580780.0,
      "step": 86540
    },
    {
      "entropy": 5.747546195983887,
      "epoch": 7.850598693759071,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001002216001320759,
      "loss": 4.888,
      "mean_token_accuracy": 0.23333926498889923,
      "num_tokens": 179591384.0,
      "step": 86545
    },
    {
      "entropy": 5.689999055862427,
      "epoch": 7.8510522496371555,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010020122528500731,
      "loss": 4.7449,
      "mean_token_accuracy": 0.2404682919383049,
      "num_tokens": 179600310.0,
      "step": 86550
    },
    {
      "entropy": 5.677389907836914,
      "epoch": 7.85150580551524,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010018085405282188,
      "loss": 4.7945,
      "mean_token_accuracy": 0.23915417939424516,
      "num_tokens": 179612703.0,
      "step": 86555
    },
    {
      "entropy": 5.682439851760864,
      "epoch": 7.851959361393323,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010016048643594094,
      "loss": 4.7894,
      "mean_token_accuracy": 0.24128830879926683,
      "num_tokens": 179623390.0,
      "step": 86560
    },
    {
      "entropy": 5.684696531295776,
      "epoch": 7.852412917271408,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010014012243478571,
      "loss": 4.7549,
      "mean_token_accuracy": 0.23215144574642183,
      "num_tokens": 179633545.0,
      "step": 86565
    },
    {
      "entropy": 5.681676292419434,
      "epoch": 7.852866473149492,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010011976204977732,
      "loss": 4.7038,
      "mean_token_accuracy": 0.23497969955205916,
      "num_tokens": 179643494.0,
      "step": 86570
    },
    {
      "entropy": 5.646165561676026,
      "epoch": 7.853320029027576,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010009940528133679,
      "loss": 4.7035,
      "mean_token_accuracy": 0.24650683104991913,
      "num_tokens": 179653750.0,
      "step": 86575
    },
    {
      "entropy": 5.689845705032349,
      "epoch": 7.85377358490566,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010007905212988507,
      "loss": 4.8061,
      "mean_token_accuracy": 0.2374645099043846,
      "num_tokens": 179664469.0,
      "step": 86580
    },
    {
      "entropy": 5.72558708190918,
      "epoch": 7.8542271407837445,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00010005870259584318,
      "loss": 4.821,
      "mean_token_accuracy": 0.23779136687517166,
      "num_tokens": 179674281.0,
      "step": 86585
    },
    {
      "entropy": 5.774060297012329,
      "epoch": 7.854680696661829,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010003835667963194,
      "loss": 4.8896,
      "mean_token_accuracy": 0.231167671084404,
      "num_tokens": 179685036.0,
      "step": 86590
    },
    {
      "entropy": 5.725205945968628,
      "epoch": 7.855134252539913,
      "grad_norm": 1.6875,
      "learning_rate": 0.00010001801438167205,
      "loss": 4.8298,
      "mean_token_accuracy": 0.2367906764149666,
      "num_tokens": 179694722.0,
      "step": 86595
    },
    {
      "entropy": 5.717504787445068,
      "epoch": 7.855587808417997,
      "grad_norm": 1.5859375,
      "learning_rate": 9.999767570238421e-05,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22615507543087005,
      "num_tokens": 179704633.0,
      "step": 86600
    },
    {
      "entropy": 5.746246862411499,
      "epoch": 7.856041364296082,
      "grad_norm": 1.421875,
      "learning_rate": 9.997734064218903e-05,
      "loss": 4.8918,
      "mean_token_accuracy": 0.23106108605861664,
      "num_tokens": 179715289.0,
      "step": 86605
    },
    {
      "entropy": 5.589977121353149,
      "epoch": 7.856494920174166,
      "grad_norm": 1.5625,
      "learning_rate": 9.995700920150698e-05,
      "loss": 4.665,
      "mean_token_accuracy": 0.2567407667636871,
      "num_tokens": 179724982.0,
      "step": 86610
    },
    {
      "entropy": 5.668941307067871,
      "epoch": 7.856948476052249,
      "grad_norm": 1.6796875,
      "learning_rate": 9.993668138075865e-05,
      "loss": 4.773,
      "mean_token_accuracy": 0.23869429677724838,
      "num_tokens": 179734753.0,
      "step": 86615
    },
    {
      "entropy": 5.702511548995972,
      "epoch": 7.8574020319303335,
      "grad_norm": 1.46875,
      "learning_rate": 9.991635718036436e-05,
      "loss": 4.7852,
      "mean_token_accuracy": 0.24054819345474243,
      "num_tokens": 179745192.0,
      "step": 86620
    },
    {
      "entropy": 5.713836050033569,
      "epoch": 7.857855587808418,
      "grad_norm": 1.3359375,
      "learning_rate": 9.989603660074443e-05,
      "loss": 4.8502,
      "mean_token_accuracy": 0.23496623784303666,
      "num_tokens": 179755967.0,
      "step": 86625
    },
    {
      "entropy": 5.693475151062012,
      "epoch": 7.858309143686502,
      "grad_norm": 1.5625,
      "learning_rate": 9.987571964231909e-05,
      "loss": 4.8114,
      "mean_token_accuracy": 0.2325301930308342,
      "num_tokens": 179765950.0,
      "step": 86630
    },
    {
      "entropy": 5.564405918121338,
      "epoch": 7.858762699564586,
      "grad_norm": 1.53125,
      "learning_rate": 9.985540630550846e-05,
      "loss": 4.6517,
      "mean_token_accuracy": 0.2534072086215019,
      "num_tokens": 179775609.0,
      "step": 86635
    },
    {
      "entropy": 5.67115044593811,
      "epoch": 7.8592162554426706,
      "grad_norm": 1.4921875,
      "learning_rate": 9.98350965907327e-05,
      "loss": 4.7931,
      "mean_token_accuracy": 0.2409462422132492,
      "num_tokens": 179785951.0,
      "step": 86640
    },
    {
      "entropy": 5.713604974746704,
      "epoch": 7.859669811320755,
      "grad_norm": 1.390625,
      "learning_rate": 9.981479049841171e-05,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22869023978710173,
      "num_tokens": 179798210.0,
      "step": 86645
    },
    {
      "entropy": 5.724179935455322,
      "epoch": 7.860123367198839,
      "grad_norm": 1.5390625,
      "learning_rate": 9.979448802896555e-05,
      "loss": 4.7644,
      "mean_token_accuracy": 0.24998607039451598,
      "num_tokens": 179808043.0,
      "step": 86650
    },
    {
      "entropy": 5.690302133560181,
      "epoch": 7.860576923076923,
      "grad_norm": 1.5859375,
      "learning_rate": 9.9774189182814e-05,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23437707573175431,
      "num_tokens": 179817793.0,
      "step": 86655
    },
    {
      "entropy": 5.753344964981079,
      "epoch": 7.861030478955008,
      "grad_norm": 1.4609375,
      "learning_rate": 9.975389396037691e-05,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22831355780363083,
      "num_tokens": 179828667.0,
      "step": 86660
    },
    {
      "entropy": 5.680123281478882,
      "epoch": 7.861484034833092,
      "grad_norm": 1.6484375,
      "learning_rate": 9.973360236207395e-05,
      "loss": 4.7324,
      "mean_token_accuracy": 0.2379102274775505,
      "num_tokens": 179838402.0,
      "step": 86665
    },
    {
      "entropy": 5.705038404464721,
      "epoch": 7.861937590711175,
      "grad_norm": 1.5859375,
      "learning_rate": 9.971331438832471e-05,
      "loss": 4.7924,
      "mean_token_accuracy": 0.23497118204832076,
      "num_tokens": 179849242.0,
      "step": 86670
    },
    {
      "entropy": 5.656812334060669,
      "epoch": 7.8623911465892595,
      "grad_norm": 1.5078125,
      "learning_rate": 9.969303003954885e-05,
      "loss": 4.7323,
      "mean_token_accuracy": 0.24991584569215775,
      "num_tokens": 179859519.0,
      "step": 86675
    },
    {
      "entropy": 5.677257251739502,
      "epoch": 7.862844702467344,
      "grad_norm": 1.71875,
      "learning_rate": 9.96727493161657e-05,
      "loss": 4.8073,
      "mean_token_accuracy": 0.2350766360759735,
      "num_tokens": 179869175.0,
      "step": 86680
    },
    {
      "entropy": 5.660198640823364,
      "epoch": 7.863298258345428,
      "grad_norm": 1.484375,
      "learning_rate": 9.965247221859488e-05,
      "loss": 4.7867,
      "mean_token_accuracy": 0.23501411080360413,
      "num_tokens": 179879515.0,
      "step": 86685
    },
    {
      "entropy": 5.636246871948242,
      "epoch": 7.863751814223512,
      "grad_norm": 1.4609375,
      "learning_rate": 9.96321987472556e-05,
      "loss": 4.7261,
      "mean_token_accuracy": 0.24689114838838577,
      "num_tokens": 179889416.0,
      "step": 86690
    },
    {
      "entropy": 5.7076146602630615,
      "epoch": 7.864205370101597,
      "grad_norm": 1.640625,
      "learning_rate": 9.961192890256712e-05,
      "loss": 4.8046,
      "mean_token_accuracy": 0.2399048000574112,
      "num_tokens": 179899638.0,
      "step": 86695
    },
    {
      "entropy": 5.692747163772583,
      "epoch": 7.864658925979681,
      "grad_norm": 1.609375,
      "learning_rate": 9.959166268494865e-05,
      "loss": 4.8149,
      "mean_token_accuracy": 0.23561485409736632,
      "num_tokens": 179909328.0,
      "step": 86700
    },
    {
      "entropy": 5.784570598602295,
      "epoch": 7.865112481857765,
      "grad_norm": 1.5546875,
      "learning_rate": 9.957140009481932e-05,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2272706612944603,
      "num_tokens": 179920352.0,
      "step": 86705
    },
    {
      "entropy": 5.7516358375549315,
      "epoch": 7.865566037735849,
      "grad_norm": 1.6328125,
      "learning_rate": 9.955114113259812e-05,
      "loss": 4.8673,
      "mean_token_accuracy": 0.2323925733566284,
      "num_tokens": 179930770.0,
      "step": 86710
    },
    {
      "entropy": 5.735660886764526,
      "epoch": 7.866019593613933,
      "grad_norm": 1.3671875,
      "learning_rate": 9.953088579870395e-05,
      "loss": 4.7991,
      "mean_token_accuracy": 0.23734381943941116,
      "num_tokens": 179941903.0,
      "step": 86715
    },
    {
      "entropy": 5.707276201248169,
      "epoch": 7.866473149492017,
      "grad_norm": 1.6484375,
      "learning_rate": 9.951063409355588e-05,
      "loss": 4.7956,
      "mean_token_accuracy": 0.24040569216012955,
      "num_tokens": 179952194.0,
      "step": 86720
    },
    {
      "entropy": 5.769540882110595,
      "epoch": 7.866926705370101,
      "grad_norm": 1.453125,
      "learning_rate": 9.949038601757264e-05,
      "loss": 4.8936,
      "mean_token_accuracy": 0.23602911531925203,
      "num_tokens": 179963111.0,
      "step": 86725
    },
    {
      "entropy": 5.7305991649627686,
      "epoch": 7.867380261248186,
      "grad_norm": 1.46875,
      "learning_rate": 9.947014157117291e-05,
      "loss": 4.8259,
      "mean_token_accuracy": 0.235191710293293,
      "num_tokens": 179973813.0,
      "step": 86730
    },
    {
      "entropy": 5.678429460525512,
      "epoch": 7.86783381712627,
      "grad_norm": 1.515625,
      "learning_rate": 9.944990075477536e-05,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23842615336179734,
      "num_tokens": 179984450.0,
      "step": 86735
    },
    {
      "entropy": 5.723262357711792,
      "epoch": 7.868287373004354,
      "grad_norm": 1.4921875,
      "learning_rate": 9.942966356879854e-05,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2285188242793083,
      "num_tokens": 179994614.0,
      "step": 86740
    },
    {
      "entropy": 5.670760154724121,
      "epoch": 7.868740928882438,
      "grad_norm": 1.578125,
      "learning_rate": 9.940943001366111e-05,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23753972947597504,
      "num_tokens": 180005259.0,
      "step": 86745
    },
    {
      "entropy": 5.627189636230469,
      "epoch": 7.869194484760523,
      "grad_norm": 1.4140625,
      "learning_rate": 9.938920008978139e-05,
      "loss": 4.7583,
      "mean_token_accuracy": 0.2378905713558197,
      "num_tokens": 180015796.0,
      "step": 86750
    },
    {
      "entropy": 5.744512605667114,
      "epoch": 7.869648040638607,
      "grad_norm": 1.4296875,
      "learning_rate": 9.936897379757776e-05,
      "loss": 4.8111,
      "mean_token_accuracy": 0.236707666516304,
      "num_tokens": 180026081.0,
      "step": 86755
    },
    {
      "entropy": 5.61801233291626,
      "epoch": 7.870101596516691,
      "grad_norm": 1.5234375,
      "learning_rate": 9.934875113746849e-05,
      "loss": 4.7267,
      "mean_token_accuracy": 0.24118055850267411,
      "num_tokens": 180036314.0,
      "step": 86760
    },
    {
      "entropy": 5.595945358276367,
      "epoch": 7.870555152394775,
      "grad_norm": 1.5234375,
      "learning_rate": 9.932853210987184e-05,
      "loss": 4.6727,
      "mean_token_accuracy": 0.24707953184843062,
      "num_tokens": 180046059.0,
      "step": 86765
    },
    {
      "entropy": 5.688156652450561,
      "epoch": 7.871008708272859,
      "grad_norm": 1.5078125,
      "learning_rate": 9.93083167152059e-05,
      "loss": 4.8294,
      "mean_token_accuracy": 0.23367003798484803,
      "num_tokens": 180056122.0,
      "step": 86770
    },
    {
      "entropy": 5.742759847640992,
      "epoch": 7.871462264150943,
      "grad_norm": 1.453125,
      "learning_rate": 9.928810495388868e-05,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22692245095968247,
      "num_tokens": 180066230.0,
      "step": 86775
    },
    {
      "entropy": 5.670346689224243,
      "epoch": 7.871915820029027,
      "grad_norm": 1.5234375,
      "learning_rate": 9.92678968263383e-05,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24195872843265534,
      "num_tokens": 180076825.0,
      "step": 86780
    },
    {
      "entropy": 5.7130927562713625,
      "epoch": 7.872369375907112,
      "grad_norm": 1.3984375,
      "learning_rate": 9.924769233297259e-05,
      "loss": 4.8523,
      "mean_token_accuracy": 0.23254120945930482,
      "num_tokens": 180087180.0,
      "step": 86785
    },
    {
      "entropy": 5.715428400039673,
      "epoch": 7.872822931785196,
      "grad_norm": 1.671875,
      "learning_rate": 9.922749147420937e-05,
      "loss": 4.7775,
      "mean_token_accuracy": 0.23550332933664322,
      "num_tokens": 180096208.0,
      "step": 86790
    },
    {
      "entropy": 5.6789785861969,
      "epoch": 7.87327648766328,
      "grad_norm": 1.7265625,
      "learning_rate": 9.920729425046645e-05,
      "loss": 4.7427,
      "mean_token_accuracy": 0.23811352699995042,
      "num_tokens": 180105769.0,
      "step": 86795
    },
    {
      "entropy": 5.697605133056641,
      "epoch": 7.873730043541364,
      "grad_norm": 1.53125,
      "learning_rate": 9.918710066216145e-05,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23459243327379226,
      "num_tokens": 180116529.0,
      "step": 86800
    },
    {
      "entropy": 5.718426704406738,
      "epoch": 7.874183599419449,
      "grad_norm": 1.546875,
      "learning_rate": 9.916691070971204e-05,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23122624605894088,
      "num_tokens": 180126543.0,
      "step": 86805
    },
    {
      "entropy": 5.647890710830689,
      "epoch": 7.874637155297533,
      "grad_norm": 1.5703125,
      "learning_rate": 9.914672439353564e-05,
      "loss": 4.7797,
      "mean_token_accuracy": 0.24586089551448823,
      "num_tokens": 180136994.0,
      "step": 86810
    },
    {
      "entropy": 5.70784068107605,
      "epoch": 7.875090711175617,
      "grad_norm": 1.4453125,
      "learning_rate": 9.912654171404986e-05,
      "loss": 4.761,
      "mean_token_accuracy": 0.25042120069265367,
      "num_tokens": 180147469.0,
      "step": 86815
    },
    {
      "entropy": 5.703917837142944,
      "epoch": 7.8755442670537015,
      "grad_norm": 1.5546875,
      "learning_rate": 9.9106362671672e-05,
      "loss": 4.8059,
      "mean_token_accuracy": 0.23906555026769638,
      "num_tokens": 180158531.0,
      "step": 86820
    },
    {
      "entropy": 5.682914113998413,
      "epoch": 7.875997822931785,
      "grad_norm": 1.515625,
      "learning_rate": 9.908618726681942e-05,
      "loss": 4.7718,
      "mean_token_accuracy": 0.24203581213951111,
      "num_tokens": 180169278.0,
      "step": 86825
    },
    {
      "entropy": 5.643332672119141,
      "epoch": 7.876451378809869,
      "grad_norm": 1.515625,
      "learning_rate": 9.906601549990932e-05,
      "loss": 4.7874,
      "mean_token_accuracy": 0.2419947162270546,
      "num_tokens": 180179505.0,
      "step": 86830
    },
    {
      "entropy": 5.640280771255493,
      "epoch": 7.876904934687953,
      "grad_norm": 1.6875,
      "learning_rate": 9.90458473713588e-05,
      "loss": 4.8017,
      "mean_token_accuracy": 0.2380651280283928,
      "num_tokens": 180190384.0,
      "step": 86835
    },
    {
      "entropy": 5.702350330352783,
      "epoch": 7.877358490566038,
      "grad_norm": 1.578125,
      "learning_rate": 9.902568288158503e-05,
      "loss": 4.7625,
      "mean_token_accuracy": 0.23565220683813096,
      "num_tokens": 180200482.0,
      "step": 86840
    },
    {
      "entropy": 5.635203409194946,
      "epoch": 7.877812046444122,
      "grad_norm": 1.5,
      "learning_rate": 9.900552203100491e-05,
      "loss": 4.6976,
      "mean_token_accuracy": 0.24165962785482406,
      "num_tokens": 180211180.0,
      "step": 86845
    },
    {
      "entropy": 5.673378801345825,
      "epoch": 7.878265602322206,
      "grad_norm": 1.5625,
      "learning_rate": 9.898536482003551e-05,
      "loss": 4.7845,
      "mean_token_accuracy": 0.2315540835261345,
      "num_tokens": 180220940.0,
      "step": 86850
    },
    {
      "entropy": 5.647364377975464,
      "epoch": 7.8787191582002905,
      "grad_norm": 1.4296875,
      "learning_rate": 9.89652112490936e-05,
      "loss": 4.7725,
      "mean_token_accuracy": 0.23208227902650833,
      "num_tokens": 180231804.0,
      "step": 86855
    },
    {
      "entropy": 5.719257974624634,
      "epoch": 7.879172714078375,
      "grad_norm": 1.515625,
      "learning_rate": 9.894506131859602e-05,
      "loss": 4.8456,
      "mean_token_accuracy": 0.23725466132164003,
      "num_tokens": 180242043.0,
      "step": 86860
    },
    {
      "entropy": 5.737079191207886,
      "epoch": 7.879626269956459,
      "grad_norm": 1.53125,
      "learning_rate": 9.892491502895943e-05,
      "loss": 4.9111,
      "mean_token_accuracy": 0.22686901539564133,
      "num_tokens": 180253420.0,
      "step": 86865
    },
    {
      "entropy": 5.699388313293457,
      "epoch": 7.880079825834542,
      "grad_norm": 1.4453125,
      "learning_rate": 9.890477238060045e-05,
      "loss": 4.8102,
      "mean_token_accuracy": 0.23495897501707078,
      "num_tokens": 180264557.0,
      "step": 86870
    },
    {
      "entropy": 5.7018270015716555,
      "epoch": 7.880533381712627,
      "grad_norm": 1.4765625,
      "learning_rate": 9.888463337393564e-05,
      "loss": 4.7919,
      "mean_token_accuracy": 0.23373782634735107,
      "num_tokens": 180274711.0,
      "step": 86875
    },
    {
      "entropy": 5.7939177513122555,
      "epoch": 7.880986937590711,
      "grad_norm": 1.453125,
      "learning_rate": 9.886449800938144e-05,
      "loss": 4.9076,
      "mean_token_accuracy": 0.22444244772195815,
      "num_tokens": 180284326.0,
      "step": 86880
    },
    {
      "entropy": 5.746591711044312,
      "epoch": 7.881440493468795,
      "grad_norm": 1.6171875,
      "learning_rate": 9.884436628735439e-05,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21983784586191177,
      "num_tokens": 180294904.0,
      "step": 86885
    },
    {
      "entropy": 5.612225198745728,
      "epoch": 7.881894049346879,
      "grad_norm": 1.3515625,
      "learning_rate": 9.882423820827071e-05,
      "loss": 4.6499,
      "mean_token_accuracy": 0.24555259495973586,
      "num_tokens": 180305438.0,
      "step": 86890
    },
    {
      "entropy": 5.742278337478638,
      "epoch": 7.882347605224964,
      "grad_norm": 1.359375,
      "learning_rate": 9.880411377254667e-05,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22825461626052856,
      "num_tokens": 180316434.0,
      "step": 86895
    },
    {
      "entropy": 5.647594404220581,
      "epoch": 7.882801161103048,
      "grad_norm": 1.4453125,
      "learning_rate": 9.878399298059848e-05,
      "loss": 4.774,
      "mean_token_accuracy": 0.23640044629573823,
      "num_tokens": 180328039.0,
      "step": 86900
    },
    {
      "entropy": 5.707698774337769,
      "epoch": 7.883254716981132,
      "grad_norm": 1.578125,
      "learning_rate": 9.876387583284218e-05,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23674561232328414,
      "num_tokens": 180337889.0,
      "step": 86905
    },
    {
      "entropy": 5.6862774848937985,
      "epoch": 7.8837082728592165,
      "grad_norm": 1.4453125,
      "learning_rate": 9.874376232969389e-05,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23799269199371337,
      "num_tokens": 180349395.0,
      "step": 86910
    },
    {
      "entropy": 5.5911292552948,
      "epoch": 7.884161828737301,
      "grad_norm": 1.5625,
      "learning_rate": 9.872365247156942e-05,
      "loss": 4.6599,
      "mean_token_accuracy": 0.2485306367278099,
      "num_tokens": 180359526.0,
      "step": 86915
    },
    {
      "entropy": 5.673538684844971,
      "epoch": 7.884615384615385,
      "grad_norm": 1.453125,
      "learning_rate": 9.870354625888484e-05,
      "loss": 4.7465,
      "mean_token_accuracy": 0.2410079687833786,
      "num_tokens": 180370803.0,
      "step": 86920
    },
    {
      "entropy": 5.693986177444458,
      "epoch": 7.885068940493468,
      "grad_norm": 1.4140625,
      "learning_rate": 9.868344369205578e-05,
      "loss": 4.8095,
      "mean_token_accuracy": 0.2347305878996849,
      "num_tokens": 180381385.0,
      "step": 86925
    },
    {
      "entropy": 5.75990662574768,
      "epoch": 7.885522496371553,
      "grad_norm": 1.5859375,
      "learning_rate": 9.86633447714981e-05,
      "loss": 4.8706,
      "mean_token_accuracy": 0.22930206805467607,
      "num_tokens": 180391887.0,
      "step": 86930
    },
    {
      "entropy": 5.656315279006958,
      "epoch": 7.885976052249637,
      "grad_norm": 1.484375,
      "learning_rate": 9.864324949762735e-05,
      "loss": 4.7505,
      "mean_token_accuracy": 0.240347421169281,
      "num_tokens": 180402676.0,
      "step": 86935
    },
    {
      "entropy": 5.684089088439942,
      "epoch": 7.886429608127721,
      "grad_norm": 1.3828125,
      "learning_rate": 9.862315787085912e-05,
      "loss": 4.7759,
      "mean_token_accuracy": 0.2337876260280609,
      "num_tokens": 180413555.0,
      "step": 86940
    },
    {
      "entropy": 5.7120500087738035,
      "epoch": 7.8868831640058055,
      "grad_norm": 1.5078125,
      "learning_rate": 9.860306989160892e-05,
      "loss": 4.8234,
      "mean_token_accuracy": 0.231971275806427,
      "num_tokens": 180424255.0,
      "step": 86945
    },
    {
      "entropy": 5.7870752811431885,
      "epoch": 7.88733671988389,
      "grad_norm": 1.5546875,
      "learning_rate": 9.858298556029224e-05,
      "loss": 4.8755,
      "mean_token_accuracy": 0.23355749100446702,
      "num_tokens": 180434625.0,
      "step": 86950
    },
    {
      "entropy": 5.694363927841186,
      "epoch": 7.887790275761974,
      "grad_norm": 1.3671875,
      "learning_rate": 9.856290487732435e-05,
      "loss": 4.7861,
      "mean_token_accuracy": 0.23962070941925048,
      "num_tokens": 180445419.0,
      "step": 86955
    },
    {
      "entropy": 5.694377708435058,
      "epoch": 7.888243831640058,
      "grad_norm": 1.578125,
      "learning_rate": 9.854282784312054e-05,
      "loss": 4.7435,
      "mean_token_accuracy": 0.24919478893280028,
      "num_tokens": 180457263.0,
      "step": 86960
    },
    {
      "entropy": 5.718230295181274,
      "epoch": 7.8886973875181425,
      "grad_norm": 1.4296875,
      "learning_rate": 9.852275445809606e-05,
      "loss": 4.8309,
      "mean_token_accuracy": 0.23803486078977584,
      "num_tokens": 180468463.0,
      "step": 86965
    },
    {
      "entropy": 5.6620911121368405,
      "epoch": 7.889150943396227,
      "grad_norm": 1.5703125,
      "learning_rate": 9.850268472266595e-05,
      "loss": 4.787,
      "mean_token_accuracy": 0.24466413110494614,
      "num_tokens": 180478321.0,
      "step": 86970
    },
    {
      "entropy": 5.716465902328491,
      "epoch": 7.889604499274311,
      "grad_norm": 1.3828125,
      "learning_rate": 9.848261863724534e-05,
      "loss": 4.7874,
      "mean_token_accuracy": 0.2369163975119591,
      "num_tokens": 180488649.0,
      "step": 86975
    },
    {
      "entropy": 5.722262620925903,
      "epoch": 7.8900580551523944,
      "grad_norm": 1.4609375,
      "learning_rate": 9.846255620224906e-05,
      "loss": 4.7673,
      "mean_token_accuracy": 0.23816208243370057,
      "num_tokens": 180499586.0,
      "step": 86980
    },
    {
      "entropy": 5.713856220245361,
      "epoch": 7.890511611030479,
      "grad_norm": 1.3828125,
      "learning_rate": 9.844249741809218e-05,
      "loss": 4.7834,
      "mean_token_accuracy": 0.24161417484283448,
      "num_tokens": 180509663.0,
      "step": 86985
    },
    {
      "entropy": 5.713913631439209,
      "epoch": 7.890965166908563,
      "grad_norm": 1.5078125,
      "learning_rate": 9.842244228518943e-05,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23925357311964035,
      "num_tokens": 180520115.0,
      "step": 86990
    },
    {
      "entropy": 5.7134356021881105,
      "epoch": 7.891418722786647,
      "grad_norm": 1.59375,
      "learning_rate": 9.840239080395557e-05,
      "loss": 4.8491,
      "mean_token_accuracy": 0.23090745955705644,
      "num_tokens": 180530781.0,
      "step": 86995
    },
    {
      "entropy": 5.719032049179077,
      "epoch": 7.8918722786647315,
      "grad_norm": 1.453125,
      "learning_rate": 9.83823429748053e-05,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2233563706278801,
      "num_tokens": 180541032.0,
      "step": 87000
    },
    {
      "epoch": 7.8918722786647315,
      "eval_entropy": 5.414384675781112,
      "eval_loss": 4.912048816680908,
      "eval_mean_token_accuracy": 0.2380072741992603,
      "eval_num_tokens": 180541032.0,
      "eval_runtime": 20.6525,
      "eval_samples_per_second": 1712.143,
      "eval_steps_per_second": 214.018,
      "step": 87000
    },
    {
      "entropy": 5.751869106292725,
      "epoch": 7.892325834542816,
      "grad_norm": 1.5078125,
      "learning_rate": 9.836229879815316e-05,
      "loss": 4.9084,
      "mean_token_accuracy": 0.22305852323770523,
      "num_tokens": 180551496.0,
      "step": 87005
    },
    {
      "entropy": 5.663478088378906,
      "epoch": 7.8927793904209,
      "grad_norm": 1.5390625,
      "learning_rate": 9.83422582744137e-05,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2395100474357605,
      "num_tokens": 180561995.0,
      "step": 87010
    },
    {
      "entropy": 5.818313121795654,
      "epoch": 7.893232946298984,
      "grad_norm": 1.53125,
      "learning_rate": 9.83222214040013e-05,
      "loss": 4.9401,
      "mean_token_accuracy": 0.22241385728120805,
      "num_tokens": 180572853.0,
      "step": 87015
    },
    {
      "entropy": 5.697114753723144,
      "epoch": 7.893686502177069,
      "grad_norm": 1.5234375,
      "learning_rate": 9.830218818733045e-05,
      "loss": 4.8032,
      "mean_token_accuracy": 0.23386497050523758,
      "num_tokens": 180584043.0,
      "step": 87020
    },
    {
      "entropy": 5.682143449783325,
      "epoch": 7.894140058055152,
      "grad_norm": 1.484375,
      "learning_rate": 9.828215862481538e-05,
      "loss": 4.7993,
      "mean_token_accuracy": 0.2377658009529114,
      "num_tokens": 180594641.0,
      "step": 87025
    },
    {
      "entropy": 5.68681583404541,
      "epoch": 7.894593613933236,
      "grad_norm": 1.4765625,
      "learning_rate": 9.826213271687028e-05,
      "loss": 4.7849,
      "mean_token_accuracy": 0.24550124108791352,
      "num_tokens": 180604734.0,
      "step": 87030
    },
    {
      "entropy": 5.743889045715332,
      "epoch": 7.8950471698113205,
      "grad_norm": 1.515625,
      "learning_rate": 9.824211046390933e-05,
      "loss": 4.8701,
      "mean_token_accuracy": 0.23066899478435515,
      "num_tokens": 180614229.0,
      "step": 87035
    },
    {
      "entropy": 5.65841965675354,
      "epoch": 7.895500725689405,
      "grad_norm": 1.515625,
      "learning_rate": 9.822209186634658e-05,
      "loss": 4.7585,
      "mean_token_accuracy": 0.24012659341096879,
      "num_tokens": 180624789.0,
      "step": 87040
    },
    {
      "entropy": 5.572323989868164,
      "epoch": 7.895954281567489,
      "grad_norm": 1.546875,
      "learning_rate": 9.820207692459601e-05,
      "loss": 4.6426,
      "mean_token_accuracy": 0.26074077785015104,
      "num_tokens": 180634801.0,
      "step": 87045
    },
    {
      "entropy": 5.694100618362427,
      "epoch": 7.896407837445573,
      "grad_norm": 1.375,
      "learning_rate": 9.818206563907148e-05,
      "loss": 4.819,
      "mean_token_accuracy": 0.24052241295576096,
      "num_tokens": 180646258.0,
      "step": 87050
    },
    {
      "entropy": 5.702416563034058,
      "epoch": 7.8968613933236576,
      "grad_norm": 1.5,
      "learning_rate": 9.816205801018696e-05,
      "loss": 4.7998,
      "mean_token_accuracy": 0.23119252324104309,
      "num_tokens": 180656518.0,
      "step": 87055
    },
    {
      "entropy": 5.669213771820068,
      "epoch": 7.897314949201742,
      "grad_norm": 1.3984375,
      "learning_rate": 9.814205403835613e-05,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2360750377178192,
      "num_tokens": 180666350.0,
      "step": 87060
    },
    {
      "entropy": 5.67708649635315,
      "epoch": 7.897768505079826,
      "grad_norm": 1.640625,
      "learning_rate": 9.812205372399268e-05,
      "loss": 4.7491,
      "mean_token_accuracy": 0.24224251508712769,
      "num_tokens": 180676265.0,
      "step": 87065
    },
    {
      "entropy": 5.684135866165161,
      "epoch": 7.89822206095791,
      "grad_norm": 1.5078125,
      "learning_rate": 9.810205706751027e-05,
      "loss": 4.7611,
      "mean_token_accuracy": 0.23781727850437165,
      "num_tokens": 180687717.0,
      "step": 87070
    },
    {
      "entropy": 5.726035642623901,
      "epoch": 7.898675616835995,
      "grad_norm": 1.609375,
      "learning_rate": 9.808206406932241e-05,
      "loss": 4.7477,
      "mean_token_accuracy": 0.24493670016527175,
      "num_tokens": 180697868.0,
      "step": 87075
    },
    {
      "entropy": 5.646252775192261,
      "epoch": 7.899129172714078,
      "grad_norm": 1.46875,
      "learning_rate": 9.806207472984252e-05,
      "loss": 4.7664,
      "mean_token_accuracy": 0.23740525543689728,
      "num_tokens": 180707912.0,
      "step": 87080
    },
    {
      "entropy": 5.628182315826416,
      "epoch": 7.899582728592162,
      "grad_norm": 1.3203125,
      "learning_rate": 9.804208904948394e-05,
      "loss": 4.7133,
      "mean_token_accuracy": 0.23370259553194045,
      "num_tokens": 180719246.0,
      "step": 87085
    },
    {
      "entropy": 5.714932537078857,
      "epoch": 7.9000362844702465,
      "grad_norm": 1.5078125,
      "learning_rate": 9.802210702866012e-05,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22626692801713943,
      "num_tokens": 180730350.0,
      "step": 87090
    },
    {
      "entropy": 5.6506232738494875,
      "epoch": 7.900489840348331,
      "grad_norm": 1.5625,
      "learning_rate": 9.800212866778426e-05,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23061112612485885,
      "num_tokens": 180739951.0,
      "step": 87095
    },
    {
      "entropy": 5.686268377304077,
      "epoch": 7.900943396226415,
      "grad_norm": 1.4765625,
      "learning_rate": 9.798215396726945e-05,
      "loss": 4.8475,
      "mean_token_accuracy": 0.22821076065301896,
      "num_tokens": 180749734.0,
      "step": 87100
    },
    {
      "entropy": 5.749236345291138,
      "epoch": 7.901396952104499,
      "grad_norm": 1.4609375,
      "learning_rate": 9.796218292752882e-05,
      "loss": 4.8582,
      "mean_token_accuracy": 0.231203992664814,
      "num_tokens": 180760174.0,
      "step": 87105
    },
    {
      "entropy": 5.71600022315979,
      "epoch": 7.901850507982584,
      "grad_norm": 1.53125,
      "learning_rate": 9.794221554897535e-05,
      "loss": 4.8442,
      "mean_token_accuracy": 0.23199087232351304,
      "num_tokens": 180770735.0,
      "step": 87110
    },
    {
      "entropy": 5.606861543655396,
      "epoch": 7.902304063860668,
      "grad_norm": 1.5078125,
      "learning_rate": 9.7922251832022e-05,
      "loss": 4.6905,
      "mean_token_accuracy": 0.2420267641544342,
      "num_tokens": 180781291.0,
      "step": 87115
    },
    {
      "entropy": 5.637450647354126,
      "epoch": 7.902757619738752,
      "grad_norm": 1.4765625,
      "learning_rate": 9.790229177708151e-05,
      "loss": 4.7412,
      "mean_token_accuracy": 0.24123049080371856,
      "num_tokens": 180791125.0,
      "step": 87120
    },
    {
      "entropy": 5.703934240341186,
      "epoch": 7.9032111756168355,
      "grad_norm": 1.484375,
      "learning_rate": 9.788233538456683e-05,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22854083329439162,
      "num_tokens": 180801615.0,
      "step": 87125
    },
    {
      "entropy": 5.770561647415161,
      "epoch": 7.903664731494921,
      "grad_norm": 1.640625,
      "learning_rate": 9.786238265489061e-05,
      "loss": 4.8412,
      "mean_token_accuracy": 0.2368207171559334,
      "num_tokens": 180812063.0,
      "step": 87130
    },
    {
      "entropy": 5.725521898269653,
      "epoch": 7.904118287373004,
      "grad_norm": 1.4609375,
      "learning_rate": 9.784243358846545e-05,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22950080186128616,
      "num_tokens": 180822329.0,
      "step": 87135
    },
    {
      "entropy": 5.7157995223999025,
      "epoch": 7.904571843251088,
      "grad_norm": 1.546875,
      "learning_rate": 9.782248818570391e-05,
      "loss": 4.838,
      "mean_token_accuracy": 0.23158010840415955,
      "num_tokens": 180832218.0,
      "step": 87140
    },
    {
      "entropy": 5.625599908828735,
      "epoch": 7.905025399129173,
      "grad_norm": 1.3515625,
      "learning_rate": 9.780254644701844e-05,
      "loss": 4.7821,
      "mean_token_accuracy": 0.2388540655374527,
      "num_tokens": 180842595.0,
      "step": 87145
    },
    {
      "entropy": 5.7169928550720215,
      "epoch": 7.905478955007257,
      "grad_norm": 1.59375,
      "learning_rate": 9.77826083728215e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2426455944776535,
      "num_tokens": 180851958.0,
      "step": 87150
    },
    {
      "entropy": 5.7193359375,
      "epoch": 7.905932510885341,
      "grad_norm": 1.5,
      "learning_rate": 9.776267396352537e-05,
      "loss": 4.8446,
      "mean_token_accuracy": 0.23052442520856858,
      "num_tokens": 180863116.0,
      "step": 87155
    },
    {
      "entropy": 5.729310083389282,
      "epoch": 7.906386066763425,
      "grad_norm": 1.640625,
      "learning_rate": 9.77427432195423e-05,
      "loss": 4.8242,
      "mean_token_accuracy": 0.23231100738048555,
      "num_tokens": 180872618.0,
      "step": 87160
    },
    {
      "entropy": 5.703055334091187,
      "epoch": 7.90683962264151,
      "grad_norm": 1.6953125,
      "learning_rate": 9.772281614128447e-05,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23450016230344772,
      "num_tokens": 180882647.0,
      "step": 87165
    },
    {
      "entropy": 5.790377473831176,
      "epoch": 7.907293178519594,
      "grad_norm": 1.5,
      "learning_rate": 9.770289272916398e-05,
      "loss": 4.9325,
      "mean_token_accuracy": 0.22049207240343094,
      "num_tokens": 180893850.0,
      "step": 87170
    },
    {
      "entropy": 5.725876808166504,
      "epoch": 7.907746734397678,
      "grad_norm": 1.5078125,
      "learning_rate": 9.768297298359286e-05,
      "loss": 4.8153,
      "mean_token_accuracy": 0.23767123222351075,
      "num_tokens": 180903721.0,
      "step": 87175
    },
    {
      "entropy": 5.732882976531982,
      "epoch": 7.9082002902757615,
      "grad_norm": 1.5,
      "learning_rate": 9.766305690498295e-05,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22995405197143554,
      "num_tokens": 180913973.0,
      "step": 87180
    },
    {
      "entropy": 5.667986822128296,
      "epoch": 7.908653846153846,
      "grad_norm": 1.578125,
      "learning_rate": 9.764314449374631e-05,
      "loss": 4.7529,
      "mean_token_accuracy": 0.24150758981704712,
      "num_tokens": 180923597.0,
      "step": 87185
    },
    {
      "entropy": 5.738833999633789,
      "epoch": 7.90910740203193,
      "grad_norm": 1.515625,
      "learning_rate": 9.762323575029464e-05,
      "loss": 4.8184,
      "mean_token_accuracy": 0.234209805727005,
      "num_tokens": 180934007.0,
      "step": 87190
    },
    {
      "entropy": 5.655106115341186,
      "epoch": 7.909560957910014,
      "grad_norm": 1.6015625,
      "learning_rate": 9.760333067503964e-05,
      "loss": 4.7429,
      "mean_token_accuracy": 0.24436591416597367,
      "num_tokens": 180943953.0,
      "step": 87195
    },
    {
      "entropy": 5.785509538650513,
      "epoch": 7.910014513788099,
      "grad_norm": 1.5546875,
      "learning_rate": 9.758342926839296e-05,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2290182113647461,
      "num_tokens": 180954397.0,
      "step": 87200
    },
    {
      "entropy": 5.711161708831787,
      "epoch": 7.910468069666183,
      "grad_norm": 1.5546875,
      "learning_rate": 9.756353153076618e-05,
      "loss": 4.782,
      "mean_token_accuracy": 0.24123273193836212,
      "num_tokens": 180964815.0,
      "step": 87205
    },
    {
      "entropy": 5.713397359848022,
      "epoch": 7.910921625544267,
      "grad_norm": 1.4296875,
      "learning_rate": 9.754363746257078e-05,
      "loss": 4.8446,
      "mean_token_accuracy": 0.23239015638828278,
      "num_tokens": 180975600.0,
      "step": 87210
    },
    {
      "entropy": 5.725415802001953,
      "epoch": 7.911375181422351,
      "grad_norm": 1.5703125,
      "learning_rate": 9.752374706421813e-05,
      "loss": 4.8686,
      "mean_token_accuracy": 0.23408686369657516,
      "num_tokens": 180985837.0,
      "step": 87215
    },
    {
      "entropy": 5.65044059753418,
      "epoch": 7.911828737300436,
      "grad_norm": 1.5546875,
      "learning_rate": 9.750386033611969e-05,
      "loss": 4.7773,
      "mean_token_accuracy": 0.23999930769205094,
      "num_tokens": 180995332.0,
      "step": 87220
    },
    {
      "entropy": 5.714193487167359,
      "epoch": 7.91228229317852,
      "grad_norm": 1.578125,
      "learning_rate": 9.748397727868663e-05,
      "loss": 4.7443,
      "mean_token_accuracy": 0.2500935852527618,
      "num_tokens": 181005727.0,
      "step": 87225
    },
    {
      "entropy": 5.737776517868042,
      "epoch": 7.912735849056604,
      "grad_norm": 1.453125,
      "learning_rate": 9.746409789233015e-05,
      "loss": 4.7951,
      "mean_token_accuracy": 0.23980753123760223,
      "num_tokens": 181017353.0,
      "step": 87230
    },
    {
      "entropy": 5.720060920715332,
      "epoch": 7.913189404934688,
      "grad_norm": 1.578125,
      "learning_rate": 9.744422217746138e-05,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23447499871253968,
      "num_tokens": 181026624.0,
      "step": 87235
    },
    {
      "entropy": 5.688614988327027,
      "epoch": 7.913642960812772,
      "grad_norm": 1.546875,
      "learning_rate": 9.742435013449131e-05,
      "loss": 4.8132,
      "mean_token_accuracy": 0.23375412523746492,
      "num_tokens": 181036813.0,
      "step": 87240
    },
    {
      "entropy": 5.6600343704223635,
      "epoch": 7.914096516690856,
      "grad_norm": 1.5859375,
      "learning_rate": 9.74044817638309e-05,
      "loss": 4.7925,
      "mean_token_accuracy": 0.23693117499351501,
      "num_tokens": 181047082.0,
      "step": 87245
    },
    {
      "entropy": 5.731364393234253,
      "epoch": 7.91455007256894,
      "grad_norm": 1.46875,
      "learning_rate": 9.738461706589113e-05,
      "loss": 4.8474,
      "mean_token_accuracy": 0.23199271261692048,
      "num_tokens": 181056719.0,
      "step": 87250
    },
    {
      "entropy": 5.751109409332275,
      "epoch": 7.915003628447025,
      "grad_norm": 1.578125,
      "learning_rate": 9.736475604108274e-05,
      "loss": 4.914,
      "mean_token_accuracy": 0.2253995880484581,
      "num_tokens": 181067898.0,
      "step": 87255
    },
    {
      "entropy": 5.7245019435882565,
      "epoch": 7.915457184325109,
      "grad_norm": 1.5234375,
      "learning_rate": 9.734489868981646e-05,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22667045146226883,
      "num_tokens": 181077506.0,
      "step": 87260
    },
    {
      "entropy": 5.650959444046021,
      "epoch": 7.915910740203193,
      "grad_norm": 1.4921875,
      "learning_rate": 9.732504501250294e-05,
      "loss": 4.7689,
      "mean_token_accuracy": 0.2402026668190956,
      "num_tokens": 181086490.0,
      "step": 87265
    },
    {
      "entropy": 5.765282011032104,
      "epoch": 7.9163642960812775,
      "grad_norm": 1.609375,
      "learning_rate": 9.730519500955275e-05,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22766055315732955,
      "num_tokens": 181096065.0,
      "step": 87270
    },
    {
      "entropy": 5.731726741790771,
      "epoch": 7.916817851959362,
      "grad_norm": 1.515625,
      "learning_rate": 9.728534868137646e-05,
      "loss": 4.8816,
      "mean_token_accuracy": 0.22509064078330993,
      "num_tokens": 181107284.0,
      "step": 87275
    },
    {
      "entropy": 5.7694385051727295,
      "epoch": 7.917271407837445,
      "grad_norm": 1.5546875,
      "learning_rate": 9.726550602838434e-05,
      "loss": 4.8284,
      "mean_token_accuracy": 0.2372206747531891,
      "num_tokens": 181117905.0,
      "step": 87280
    },
    {
      "entropy": 5.632565832138061,
      "epoch": 7.91772496371553,
      "grad_norm": 1.4296875,
      "learning_rate": 9.724566705098692e-05,
      "loss": 4.7196,
      "mean_token_accuracy": 0.23734648078680037,
      "num_tokens": 181128471.0,
      "step": 87285
    },
    {
      "entropy": 5.7216408252716064,
      "epoch": 7.918178519593614,
      "grad_norm": 1.578125,
      "learning_rate": 9.72258317495944e-05,
      "loss": 4.8398,
      "mean_token_accuracy": 0.230770742893219,
      "num_tokens": 181138221.0,
      "step": 87290
    },
    {
      "entropy": 5.745098543167114,
      "epoch": 7.918632075471698,
      "grad_norm": 1.4140625,
      "learning_rate": 9.720600012461697e-05,
      "loss": 4.8675,
      "mean_token_accuracy": 0.23454882353544235,
      "num_tokens": 181150744.0,
      "step": 87295
    },
    {
      "entropy": 5.715253686904907,
      "epoch": 7.919085631349782,
      "grad_norm": 1.515625,
      "learning_rate": 9.718617217646479e-05,
      "loss": 4.8158,
      "mean_token_accuracy": 0.23500705063343047,
      "num_tokens": 181160642.0,
      "step": 87300
    },
    {
      "entropy": 5.6652259349823,
      "epoch": 7.919539187227866,
      "grad_norm": 1.5546875,
      "learning_rate": 9.716634790554787e-05,
      "loss": 4.7765,
      "mean_token_accuracy": 0.24087301641702652,
      "num_tokens": 181170166.0,
      "step": 87305
    },
    {
      "entropy": 5.66773042678833,
      "epoch": 7.919992743105951,
      "grad_norm": 1.5390625,
      "learning_rate": 9.714652731227622e-05,
      "loss": 4.7767,
      "mean_token_accuracy": 0.2385898068547249,
      "num_tokens": 181180410.0,
      "step": 87310
    },
    {
      "entropy": 5.654897546768188,
      "epoch": 7.920446298984035,
      "grad_norm": 1.640625,
      "learning_rate": 9.712671039705961e-05,
      "loss": 4.7718,
      "mean_token_accuracy": 0.23883384317159653,
      "num_tokens": 181190507.0,
      "step": 87315
    },
    {
      "entropy": 5.675217533111573,
      "epoch": 7.920899854862119,
      "grad_norm": 1.6015625,
      "learning_rate": 9.710689716030801e-05,
      "loss": 4.7884,
      "mean_token_accuracy": 0.23187389075756074,
      "num_tokens": 181200973.0,
      "step": 87320
    },
    {
      "entropy": 5.698868083953857,
      "epoch": 7.9213534107402035,
      "grad_norm": 1.578125,
      "learning_rate": 9.708708760243114e-05,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2533221676945686,
      "num_tokens": 181210246.0,
      "step": 87325
    },
    {
      "entropy": 5.6866161823272705,
      "epoch": 7.921806966618288,
      "grad_norm": 1.453125,
      "learning_rate": 9.706728172383863e-05,
      "loss": 4.7664,
      "mean_token_accuracy": 0.23617787212133406,
      "num_tokens": 181219545.0,
      "step": 87330
    },
    {
      "entropy": 5.706865501403809,
      "epoch": 7.922260522496371,
      "grad_norm": 1.4609375,
      "learning_rate": 9.704747952494007e-05,
      "loss": 4.8357,
      "mean_token_accuracy": 0.24003857523202896,
      "num_tokens": 181229420.0,
      "step": 87335
    },
    {
      "entropy": 5.757613611221314,
      "epoch": 7.922714078374455,
      "grad_norm": 1.5234375,
      "learning_rate": 9.702768100614498e-05,
      "loss": 4.886,
      "mean_token_accuracy": 0.23249874860048295,
      "num_tokens": 181239603.0,
      "step": 87340
    },
    {
      "entropy": 5.715777206420898,
      "epoch": 7.92316763425254,
      "grad_norm": 1.359375,
      "learning_rate": 9.700788616786279e-05,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23607319295406343,
      "num_tokens": 181249445.0,
      "step": 87345
    },
    {
      "entropy": 5.751061582565308,
      "epoch": 7.923621190130624,
      "grad_norm": 1.5859375,
      "learning_rate": 9.698809501050282e-05,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22486638724803926,
      "num_tokens": 181259906.0,
      "step": 87350
    },
    {
      "entropy": 5.7338831424713135,
      "epoch": 7.924074746008708,
      "grad_norm": 1.578125,
      "learning_rate": 9.696830753447449e-05,
      "loss": 4.8775,
      "mean_token_accuracy": 0.23410455584526063,
      "num_tokens": 181269931.0,
      "step": 87355
    },
    {
      "entropy": 5.704131364822388,
      "epoch": 7.9245283018867925,
      "grad_norm": 1.390625,
      "learning_rate": 9.694852374018689e-05,
      "loss": 4.8054,
      "mean_token_accuracy": 0.22826474159955978,
      "num_tokens": 181279794.0,
      "step": 87360
    },
    {
      "entropy": 5.760335397720337,
      "epoch": 7.924981857764877,
      "grad_norm": 1.3125,
      "learning_rate": 9.692874362804922e-05,
      "loss": 4.9236,
      "mean_token_accuracy": 0.22448900640010833,
      "num_tokens": 181290297.0,
      "step": 87365
    },
    {
      "entropy": 5.723027181625366,
      "epoch": 7.925435413642961,
      "grad_norm": 1.640625,
      "learning_rate": 9.69089671984705e-05,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23789407610893248,
      "num_tokens": 181299994.0,
      "step": 87370
    },
    {
      "entropy": 5.690699911117553,
      "epoch": 7.925888969521045,
      "grad_norm": 1.46875,
      "learning_rate": 9.688919445185972e-05,
      "loss": 4.7992,
      "mean_token_accuracy": 0.23567242920398712,
      "num_tokens": 181310958.0,
      "step": 87375
    },
    {
      "entropy": 5.694498777389526,
      "epoch": 7.9263425253991295,
      "grad_norm": 1.546875,
      "learning_rate": 9.686942538862581e-05,
      "loss": 4.7889,
      "mean_token_accuracy": 0.2332756921648979,
      "num_tokens": 181321459.0,
      "step": 87380
    },
    {
      "entropy": 5.750507831573486,
      "epoch": 7.926796081277214,
      "grad_norm": 1.671875,
      "learning_rate": 9.684966000917747e-05,
      "loss": 4.8338,
      "mean_token_accuracy": 0.2380887359380722,
      "num_tokens": 181331169.0,
      "step": 87385
    },
    {
      "entropy": 5.606078195571899,
      "epoch": 7.927249637155297,
      "grad_norm": 1.5,
      "learning_rate": 9.682989831392366e-05,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2436349257826805,
      "num_tokens": 181341338.0,
      "step": 87390
    },
    {
      "entropy": 5.692495727539063,
      "epoch": 7.9277031930333814,
      "grad_norm": 1.40625,
      "learning_rate": 9.681014030327294e-05,
      "loss": 4.7821,
      "mean_token_accuracy": 0.240826877951622,
      "num_tokens": 181351578.0,
      "step": 87395
    },
    {
      "entropy": 5.676623249053955,
      "epoch": 7.928156748911466,
      "grad_norm": 1.515625,
      "learning_rate": 9.679038597763391e-05,
      "loss": 4.785,
      "mean_token_accuracy": 0.2373472347855568,
      "num_tokens": 181361795.0,
      "step": 87400
    },
    {
      "entropy": 5.653452968597412,
      "epoch": 7.92861030478955,
      "grad_norm": 1.71875,
      "learning_rate": 9.677063533741512e-05,
      "loss": 4.7325,
      "mean_token_accuracy": 0.24488386809825896,
      "num_tokens": 181371843.0,
      "step": 87405
    },
    {
      "entropy": 5.69228892326355,
      "epoch": 7.929063860667634,
      "grad_norm": 1.578125,
      "learning_rate": 9.675088838302504e-05,
      "loss": 4.7743,
      "mean_token_accuracy": 0.2379530280828476,
      "num_tokens": 181381578.0,
      "step": 87410
    },
    {
      "entropy": 5.743988800048828,
      "epoch": 7.9295174165457185,
      "grad_norm": 1.6015625,
      "learning_rate": 9.673114511487199e-05,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2308070883154869,
      "num_tokens": 181391736.0,
      "step": 87415
    },
    {
      "entropy": 5.707587003707886,
      "epoch": 7.929970972423803,
      "grad_norm": 1.53125,
      "learning_rate": 9.671140553336423e-05,
      "loss": 4.7507,
      "mean_token_accuracy": 0.2446032077074051,
      "num_tokens": 181401250.0,
      "step": 87420
    },
    {
      "entropy": 5.6486891746521,
      "epoch": 7.930424528301887,
      "grad_norm": 1.5234375,
      "learning_rate": 9.669166963891008e-05,
      "loss": 4.7671,
      "mean_token_accuracy": 0.2360657200217247,
      "num_tokens": 181411662.0,
      "step": 87425
    },
    {
      "entropy": 5.667290925979614,
      "epoch": 7.930878084179971,
      "grad_norm": 1.5234375,
      "learning_rate": 9.667193743191766e-05,
      "loss": 4.7112,
      "mean_token_accuracy": 0.24959697127342223,
      "num_tokens": 181422470.0,
      "step": 87430
    },
    {
      "entropy": 5.6852611064910885,
      "epoch": 7.931331640058055,
      "grad_norm": 1.65625,
      "learning_rate": 9.665220891279503e-05,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23091568797826767,
      "num_tokens": 181433153.0,
      "step": 87435
    },
    {
      "entropy": 5.640421152114868,
      "epoch": 7.931785195936139,
      "grad_norm": 1.5,
      "learning_rate": 9.663248408195012e-05,
      "loss": 4.693,
      "mean_token_accuracy": 0.24480426758527757,
      "num_tokens": 181442199.0,
      "step": 87440
    },
    {
      "entropy": 5.722222805023193,
      "epoch": 7.932238751814223,
      "grad_norm": 1.4375,
      "learning_rate": 9.661276293979094e-05,
      "loss": 4.8826,
      "mean_token_accuracy": 0.23163807690143584,
      "num_tokens": 181453092.0,
      "step": 87445
    },
    {
      "entropy": 5.678737545013428,
      "epoch": 7.9326923076923075,
      "grad_norm": 1.609375,
      "learning_rate": 9.65930454867252e-05,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2393839478492737,
      "num_tokens": 181462584.0,
      "step": 87450
    },
    {
      "entropy": 5.700380563735962,
      "epoch": 7.933145863570392,
      "grad_norm": 1.5390625,
      "learning_rate": 9.657333172316079e-05,
      "loss": 4.8213,
      "mean_token_accuracy": 0.22974174171686174,
      "num_tokens": 181472618.0,
      "step": 87455
    },
    {
      "entropy": 5.719121408462525,
      "epoch": 7.933599419448476,
      "grad_norm": 1.390625,
      "learning_rate": 9.655362164950537e-05,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22541722655296326,
      "num_tokens": 181484130.0,
      "step": 87460
    },
    {
      "entropy": 5.700156545639038,
      "epoch": 7.93405297532656,
      "grad_norm": 1.6953125,
      "learning_rate": 9.653391526616653e-05,
      "loss": 4.7959,
      "mean_token_accuracy": 0.22927391678094863,
      "num_tokens": 181495055.0,
      "step": 87465
    },
    {
      "entropy": 5.7421070575714115,
      "epoch": 7.9345065312046446,
      "grad_norm": 1.4140625,
      "learning_rate": 9.651421257355178e-05,
      "loss": 4.9005,
      "mean_token_accuracy": 0.22608863562345505,
      "num_tokens": 181506948.0,
      "step": 87470
    },
    {
      "entropy": 5.689656162261963,
      "epoch": 7.934960087082729,
      "grad_norm": 1.4453125,
      "learning_rate": 9.649451357206862e-05,
      "loss": 4.7664,
      "mean_token_accuracy": 0.24113565534353257,
      "num_tokens": 181518014.0,
      "step": 87475
    },
    {
      "entropy": 5.687792158126831,
      "epoch": 7.935413642960813,
      "grad_norm": 1.40625,
      "learning_rate": 9.64748182621244e-05,
      "loss": 4.7816,
      "mean_token_accuracy": 0.23996087908744812,
      "num_tokens": 181529387.0,
      "step": 87480
    },
    {
      "entropy": 5.670304250717163,
      "epoch": 7.935867198838897,
      "grad_norm": 1.4453125,
      "learning_rate": 9.645512664412636e-05,
      "loss": 4.769,
      "mean_token_accuracy": 0.23983875513076783,
      "num_tokens": 181539230.0,
      "step": 87485
    },
    {
      "entropy": 5.6974681377410885,
      "epoch": 7.936320754716981,
      "grad_norm": 1.5078125,
      "learning_rate": 9.643543871848187e-05,
      "loss": 4.87,
      "mean_token_accuracy": 0.23629872649908065,
      "num_tokens": 181550399.0,
      "step": 87490
    },
    {
      "entropy": 5.61424765586853,
      "epoch": 7.936774310595065,
      "grad_norm": 1.4765625,
      "learning_rate": 9.641575448559803e-05,
      "loss": 4.7718,
      "mean_token_accuracy": 0.24039523899555207,
      "num_tokens": 181561363.0,
      "step": 87495
    },
    {
      "entropy": 5.70606746673584,
      "epoch": 7.937227866473149,
      "grad_norm": 1.5,
      "learning_rate": 9.639607394588188e-05,
      "loss": 4.7816,
      "mean_token_accuracy": 0.23310442715883256,
      "num_tokens": 181572615.0,
      "step": 87500
    },
    {
      "entropy": 5.739350032806397,
      "epoch": 7.9376814223512335,
      "grad_norm": 1.640625,
      "learning_rate": 9.637639709974042e-05,
      "loss": 4.8312,
      "mean_token_accuracy": 0.2374472662806511,
      "num_tokens": 181582030.0,
      "step": 87505
    },
    {
      "entropy": 5.646817779541015,
      "epoch": 7.938134978229318,
      "grad_norm": 1.46875,
      "learning_rate": 9.63567239475806e-05,
      "loss": 4.7256,
      "mean_token_accuracy": 0.24085261970758437,
      "num_tokens": 181591689.0,
      "step": 87510
    },
    {
      "entropy": 5.673170709609986,
      "epoch": 7.938588534107402,
      "grad_norm": 1.4609375,
      "learning_rate": 9.633705448980923e-05,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2334000825881958,
      "num_tokens": 181602512.0,
      "step": 87515
    },
    {
      "entropy": 5.694848680496216,
      "epoch": 7.939042089985486,
      "grad_norm": 1.4140625,
      "learning_rate": 9.631738872683308e-05,
      "loss": 4.794,
      "mean_token_accuracy": 0.23949141353368758,
      "num_tokens": 181612421.0,
      "step": 87520
    },
    {
      "entropy": 5.693871688842774,
      "epoch": 7.939495645863571,
      "grad_norm": 1.6015625,
      "learning_rate": 9.62977266590589e-05,
      "loss": 4.8582,
      "mean_token_accuracy": 0.23418801575899123,
      "num_tokens": 181622343.0,
      "step": 87525
    },
    {
      "entropy": 5.752273607254028,
      "epoch": 7.939949201741655,
      "grad_norm": 1.4453125,
      "learning_rate": 9.627806828689326e-05,
      "loss": 4.8364,
      "mean_token_accuracy": 0.23503201603889465,
      "num_tokens": 181633035.0,
      "step": 87530
    },
    {
      "entropy": 5.645566797256469,
      "epoch": 7.940402757619739,
      "grad_norm": 1.421875,
      "learning_rate": 9.625841361074275e-05,
      "loss": 4.7074,
      "mean_token_accuracy": 0.24408502280712127,
      "num_tokens": 181643992.0,
      "step": 87535
    },
    {
      "entropy": 5.694719028472901,
      "epoch": 7.940856313497823,
      "grad_norm": 1.4375,
      "learning_rate": 9.623876263101378e-05,
      "loss": 4.783,
      "mean_token_accuracy": 0.24111449718475342,
      "num_tokens": 181655907.0,
      "step": 87540
    },
    {
      "entropy": 5.76724419593811,
      "epoch": 7.941309869375907,
      "grad_norm": 1.53125,
      "learning_rate": 9.621911534811276e-05,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2299605816602707,
      "num_tokens": 181666849.0,
      "step": 87545
    },
    {
      "entropy": 5.746979713439941,
      "epoch": 7.941763425253991,
      "grad_norm": 1.5234375,
      "learning_rate": 9.619947176244597e-05,
      "loss": 4.9229,
      "mean_token_accuracy": 0.2332722544670105,
      "num_tokens": 181677098.0,
      "step": 87550
    },
    {
      "entropy": 5.683475017547607,
      "epoch": 7.942216981132075,
      "grad_norm": 1.5078125,
      "learning_rate": 9.61798318744196e-05,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24780346155166627,
      "num_tokens": 181687065.0,
      "step": 87555
    },
    {
      "entropy": 5.651852130889893,
      "epoch": 7.94267053701016,
      "grad_norm": 1.5234375,
      "learning_rate": 9.616019568443995e-05,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2446806475520134,
      "num_tokens": 181698085.0,
      "step": 87560
    },
    {
      "entropy": 5.634235143661499,
      "epoch": 7.943124092888244,
      "grad_norm": 1.515625,
      "learning_rate": 9.614056319291307e-05,
      "loss": 4.7129,
      "mean_token_accuracy": 0.24854480922222139,
      "num_tokens": 181708397.0,
      "step": 87565
    },
    {
      "entropy": 5.745596027374267,
      "epoch": 7.943577648766328,
      "grad_norm": 1.5625,
      "learning_rate": 9.612093440024486e-05,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2342215359210968,
      "num_tokens": 181718984.0,
      "step": 87570
    },
    {
      "entropy": 5.700041389465332,
      "epoch": 7.944031204644412,
      "grad_norm": 1.546875,
      "learning_rate": 9.610130930684142e-05,
      "loss": 4.8462,
      "mean_token_accuracy": 0.22924493998289108,
      "num_tokens": 181729269.0,
      "step": 87575
    },
    {
      "entropy": 5.727577447891235,
      "epoch": 7.944484760522497,
      "grad_norm": 1.375,
      "learning_rate": 9.608168791310832e-05,
      "loss": 4.801,
      "mean_token_accuracy": 0.23631082773208617,
      "num_tokens": 181739755.0,
      "step": 87580
    },
    {
      "entropy": 5.703035306930542,
      "epoch": 7.944938316400581,
      "grad_norm": 1.4453125,
      "learning_rate": 9.60620702194516e-05,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23230782449245452,
      "num_tokens": 181749357.0,
      "step": 87585
    },
    {
      "entropy": 5.69704532623291,
      "epoch": 7.945391872278664,
      "grad_norm": 1.5625,
      "learning_rate": 9.604245622627686e-05,
      "loss": 4.8036,
      "mean_token_accuracy": 0.24238087385892867,
      "num_tokens": 181758939.0,
      "step": 87590
    },
    {
      "entropy": 5.758084774017334,
      "epoch": 7.9458454281567485,
      "grad_norm": 1.515625,
      "learning_rate": 9.602284593398972e-05,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2340521454811096,
      "num_tokens": 181768420.0,
      "step": 87595
    },
    {
      "entropy": 5.653451633453369,
      "epoch": 7.946298984034833,
      "grad_norm": 1.4765625,
      "learning_rate": 9.600323934299574e-05,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23075223863124847,
      "num_tokens": 181777705.0,
      "step": 87600
    },
    {
      "entropy": 5.7083135604858395,
      "epoch": 7.946752539912917,
      "grad_norm": 1.6015625,
      "learning_rate": 9.59836364537004e-05,
      "loss": 4.8218,
      "mean_token_accuracy": 0.23609929978847505,
      "num_tokens": 181788366.0,
      "step": 87605
    },
    {
      "entropy": 5.721961593627929,
      "epoch": 7.947206095791001,
      "grad_norm": 1.4296875,
      "learning_rate": 9.596403726650907e-05,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2353658437728882,
      "num_tokens": 181799539.0,
      "step": 87610
    },
    {
      "entropy": 5.733346176147461,
      "epoch": 7.947659651669086,
      "grad_norm": 1.578125,
      "learning_rate": 9.594444178182701e-05,
      "loss": 4.8562,
      "mean_token_accuracy": 0.2341957077383995,
      "num_tokens": 181809457.0,
      "step": 87615
    },
    {
      "entropy": 5.741622638702393,
      "epoch": 7.94811320754717,
      "grad_norm": 1.609375,
      "learning_rate": 9.592485000005957e-05,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2240719184279442,
      "num_tokens": 181819028.0,
      "step": 87620
    },
    {
      "entropy": 5.686750411987305,
      "epoch": 7.948566763425254,
      "grad_norm": 1.5625,
      "learning_rate": 9.590526192161188e-05,
      "loss": 4.7817,
      "mean_token_accuracy": 0.23988739252090455,
      "num_tokens": 181829667.0,
      "step": 87625
    },
    {
      "entropy": 5.733353948593139,
      "epoch": 7.949020319303338,
      "grad_norm": 1.515625,
      "learning_rate": 9.588567754688902e-05,
      "loss": 4.792,
      "mean_token_accuracy": 0.2434057906270027,
      "num_tokens": 181839850.0,
      "step": 87630
    },
    {
      "entropy": 5.696505212783814,
      "epoch": 7.949473875181423,
      "grad_norm": 1.4453125,
      "learning_rate": 9.5866096876296e-05,
      "loss": 4.7921,
      "mean_token_accuracy": 0.237067149579525,
      "num_tokens": 181850287.0,
      "step": 87635
    },
    {
      "entropy": 5.729324722290039,
      "epoch": 7.949927431059507,
      "grad_norm": 1.3984375,
      "learning_rate": 9.584651991023771e-05,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2320139765739441,
      "num_tokens": 181861842.0,
      "step": 87640
    },
    {
      "entropy": 5.680114841461181,
      "epoch": 7.95038098693759,
      "grad_norm": 1.4375,
      "learning_rate": 9.582694664911907e-05,
      "loss": 4.7356,
      "mean_token_accuracy": 0.24751234501600267,
      "num_tokens": 181872738.0,
      "step": 87645
    },
    {
      "entropy": 5.63181700706482,
      "epoch": 7.950834542815675,
      "grad_norm": 1.5625,
      "learning_rate": 9.580737709334474e-05,
      "loss": 4.7681,
      "mean_token_accuracy": 0.2349335178732872,
      "num_tokens": 181883535.0,
      "step": 87650
    },
    {
      "entropy": 5.695129013061523,
      "epoch": 7.951288098693759,
      "grad_norm": 1.4453125,
      "learning_rate": 9.57878112433196e-05,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23326242119073867,
      "num_tokens": 181893667.0,
      "step": 87655
    },
    {
      "entropy": 5.643803644180298,
      "epoch": 7.951741654571843,
      "grad_norm": 1.421875,
      "learning_rate": 9.57682490994482e-05,
      "loss": 4.7634,
      "mean_token_accuracy": 0.24369659125804902,
      "num_tokens": 181904974.0,
      "step": 87660
    },
    {
      "entropy": 5.721652936935425,
      "epoch": 7.952195210449927,
      "grad_norm": 1.4765625,
      "learning_rate": 9.574869066213507e-05,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2304323971271515,
      "num_tokens": 181915984.0,
      "step": 87665
    },
    {
      "entropy": 5.709003162384033,
      "epoch": 7.952648766328012,
      "grad_norm": 1.4140625,
      "learning_rate": 9.572913593178467e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.24170975238084794,
      "num_tokens": 181926332.0,
      "step": 87670
    },
    {
      "entropy": 5.715934991836548,
      "epoch": 7.953102322206096,
      "grad_norm": 1.4921875,
      "learning_rate": 9.570958490880146e-05,
      "loss": 4.78,
      "mean_token_accuracy": 0.2349340483546257,
      "num_tokens": 181936844.0,
      "step": 87675
    },
    {
      "entropy": 5.742207145690918,
      "epoch": 7.95355587808418,
      "grad_norm": 1.6328125,
      "learning_rate": 9.569003759358974e-05,
      "loss": 4.861,
      "mean_token_accuracy": 0.2296053186058998,
      "num_tokens": 181946626.0,
      "step": 87680
    },
    {
      "entropy": 5.674759244918823,
      "epoch": 7.9540094339622645,
      "grad_norm": 1.4609375,
      "learning_rate": 9.567049398655363e-05,
      "loss": 4.7628,
      "mean_token_accuracy": 0.24016064703464507,
      "num_tokens": 181956728.0,
      "step": 87685
    },
    {
      "entropy": 5.677335977554321,
      "epoch": 7.954462989840349,
      "grad_norm": 1.53125,
      "learning_rate": 9.565095408809749e-05,
      "loss": 4.746,
      "mean_token_accuracy": 0.2369026318192482,
      "num_tokens": 181967147.0,
      "step": 87690
    },
    {
      "entropy": 5.657222318649292,
      "epoch": 7.954916545718433,
      "grad_norm": 1.46875,
      "learning_rate": 9.563141789862531e-05,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2434321641921997,
      "num_tokens": 181977217.0,
      "step": 87695
    },
    {
      "entropy": 5.707433843612671,
      "epoch": 7.955370101596516,
      "grad_norm": 1.578125,
      "learning_rate": 9.561188541854111e-05,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2305363282561302,
      "num_tokens": 181986863.0,
      "step": 87700
    },
    {
      "entropy": 5.67892050743103,
      "epoch": 7.955823657474601,
      "grad_norm": 1.4765625,
      "learning_rate": 9.559235664824886e-05,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23232758045196533,
      "num_tokens": 181997447.0,
      "step": 87705
    },
    {
      "entropy": 5.696907472610474,
      "epoch": 7.956277213352685,
      "grad_norm": 1.421875,
      "learning_rate": 9.557283158815236e-05,
      "loss": 4.7475,
      "mean_token_accuracy": 0.24599432051181794,
      "num_tokens": 182007517.0,
      "step": 87710
    },
    {
      "entropy": 5.710035800933838,
      "epoch": 7.956730769230769,
      "grad_norm": 1.4921875,
      "learning_rate": 9.555331023865546e-05,
      "loss": 4.7922,
      "mean_token_accuracy": 0.24006603360176088,
      "num_tokens": 182017050.0,
      "step": 87715
    },
    {
      "entropy": 5.6860771656036375,
      "epoch": 7.957184325108853,
      "grad_norm": 1.6328125,
      "learning_rate": 9.553379260016174e-05,
      "loss": 4.8017,
      "mean_token_accuracy": 0.23284506797790527,
      "num_tokens": 182026546.0,
      "step": 87720
    },
    {
      "entropy": 5.700920057296753,
      "epoch": 7.957637880986938,
      "grad_norm": 1.4140625,
      "learning_rate": 9.551427867307501e-05,
      "loss": 4.8079,
      "mean_token_accuracy": 0.2360449880361557,
      "num_tokens": 182036776.0,
      "step": 87725
    },
    {
      "entropy": 5.594114303588867,
      "epoch": 7.958091436865022,
      "grad_norm": 1.6953125,
      "learning_rate": 9.549476845779873e-05,
      "loss": 4.6503,
      "mean_token_accuracy": 0.24551893919706344,
      "num_tokens": 182046552.0,
      "step": 87730
    },
    {
      "entropy": 5.692222309112549,
      "epoch": 7.958544992743106,
      "grad_norm": 1.65625,
      "learning_rate": 9.547526195473635e-05,
      "loss": 4.7746,
      "mean_token_accuracy": 0.24344618767499923,
      "num_tokens": 182056496.0,
      "step": 87735
    },
    {
      "entropy": 5.706663465499878,
      "epoch": 7.9589985486211905,
      "grad_norm": 1.6484375,
      "learning_rate": 9.545575916429134e-05,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23208450227975846,
      "num_tokens": 182065888.0,
      "step": 87740
    },
    {
      "entropy": 5.721048593521118,
      "epoch": 7.959452104499274,
      "grad_norm": 1.6875,
      "learning_rate": 9.543626008686699e-05,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23718950897455215,
      "num_tokens": 182075945.0,
      "step": 87745
    },
    {
      "entropy": 5.684735727310181,
      "epoch": 7.959905660377358,
      "grad_norm": 1.71875,
      "learning_rate": 9.541676472286645e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.24222783297300338,
      "num_tokens": 182085836.0,
      "step": 87750
    },
    {
      "entropy": 5.693691110610962,
      "epoch": 7.960359216255442,
      "grad_norm": 1.4140625,
      "learning_rate": 9.539727307269305e-05,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23382744789123536,
      "num_tokens": 182096891.0,
      "step": 87755
    },
    {
      "entropy": 5.681887531280518,
      "epoch": 7.960812772133527,
      "grad_norm": 1.4765625,
      "learning_rate": 9.53777851367498e-05,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23892112374305724,
      "num_tokens": 182106825.0,
      "step": 87760
    },
    {
      "entropy": 5.710063123703003,
      "epoch": 7.961266328011611,
      "grad_norm": 1.4765625,
      "learning_rate": 9.535830091543973e-05,
      "loss": 4.7885,
      "mean_token_accuracy": 0.23587094992399216,
      "num_tokens": 182116593.0,
      "step": 87765
    },
    {
      "entropy": 5.7011669158935545,
      "epoch": 7.961719883889695,
      "grad_norm": 1.625,
      "learning_rate": 9.533882040916579e-05,
      "loss": 4.7809,
      "mean_token_accuracy": 0.23716637641191482,
      "num_tokens": 182127521.0,
      "step": 87770
    },
    {
      "entropy": 5.684771156311035,
      "epoch": 7.9621734397677795,
      "grad_norm": 1.4921875,
      "learning_rate": 9.531934361833084e-05,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23633260428905487,
      "num_tokens": 182139146.0,
      "step": 87775
    },
    {
      "entropy": 5.6924357414245605,
      "epoch": 7.962626995645864,
      "grad_norm": 1.5546875,
      "learning_rate": 9.529987054333762e-05,
      "loss": 4.8227,
      "mean_token_accuracy": 0.2320894256234169,
      "num_tokens": 182149551.0,
      "step": 87780
    },
    {
      "entropy": 5.7095372676849365,
      "epoch": 7.963080551523948,
      "grad_norm": 1.421875,
      "learning_rate": 9.52804011845888e-05,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23597839027643203,
      "num_tokens": 182160290.0,
      "step": 87785
    },
    {
      "entropy": 5.662941122055054,
      "epoch": 7.963534107402032,
      "grad_norm": 1.4453125,
      "learning_rate": 9.526093554248718e-05,
      "loss": 4.7621,
      "mean_token_accuracy": 0.24428500533103942,
      "num_tokens": 182170389.0,
      "step": 87790
    },
    {
      "entropy": 5.636639833450317,
      "epoch": 7.9639876632801165,
      "grad_norm": 1.5546875,
      "learning_rate": 9.524147361743519e-05,
      "loss": 4.6597,
      "mean_token_accuracy": 0.25008744448423387,
      "num_tokens": 182180138.0,
      "step": 87795
    },
    {
      "entropy": 5.631690120697021,
      "epoch": 7.9644412191582,
      "grad_norm": 1.53125,
      "learning_rate": 9.522201540983536e-05,
      "loss": 4.6998,
      "mean_token_accuracy": 0.23984672129154205,
      "num_tokens": 182189485.0,
      "step": 87800
    },
    {
      "entropy": 5.568362760543823,
      "epoch": 7.964894775036284,
      "grad_norm": 1.46875,
      "learning_rate": 9.520256092009005e-05,
      "loss": 4.6347,
      "mean_token_accuracy": 0.2515346944332123,
      "num_tokens": 182199999.0,
      "step": 87805
    },
    {
      "entropy": 5.605184936523438,
      "epoch": 7.9653483309143684,
      "grad_norm": 1.5859375,
      "learning_rate": 9.51831101486016e-05,
      "loss": 4.6736,
      "mean_token_accuracy": 0.2523601219058037,
      "num_tokens": 182209952.0,
      "step": 87810
    },
    {
      "entropy": 5.623497867584229,
      "epoch": 7.965801886792453,
      "grad_norm": 1.5,
      "learning_rate": 9.516366309577223e-05,
      "loss": 4.7521,
      "mean_token_accuracy": 0.23946987688541413,
      "num_tokens": 182220680.0,
      "step": 87815
    },
    {
      "entropy": 5.724766254425049,
      "epoch": 7.966255442670537,
      "grad_norm": 1.6640625,
      "learning_rate": 9.514421976200408e-05,
      "loss": 4.8935,
      "mean_token_accuracy": 0.23106556832790376,
      "num_tokens": 182231620.0,
      "step": 87820
    },
    {
      "entropy": 5.746888923645019,
      "epoch": 7.966708998548621,
      "grad_norm": 1.46875,
      "learning_rate": 9.512478014769938e-05,
      "loss": 4.9392,
      "mean_token_accuracy": 0.22652229964733123,
      "num_tokens": 182242548.0,
      "step": 87825
    },
    {
      "entropy": 5.72778730392456,
      "epoch": 7.9671625544267055,
      "grad_norm": 1.515625,
      "learning_rate": 9.510534425326006e-05,
      "loss": 4.8207,
      "mean_token_accuracy": 0.24160379767417908,
      "num_tokens": 182252517.0,
      "step": 87830
    },
    {
      "entropy": 5.676494836807251,
      "epoch": 7.96761611030479,
      "grad_norm": 1.5625,
      "learning_rate": 9.508591207908807e-05,
      "loss": 4.7743,
      "mean_token_accuracy": 0.23492642045021056,
      "num_tokens": 182263258.0,
      "step": 87835
    },
    {
      "entropy": 5.7228250980377195,
      "epoch": 7.968069666182874,
      "grad_norm": 1.4375,
      "learning_rate": 9.506648362558526e-05,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2420066624879837,
      "num_tokens": 182274329.0,
      "step": 87840
    },
    {
      "entropy": 5.69651575088501,
      "epoch": 7.968523222060957,
      "grad_norm": 1.375,
      "learning_rate": 9.50470588931534e-05,
      "loss": 4.7606,
      "mean_token_accuracy": 0.23389952927827834,
      "num_tokens": 182284258.0,
      "step": 87845
    },
    {
      "entropy": 5.6562258243560795,
      "epoch": 7.968976777939043,
      "grad_norm": 1.40625,
      "learning_rate": 9.502763788219423e-05,
      "loss": 4.7919,
      "mean_token_accuracy": 0.23742133975028992,
      "num_tokens": 182295823.0,
      "step": 87850
    },
    {
      "entropy": 5.742100429534912,
      "epoch": 7.969430333817126,
      "grad_norm": 1.6015625,
      "learning_rate": 9.500822059310931e-05,
      "loss": 4.8811,
      "mean_token_accuracy": 0.22914873957633972,
      "num_tokens": 182306034.0,
      "step": 87855
    },
    {
      "entropy": 5.691501474380493,
      "epoch": 7.96988388969521,
      "grad_norm": 1.5703125,
      "learning_rate": 9.498880702630037e-05,
      "loss": 4.7287,
      "mean_token_accuracy": 0.24095435589551925,
      "num_tokens": 182316048.0,
      "step": 87860
    },
    {
      "entropy": 5.7109949588775635,
      "epoch": 7.9703374455732945,
      "grad_norm": 1.4921875,
      "learning_rate": 9.49693971821687e-05,
      "loss": 4.8736,
      "mean_token_accuracy": 0.223187156021595,
      "num_tokens": 182326404.0,
      "step": 87865
    },
    {
      "entropy": 5.715108108520508,
      "epoch": 7.970791001451379,
      "grad_norm": 1.4140625,
      "learning_rate": 9.494999106111583e-05,
      "loss": 4.8499,
      "mean_token_accuracy": 0.23061989545822142,
      "num_tokens": 182336704.0,
      "step": 87870
    },
    {
      "entropy": 5.708465957641602,
      "epoch": 7.971244557329463,
      "grad_norm": 1.5,
      "learning_rate": 9.493058866354299e-05,
      "loss": 4.799,
      "mean_token_accuracy": 0.23742962777614593,
      "num_tokens": 182347216.0,
      "step": 87875
    },
    {
      "entropy": 5.628433465957642,
      "epoch": 7.971698113207547,
      "grad_norm": 1.46875,
      "learning_rate": 9.491118998985143e-05,
      "loss": 4.7156,
      "mean_token_accuracy": 0.24024857729673385,
      "num_tokens": 182357927.0,
      "step": 87880
    },
    {
      "entropy": 5.68743486404419,
      "epoch": 7.9721516690856316,
      "grad_norm": 1.3828125,
      "learning_rate": 9.48917950404424e-05,
      "loss": 4.8302,
      "mean_token_accuracy": 0.2377151593565941,
      "num_tokens": 182367964.0,
      "step": 87885
    },
    {
      "entropy": 5.688198804855347,
      "epoch": 7.972605224963716,
      "grad_norm": 1.5078125,
      "learning_rate": 9.48724038157169e-05,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23709812611341477,
      "num_tokens": 182378665.0,
      "step": 87890
    },
    {
      "entropy": 5.64967565536499,
      "epoch": 7.9730587808418,
      "grad_norm": 1.484375,
      "learning_rate": 9.485301631607601e-05,
      "loss": 4.7839,
      "mean_token_accuracy": 0.23658324629068375,
      "num_tokens": 182388605.0,
      "step": 87895
    },
    {
      "entropy": 5.727549886703491,
      "epoch": 7.9735123367198835,
      "grad_norm": 1.5625,
      "learning_rate": 9.483363254192068e-05,
      "loss": 4.8813,
      "mean_token_accuracy": 0.23269244581460952,
      "num_tokens": 182399266.0,
      "step": 87900
    },
    {
      "entropy": 5.736081838607788,
      "epoch": 7.973965892597968,
      "grad_norm": 1.4453125,
      "learning_rate": 9.481425249365175e-05,
      "loss": 4.8356,
      "mean_token_accuracy": 0.2309538722038269,
      "num_tokens": 182410312.0,
      "step": 87905
    },
    {
      "entropy": 5.709744644165039,
      "epoch": 7.974419448476052,
      "grad_norm": 1.515625,
      "learning_rate": 9.479487617166995e-05,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23470914214849473,
      "num_tokens": 182420220.0,
      "step": 87910
    },
    {
      "entropy": 5.7063148021698,
      "epoch": 7.974873004354136,
      "grad_norm": 1.4140625,
      "learning_rate": 9.477550357637607e-05,
      "loss": 4.8396,
      "mean_token_accuracy": 0.22992817163467408,
      "num_tokens": 182432893.0,
      "step": 87915
    },
    {
      "entropy": 5.766660976409912,
      "epoch": 7.9753265602322205,
      "grad_norm": 1.4375,
      "learning_rate": 9.475613470817068e-05,
      "loss": 4.9833,
      "mean_token_accuracy": 0.23242629170417786,
      "num_tokens": 182444578.0,
      "step": 87920
    },
    {
      "entropy": 5.68111138343811,
      "epoch": 7.975780116110305,
      "grad_norm": 1.4765625,
      "learning_rate": 9.47367695674543e-05,
      "loss": 4.7779,
      "mean_token_accuracy": 0.23525894433259964,
      "num_tokens": 182455752.0,
      "step": 87925
    },
    {
      "entropy": 5.744022464752197,
      "epoch": 7.976233671988389,
      "grad_norm": 1.484375,
      "learning_rate": 9.471740815462754e-05,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2378506377339363,
      "num_tokens": 182465661.0,
      "step": 87930
    },
    {
      "entropy": 5.706497001647949,
      "epoch": 7.976687227866473,
      "grad_norm": 1.484375,
      "learning_rate": 9.469805047009069e-05,
      "loss": 4.8448,
      "mean_token_accuracy": 0.23694469183683395,
      "num_tokens": 182476612.0,
      "step": 87935
    },
    {
      "entropy": 5.676213073730469,
      "epoch": 7.977140783744558,
      "grad_norm": 1.59375,
      "learning_rate": 9.467869651424414e-05,
      "loss": 4.8076,
      "mean_token_accuracy": 0.230433189868927,
      "num_tokens": 182486962.0,
      "step": 87940
    },
    {
      "entropy": 5.696678400039673,
      "epoch": 7.977594339622642,
      "grad_norm": 1.6171875,
      "learning_rate": 9.465934628748807e-05,
      "loss": 4.837,
      "mean_token_accuracy": 0.23751200288534163,
      "num_tokens": 182497516.0,
      "step": 87945
    },
    {
      "entropy": 5.762691307067871,
      "epoch": 7.978047895500726,
      "grad_norm": 1.5859375,
      "learning_rate": 9.46399997902227e-05,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2310766652226448,
      "num_tokens": 182507829.0,
      "step": 87950
    },
    {
      "entropy": 5.699523687362671,
      "epoch": 7.9785014513788095,
      "grad_norm": 1.46875,
      "learning_rate": 9.4620657022848e-05,
      "loss": 4.838,
      "mean_token_accuracy": 0.23526751399040222,
      "num_tokens": 182519181.0,
      "step": 87955
    },
    {
      "entropy": 5.6380115985870365,
      "epoch": 7.978955007256894,
      "grad_norm": 1.46875,
      "learning_rate": 9.460131798576418e-05,
      "loss": 4.7782,
      "mean_token_accuracy": 0.23957246392965317,
      "num_tokens": 182529123.0,
      "step": 87960
    },
    {
      "entropy": 5.669067907333374,
      "epoch": 7.979408563134978,
      "grad_norm": 1.703125,
      "learning_rate": 9.458198267937104e-05,
      "loss": 4.7811,
      "mean_token_accuracy": 0.2415493682026863,
      "num_tokens": 182539179.0,
      "step": 87965
    },
    {
      "entropy": 5.68282036781311,
      "epoch": 7.979862119013062,
      "grad_norm": 1.515625,
      "learning_rate": 9.456265110406851e-05,
      "loss": 4.8471,
      "mean_token_accuracy": 0.23391018211841583,
      "num_tokens": 182550421.0,
      "step": 87970
    },
    {
      "entropy": 5.727406692504883,
      "epoch": 7.980315674891147,
      "grad_norm": 1.3671875,
      "learning_rate": 9.45433232602563e-05,
      "loss": 4.8887,
      "mean_token_accuracy": 0.2292720705270767,
      "num_tokens": 182561294.0,
      "step": 87975
    },
    {
      "entropy": 5.71059889793396,
      "epoch": 7.980769230769231,
      "grad_norm": 1.4140625,
      "learning_rate": 9.452399914833417e-05,
      "loss": 4.8359,
      "mean_token_accuracy": 0.240176160633564,
      "num_tokens": 182572391.0,
      "step": 87980
    },
    {
      "entropy": 5.741012859344482,
      "epoch": 7.981222786647315,
      "grad_norm": 1.53125,
      "learning_rate": 9.450467876870173e-05,
      "loss": 4.8582,
      "mean_token_accuracy": 0.22943118959665298,
      "num_tokens": 182582845.0,
      "step": 87985
    },
    {
      "entropy": 5.617027235031128,
      "epoch": 7.981676342525399,
      "grad_norm": 1.546875,
      "learning_rate": 9.448536212175846e-05,
      "loss": 4.7105,
      "mean_token_accuracy": 0.24101076871156693,
      "num_tokens": 182593328.0,
      "step": 87990
    },
    {
      "entropy": 5.698042345046997,
      "epoch": 7.982129898403484,
      "grad_norm": 1.7421875,
      "learning_rate": 9.446604920790395e-05,
      "loss": 4.7737,
      "mean_token_accuracy": 0.23826091587543488,
      "num_tokens": 182602634.0,
      "step": 87995
    },
    {
      "entropy": 5.6513019561767575,
      "epoch": 7.982583454281567,
      "grad_norm": 1.46875,
      "learning_rate": 9.444674002753756e-05,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23912206143140793,
      "num_tokens": 182612907.0,
      "step": 88000
    },
    {
      "entropy": 5.746549272537232,
      "epoch": 7.983037010159651,
      "grad_norm": 1.5390625,
      "learning_rate": 9.442743458105868e-05,
      "loss": 4.8875,
      "mean_token_accuracy": 0.23034819066524506,
      "num_tokens": 182624252.0,
      "step": 88005
    },
    {
      "entropy": 5.652924346923828,
      "epoch": 7.9834905660377355,
      "grad_norm": 1.4453125,
      "learning_rate": 9.440813286886638e-05,
      "loss": 4.7823,
      "mean_token_accuracy": 0.23674494475126268,
      "num_tokens": 182633378.0,
      "step": 88010
    },
    {
      "entropy": 5.741404914855957,
      "epoch": 7.98394412191582,
      "grad_norm": 1.390625,
      "learning_rate": 9.438883489135991e-05,
      "loss": 4.9272,
      "mean_token_accuracy": 0.23461758047342302,
      "num_tokens": 182644691.0,
      "step": 88015
    },
    {
      "entropy": 5.7141142845153805,
      "epoch": 7.984397677793904,
      "grad_norm": 1.4921875,
      "learning_rate": 9.436954064893833e-05,
      "loss": 4.8341,
      "mean_token_accuracy": 0.2320602849125862,
      "num_tokens": 182655775.0,
      "step": 88020
    },
    {
      "entropy": 5.739486122131348,
      "epoch": 7.984851233671988,
      "grad_norm": 1.4296875,
      "learning_rate": 9.435025014200069e-05,
      "loss": 4.8442,
      "mean_token_accuracy": 0.22848732322454451,
      "num_tokens": 182665471.0,
      "step": 88025
    },
    {
      "entropy": 5.724664926528931,
      "epoch": 7.985304789550073,
      "grad_norm": 1.4921875,
      "learning_rate": 9.433096337094595e-05,
      "loss": 4.7768,
      "mean_token_accuracy": 0.23432972580194472,
      "num_tokens": 182675056.0,
      "step": 88030
    },
    {
      "entropy": 5.637815046310425,
      "epoch": 7.985758345428157,
      "grad_norm": 1.4296875,
      "learning_rate": 9.431168033617294e-05,
      "loss": 4.7751,
      "mean_token_accuracy": 0.24268036037683488,
      "num_tokens": 182685779.0,
      "step": 88035
    },
    {
      "entropy": 5.748349189758301,
      "epoch": 7.986211901306241,
      "grad_norm": 1.5,
      "learning_rate": 9.42924010380804e-05,
      "loss": 4.8126,
      "mean_token_accuracy": 0.24326829314231874,
      "num_tokens": 182696263.0,
      "step": 88040
    },
    {
      "entropy": 5.727888679504394,
      "epoch": 7.986665457184325,
      "grad_norm": 1.5078125,
      "learning_rate": 9.427312547706703e-05,
      "loss": 4.837,
      "mean_token_accuracy": 0.23079730123281478,
      "num_tokens": 182705466.0,
      "step": 88045
    },
    {
      "entropy": 5.649881172180176,
      "epoch": 7.98711901306241,
      "grad_norm": 1.6015625,
      "learning_rate": 9.425385365353146e-05,
      "loss": 4.745,
      "mean_token_accuracy": 0.23921107947826387,
      "num_tokens": 182716224.0,
      "step": 88050
    },
    {
      "entropy": 5.662356662750244,
      "epoch": 7.987572568940493,
      "grad_norm": 1.4921875,
      "learning_rate": 9.423458556787231e-05,
      "loss": 4.8474,
      "mean_token_accuracy": 0.22689884305000305,
      "num_tokens": 182726452.0,
      "step": 88055
    },
    {
      "entropy": 5.7440516471862795,
      "epoch": 7.988026124818577,
      "grad_norm": 1.5625,
      "learning_rate": 9.421532122048799e-05,
      "loss": 4.8095,
      "mean_token_accuracy": 0.2317022815346718,
      "num_tokens": 182736601.0,
      "step": 88060
    },
    {
      "entropy": 5.756242465972901,
      "epoch": 7.988479680696662,
      "grad_norm": 1.546875,
      "learning_rate": 9.419606061177691e-05,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2206077128648758,
      "num_tokens": 182747426.0,
      "step": 88065
    },
    {
      "entropy": 5.706434822082519,
      "epoch": 7.988933236574746,
      "grad_norm": 1.453125,
      "learning_rate": 9.417680374213733e-05,
      "loss": 4.8611,
      "mean_token_accuracy": 0.23258135467767715,
      "num_tokens": 182757987.0,
      "step": 88070
    },
    {
      "entropy": 5.768540382385254,
      "epoch": 7.98938679245283,
      "grad_norm": 1.6171875,
      "learning_rate": 9.41575506119676e-05,
      "loss": 4.8934,
      "mean_token_accuracy": 0.23206407576799393,
      "num_tokens": 182768477.0,
      "step": 88075
    },
    {
      "entropy": 5.701847648620605,
      "epoch": 7.989840348330914,
      "grad_norm": 1.3046875,
      "learning_rate": 9.413830122166572e-05,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23991002440452575,
      "num_tokens": 182779064.0,
      "step": 88080
    },
    {
      "entropy": 5.722496557235718,
      "epoch": 7.990293904208999,
      "grad_norm": 1.7265625,
      "learning_rate": 9.411905557162987e-05,
      "loss": 4.8747,
      "mean_token_accuracy": 0.2313242346048355,
      "num_tokens": 182789407.0,
      "step": 88085
    },
    {
      "entropy": 5.734282159805298,
      "epoch": 7.990747460087083,
      "grad_norm": 1.4375,
      "learning_rate": 9.409981366225809e-05,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2353300780057907,
      "num_tokens": 182800238.0,
      "step": 88090
    },
    {
      "entropy": 5.687178373336792,
      "epoch": 7.991201015965167,
      "grad_norm": 1.546875,
      "learning_rate": 9.408057549394824e-05,
      "loss": 4.7696,
      "mean_token_accuracy": 0.24075304567813874,
      "num_tokens": 182809980.0,
      "step": 88095
    },
    {
      "entropy": 5.651577472686768,
      "epoch": 7.9916545718432515,
      "grad_norm": 1.40625,
      "learning_rate": 9.406134106709821e-05,
      "loss": 4.8153,
      "mean_token_accuracy": 0.22634083330631255,
      "num_tokens": 182820273.0,
      "step": 88100
    },
    {
      "entropy": 5.7504637241363525,
      "epoch": 7.992108127721336,
      "grad_norm": 1.5625,
      "learning_rate": 9.404211038210575e-05,
      "loss": 4.8302,
      "mean_token_accuracy": 0.23993983417749404,
      "num_tokens": 182830569.0,
      "step": 88105
    },
    {
      "entropy": 5.6735267162323,
      "epoch": 7.992561683599419,
      "grad_norm": 1.53125,
      "learning_rate": 9.402288343936855e-05,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23335183262825013,
      "num_tokens": 182841925.0,
      "step": 88110
    },
    {
      "entropy": 5.701022481918335,
      "epoch": 7.993015239477503,
      "grad_norm": 1.5,
      "learning_rate": 9.400366023928425e-05,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23838406354188918,
      "num_tokens": 182851835.0,
      "step": 88115
    },
    {
      "entropy": 5.660261011123657,
      "epoch": 7.993468795355588,
      "grad_norm": 1.40625,
      "learning_rate": 9.398444078225031e-05,
      "loss": 4.7951,
      "mean_token_accuracy": 0.22995390444993974,
      "num_tokens": 182863727.0,
      "step": 88120
    },
    {
      "entropy": 5.733393001556396,
      "epoch": 7.993922351233672,
      "grad_norm": 1.578125,
      "learning_rate": 9.396522506866433e-05,
      "loss": 4.8012,
      "mean_token_accuracy": 0.23606881946325303,
      "num_tokens": 182874075.0,
      "step": 88125
    },
    {
      "entropy": 5.648679876327515,
      "epoch": 7.994375907111756,
      "grad_norm": 1.5390625,
      "learning_rate": 9.394601309892364e-05,
      "loss": 4.7293,
      "mean_token_accuracy": 0.240378238260746,
      "num_tokens": 182884468.0,
      "step": 88130
    },
    {
      "entropy": 5.733138704299927,
      "epoch": 7.99482946298984,
      "grad_norm": 1.640625,
      "learning_rate": 9.39268048734255e-05,
      "loss": 4.8129,
      "mean_token_accuracy": 0.22977976948022844,
      "num_tokens": 182895940.0,
      "step": 88135
    },
    {
      "entropy": 5.630243635177612,
      "epoch": 7.995283018867925,
      "grad_norm": 1.359375,
      "learning_rate": 9.390760039256721e-05,
      "loss": 4.7163,
      "mean_token_accuracy": 0.24443280100822448,
      "num_tokens": 182906280.0,
      "step": 88140
    },
    {
      "entropy": 5.764097118377686,
      "epoch": 7.995736574746009,
      "grad_norm": 1.5078125,
      "learning_rate": 9.38883996567459e-05,
      "loss": 4.8519,
      "mean_token_accuracy": 0.23180177062749863,
      "num_tokens": 182916988.0,
      "step": 88145
    },
    {
      "entropy": 5.655654430389404,
      "epoch": 7.996190130624093,
      "grad_norm": 1.6171875,
      "learning_rate": 9.386920266635862e-05,
      "loss": 4.7236,
      "mean_token_accuracy": 0.24526562839746474,
      "num_tokens": 182927127.0,
      "step": 88150
    },
    {
      "entropy": 5.642948150634766,
      "epoch": 7.996643686502177,
      "grad_norm": 1.59375,
      "learning_rate": 9.385000942180235e-05,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23502461314201356,
      "num_tokens": 182937760.0,
      "step": 88155
    },
    {
      "entropy": 5.662493658065796,
      "epoch": 7.997097242380261,
      "grad_norm": 1.4375,
      "learning_rate": 9.383081992347406e-05,
      "loss": 4.7654,
      "mean_token_accuracy": 0.23929101526737212,
      "num_tokens": 182949554.0,
      "step": 88160
    },
    {
      "entropy": 5.737311506271363,
      "epoch": 7.997550798258345,
      "grad_norm": 1.5,
      "learning_rate": 9.381163417177063e-05,
      "loss": 4.8449,
      "mean_token_accuracy": 0.238260155916214,
      "num_tokens": 182960200.0,
      "step": 88165
    },
    {
      "entropy": 5.696728992462158,
      "epoch": 7.998004354136429,
      "grad_norm": 1.6171875,
      "learning_rate": 9.379245216708875e-05,
      "loss": 4.811,
      "mean_token_accuracy": 0.23700239211320878,
      "num_tokens": 182969964.0,
      "step": 88170
    },
    {
      "entropy": 5.706399393081665,
      "epoch": 7.998457910014514,
      "grad_norm": 1.359375,
      "learning_rate": 9.377327390982516e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2312012195587158,
      "num_tokens": 182980291.0,
      "step": 88175
    },
    {
      "entropy": 5.681345987319946,
      "epoch": 7.998911465892598,
      "grad_norm": 1.421875,
      "learning_rate": 9.375409940037641e-05,
      "loss": 4.8218,
      "mean_token_accuracy": 0.2374522179365158,
      "num_tokens": 182991146.0,
      "step": 88180
    },
    {
      "entropy": 5.711847925186158,
      "epoch": 7.999365021770682,
      "grad_norm": 1.546875,
      "learning_rate": 9.37349286391391e-05,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23589852899312974,
      "num_tokens": 183001877.0,
      "step": 88185
    },
    {
      "entropy": 5.772753381729126,
      "epoch": 7.9998185776487665,
      "grad_norm": 1.5703125,
      "learning_rate": 9.371576162650957e-05,
      "loss": 4.9529,
      "mean_token_accuracy": 0.22441859394311905,
      "num_tokens": 183013478.0,
      "step": 88190
    },
    {
      "entropy": 5.715186500549317,
      "epoch": 8.00027213352685,
      "grad_norm": 1.5625,
      "learning_rate": 9.369659836288438e-05,
      "loss": 4.8025,
      "mean_token_accuracy": 0.24209764003753662,
      "num_tokens": 183023741.0,
      "step": 88195
    },
    {
      "entropy": 5.646505117416382,
      "epoch": 8.000725689404934,
      "grad_norm": 1.5,
      "learning_rate": 9.367743884865973e-05,
      "loss": 4.7401,
      "mean_token_accuracy": 0.2424220100045204,
      "num_tokens": 183033632.0,
      "step": 88200
    },
    {
      "entropy": 5.70404314994812,
      "epoch": 8.00117924528302,
      "grad_norm": 1.515625,
      "learning_rate": 9.365828308423185e-05,
      "loss": 4.8339,
      "mean_token_accuracy": 0.2331715852022171,
      "num_tokens": 183044360.0,
      "step": 88205
    },
    {
      "entropy": 5.657847499847412,
      "epoch": 8.001632801161103,
      "grad_norm": 1.59375,
      "learning_rate": 9.363913106999689e-05,
      "loss": 4.7508,
      "mean_token_accuracy": 0.24239543676376343,
      "num_tokens": 183054830.0,
      "step": 88210
    },
    {
      "entropy": 5.667177820205689,
      "epoch": 8.002086357039188,
      "grad_norm": 1.578125,
      "learning_rate": 9.36199828063509e-05,
      "loss": 4.7845,
      "mean_token_accuracy": 0.24103267192840577,
      "num_tokens": 183065932.0,
      "step": 88215
    },
    {
      "entropy": 5.741944980621338,
      "epoch": 8.002539912917271,
      "grad_norm": 1.5859375,
      "learning_rate": 9.36008382936899e-05,
      "loss": 4.8497,
      "mean_token_accuracy": 0.22576398849487306,
      "num_tokens": 183076696.0,
      "step": 88220
    },
    {
      "entropy": 5.689089918136597,
      "epoch": 8.002993468795356,
      "grad_norm": 1.5625,
      "learning_rate": 9.358169753240972e-05,
      "loss": 4.7826,
      "mean_token_accuracy": 0.24295923560857774,
      "num_tokens": 183087170.0,
      "step": 88225
    },
    {
      "entropy": 5.663851547241211,
      "epoch": 8.00344702467344,
      "grad_norm": 1.5625,
      "learning_rate": 9.356256052290634e-05,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23823729604482652,
      "num_tokens": 183097409.0,
      "step": 88230
    },
    {
      "entropy": 5.6409478187561035,
      "epoch": 8.003900580551523,
      "grad_norm": 1.4921875,
      "learning_rate": 9.354342726557544e-05,
      "loss": 4.75,
      "mean_token_accuracy": 0.23990684300661086,
      "num_tokens": 183107974.0,
      "step": 88235
    },
    {
      "entropy": 5.6629303932189945,
      "epoch": 8.004354136429608,
      "grad_norm": 1.4921875,
      "learning_rate": 9.352429776081268e-05,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23653139173984528,
      "num_tokens": 183118759.0,
      "step": 88240
    },
    {
      "entropy": 5.7368391990661625,
      "epoch": 8.004807692307692,
      "grad_norm": 1.5234375,
      "learning_rate": 9.350517200901373e-05,
      "loss": 4.8723,
      "mean_token_accuracy": 0.2295817941427231,
      "num_tokens": 183129404.0,
      "step": 88245
    },
    {
      "entropy": 5.710756635665893,
      "epoch": 8.005261248185777,
      "grad_norm": 1.546875,
      "learning_rate": 9.348605001057403e-05,
      "loss": 4.7843,
      "mean_token_accuracy": 0.23809927999973296,
      "num_tokens": 183139958.0,
      "step": 88250
    },
    {
      "entropy": 5.668074369430542,
      "epoch": 8.00571480406386,
      "grad_norm": 1.453125,
      "learning_rate": 9.346693176588903e-05,
      "loss": 4.7862,
      "mean_token_accuracy": 0.24206500202417375,
      "num_tokens": 183150147.0,
      "step": 88255
    },
    {
      "entropy": 5.739777231216431,
      "epoch": 8.006168359941945,
      "grad_norm": 1.6015625,
      "learning_rate": 9.34478172753542e-05,
      "loss": 4.7502,
      "mean_token_accuracy": 0.24019258767366408,
      "num_tokens": 183160257.0,
      "step": 88260
    },
    {
      "entropy": 5.652194404602051,
      "epoch": 8.006621915820029,
      "grad_norm": 1.6015625,
      "learning_rate": 9.342870653936478e-05,
      "loss": 4.772,
      "mean_token_accuracy": 0.23850998282432556,
      "num_tokens": 183169555.0,
      "step": 88265
    },
    {
      "entropy": 5.70416579246521,
      "epoch": 8.007075471698114,
      "grad_norm": 1.5703125,
      "learning_rate": 9.340959955831599e-05,
      "loss": 4.8177,
      "mean_token_accuracy": 0.23292104303836822,
      "num_tokens": 183180192.0,
      "step": 88270
    },
    {
      "entropy": 5.664565563201904,
      "epoch": 8.007529027576197,
      "grad_norm": 1.6953125,
      "learning_rate": 9.339049633260293e-05,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23261932879686356,
      "num_tokens": 183190209.0,
      "step": 88275
    },
    {
      "entropy": 5.701813888549805,
      "epoch": 8.007982583454282,
      "grad_norm": 1.5078125,
      "learning_rate": 9.337139686262071e-05,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2297796428203583,
      "num_tokens": 183199566.0,
      "step": 88280
    },
    {
      "entropy": 5.738393497467041,
      "epoch": 8.008436139332366,
      "grad_norm": 1.4609375,
      "learning_rate": 9.335230114876426e-05,
      "loss": 4.797,
      "mean_token_accuracy": 0.23655875027179718,
      "num_tokens": 183209063.0,
      "step": 88285
    },
    {
      "entropy": 5.670731163024902,
      "epoch": 8.00888969521045,
      "grad_norm": 1.4609375,
      "learning_rate": 9.333320919142847e-05,
      "loss": 4.7621,
      "mean_token_accuracy": 0.23458757400512695,
      "num_tokens": 183219843.0,
      "step": 88290
    },
    {
      "entropy": 5.695229721069336,
      "epoch": 8.009343251088534,
      "grad_norm": 1.3984375,
      "learning_rate": 9.331412099100828e-05,
      "loss": 4.8124,
      "mean_token_accuracy": 0.24537490904331208,
      "num_tokens": 183230070.0,
      "step": 88295
    },
    {
      "entropy": 5.717981481552124,
      "epoch": 8.009796806966618,
      "grad_norm": 1.609375,
      "learning_rate": 9.329503654789835e-05,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23934076130390167,
      "num_tokens": 183241165.0,
      "step": 88300
    },
    {
      "entropy": 5.679701089859009,
      "epoch": 8.010250362844703,
      "grad_norm": 1.4140625,
      "learning_rate": 9.327595586249338e-05,
      "loss": 4.7723,
      "mean_token_accuracy": 0.24253403693437575,
      "num_tokens": 183251488.0,
      "step": 88305
    },
    {
      "entropy": 5.675515985488891,
      "epoch": 8.010703918722786,
      "grad_norm": 1.4140625,
      "learning_rate": 9.325687893518791e-05,
      "loss": 4.807,
      "mean_token_accuracy": 0.24265253394842148,
      "num_tokens": 183260839.0,
      "step": 88310
    },
    {
      "entropy": 5.685131406784057,
      "epoch": 8.011157474600871,
      "grad_norm": 1.546875,
      "learning_rate": 9.323780576637653e-05,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23078867197036743,
      "num_tokens": 183271089.0,
      "step": 88315
    },
    {
      "entropy": 5.689740180969238,
      "epoch": 8.011611030478955,
      "grad_norm": 1.4765625,
      "learning_rate": 9.321873635645364e-05,
      "loss": 4.833,
      "mean_token_accuracy": 0.23064540177583695,
      "num_tokens": 183281582.0,
      "step": 88320
    },
    {
      "entropy": 5.728508710861206,
      "epoch": 8.01206458635704,
      "grad_norm": 1.40625,
      "learning_rate": 9.319967070581356e-05,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23349650651216508,
      "num_tokens": 183292356.0,
      "step": 88325
    },
    {
      "entropy": 5.598227453231812,
      "epoch": 8.012518142235123,
      "grad_norm": 1.6953125,
      "learning_rate": 9.318060881485068e-05,
      "loss": 4.6408,
      "mean_token_accuracy": 0.2520205765962601,
      "num_tokens": 183301253.0,
      "step": 88330
    },
    {
      "entropy": 5.658331108093262,
      "epoch": 8.012971698113208,
      "grad_norm": 1.515625,
      "learning_rate": 9.316155068395912e-05,
      "loss": 4.738,
      "mean_token_accuracy": 0.24567190557718277,
      "num_tokens": 183311245.0,
      "step": 88335
    },
    {
      "entropy": 5.631287479400635,
      "epoch": 8.013425253991292,
      "grad_norm": 1.5078125,
      "learning_rate": 9.314249631353302e-05,
      "loss": 4.7179,
      "mean_token_accuracy": 0.23868029713630676,
      "num_tokens": 183321124.0,
      "step": 88340
    },
    {
      "entropy": 5.7045759677886965,
      "epoch": 8.013878809869375,
      "grad_norm": 1.53125,
      "learning_rate": 9.312344570396647e-05,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23661972731351852,
      "num_tokens": 183331418.0,
      "step": 88345
    },
    {
      "entropy": 5.661901330947876,
      "epoch": 8.01433236574746,
      "grad_norm": 1.4921875,
      "learning_rate": 9.310439885565339e-05,
      "loss": 4.7914,
      "mean_token_accuracy": 0.2376572921872139,
      "num_tokens": 183342387.0,
      "step": 88350
    },
    {
      "entropy": 5.672228384017944,
      "epoch": 8.014785921625544,
      "grad_norm": 1.46875,
      "learning_rate": 9.308535576898772e-05,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2363171696662903,
      "num_tokens": 183353103.0,
      "step": 88355
    },
    {
      "entropy": 5.6636378288269045,
      "epoch": 8.015239477503629,
      "grad_norm": 1.4765625,
      "learning_rate": 9.306631644436316e-05,
      "loss": 4.7875,
      "mean_token_accuracy": 0.2424418583512306,
      "num_tokens": 183363669.0,
      "step": 88360
    },
    {
      "entropy": 5.610389518737793,
      "epoch": 8.015693033381712,
      "grad_norm": 1.5546875,
      "learning_rate": 9.30472808821736e-05,
      "loss": 4.7167,
      "mean_token_accuracy": 0.2385832279920578,
      "num_tokens": 183374035.0,
      "step": 88365
    },
    {
      "entropy": 5.714199542999268,
      "epoch": 8.016146589259797,
      "grad_norm": 1.5234375,
      "learning_rate": 9.302824908281268e-05,
      "loss": 4.8281,
      "mean_token_accuracy": 0.22912790030241012,
      "num_tokens": 183384787.0,
      "step": 88370
    },
    {
      "entropy": 5.673964166641236,
      "epoch": 8.01660014513788,
      "grad_norm": 1.5390625,
      "learning_rate": 9.30092210466739e-05,
      "loss": 4.7092,
      "mean_token_accuracy": 0.2524342596530914,
      "num_tokens": 183395064.0,
      "step": 88375
    },
    {
      "entropy": 5.669791269302368,
      "epoch": 8.017053701015966,
      "grad_norm": 1.515625,
      "learning_rate": 9.299019677415082e-05,
      "loss": 4.8426,
      "mean_token_accuracy": 0.2349270075559616,
      "num_tokens": 183404780.0,
      "step": 88380
    },
    {
      "entropy": 5.661404609680176,
      "epoch": 8.01750725689405,
      "grad_norm": 1.5078125,
      "learning_rate": 9.297117626563687e-05,
      "loss": 4.7637,
      "mean_token_accuracy": 0.23467233031988144,
      "num_tokens": 183416502.0,
      "step": 88385
    },
    {
      "entropy": 5.621473026275635,
      "epoch": 8.017960812772133,
      "grad_norm": 1.453125,
      "learning_rate": 9.295215952152533e-05,
      "loss": 4.7276,
      "mean_token_accuracy": 0.2485438421368599,
      "num_tokens": 183426961.0,
      "step": 88390
    },
    {
      "entropy": 5.725000476837158,
      "epoch": 8.018414368650218,
      "grad_norm": 1.578125,
      "learning_rate": 9.29331465422095e-05,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23878088295459748,
      "num_tokens": 183437905.0,
      "step": 88395
    },
    {
      "entropy": 5.670975208282471,
      "epoch": 8.018867924528301,
      "grad_norm": 1.53125,
      "learning_rate": 9.291413732808266e-05,
      "loss": 4.7018,
      "mean_token_accuracy": 0.23917969316244125,
      "num_tokens": 183447718.0,
      "step": 88400
    },
    {
      "entropy": 5.620235443115234,
      "epoch": 8.019321480406386,
      "grad_norm": 1.46875,
      "learning_rate": 9.289513187953782e-05,
      "loss": 4.6669,
      "mean_token_accuracy": 0.23775699138641357,
      "num_tokens": 183457463.0,
      "step": 88405
    },
    {
      "entropy": 5.652186632156372,
      "epoch": 8.01977503628447,
      "grad_norm": 1.4921875,
      "learning_rate": 9.287613019696811e-05,
      "loss": 4.7159,
      "mean_token_accuracy": 0.24557536840438843,
      "num_tokens": 183467339.0,
      "step": 88410
    },
    {
      "entropy": 5.669899415969849,
      "epoch": 8.020228592162555,
      "grad_norm": 1.515625,
      "learning_rate": 9.285713228076643e-05,
      "loss": 4.8149,
      "mean_token_accuracy": 0.23412976711988448,
      "num_tokens": 183478123.0,
      "step": 88415
    },
    {
      "entropy": 5.73772439956665,
      "epoch": 8.020682148040638,
      "grad_norm": 1.5,
      "learning_rate": 9.283813813132565e-05,
      "loss": 4.8896,
      "mean_token_accuracy": 0.23269892930984498,
      "num_tokens": 183489891.0,
      "step": 88420
    },
    {
      "entropy": 5.68402795791626,
      "epoch": 8.021135703918723,
      "grad_norm": 1.4375,
      "learning_rate": 9.281914774903857e-05,
      "loss": 4.7733,
      "mean_token_accuracy": 0.2439860224723816,
      "num_tokens": 183500704.0,
      "step": 88425
    },
    {
      "entropy": 5.602026128768921,
      "epoch": 8.021589259796807,
      "grad_norm": 1.6796875,
      "learning_rate": 9.280016113429798e-05,
      "loss": 4.6538,
      "mean_token_accuracy": 0.25655538141727446,
      "num_tokens": 183510669.0,
      "step": 88430
    },
    {
      "entropy": 5.7112476348876955,
      "epoch": 8.022042815674892,
      "grad_norm": 1.5,
      "learning_rate": 9.278117828749649e-05,
      "loss": 4.7798,
      "mean_token_accuracy": 0.23378436714410783,
      "num_tokens": 183520943.0,
      "step": 88435
    },
    {
      "entropy": 5.734371328353882,
      "epoch": 8.022496371552975,
      "grad_norm": 1.4609375,
      "learning_rate": 9.276219920902665e-05,
      "loss": 4.8959,
      "mean_token_accuracy": 0.2299284964799881,
      "num_tokens": 183531351.0,
      "step": 88440
    },
    {
      "entropy": 5.643964195251465,
      "epoch": 8.022949927431059,
      "grad_norm": 1.5546875,
      "learning_rate": 9.274322389928097e-05,
      "loss": 4.7373,
      "mean_token_accuracy": 0.23890479207038878,
      "num_tokens": 183541007.0,
      "step": 88445
    },
    {
      "entropy": 5.69184365272522,
      "epoch": 8.023403483309144,
      "grad_norm": 1.515625,
      "learning_rate": 9.272425235865187e-05,
      "loss": 4.791,
      "mean_token_accuracy": 0.24737387001514435,
      "num_tokens": 183550352.0,
      "step": 88450
    },
    {
      "entropy": 5.672103786468506,
      "epoch": 8.023857039187227,
      "grad_norm": 1.453125,
      "learning_rate": 9.27052845875316e-05,
      "loss": 4.786,
      "mean_token_accuracy": 0.23671441674232482,
      "num_tokens": 183559494.0,
      "step": 88455
    },
    {
      "entropy": 5.69144983291626,
      "epoch": 8.024310595065312,
      "grad_norm": 1.5234375,
      "learning_rate": 9.26863205863126e-05,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23363403230905533,
      "num_tokens": 183569855.0,
      "step": 88460
    },
    {
      "entropy": 5.675933694839477,
      "epoch": 8.024764150943396,
      "grad_norm": 1.46875,
      "learning_rate": 9.266736035538691e-05,
      "loss": 4.8266,
      "mean_token_accuracy": 0.228882859647274,
      "num_tokens": 183580777.0,
      "step": 88465
    },
    {
      "entropy": 5.701038026809693,
      "epoch": 8.025217706821481,
      "grad_norm": 1.6171875,
      "learning_rate": 9.26484038951467e-05,
      "loss": 4.7849,
      "mean_token_accuracy": 0.24200354814529418,
      "num_tokens": 183591505.0,
      "step": 88470
    },
    {
      "entropy": 5.6789978504180905,
      "epoch": 8.025671262699564,
      "grad_norm": 1.5390625,
      "learning_rate": 9.262945120598398e-05,
      "loss": 4.7327,
      "mean_token_accuracy": 0.24926633089780809,
      "num_tokens": 183602194.0,
      "step": 88475
    },
    {
      "entropy": 5.692849445343017,
      "epoch": 8.02612481857765,
      "grad_norm": 1.5625,
      "learning_rate": 9.261050228829063e-05,
      "loss": 4.8157,
      "mean_token_accuracy": 0.2305000603199005,
      "num_tokens": 183611130.0,
      "step": 88480
    },
    {
      "entropy": 5.731398677825927,
      "epoch": 8.026578374455733,
      "grad_norm": 1.53125,
      "learning_rate": 9.259155714245863e-05,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23014546781778336,
      "num_tokens": 183620438.0,
      "step": 88485
    },
    {
      "entropy": 5.73216404914856,
      "epoch": 8.027031930333818,
      "grad_norm": 1.4453125,
      "learning_rate": 9.25726157688796e-05,
      "loss": 4.8581,
      "mean_token_accuracy": 0.23354374021291732,
      "num_tokens": 183630713.0,
      "step": 88490
    },
    {
      "entropy": 5.700793027877808,
      "epoch": 8.027485486211901,
      "grad_norm": 1.7734375,
      "learning_rate": 9.255367816794545e-05,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2363242357969284,
      "num_tokens": 183640739.0,
      "step": 88495
    },
    {
      "entropy": 5.699876928329468,
      "epoch": 8.027939042089985,
      "grad_norm": 1.5,
      "learning_rate": 9.253474434004772e-05,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23172446340322495,
      "num_tokens": 183651607.0,
      "step": 88500
    },
    {
      "entropy": 5.801557970046997,
      "epoch": 8.02839259796807,
      "grad_norm": 1.78125,
      "learning_rate": 9.251581428557799e-05,
      "loss": 4.9094,
      "mean_token_accuracy": 0.23136242032051085,
      "num_tokens": 183661250.0,
      "step": 88505
    },
    {
      "entropy": 5.6889947891235355,
      "epoch": 8.028846153846153,
      "grad_norm": 1.5859375,
      "learning_rate": 9.249688800492773e-05,
      "loss": 4.7786,
      "mean_token_accuracy": 0.24274313896894456,
      "num_tokens": 183671214.0,
      "step": 88510
    },
    {
      "entropy": 5.720318984985352,
      "epoch": 8.029299709724238,
      "grad_norm": 1.546875,
      "learning_rate": 9.247796549848829e-05,
      "loss": 4.8167,
      "mean_token_accuracy": 0.2355399549007416,
      "num_tokens": 183681454.0,
      "step": 88515
    },
    {
      "entropy": 5.780133819580078,
      "epoch": 8.029753265602322,
      "grad_norm": 1.5,
      "learning_rate": 9.245904676665106e-05,
      "loss": 4.9006,
      "mean_token_accuracy": 0.22557660043239594,
      "num_tokens": 183692368.0,
      "step": 88520
    },
    {
      "entropy": 5.628642416000366,
      "epoch": 8.030206821480407,
      "grad_norm": 1.625,
      "learning_rate": 9.24401318098072e-05,
      "loss": 4.7915,
      "mean_token_accuracy": 0.24024482667446137,
      "num_tokens": 183702748.0,
      "step": 88525
    },
    {
      "entropy": 5.668634462356567,
      "epoch": 8.03066037735849,
      "grad_norm": 1.5703125,
      "learning_rate": 9.242122062834802e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.23834910243749619,
      "num_tokens": 183712870.0,
      "step": 88530
    },
    {
      "entropy": 5.73281831741333,
      "epoch": 8.031113933236576,
      "grad_norm": 1.5234375,
      "learning_rate": 9.240231322266448e-05,
      "loss": 4.8248,
      "mean_token_accuracy": 0.23371611088514327,
      "num_tokens": 183723117.0,
      "step": 88535
    },
    {
      "entropy": 5.708947086334229,
      "epoch": 8.031567489114659,
      "grad_norm": 1.546875,
      "learning_rate": 9.238340959314767e-05,
      "loss": 4.7981,
      "mean_token_accuracy": 0.22813817858695984,
      "num_tokens": 183734045.0,
      "step": 88540
    },
    {
      "entropy": 5.71813588142395,
      "epoch": 8.032021044992742,
      "grad_norm": 1.546875,
      "learning_rate": 9.236450974018845e-05,
      "loss": 4.7856,
      "mean_token_accuracy": 0.2404692143201828,
      "num_tokens": 183744346.0,
      "step": 88545
    },
    {
      "entropy": 5.710716676712036,
      "epoch": 8.032474600870827,
      "grad_norm": 1.5390625,
      "learning_rate": 9.234561366417774e-05,
      "loss": 4.8477,
      "mean_token_accuracy": 0.23034248054027556,
      "num_tokens": 183755612.0,
      "step": 88550
    },
    {
      "entropy": 5.678199911117554,
      "epoch": 8.03292815674891,
      "grad_norm": 1.46875,
      "learning_rate": 9.232672136550621e-05,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2348870575428009,
      "num_tokens": 183766296.0,
      "step": 88555
    },
    {
      "entropy": 5.748441314697265,
      "epoch": 8.033381712626996,
      "grad_norm": 1.4453125,
      "learning_rate": 9.23078328445647e-05,
      "loss": 4.8535,
      "mean_token_accuracy": 0.23239357471466066,
      "num_tokens": 183776820.0,
      "step": 88560
    },
    {
      "entropy": 5.706440210342407,
      "epoch": 8.03383526850508,
      "grad_norm": 1.390625,
      "learning_rate": 9.228894810174377e-05,
      "loss": 4.7836,
      "mean_token_accuracy": 0.23706794381141663,
      "num_tokens": 183786746.0,
      "step": 88565
    },
    {
      "entropy": 5.691047811508179,
      "epoch": 8.034288824383164,
      "grad_norm": 1.4765625,
      "learning_rate": 9.227006713743394e-05,
      "loss": 4.7624,
      "mean_token_accuracy": 0.23711108416318893,
      "num_tokens": 183797173.0,
      "step": 88570
    },
    {
      "entropy": 5.665773630142212,
      "epoch": 8.034742380261248,
      "grad_norm": 1.4921875,
      "learning_rate": 9.225118995202567e-05,
      "loss": 4.8053,
      "mean_token_accuracy": 0.23113427460193633,
      "num_tokens": 183807512.0,
      "step": 88575
    },
    {
      "entropy": 5.71904821395874,
      "epoch": 8.035195936139333,
      "grad_norm": 1.421875,
      "learning_rate": 9.22323165459094e-05,
      "loss": 4.8266,
      "mean_token_accuracy": 0.23099115788936614,
      "num_tokens": 183818001.0,
      "step": 88580
    },
    {
      "entropy": 5.652286386489868,
      "epoch": 8.035649492017416,
      "grad_norm": 1.40625,
      "learning_rate": 9.221344691947538e-05,
      "loss": 4.7277,
      "mean_token_accuracy": 0.24822088181972504,
      "num_tokens": 183828467.0,
      "step": 88585
    },
    {
      "entropy": 5.66711540222168,
      "epoch": 8.036103047895502,
      "grad_norm": 1.484375,
      "learning_rate": 9.21945810731138e-05,
      "loss": 4.728,
      "mean_token_accuracy": 0.2455317884683609,
      "num_tokens": 183838842.0,
      "step": 88590
    },
    {
      "entropy": 5.699860429763794,
      "epoch": 8.036556603773585,
      "grad_norm": 1.453125,
      "learning_rate": 9.217571900721491e-05,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23736523240804672,
      "num_tokens": 183849538.0,
      "step": 88595
    },
    {
      "entropy": 5.6844895362854,
      "epoch": 8.037010159651668,
      "grad_norm": 1.5390625,
      "learning_rate": 9.215686072216877e-05,
      "loss": 4.7701,
      "mean_token_accuracy": 0.24081097543239594,
      "num_tokens": 183859178.0,
      "step": 88600
    },
    {
      "entropy": 5.678664970397949,
      "epoch": 8.037463715529753,
      "grad_norm": 1.421875,
      "learning_rate": 9.213800621836532e-05,
      "loss": 4.8251,
      "mean_token_accuracy": 0.24073920249938965,
      "num_tokens": 183869920.0,
      "step": 88605
    },
    {
      "entropy": 5.695208263397217,
      "epoch": 8.037917271407837,
      "grad_norm": 1.515625,
      "learning_rate": 9.211915549619452e-05,
      "loss": 4.7938,
      "mean_token_accuracy": 0.24934066981077194,
      "num_tokens": 183879890.0,
      "step": 88610
    },
    {
      "entropy": 5.7102234840393065,
      "epoch": 8.038370827285922,
      "grad_norm": 1.53125,
      "learning_rate": 9.210030855604618e-05,
      "loss": 4.8188,
      "mean_token_accuracy": 0.2371706485748291,
      "num_tokens": 183890127.0,
      "step": 88615
    },
    {
      "entropy": 5.756572437286377,
      "epoch": 8.038824383164005,
      "grad_norm": 1.53125,
      "learning_rate": 9.208146539831009e-05,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2298090025782585,
      "num_tokens": 183901256.0,
      "step": 88620
    },
    {
      "entropy": 5.619883108139038,
      "epoch": 8.03927793904209,
      "grad_norm": 1.515625,
      "learning_rate": 9.20626260233758e-05,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2464712753891945,
      "num_tokens": 183911820.0,
      "step": 88625
    },
    {
      "entropy": 5.637634563446045,
      "epoch": 8.039731494920174,
      "grad_norm": 1.4921875,
      "learning_rate": 9.204379043163311e-05,
      "loss": 4.7681,
      "mean_token_accuracy": 0.2362074762582779,
      "num_tokens": 183921935.0,
      "step": 88630
    },
    {
      "entropy": 5.685691070556641,
      "epoch": 8.040185050798259,
      "grad_norm": 1.5078125,
      "learning_rate": 9.202495862347145e-05,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23913394510746003,
      "num_tokens": 183932123.0,
      "step": 88635
    },
    {
      "entropy": 5.680587816238403,
      "epoch": 8.040638606676342,
      "grad_norm": 1.5546875,
      "learning_rate": 9.200613059928029e-05,
      "loss": 4.8218,
      "mean_token_accuracy": 0.23506616652011872,
      "num_tokens": 183941716.0,
      "step": 88640
    },
    {
      "entropy": 5.684586477279663,
      "epoch": 8.041092162554426,
      "grad_norm": 1.5625,
      "learning_rate": 9.198730635944894e-05,
      "loss": 4.7712,
      "mean_token_accuracy": 0.23652932941913604,
      "num_tokens": 183952139.0,
      "step": 88645
    },
    {
      "entropy": 5.6992861270904545,
      "epoch": 8.041545718432511,
      "grad_norm": 1.625,
      "learning_rate": 9.196848590436679e-05,
      "loss": 4.7959,
      "mean_token_accuracy": 0.2315707817673683,
      "num_tokens": 183962383.0,
      "step": 88650
    },
    {
      "entropy": 5.675466871261596,
      "epoch": 8.041999274310594,
      "grad_norm": 1.4921875,
      "learning_rate": 9.194966923442293e-05,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2349365994334221,
      "num_tokens": 183972799.0,
      "step": 88655
    },
    {
      "entropy": 5.641862440109253,
      "epoch": 8.04245283018868,
      "grad_norm": 1.59375,
      "learning_rate": 9.193085635000655e-05,
      "loss": 4.7709,
      "mean_token_accuracy": 0.24443571269512177,
      "num_tokens": 183983290.0,
      "step": 88660
    },
    {
      "entropy": 5.705705165863037,
      "epoch": 8.042906386066763,
      "grad_norm": 1.4375,
      "learning_rate": 9.191204725150675e-05,
      "loss": 4.7854,
      "mean_token_accuracy": 0.24102916419506074,
      "num_tokens": 183993624.0,
      "step": 88665
    },
    {
      "entropy": 5.639498567581176,
      "epoch": 8.043359941944848,
      "grad_norm": 1.6171875,
      "learning_rate": 9.189324193931248e-05,
      "loss": 4.7472,
      "mean_token_accuracy": 0.24344746321439742,
      "num_tokens": 184003074.0,
      "step": 88670
    },
    {
      "entropy": 5.693265581130982,
      "epoch": 8.043813497822931,
      "grad_norm": 1.453125,
      "learning_rate": 9.187444041381263e-05,
      "loss": 4.8376,
      "mean_token_accuracy": 0.2378740891814232,
      "num_tokens": 184013691.0,
      "step": 88675
    },
    {
      "entropy": 5.78386082649231,
      "epoch": 8.044267053701017,
      "grad_norm": 1.421875,
      "learning_rate": 9.185564267539601e-05,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22135929614305497,
      "num_tokens": 184024952.0,
      "step": 88680
    },
    {
      "entropy": 5.703906106948852,
      "epoch": 8.0447206095791,
      "grad_norm": 1.640625,
      "learning_rate": 9.183684872445135e-05,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23543348163366318,
      "num_tokens": 184035525.0,
      "step": 88685
    },
    {
      "entropy": 5.658564805984497,
      "epoch": 8.045174165457185,
      "grad_norm": 1.640625,
      "learning_rate": 9.181805856136734e-05,
      "loss": 4.7574,
      "mean_token_accuracy": 0.2433038517832756,
      "num_tokens": 184044870.0,
      "step": 88690
    },
    {
      "entropy": 5.676559543609619,
      "epoch": 8.045627721335268,
      "grad_norm": 1.6328125,
      "learning_rate": 9.179927218653253e-05,
      "loss": 4.8117,
      "mean_token_accuracy": 0.23702295273542404,
      "num_tokens": 184054913.0,
      "step": 88695
    },
    {
      "entropy": 5.677186298370361,
      "epoch": 8.046081277213352,
      "grad_norm": 1.4921875,
      "learning_rate": 9.178048960033549e-05,
      "loss": 4.7865,
      "mean_token_accuracy": 0.24147088825702667,
      "num_tokens": 184065242.0,
      "step": 88700
    },
    {
      "entropy": 5.560853433609009,
      "epoch": 8.046534833091437,
      "grad_norm": 1.4453125,
      "learning_rate": 9.176171080316462e-05,
      "loss": 4.6573,
      "mean_token_accuracy": 0.2605211943387985,
      "num_tokens": 184075569.0,
      "step": 88705
    },
    {
      "entropy": 5.739537572860717,
      "epoch": 8.04698838896952,
      "grad_norm": 1.4921875,
      "learning_rate": 9.174293579540826e-05,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22855354994535446,
      "num_tokens": 184086260.0,
      "step": 88710
    },
    {
      "entropy": 5.689338493347168,
      "epoch": 8.047441944847606,
      "grad_norm": 1.625,
      "learning_rate": 9.172416457745471e-05,
      "loss": 4.7712,
      "mean_token_accuracy": 0.24145904183387756,
      "num_tokens": 184095637.0,
      "step": 88715
    },
    {
      "entropy": 5.641110610961914,
      "epoch": 8.047895500725689,
      "grad_norm": 1.40625,
      "learning_rate": 9.170539714969211e-05,
      "loss": 4.7662,
      "mean_token_accuracy": 0.2402218386530876,
      "num_tokens": 184106343.0,
      "step": 88720
    },
    {
      "entropy": 5.707390308380127,
      "epoch": 8.048349056603774,
      "grad_norm": 1.5546875,
      "learning_rate": 9.168663351250861e-05,
      "loss": 4.7772,
      "mean_token_accuracy": 0.240192411839962,
      "num_tokens": 184116407.0,
      "step": 88725
    },
    {
      "entropy": 5.685496997833252,
      "epoch": 8.048802612481857,
      "grad_norm": 1.5078125,
      "learning_rate": 9.166787366629223e-05,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23902878910303116,
      "num_tokens": 184127925.0,
      "step": 88730
    },
    {
      "entropy": 5.83755784034729,
      "epoch": 8.049256168359943,
      "grad_norm": 1.59375,
      "learning_rate": 9.164911761143093e-05,
      "loss": 5.0366,
      "mean_token_accuracy": 0.21157211661338807,
      "num_tokens": 184139542.0,
      "step": 88735
    },
    {
      "entropy": 5.741742086410523,
      "epoch": 8.049709724238026,
      "grad_norm": 1.65625,
      "learning_rate": 9.163036534831265e-05,
      "loss": 4.8324,
      "mean_token_accuracy": 0.24097893089056016,
      "num_tokens": 184150311.0,
      "step": 88740
    },
    {
      "entropy": 5.7208198547363285,
      "epoch": 8.050163280116111,
      "grad_norm": 1.7109375,
      "learning_rate": 9.161161687732514e-05,
      "loss": 4.7729,
      "mean_token_accuracy": 0.2434193566441536,
      "num_tokens": 184160698.0,
      "step": 88745
    },
    {
      "entropy": 5.663554048538208,
      "epoch": 8.050616835994195,
      "grad_norm": 1.59375,
      "learning_rate": 9.159287219885612e-05,
      "loss": 4.7716,
      "mean_token_accuracy": 0.22840117961168288,
      "num_tokens": 184170194.0,
      "step": 88750
    },
    {
      "entropy": 5.728456115722656,
      "epoch": 8.051070391872278,
      "grad_norm": 1.703125,
      "learning_rate": 9.157413131329323e-05,
      "loss": 4.792,
      "mean_token_accuracy": 0.24239365011453629,
      "num_tokens": 184180129.0,
      "step": 88755
    },
    {
      "entropy": 5.71080584526062,
      "epoch": 8.051523947750363,
      "grad_norm": 1.4765625,
      "learning_rate": 9.155539422102399e-05,
      "loss": 4.7482,
      "mean_token_accuracy": 0.23962612003087996,
      "num_tokens": 184189992.0,
      "step": 88760
    },
    {
      "entropy": 5.6686704635620115,
      "epoch": 8.051977503628446,
      "grad_norm": 1.59375,
      "learning_rate": 9.153666092243601e-05,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23781568109989165,
      "num_tokens": 184200190.0,
      "step": 88765
    },
    {
      "entropy": 5.697470617294312,
      "epoch": 8.052431059506532,
      "grad_norm": 1.5703125,
      "learning_rate": 9.151793141791663e-05,
      "loss": 4.8304,
      "mean_token_accuracy": 0.2352663740515709,
      "num_tokens": 184210770.0,
      "step": 88770
    },
    {
      "entropy": 5.616655349731445,
      "epoch": 8.052884615384615,
      "grad_norm": 1.4609375,
      "learning_rate": 9.149920570785319e-05,
      "loss": 4.7318,
      "mean_token_accuracy": 0.23856785893440247,
      "num_tokens": 184221953.0,
      "step": 88775
    },
    {
      "entropy": 5.7251605033874515,
      "epoch": 8.0533381712627,
      "grad_norm": 1.484375,
      "learning_rate": 9.148048379263295e-05,
      "loss": 4.7744,
      "mean_token_accuracy": 0.2422219604253769,
      "num_tokens": 184232417.0,
      "step": 88780
    },
    {
      "entropy": 5.761077547073365,
      "epoch": 8.053791727140784,
      "grad_norm": 1.6484375,
      "learning_rate": 9.146176567264303e-05,
      "loss": 4.7985,
      "mean_token_accuracy": 0.24048984944820403,
      "num_tokens": 184243183.0,
      "step": 88785
    },
    {
      "entropy": 5.694214677810669,
      "epoch": 8.054245283018869,
      "grad_norm": 1.5625,
      "learning_rate": 9.14430513482706e-05,
      "loss": 4.7222,
      "mean_token_accuracy": 0.23891605585813522,
      "num_tokens": 184253259.0,
      "step": 88790
    },
    {
      "entropy": 5.65068736076355,
      "epoch": 8.054698838896952,
      "grad_norm": 1.484375,
      "learning_rate": 9.142434081990258e-05,
      "loss": 4.7374,
      "mean_token_accuracy": 0.2435229688882828,
      "num_tokens": 184263622.0,
      "step": 88795
    },
    {
      "entropy": 5.771281719207764,
      "epoch": 8.055152394775035,
      "grad_norm": 1.484375,
      "learning_rate": 9.140563408792604e-05,
      "loss": 4.8409,
      "mean_token_accuracy": 0.23319167047739028,
      "num_tokens": 184274095.0,
      "step": 88800
    },
    {
      "entropy": 5.658109903335571,
      "epoch": 8.05560595065312,
      "grad_norm": 1.5703125,
      "learning_rate": 9.138693115272775e-05,
      "loss": 4.7933,
      "mean_token_accuracy": 0.23921722918748856,
      "num_tokens": 184284296.0,
      "step": 88805
    },
    {
      "entropy": 5.695224285125732,
      "epoch": 8.056059506531204,
      "grad_norm": 1.40625,
      "learning_rate": 9.136823201469455e-05,
      "loss": 4.8318,
      "mean_token_accuracy": 0.23997295945882796,
      "num_tokens": 184295176.0,
      "step": 88810
    },
    {
      "entropy": 5.6689777851104735,
      "epoch": 8.056513062409289,
      "grad_norm": 1.59375,
      "learning_rate": 9.134953667421305e-05,
      "loss": 4.789,
      "mean_token_accuracy": 0.23251079618930817,
      "num_tokens": 184304402.0,
      "step": 88815
    },
    {
      "entropy": 5.65265474319458,
      "epoch": 8.056966618287372,
      "grad_norm": 1.5625,
      "learning_rate": 9.133084513166998e-05,
      "loss": 4.7566,
      "mean_token_accuracy": 0.2435720220208168,
      "num_tokens": 184314450.0,
      "step": 88820
    },
    {
      "entropy": 5.67157883644104,
      "epoch": 8.057420174165458,
      "grad_norm": 1.484375,
      "learning_rate": 9.131215738745181e-05,
      "loss": 4.7298,
      "mean_token_accuracy": 0.24169321060180665,
      "num_tokens": 184324493.0,
      "step": 88825
    },
    {
      "entropy": 5.632923555374146,
      "epoch": 8.057873730043541,
      "grad_norm": 1.6484375,
      "learning_rate": 9.1293473441945e-05,
      "loss": 4.74,
      "mean_token_accuracy": 0.23957567662000656,
      "num_tokens": 184334938.0,
      "step": 88830
    },
    {
      "entropy": 5.661075115203857,
      "epoch": 8.058327285921626,
      "grad_norm": 1.5234375,
      "learning_rate": 9.127479329553601e-05,
      "loss": 4.7493,
      "mean_token_accuracy": 0.2402481123805046,
      "num_tokens": 184345126.0,
      "step": 88835
    },
    {
      "entropy": 5.745878458023071,
      "epoch": 8.05878084179971,
      "grad_norm": 1.5625,
      "learning_rate": 9.125611694861112e-05,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23454851657152176,
      "num_tokens": 184355219.0,
      "step": 88840
    },
    {
      "entropy": 5.626393604278564,
      "epoch": 8.059234397677795,
      "grad_norm": 1.5625,
      "learning_rate": 9.123744440155655e-05,
      "loss": 4.7724,
      "mean_token_accuracy": 0.24578992575407027,
      "num_tokens": 184366124.0,
      "step": 88845
    },
    {
      "entropy": 5.662112045288086,
      "epoch": 8.059687953555878,
      "grad_norm": 1.578125,
      "learning_rate": 9.121877565475847e-05,
      "loss": 4.77,
      "mean_token_accuracy": 0.24416709542274476,
      "num_tokens": 184376047.0,
      "step": 88850
    },
    {
      "entropy": 5.6440110206604,
      "epoch": 8.060141509433961,
      "grad_norm": 1.5,
      "learning_rate": 9.120011070860298e-05,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2382199376821518,
      "num_tokens": 184386147.0,
      "step": 88855
    },
    {
      "entropy": 5.668408012390136,
      "epoch": 8.060595065312047,
      "grad_norm": 1.5703125,
      "learning_rate": 9.118144956347596e-05,
      "loss": 4.7852,
      "mean_token_accuracy": 0.2400616869330406,
      "num_tokens": 184397419.0,
      "step": 88860
    },
    {
      "entropy": 5.743520927429199,
      "epoch": 8.06104862119013,
      "grad_norm": 1.5,
      "learning_rate": 9.116279221976346e-05,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2330465704202652,
      "num_tokens": 184407750.0,
      "step": 88865
    },
    {
      "entropy": 5.71112093925476,
      "epoch": 8.061502177068215,
      "grad_norm": 1.7734375,
      "learning_rate": 9.11441386778512e-05,
      "loss": 4.828,
      "mean_token_accuracy": 0.23672075122594832,
      "num_tokens": 184417144.0,
      "step": 88870
    },
    {
      "entropy": 5.734814453125,
      "epoch": 8.061955732946299,
      "grad_norm": 1.421875,
      "learning_rate": 9.112548893812509e-05,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2315828874707222,
      "num_tokens": 184428465.0,
      "step": 88875
    },
    {
      "entropy": 5.689841842651367,
      "epoch": 8.062409288824384,
      "grad_norm": 1.671875,
      "learning_rate": 9.110684300097066e-05,
      "loss": 4.7589,
      "mean_token_accuracy": 0.2486005663871765,
      "num_tokens": 184438047.0,
      "step": 88880
    },
    {
      "entropy": 5.677503061294556,
      "epoch": 8.062862844702467,
      "grad_norm": 1.578125,
      "learning_rate": 9.10882008667736e-05,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23906761407852173,
      "num_tokens": 184448817.0,
      "step": 88885
    },
    {
      "entropy": 5.679182863235473,
      "epoch": 8.063316400580552,
      "grad_norm": 1.8671875,
      "learning_rate": 9.106956253591935e-05,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24036614298820497,
      "num_tokens": 184459709.0,
      "step": 88890
    },
    {
      "entropy": 5.672721815109253,
      "epoch": 8.063769956458636,
      "grad_norm": 1.3203125,
      "learning_rate": 9.10509280087935e-05,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23774975538253784,
      "num_tokens": 184470464.0,
      "step": 88895
    },
    {
      "entropy": 5.6767439365386965,
      "epoch": 8.06422351233672,
      "grad_norm": 1.4921875,
      "learning_rate": 9.103229728578132e-05,
      "loss": 4.7785,
      "mean_token_accuracy": 0.23626512587070464,
      "num_tokens": 184480225.0,
      "step": 88900
    },
    {
      "entropy": 5.697042846679688,
      "epoch": 8.064677068214804,
      "grad_norm": 1.453125,
      "learning_rate": 9.101367036726813e-05,
      "loss": 4.7795,
      "mean_token_accuracy": 0.23901121318340302,
      "num_tokens": 184490179.0,
      "step": 88905
    },
    {
      "entropy": 5.71066198348999,
      "epoch": 8.065130624092888,
      "grad_norm": 1.5546875,
      "learning_rate": 9.099504725363908e-05,
      "loss": 4.8549,
      "mean_token_accuracy": 0.23941620588302612,
      "num_tokens": 184499845.0,
      "step": 88910
    },
    {
      "entropy": 5.695357608795166,
      "epoch": 8.065584179970973,
      "grad_norm": 1.5078125,
      "learning_rate": 9.097642794527938e-05,
      "loss": 4.7798,
      "mean_token_accuracy": 0.2355499416589737,
      "num_tokens": 184509802.0,
      "step": 88915
    },
    {
      "entropy": 5.773023462295532,
      "epoch": 8.066037735849056,
      "grad_norm": 1.4921875,
      "learning_rate": 9.095781244257404e-05,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22920386195182801,
      "num_tokens": 184519896.0,
      "step": 88920
    },
    {
      "entropy": 5.819915151596069,
      "epoch": 8.066491291727141,
      "grad_norm": 1.4296875,
      "learning_rate": 9.093920074590798e-05,
      "loss": 5.0276,
      "mean_token_accuracy": 0.21949193328619004,
      "num_tokens": 184531208.0,
      "step": 88925
    },
    {
      "entropy": 5.674592399597168,
      "epoch": 8.066944847605225,
      "grad_norm": 1.5859375,
      "learning_rate": 9.092059285566617e-05,
      "loss": 4.81,
      "mean_token_accuracy": 0.23307133615016937,
      "num_tokens": 184541887.0,
      "step": 88930
    },
    {
      "entropy": 5.77648138999939,
      "epoch": 8.06739840348331,
      "grad_norm": 1.5078125,
      "learning_rate": 9.090198877223346e-05,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2343248873949051,
      "num_tokens": 184553010.0,
      "step": 88935
    },
    {
      "entropy": 5.6992381572723385,
      "epoch": 8.067851959361393,
      "grad_norm": 1.4296875,
      "learning_rate": 9.08833884959945e-05,
      "loss": 4.7947,
      "mean_token_accuracy": 0.22851759642362596,
      "num_tokens": 184563583.0,
      "step": 88940
    },
    {
      "entropy": 5.679094743728638,
      "epoch": 8.068305515239478,
      "grad_norm": 1.484375,
      "learning_rate": 9.086479202733402e-05,
      "loss": 4.7298,
      "mean_token_accuracy": 0.24328063279390336,
      "num_tokens": 184572676.0,
      "step": 88945
    },
    {
      "entropy": 5.680402755737305,
      "epoch": 8.068759071117562,
      "grad_norm": 1.3828125,
      "learning_rate": 9.084619936663654e-05,
      "loss": 4.7794,
      "mean_token_accuracy": 0.23439575731754303,
      "num_tokens": 184583585.0,
      "step": 88950
    },
    {
      "entropy": 5.736190271377564,
      "epoch": 8.069212626995645,
      "grad_norm": 1.5703125,
      "learning_rate": 9.082761051428659e-05,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2404959037899971,
      "num_tokens": 184593110.0,
      "step": 88955
    },
    {
      "entropy": 5.708775520324707,
      "epoch": 8.06966618287373,
      "grad_norm": 1.5,
      "learning_rate": 9.080902547066852e-05,
      "loss": 4.7723,
      "mean_token_accuracy": 0.23617960661649703,
      "num_tokens": 184604053.0,
      "step": 88960
    },
    {
      "entropy": 5.6418252944946286,
      "epoch": 8.070119738751814,
      "grad_norm": 1.5390625,
      "learning_rate": 9.079044423616682e-05,
      "loss": 4.77,
      "mean_token_accuracy": 0.23547545969486236,
      "num_tokens": 184615493.0,
      "step": 88965
    },
    {
      "entropy": 5.634408664703369,
      "epoch": 8.070573294629899,
      "grad_norm": 1.5546875,
      "learning_rate": 9.077186681116567e-05,
      "loss": 4.7583,
      "mean_token_accuracy": 0.241566464304924,
      "num_tokens": 184625621.0,
      "step": 88970
    },
    {
      "entropy": 5.732286834716797,
      "epoch": 8.071026850507982,
      "grad_norm": 1.484375,
      "learning_rate": 9.075329319604924e-05,
      "loss": 4.8237,
      "mean_token_accuracy": 0.232633538544178,
      "num_tokens": 184635846.0,
      "step": 88975
    },
    {
      "entropy": 5.718103837966919,
      "epoch": 8.071480406386067,
      "grad_norm": 1.4453125,
      "learning_rate": 9.07347233912017e-05,
      "loss": 4.752,
      "mean_token_accuracy": 0.24072131067514418,
      "num_tokens": 184646254.0,
      "step": 88980
    },
    {
      "entropy": 5.70264744758606,
      "epoch": 8.07193396226415,
      "grad_norm": 1.5859375,
      "learning_rate": 9.071615739700697e-05,
      "loss": 4.8219,
      "mean_token_accuracy": 0.23307468146085739,
      "num_tokens": 184656672.0,
      "step": 88985
    },
    {
      "entropy": 5.698955154418945,
      "epoch": 8.072387518142236,
      "grad_norm": 1.46875,
      "learning_rate": 9.069759521384913e-05,
      "loss": 4.8004,
      "mean_token_accuracy": 0.2395714983344078,
      "num_tokens": 184666938.0,
      "step": 88990
    },
    {
      "entropy": 5.711404037475586,
      "epoch": 8.07284107402032,
      "grad_norm": 1.5390625,
      "learning_rate": 9.06790368421119e-05,
      "loss": 4.8051,
      "mean_token_accuracy": 0.239711731672287,
      "num_tokens": 184677092.0,
      "step": 88995
    },
    {
      "entropy": 5.7073667526245115,
      "epoch": 8.073294629898404,
      "grad_norm": 1.53125,
      "learning_rate": 9.066048228217924e-05,
      "loss": 4.8479,
      "mean_token_accuracy": 0.2289802297949791,
      "num_tokens": 184688371.0,
      "step": 89000
    },
    {
      "entropy": 5.634005737304688,
      "epoch": 8.073748185776488,
      "grad_norm": 1.4921875,
      "learning_rate": 9.064193153443476e-05,
      "loss": 4.725,
      "mean_token_accuracy": 0.23961655497550965,
      "num_tokens": 184698403.0,
      "step": 89005
    },
    {
      "entropy": 5.7286382675170895,
      "epoch": 8.074201741654571,
      "grad_norm": 1.5859375,
      "learning_rate": 9.062338459926211e-05,
      "loss": 4.8082,
      "mean_token_accuracy": 0.23816193491220475,
      "num_tokens": 184708641.0,
      "step": 89010
    },
    {
      "entropy": 5.68962664604187,
      "epoch": 8.074655297532656,
      "grad_norm": 1.484375,
      "learning_rate": 9.060484147704487e-05,
      "loss": 4.7926,
      "mean_token_accuracy": 0.23962654024362565,
      "num_tokens": 184719675.0,
      "step": 89015
    },
    {
      "entropy": 5.709149169921875,
      "epoch": 8.07510885341074,
      "grad_norm": 1.5546875,
      "learning_rate": 9.058630216816647e-05,
      "loss": 4.8464,
      "mean_token_accuracy": 0.23370870500802993,
      "num_tokens": 184728979.0,
      "step": 89020
    },
    {
      "entropy": 5.734915828704834,
      "epoch": 8.075562409288825,
      "grad_norm": 1.5625,
      "learning_rate": 9.056776667301037e-05,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23198716193437577,
      "num_tokens": 184738965.0,
      "step": 89025
    },
    {
      "entropy": 5.675928115844727,
      "epoch": 8.076015965166908,
      "grad_norm": 1.5546875,
      "learning_rate": 9.054923499195977e-05,
      "loss": 4.7962,
      "mean_token_accuracy": 0.23705653995275497,
      "num_tokens": 184748946.0,
      "step": 89030
    },
    {
      "entropy": 5.697159385681152,
      "epoch": 8.076469521044993,
      "grad_norm": 1.46875,
      "learning_rate": 9.053070712539805e-05,
      "loss": 4.7799,
      "mean_token_accuracy": 0.2366035744547844,
      "num_tokens": 184759840.0,
      "step": 89035
    },
    {
      "entropy": 5.700419616699219,
      "epoch": 8.076923076923077,
      "grad_norm": 1.53125,
      "learning_rate": 9.051218307370834e-05,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22823619842529297,
      "num_tokens": 184770429.0,
      "step": 89040
    },
    {
      "entropy": 5.631366539001465,
      "epoch": 8.077376632801162,
      "grad_norm": 1.3828125,
      "learning_rate": 9.04936628372737e-05,
      "loss": 4.7306,
      "mean_token_accuracy": 0.2492005705833435,
      "num_tokens": 184781064.0,
      "step": 89045
    },
    {
      "entropy": 5.662292766571045,
      "epoch": 8.077830188679245,
      "grad_norm": 1.5078125,
      "learning_rate": 9.047514641647709e-05,
      "loss": 4.7423,
      "mean_token_accuracy": 0.24779132157564163,
      "num_tokens": 184791301.0,
      "step": 89050
    },
    {
      "entropy": 5.625855350494385,
      "epoch": 8.07828374455733,
      "grad_norm": 1.4140625,
      "learning_rate": 9.045663381170149e-05,
      "loss": 4.7498,
      "mean_token_accuracy": 0.2401564747095108,
      "num_tokens": 184802067.0,
      "step": 89055
    },
    {
      "entropy": 5.693572187423706,
      "epoch": 8.078737300435414,
      "grad_norm": 1.484375,
      "learning_rate": 9.043812502332975e-05,
      "loss": 4.8312,
      "mean_token_accuracy": 0.2431740939617157,
      "num_tokens": 184811735.0,
      "step": 89060
    },
    {
      "entropy": 5.664408874511719,
      "epoch": 8.079190856313497,
      "grad_norm": 1.625,
      "learning_rate": 9.041962005174456e-05,
      "loss": 4.8016,
      "mean_token_accuracy": 0.23787269592285157,
      "num_tokens": 184821430.0,
      "step": 89065
    },
    {
      "entropy": 5.636523866653443,
      "epoch": 8.079644412191582,
      "grad_norm": 1.6875,
      "learning_rate": 9.04011188973287e-05,
      "loss": 4.6941,
      "mean_token_accuracy": 0.24657850563526154,
      "num_tokens": 184831022.0,
      "step": 89070
    },
    {
      "entropy": 5.687421131134033,
      "epoch": 8.080097968069666,
      "grad_norm": 1.484375,
      "learning_rate": 9.038262156046475e-05,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2355932652950287,
      "num_tokens": 184841951.0,
      "step": 89075
    },
    {
      "entropy": 5.700655841827393,
      "epoch": 8.08055152394775,
      "grad_norm": 1.53125,
      "learning_rate": 9.036412804153522e-05,
      "loss": 4.6975,
      "mean_token_accuracy": 0.242298686504364,
      "num_tokens": 184851564.0,
      "step": 89080
    },
    {
      "entropy": 5.7249672412872314,
      "epoch": 8.081005079825834,
      "grad_norm": 1.6328125,
      "learning_rate": 9.03456383409226e-05,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2245328739285469,
      "num_tokens": 184862218.0,
      "step": 89085
    },
    {
      "entropy": 5.6476788997650145,
      "epoch": 8.08145863570392,
      "grad_norm": 1.4921875,
      "learning_rate": 9.032715245900922e-05,
      "loss": 4.7019,
      "mean_token_accuracy": 0.24718987494707106,
      "num_tokens": 184872658.0,
      "step": 89090
    },
    {
      "entropy": 5.664550113677978,
      "epoch": 8.081912191582003,
      "grad_norm": 1.4765625,
      "learning_rate": 9.030867039617732e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.2402094319462776,
      "num_tokens": 184883242.0,
      "step": 89095
    },
    {
      "entropy": 5.625793361663819,
      "epoch": 8.082365747460088,
      "grad_norm": 1.515625,
      "learning_rate": 9.029019215280928e-05,
      "loss": 4.7013,
      "mean_token_accuracy": 0.2448785722255707,
      "num_tokens": 184894491.0,
      "step": 89100
    },
    {
      "entropy": 5.6579986095428465,
      "epoch": 8.082819303338171,
      "grad_norm": 1.6171875,
      "learning_rate": 9.02717177292871e-05,
      "loss": 4.768,
      "mean_token_accuracy": 0.23947569727897644,
      "num_tokens": 184904423.0,
      "step": 89105
    },
    {
      "entropy": 5.678820991516114,
      "epoch": 8.083272859216255,
      "grad_norm": 1.546875,
      "learning_rate": 9.025324712599286e-05,
      "loss": 4.8004,
      "mean_token_accuracy": 0.23370856046676636,
      "num_tokens": 184914883.0,
      "step": 89110
    },
    {
      "entropy": 5.748437833786011,
      "epoch": 8.08372641509434,
      "grad_norm": 1.484375,
      "learning_rate": 9.023478034330857e-05,
      "loss": 4.9031,
      "mean_token_accuracy": 0.2281048759818077,
      "num_tokens": 184925284.0,
      "step": 89115
    },
    {
      "entropy": 5.619724559783935,
      "epoch": 8.084179970972423,
      "grad_norm": 1.4296875,
      "learning_rate": 9.021631738161612e-05,
      "loss": 4.6946,
      "mean_token_accuracy": 0.23424165248870848,
      "num_tokens": 184935475.0,
      "step": 89120
    },
    {
      "entropy": 5.639765739440918,
      "epoch": 8.084633526850508,
      "grad_norm": 1.5625,
      "learning_rate": 9.019785824129728e-05,
      "loss": 4.7713,
      "mean_token_accuracy": 0.2414167433977127,
      "num_tokens": 184946514.0,
      "step": 89125
    },
    {
      "entropy": 5.647387456893921,
      "epoch": 8.085087082728592,
      "grad_norm": 1.5078125,
      "learning_rate": 9.01794029227338e-05,
      "loss": 4.7343,
      "mean_token_accuracy": 0.24221238791942595,
      "num_tokens": 184957264.0,
      "step": 89130
    },
    {
      "entropy": 5.733387660980225,
      "epoch": 8.085540638606677,
      "grad_norm": 1.5078125,
      "learning_rate": 9.01609514263074e-05,
      "loss": 4.9271,
      "mean_token_accuracy": 0.23073431700468064,
      "num_tokens": 184968387.0,
      "step": 89135
    },
    {
      "entropy": 5.677318000793457,
      "epoch": 8.08599419448476,
      "grad_norm": 1.5390625,
      "learning_rate": 9.014250375239964e-05,
      "loss": 4.7679,
      "mean_token_accuracy": 0.2396603509783745,
      "num_tokens": 184977861.0,
      "step": 89140
    },
    {
      "entropy": 5.735145711898804,
      "epoch": 8.086447750362845,
      "grad_norm": 1.6171875,
      "learning_rate": 9.0124059901392e-05,
      "loss": 4.7971,
      "mean_token_accuracy": 0.2469117134809494,
      "num_tokens": 184987922.0,
      "step": 89145
    },
    {
      "entropy": 5.648133897781372,
      "epoch": 8.086901306240929,
      "grad_norm": 1.4375,
      "learning_rate": 9.010561987366591e-05,
      "loss": 4.7556,
      "mean_token_accuracy": 0.23937630504369736,
      "num_tokens": 184999294.0,
      "step": 89150
    },
    {
      "entropy": 5.697195243835449,
      "epoch": 8.087354862119014,
      "grad_norm": 1.5390625,
      "learning_rate": 9.008718366960274e-05,
      "loss": 4.8839,
      "mean_token_accuracy": 0.23131947517395018,
      "num_tokens": 185009111.0,
      "step": 89155
    },
    {
      "entropy": 5.704985046386719,
      "epoch": 8.087808417997097,
      "grad_norm": 1.5078125,
      "learning_rate": 9.006875128958369e-05,
      "loss": 4.7937,
      "mean_token_accuracy": 0.24174809604883193,
      "num_tokens": 185019549.0,
      "step": 89160
    },
    {
      "entropy": 5.64947099685669,
      "epoch": 8.08826197387518,
      "grad_norm": 1.5625,
      "learning_rate": 9.005032273399e-05,
      "loss": 4.7155,
      "mean_token_accuracy": 0.24529820680618286,
      "num_tokens": 185029264.0,
      "step": 89165
    },
    {
      "entropy": 5.67053189277649,
      "epoch": 8.088715529753266,
      "grad_norm": 1.3671875,
      "learning_rate": 9.003189800320277e-05,
      "loss": 4.7949,
      "mean_token_accuracy": 0.2371310904622078,
      "num_tokens": 185040214.0,
      "step": 89170
    },
    {
      "entropy": 5.665576410293579,
      "epoch": 8.08916908563135,
      "grad_norm": 1.546875,
      "learning_rate": 9.001347709760305e-05,
      "loss": 4.7387,
      "mean_token_accuracy": 0.23648815155029296,
      "num_tokens": 185050629.0,
      "step": 89175
    },
    {
      "entropy": 5.708168792724609,
      "epoch": 8.089622641509434,
      "grad_norm": 1.671875,
      "learning_rate": 8.999506001757174e-05,
      "loss": 4.8317,
      "mean_token_accuracy": 0.23571875244379042,
      "num_tokens": 185061851.0,
      "step": 89180
    },
    {
      "entropy": 5.728915786743164,
      "epoch": 8.090076197387518,
      "grad_norm": 1.5390625,
      "learning_rate": 8.997664676348974e-05,
      "loss": 4.877,
      "mean_token_accuracy": 0.22821419984102248,
      "num_tokens": 185072937.0,
      "step": 89185
    },
    {
      "entropy": 5.68441858291626,
      "epoch": 8.090529753265603,
      "grad_norm": 1.4453125,
      "learning_rate": 8.995823733573785e-05,
      "loss": 4.8504,
      "mean_token_accuracy": 0.22762657552957535,
      "num_tokens": 185083769.0,
      "step": 89190
    },
    {
      "entropy": 5.6146924018859865,
      "epoch": 8.090983309143686,
      "grad_norm": 1.546875,
      "learning_rate": 8.993983173469675e-05,
      "loss": 4.7224,
      "mean_token_accuracy": 0.24021857529878615,
      "num_tokens": 185094408.0,
      "step": 89195
    },
    {
      "entropy": 5.713783407211304,
      "epoch": 8.091436865021771,
      "grad_norm": 1.546875,
      "learning_rate": 8.992142996074705e-05,
      "loss": 4.8395,
      "mean_token_accuracy": 0.22884520888328552,
      "num_tokens": 185105411.0,
      "step": 89200
    },
    {
      "entropy": 5.6462188243865965,
      "epoch": 8.091890420899855,
      "grad_norm": 1.609375,
      "learning_rate": 8.990303201426939e-05,
      "loss": 4.715,
      "mean_token_accuracy": 0.24122560769319534,
      "num_tokens": 185114243.0,
      "step": 89205
    },
    {
      "entropy": 5.6803981304168705,
      "epoch": 8.09234397677794,
      "grad_norm": 1.5546875,
      "learning_rate": 8.988463789564419e-05,
      "loss": 4.7125,
      "mean_token_accuracy": 0.24358083605766295,
      "num_tokens": 185123395.0,
      "step": 89210
    },
    {
      "entropy": 5.708551454544067,
      "epoch": 8.092797532656023,
      "grad_norm": 1.4765625,
      "learning_rate": 8.986624760525186e-05,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2414117708802223,
      "num_tokens": 185133495.0,
      "step": 89215
    },
    {
      "entropy": 5.74036717414856,
      "epoch": 8.093251088534107,
      "grad_norm": 1.515625,
      "learning_rate": 8.984786114347271e-05,
      "loss": 4.841,
      "mean_token_accuracy": 0.24228350669145585,
      "num_tokens": 185143363.0,
      "step": 89220
    },
    {
      "entropy": 5.737514352798462,
      "epoch": 8.093704644412192,
      "grad_norm": 1.3828125,
      "learning_rate": 8.982947851068696e-05,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2332490637898445,
      "num_tokens": 185154171.0,
      "step": 89225
    },
    {
      "entropy": 5.727664756774902,
      "epoch": 8.094158200290275,
      "grad_norm": 1.6015625,
      "learning_rate": 8.981109970727478e-05,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2409597098827362,
      "num_tokens": 185164100.0,
      "step": 89230
    },
    {
      "entropy": 5.657065677642822,
      "epoch": 8.09461175616836,
      "grad_norm": 1.53125,
      "learning_rate": 8.979272473361622e-05,
      "loss": 4.7082,
      "mean_token_accuracy": 0.24686072617769242,
      "num_tokens": 185175791.0,
      "step": 89235
    },
    {
      "entropy": 5.660937118530273,
      "epoch": 8.095065312046444,
      "grad_norm": 1.4375,
      "learning_rate": 8.977435359009134e-05,
      "loss": 4.721,
      "mean_token_accuracy": 0.2426997020840645,
      "num_tokens": 185186652.0,
      "step": 89240
    },
    {
      "entropy": 5.6888936996459964,
      "epoch": 8.095518867924529,
      "grad_norm": 1.4765625,
      "learning_rate": 8.975598627708004e-05,
      "loss": 4.7702,
      "mean_token_accuracy": 0.2427448883652687,
      "num_tokens": 185198604.0,
      "step": 89245
    },
    {
      "entropy": 5.632753181457519,
      "epoch": 8.095972423802612,
      "grad_norm": 1.53125,
      "learning_rate": 8.973762279496214e-05,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2439724326133728,
      "num_tokens": 185208997.0,
      "step": 89250
    },
    {
      "entropy": 5.593669748306274,
      "epoch": 8.096425979680697,
      "grad_norm": 1.5625,
      "learning_rate": 8.97192631441174e-05,
      "loss": 4.7258,
      "mean_token_accuracy": 0.24570396840572356,
      "num_tokens": 185218722.0,
      "step": 89255
    },
    {
      "entropy": 5.658975601196289,
      "epoch": 8.09687953555878,
      "grad_norm": 1.640625,
      "learning_rate": 8.970090732492556e-05,
      "loss": 4.7664,
      "mean_token_accuracy": 0.24130937606096267,
      "num_tokens": 185228570.0,
      "step": 89260
    },
    {
      "entropy": 5.681626892089843,
      "epoch": 8.097333091436864,
      "grad_norm": 1.6015625,
      "learning_rate": 8.968255533776607e-05,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23633079081773758,
      "num_tokens": 185239047.0,
      "step": 89265
    },
    {
      "entropy": 5.708721733093261,
      "epoch": 8.09778664731495,
      "grad_norm": 1.53125,
      "learning_rate": 8.966420718301865e-05,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22242368757724762,
      "num_tokens": 185248211.0,
      "step": 89270
    },
    {
      "entropy": 5.7149416446685795,
      "epoch": 8.098240203193033,
      "grad_norm": 1.5859375,
      "learning_rate": 8.964586286106263e-05,
      "loss": 4.8553,
      "mean_token_accuracy": 0.23374096900224686,
      "num_tokens": 185258883.0,
      "step": 89275
    },
    {
      "entropy": 5.741793918609619,
      "epoch": 8.098693759071118,
      "grad_norm": 1.53125,
      "learning_rate": 8.962752237227742e-05,
      "loss": 4.8232,
      "mean_token_accuracy": 0.22964415252208709,
      "num_tokens": 185268263.0,
      "step": 89280
    },
    {
      "entropy": 5.668920850753784,
      "epoch": 8.099147314949201,
      "grad_norm": 1.578125,
      "learning_rate": 8.960918571704235e-05,
      "loss": 4.7838,
      "mean_token_accuracy": 0.23635635673999786,
      "num_tokens": 185278851.0,
      "step": 89285
    },
    {
      "entropy": 5.687656116485596,
      "epoch": 8.099600870827286,
      "grad_norm": 1.5546875,
      "learning_rate": 8.959085289573649e-05,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2297337234020233,
      "num_tokens": 185289205.0,
      "step": 89290
    },
    {
      "entropy": 5.73024091720581,
      "epoch": 8.10005442670537,
      "grad_norm": 1.5859375,
      "learning_rate": 8.957252390873897e-05,
      "loss": 4.785,
      "mean_token_accuracy": 0.24525291919708253,
      "num_tokens": 185298690.0,
      "step": 89295
    },
    {
      "entropy": 5.699236106872559,
      "epoch": 8.100507982583455,
      "grad_norm": 1.4375,
      "learning_rate": 8.955419875642898e-05,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22627556473016738,
      "num_tokens": 185309164.0,
      "step": 89300
    },
    {
      "entropy": 5.716329431533813,
      "epoch": 8.100961538461538,
      "grad_norm": 1.5703125,
      "learning_rate": 8.953587743918543e-05,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23891621530056,
      "num_tokens": 185319190.0,
      "step": 89305
    },
    {
      "entropy": 5.7009416103363035,
      "epoch": 8.101415094339623,
      "grad_norm": 1.6953125,
      "learning_rate": 8.951755995738716e-05,
      "loss": 4.7878,
      "mean_token_accuracy": 0.24279410243034363,
      "num_tokens": 185329353.0,
      "step": 89310
    },
    {
      "entropy": 5.777675008773803,
      "epoch": 8.101868650217707,
      "grad_norm": 1.46875,
      "learning_rate": 8.949924631141304e-05,
      "loss": 4.8617,
      "mean_token_accuracy": 0.23790584057569503,
      "num_tokens": 185339991.0,
      "step": 89315
    },
    {
      "entropy": 5.733343124389648,
      "epoch": 8.10232220609579,
      "grad_norm": 1.4453125,
      "learning_rate": 8.94809365016418e-05,
      "loss": 4.859,
      "mean_token_accuracy": 0.23597554117441177,
      "num_tokens": 185351605.0,
      "step": 89320
    },
    {
      "entropy": 5.633254432678223,
      "epoch": 8.102775761973875,
      "grad_norm": 1.375,
      "learning_rate": 8.946263052845202e-05,
      "loss": 4.6946,
      "mean_token_accuracy": 0.24437943249940872,
      "num_tokens": 185362414.0,
      "step": 89325
    },
    {
      "entropy": 5.713235807418823,
      "epoch": 8.103229317851959,
      "grad_norm": 1.484375,
      "learning_rate": 8.944432839222228e-05,
      "loss": 4.8242,
      "mean_token_accuracy": 0.24062219262123108,
      "num_tokens": 185372959.0,
      "step": 89330
    },
    {
      "entropy": 5.67259578704834,
      "epoch": 8.103682873730044,
      "grad_norm": 1.5390625,
      "learning_rate": 8.94260300933312e-05,
      "loss": 4.7695,
      "mean_token_accuracy": 0.2409458801150322,
      "num_tokens": 185382880.0,
      "step": 89335
    },
    {
      "entropy": 5.6586606979370115,
      "epoch": 8.104136429608127,
      "grad_norm": 1.5234375,
      "learning_rate": 8.94077356321571e-05,
      "loss": 4.8575,
      "mean_token_accuracy": 0.23404385447502135,
      "num_tokens": 185392574.0,
      "step": 89340
    },
    {
      "entropy": 5.7516499042510985,
      "epoch": 8.104589985486212,
      "grad_norm": 1.5234375,
      "learning_rate": 8.938944500907829e-05,
      "loss": 4.8639,
      "mean_token_accuracy": 0.23929130136966706,
      "num_tokens": 185402915.0,
      "step": 89345
    },
    {
      "entropy": 5.668077659606934,
      "epoch": 8.105043541364296,
      "grad_norm": 1.4609375,
      "learning_rate": 8.937115822447306e-05,
      "loss": 4.7922,
      "mean_token_accuracy": 0.2332243874669075,
      "num_tokens": 185413861.0,
      "step": 89350
    },
    {
      "entropy": 5.7491003513336185,
      "epoch": 8.105497097242381,
      "grad_norm": 1.6484375,
      "learning_rate": 8.935287527871961e-05,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2283594697713852,
      "num_tokens": 185423684.0,
      "step": 89355
    },
    {
      "entropy": 5.7050111293792725,
      "epoch": 8.105950653120464,
      "grad_norm": 1.453125,
      "learning_rate": 8.9334596172196e-05,
      "loss": 4.7281,
      "mean_token_accuracy": 0.2469437062740326,
      "num_tokens": 185433571.0,
      "step": 89360
    },
    {
      "entropy": 5.675871419906616,
      "epoch": 8.106404208998548,
      "grad_norm": 1.46875,
      "learning_rate": 8.931632090528018e-05,
      "loss": 4.804,
      "mean_token_accuracy": 0.23672958165407182,
      "num_tokens": 185443757.0,
      "step": 89365
    },
    {
      "entropy": 5.699294137954712,
      "epoch": 8.106857764876633,
      "grad_norm": 1.484375,
      "learning_rate": 8.929804947835022e-05,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23314024060964583,
      "num_tokens": 185453072.0,
      "step": 89370
    },
    {
      "entropy": 5.681890773773193,
      "epoch": 8.107311320754716,
      "grad_norm": 1.6640625,
      "learning_rate": 8.927978189178395e-05,
      "loss": 4.8416,
      "mean_token_accuracy": 0.23526689708232879,
      "num_tokens": 185463537.0,
      "step": 89375
    },
    {
      "entropy": 5.660835838317871,
      "epoch": 8.107764876632801,
      "grad_norm": 1.6015625,
      "learning_rate": 8.92615181459591e-05,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23723675161600113,
      "num_tokens": 185474042.0,
      "step": 89380
    },
    {
      "entropy": 5.7123548030853275,
      "epoch": 8.108218432510885,
      "grad_norm": 1.4609375,
      "learning_rate": 8.924325824125337e-05,
      "loss": 4.8545,
      "mean_token_accuracy": 0.2339087978005409,
      "num_tokens": 185484322.0,
      "step": 89385
    },
    {
      "entropy": 5.830361318588257,
      "epoch": 8.10867198838897,
      "grad_norm": 1.609375,
      "learning_rate": 8.922500217804442e-05,
      "loss": 4.8869,
      "mean_token_accuracy": 0.23396568596363068,
      "num_tokens": 185494113.0,
      "step": 89390
    },
    {
      "entropy": 5.69447169303894,
      "epoch": 8.109125544267053,
      "grad_norm": 1.3515625,
      "learning_rate": 8.920674995670969e-05,
      "loss": 4.7684,
      "mean_token_accuracy": 0.24247500747442247,
      "num_tokens": 185504813.0,
      "step": 89395
    },
    {
      "entropy": 5.688111591339111,
      "epoch": 8.109579100145138,
      "grad_norm": 1.7421875,
      "learning_rate": 8.91885015776268e-05,
      "loss": 4.7898,
      "mean_token_accuracy": 0.23887424021959305,
      "num_tokens": 185515164.0,
      "step": 89400
    },
    {
      "entropy": 5.701454114913941,
      "epoch": 8.110032656023222,
      "grad_norm": 1.5703125,
      "learning_rate": 8.917025704117302e-05,
      "loss": 4.8342,
      "mean_token_accuracy": 0.23463582694530488,
      "num_tokens": 185524463.0,
      "step": 89405
    },
    {
      "entropy": 5.658344745635986,
      "epoch": 8.110486211901307,
      "grad_norm": 1.484375,
      "learning_rate": 8.915201634772571e-05,
      "loss": 4.7193,
      "mean_token_accuracy": 0.24628141522407532,
      "num_tokens": 185535515.0,
      "step": 89410
    },
    {
      "entropy": 5.692699527740478,
      "epoch": 8.11093976777939,
      "grad_norm": 1.46875,
      "learning_rate": 8.913377949766202e-05,
      "loss": 4.7704,
      "mean_token_accuracy": 0.24527401179075242,
      "num_tokens": 185546363.0,
      "step": 89415
    },
    {
      "entropy": 5.745241308212281,
      "epoch": 8.111393323657474,
      "grad_norm": 1.6484375,
      "learning_rate": 8.911554649135917e-05,
      "loss": 4.8551,
      "mean_token_accuracy": 0.2272896483540535,
      "num_tokens": 185557230.0,
      "step": 89420
    },
    {
      "entropy": 5.695393705368042,
      "epoch": 8.111846879535559,
      "grad_norm": 1.5,
      "learning_rate": 8.909731732919415e-05,
      "loss": 4.8041,
      "mean_token_accuracy": 0.23711320906877517,
      "num_tokens": 185567833.0,
      "step": 89425
    },
    {
      "entropy": 5.683047437667847,
      "epoch": 8.112300435413642,
      "grad_norm": 1.5234375,
      "learning_rate": 8.907909201154397e-05,
      "loss": 4.7997,
      "mean_token_accuracy": 0.23706262409687043,
      "num_tokens": 185577247.0,
      "step": 89430
    },
    {
      "entropy": 5.652046251296997,
      "epoch": 8.112753991291727,
      "grad_norm": 1.484375,
      "learning_rate": 8.90608705387856e-05,
      "loss": 4.7207,
      "mean_token_accuracy": 0.24440715759992598,
      "num_tokens": 185587233.0,
      "step": 89435
    },
    {
      "entropy": 5.702265739440918,
      "epoch": 8.11320754716981,
      "grad_norm": 1.4296875,
      "learning_rate": 8.904265291129577e-05,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23495918959379197,
      "num_tokens": 185598530.0,
      "step": 89440
    },
    {
      "entropy": 5.6863791942596436,
      "epoch": 8.113661103047896,
      "grad_norm": 1.5,
      "learning_rate": 8.90244391294513e-05,
      "loss": 4.7445,
      "mean_token_accuracy": 0.2422808066010475,
      "num_tokens": 185608547.0,
      "step": 89445
    },
    {
      "entropy": 5.6876208782196045,
      "epoch": 8.11411465892598,
      "grad_norm": 1.296875,
      "learning_rate": 8.90062291936288e-05,
      "loss": 4.7883,
      "mean_token_accuracy": 0.23916348218917846,
      "num_tokens": 185619751.0,
      "step": 89450
    },
    {
      "entropy": 5.634572982788086,
      "epoch": 8.114568214804065,
      "grad_norm": 1.5234375,
      "learning_rate": 8.898802310420486e-05,
      "loss": 4.6567,
      "mean_token_accuracy": 0.24047423303127288,
      "num_tokens": 185630483.0,
      "step": 89455
    },
    {
      "entropy": 5.729820013046265,
      "epoch": 8.115021770682148,
      "grad_norm": 1.6484375,
      "learning_rate": 8.896982086155602e-05,
      "loss": 4.8138,
      "mean_token_accuracy": 0.2338313102722168,
      "num_tokens": 185640640.0,
      "step": 89460
    },
    {
      "entropy": 5.715052223205566,
      "epoch": 8.115475326560233,
      "grad_norm": 1.4921875,
      "learning_rate": 8.895162246605866e-05,
      "loss": 4.8396,
      "mean_token_accuracy": 0.23480065762996674,
      "num_tokens": 185651741.0,
      "step": 89465
    },
    {
      "entropy": 5.686949825286865,
      "epoch": 8.115928882438316,
      "grad_norm": 1.5078125,
      "learning_rate": 8.89334279180892e-05,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23417022824287415,
      "num_tokens": 185662335.0,
      "step": 89470
    },
    {
      "entropy": 5.66838583946228,
      "epoch": 8.1163824383164,
      "grad_norm": 1.6171875,
      "learning_rate": 8.891523721802387e-05,
      "loss": 4.7939,
      "mean_token_accuracy": 0.23998498171567917,
      "num_tokens": 185672711.0,
      "step": 89475
    },
    {
      "entropy": 5.73431806564331,
      "epoch": 8.116835994194485,
      "grad_norm": 1.53125,
      "learning_rate": 8.889705036623883e-05,
      "loss": 4.8297,
      "mean_token_accuracy": 0.23357219249010086,
      "num_tokens": 185682352.0,
      "step": 89480
    },
    {
      "entropy": 5.709272146224976,
      "epoch": 8.117289550072568,
      "grad_norm": 1.609375,
      "learning_rate": 8.887886736311023e-05,
      "loss": 4.8232,
      "mean_token_accuracy": 0.23515274077653886,
      "num_tokens": 185691700.0,
      "step": 89485
    },
    {
      "entropy": 5.6883868217468265,
      "epoch": 8.117743105950654,
      "grad_norm": 1.546875,
      "learning_rate": 8.88606882090141e-05,
      "loss": 4.8248,
      "mean_token_accuracy": 0.23279266506433488,
      "num_tokens": 185701412.0,
      "step": 89490
    },
    {
      "entropy": 5.719750118255615,
      "epoch": 8.118196661828737,
      "grad_norm": 1.34375,
      "learning_rate": 8.884251290432635e-05,
      "loss": 4.8234,
      "mean_token_accuracy": 0.24134237170219422,
      "num_tokens": 185711568.0,
      "step": 89495
    },
    {
      "entropy": 5.65893588066101,
      "epoch": 8.118650217706822,
      "grad_norm": 1.4765625,
      "learning_rate": 8.88243414494228e-05,
      "loss": 4.7522,
      "mean_token_accuracy": 0.23870896846055983,
      "num_tokens": 185722015.0,
      "step": 89500
    },
    {
      "entropy": 5.705086755752563,
      "epoch": 8.119103773584905,
      "grad_norm": 1.5078125,
      "learning_rate": 8.880617384467942e-05,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23918445855379106,
      "num_tokens": 185732671.0,
      "step": 89505
    },
    {
      "entropy": 5.696166753768921,
      "epoch": 8.11955732946299,
      "grad_norm": 1.6015625,
      "learning_rate": 8.878801009047177e-05,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23964374214410783,
      "num_tokens": 185742939.0,
      "step": 89510
    },
    {
      "entropy": 5.735066509246826,
      "epoch": 8.120010885341074,
      "grad_norm": 1.5234375,
      "learning_rate": 8.876985018717553e-05,
      "loss": 4.8374,
      "mean_token_accuracy": 0.24136148244142533,
      "num_tokens": 185753734.0,
      "step": 89515
    },
    {
      "entropy": 5.729222822189331,
      "epoch": 8.120464441219157,
      "grad_norm": 1.46875,
      "learning_rate": 8.875169413516625e-05,
      "loss": 4.8703,
      "mean_token_accuracy": 0.23022842705249785,
      "num_tokens": 185764860.0,
      "step": 89520
    },
    {
      "entropy": 5.65214991569519,
      "epoch": 8.120917997097242,
      "grad_norm": 1.484375,
      "learning_rate": 8.87335419348194e-05,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2312069609761238,
      "num_tokens": 185775608.0,
      "step": 89525
    },
    {
      "entropy": 5.6789531230926515,
      "epoch": 8.121371552975326,
      "grad_norm": 1.6484375,
      "learning_rate": 8.871539358651037e-05,
      "loss": 4.707,
      "mean_token_accuracy": 0.23685746937990187,
      "num_tokens": 185785669.0,
      "step": 89530
    },
    {
      "entropy": 5.6106141090393065,
      "epoch": 8.121825108853411,
      "grad_norm": 1.5234375,
      "learning_rate": 8.869724909061441e-05,
      "loss": 4.7199,
      "mean_token_accuracy": 0.24623219072818756,
      "num_tokens": 185795280.0,
      "step": 89535
    },
    {
      "entropy": 5.697428464889526,
      "epoch": 8.122278664731494,
      "grad_norm": 1.5,
      "learning_rate": 8.867910844750687e-05,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23055831640958785,
      "num_tokens": 185804678.0,
      "step": 89540
    },
    {
      "entropy": 5.717268371582032,
      "epoch": 8.12273222060958,
      "grad_norm": 1.4921875,
      "learning_rate": 8.866097165756288e-05,
      "loss": 4.7989,
      "mean_token_accuracy": 0.2345537781715393,
      "num_tokens": 185813932.0,
      "step": 89545
    },
    {
      "entropy": 5.692655086517334,
      "epoch": 8.123185776487663,
      "grad_norm": 1.5546875,
      "learning_rate": 8.864283872115741e-05,
      "loss": 4.7549,
      "mean_token_accuracy": 0.23273192644119262,
      "num_tokens": 185823817.0,
      "step": 89550
    },
    {
      "entropy": 5.591111660003662,
      "epoch": 8.123639332365748,
      "grad_norm": 1.4921875,
      "learning_rate": 8.862470963866556e-05,
      "loss": 4.715,
      "mean_token_accuracy": 0.24355630129575728,
      "num_tokens": 185833984.0,
      "step": 89555
    },
    {
      "entropy": 5.620126056671142,
      "epoch": 8.124092888243831,
      "grad_norm": 1.6171875,
      "learning_rate": 8.860658441046223e-05,
      "loss": 4.7348,
      "mean_token_accuracy": 0.2356054052710533,
      "num_tokens": 185843519.0,
      "step": 89560
    },
    {
      "entropy": 5.68628625869751,
      "epoch": 8.124546444121917,
      "grad_norm": 1.53125,
      "learning_rate": 8.858846303692221e-05,
      "loss": 4.7981,
      "mean_token_accuracy": 0.23522736728191376,
      "num_tokens": 185854003.0,
      "step": 89565
    },
    {
      "entropy": 5.644939136505127,
      "epoch": 8.125,
      "grad_norm": 1.5234375,
      "learning_rate": 8.857034551842025e-05,
      "loss": 4.7422,
      "mean_token_accuracy": 0.23392013907432557,
      "num_tokens": 185864121.0,
      "step": 89570
    },
    {
      "entropy": 5.654284715652466,
      "epoch": 8.125453555878083,
      "grad_norm": 1.3515625,
      "learning_rate": 8.855223185533108e-05,
      "loss": 4.7574,
      "mean_token_accuracy": 0.23631961792707443,
      "num_tokens": 185875075.0,
      "step": 89575
    },
    {
      "entropy": 5.662505865097046,
      "epoch": 8.125907111756169,
      "grad_norm": 1.5078125,
      "learning_rate": 8.853412204802927e-05,
      "loss": 4.7724,
      "mean_token_accuracy": 0.23804159313440323,
      "num_tokens": 185885037.0,
      "step": 89580
    },
    {
      "entropy": 5.689383125305175,
      "epoch": 8.126360667634252,
      "grad_norm": 1.4921875,
      "learning_rate": 8.851601609688933e-05,
      "loss": 4.8485,
      "mean_token_accuracy": 0.2317654386162758,
      "num_tokens": 185896097.0,
      "step": 89585
    },
    {
      "entropy": 5.693864345550537,
      "epoch": 8.126814223512337,
      "grad_norm": 1.7734375,
      "learning_rate": 8.849791400228571e-05,
      "loss": 4.8243,
      "mean_token_accuracy": 0.24052711129188536,
      "num_tokens": 185905840.0,
      "step": 89590
    },
    {
      "entropy": 5.690061378479004,
      "epoch": 8.12726777939042,
      "grad_norm": 1.4765625,
      "learning_rate": 8.847981576459278e-05,
      "loss": 4.8285,
      "mean_token_accuracy": 0.23668936342000962,
      "num_tokens": 185916579.0,
      "step": 89595
    },
    {
      "entropy": 5.689873552322387,
      "epoch": 8.127721335268506,
      "grad_norm": 1.4765625,
      "learning_rate": 8.846172138418468e-05,
      "loss": 4.836,
      "mean_token_accuracy": 0.22955555468797684,
      "num_tokens": 185926048.0,
      "step": 89600
    },
    {
      "entropy": 5.711649465560913,
      "epoch": 8.128174891146589,
      "grad_norm": 1.5546875,
      "learning_rate": 8.84436308614358e-05,
      "loss": 4.7744,
      "mean_token_accuracy": 0.24577468931674956,
      "num_tokens": 185936385.0,
      "step": 89605
    },
    {
      "entropy": 5.685956144332886,
      "epoch": 8.128628447024674,
      "grad_norm": 1.4609375,
      "learning_rate": 8.842554419672017e-05,
      "loss": 4.8244,
      "mean_token_accuracy": 0.23823851495981216,
      "num_tokens": 185946194.0,
      "step": 89610
    },
    {
      "entropy": 5.686057662963867,
      "epoch": 8.129082002902758,
      "grad_norm": 1.7578125,
      "learning_rate": 8.840746139041183e-05,
      "loss": 4.793,
      "mean_token_accuracy": 0.2318787917494774,
      "num_tokens": 185955647.0,
      "step": 89615
    },
    {
      "entropy": 5.712392044067383,
      "epoch": 8.129535558780843,
      "grad_norm": 1.421875,
      "learning_rate": 8.838938244288476e-05,
      "loss": 4.7882,
      "mean_token_accuracy": 0.23531757444143295,
      "num_tokens": 185966108.0,
      "step": 89620
    },
    {
      "entropy": 5.68887734413147,
      "epoch": 8.129989114658926,
      "grad_norm": 1.421875,
      "learning_rate": 8.83713073545128e-05,
      "loss": 4.8068,
      "mean_token_accuracy": 0.2356111600995064,
      "num_tokens": 185975736.0,
      "step": 89625
    },
    {
      "entropy": 5.736539363861084,
      "epoch": 8.13044267053701,
      "grad_norm": 1.546875,
      "learning_rate": 8.835323612566976e-05,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23708042949438096,
      "num_tokens": 185985233.0,
      "step": 89630
    },
    {
      "entropy": 5.650072717666626,
      "epoch": 8.130896226415095,
      "grad_norm": 1.453125,
      "learning_rate": 8.833516875672931e-05,
      "loss": 4.6777,
      "mean_token_accuracy": 0.2518389344215393,
      "num_tokens": 185995938.0,
      "step": 89635
    },
    {
      "entropy": 5.6717212200164795,
      "epoch": 8.131349782293178,
      "grad_norm": 1.515625,
      "learning_rate": 8.831710524806519e-05,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23518614172935487,
      "num_tokens": 186006651.0,
      "step": 89640
    },
    {
      "entropy": 5.6198876857757565,
      "epoch": 8.131803338171263,
      "grad_norm": 1.4609375,
      "learning_rate": 8.829904560005089e-05,
      "loss": 4.75,
      "mean_token_accuracy": 0.2408038228750229,
      "num_tokens": 186017933.0,
      "step": 89645
    },
    {
      "entropy": 5.699162483215332,
      "epoch": 8.132256894049346,
      "grad_norm": 1.578125,
      "learning_rate": 8.828098981305993e-05,
      "loss": 4.8236,
      "mean_token_accuracy": 0.22942500561475754,
      "num_tokens": 186028796.0,
      "step": 89650
    },
    {
      "entropy": 5.70589599609375,
      "epoch": 8.132710449927432,
      "grad_norm": 1.421875,
      "learning_rate": 8.826293788746566e-05,
      "loss": 4.863,
      "mean_token_accuracy": 0.2356010541319847,
      "num_tokens": 186039967.0,
      "step": 89655
    },
    {
      "entropy": 5.662939643859863,
      "epoch": 8.133164005805515,
      "grad_norm": 1.5546875,
      "learning_rate": 8.824488982364145e-05,
      "loss": 4.7337,
      "mean_token_accuracy": 0.24009094536304473,
      "num_tokens": 186050827.0,
      "step": 89660
    },
    {
      "entropy": 5.730684566497803,
      "epoch": 8.1336175616836,
      "grad_norm": 1.40625,
      "learning_rate": 8.822684562196048e-05,
      "loss": 4.8358,
      "mean_token_accuracy": 0.23083646595478058,
      "num_tokens": 186062158.0,
      "step": 89665
    },
    {
      "entropy": 5.720304870605469,
      "epoch": 8.134071117561684,
      "grad_norm": 1.4765625,
      "learning_rate": 8.820880528279588e-05,
      "loss": 4.7808,
      "mean_token_accuracy": 0.23295773863792418,
      "num_tokens": 186072451.0,
      "step": 89670
    },
    {
      "entropy": 5.745318603515625,
      "epoch": 8.134524673439767,
      "grad_norm": 1.484375,
      "learning_rate": 8.819076880652087e-05,
      "loss": 4.8426,
      "mean_token_accuracy": 0.23837959468364717,
      "num_tokens": 186083005.0,
      "step": 89675
    },
    {
      "entropy": 5.755396032333374,
      "epoch": 8.134978229317852,
      "grad_norm": 1.4296875,
      "learning_rate": 8.817273619350834e-05,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2281452640891075,
      "num_tokens": 186093938.0,
      "step": 89680
    },
    {
      "entropy": 5.689357757568359,
      "epoch": 8.135431785195935,
      "grad_norm": 1.4765625,
      "learning_rate": 8.815470744413125e-05,
      "loss": 4.762,
      "mean_token_accuracy": 0.24225643873214722,
      "num_tokens": 186104116.0,
      "step": 89685
    },
    {
      "entropy": 5.649821710586548,
      "epoch": 8.13588534107402,
      "grad_norm": 1.4375,
      "learning_rate": 8.813668255876242e-05,
      "loss": 4.6999,
      "mean_token_accuracy": 0.2388089582324028,
      "num_tokens": 186113793.0,
      "step": 89690
    },
    {
      "entropy": 5.718012094497681,
      "epoch": 8.136338896952104,
      "grad_norm": 1.4765625,
      "learning_rate": 8.811866153777463e-05,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2343737930059433,
      "num_tokens": 186124486.0,
      "step": 89695
    },
    {
      "entropy": 5.6485595703125,
      "epoch": 8.13679245283019,
      "grad_norm": 1.484375,
      "learning_rate": 8.81006443815405e-05,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24158625900745392,
      "num_tokens": 186135140.0,
      "step": 89700
    },
    {
      "entropy": 5.622825860977173,
      "epoch": 8.137246008708273,
      "grad_norm": 1.5859375,
      "learning_rate": 8.808263109043265e-05,
      "loss": 4.7831,
      "mean_token_accuracy": 0.24242624789476394,
      "num_tokens": 186145402.0,
      "step": 89705
    },
    {
      "entropy": 5.598331069946289,
      "epoch": 8.137699564586358,
      "grad_norm": 1.734375,
      "learning_rate": 8.806462166482373e-05,
      "loss": 4.6595,
      "mean_token_accuracy": 0.24944123923778533,
      "num_tokens": 186155264.0,
      "step": 89710
    },
    {
      "entropy": 5.675222873687744,
      "epoch": 8.138153120464441,
      "grad_norm": 1.59375,
      "learning_rate": 8.804661610508602e-05,
      "loss": 4.7554,
      "mean_token_accuracy": 0.2394178882241249,
      "num_tokens": 186165190.0,
      "step": 89715
    },
    {
      "entropy": 5.689668893814087,
      "epoch": 8.138606676342526,
      "grad_norm": 1.65625,
      "learning_rate": 8.802861441159191e-05,
      "loss": 4.8053,
      "mean_token_accuracy": 0.23495351672172546,
      "num_tokens": 186174938.0,
      "step": 89720
    },
    {
      "entropy": 5.7267961502075195,
      "epoch": 8.13906023222061,
      "grad_norm": 1.5546875,
      "learning_rate": 8.801061658471374e-05,
      "loss": 4.8474,
      "mean_token_accuracy": 0.2335069075226784,
      "num_tokens": 186185613.0,
      "step": 89725
    },
    {
      "entropy": 5.666635990142822,
      "epoch": 8.139513788098693,
      "grad_norm": 1.515625,
      "learning_rate": 8.799262262482357e-05,
      "loss": 4.7341,
      "mean_token_accuracy": 0.24226075857877732,
      "num_tokens": 186195525.0,
      "step": 89730
    },
    {
      "entropy": 5.625405740737915,
      "epoch": 8.139967343976778,
      "grad_norm": 1.3828125,
      "learning_rate": 8.797463253229368e-05,
      "loss": 4.7561,
      "mean_token_accuracy": 0.24490258544683458,
      "num_tokens": 186206594.0,
      "step": 89735
    },
    {
      "entropy": 5.694393539428711,
      "epoch": 8.140420899854862,
      "grad_norm": 1.484375,
      "learning_rate": 8.795664630749606e-05,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2410702332854271,
      "num_tokens": 186216640.0,
      "step": 89740
    },
    {
      "entropy": 5.74392786026001,
      "epoch": 8.140874455732947,
      "grad_norm": 1.4609375,
      "learning_rate": 8.793866395080266e-05,
      "loss": 4.8939,
      "mean_token_accuracy": 0.23235346525907516,
      "num_tokens": 186227573.0,
      "step": 89745
    },
    {
      "entropy": 5.7161647319793705,
      "epoch": 8.14132801161103,
      "grad_norm": 1.7890625,
      "learning_rate": 8.792068546258535e-05,
      "loss": 4.7795,
      "mean_token_accuracy": 0.24592860639095307,
      "num_tokens": 186237520.0,
      "step": 89750
    },
    {
      "entropy": 5.669751739501953,
      "epoch": 8.141781567489115,
      "grad_norm": 1.4296875,
      "learning_rate": 8.790271084321594e-05,
      "loss": 4.7727,
      "mean_token_accuracy": 0.24219569116830825,
      "num_tokens": 186247667.0,
      "step": 89755
    },
    {
      "entropy": 5.640280771255493,
      "epoch": 8.142235123367199,
      "grad_norm": 1.625,
      "learning_rate": 8.788474009306615e-05,
      "loss": 4.7638,
      "mean_token_accuracy": 0.24317547529935837,
      "num_tokens": 186258109.0,
      "step": 89760
    },
    {
      "entropy": 5.66305251121521,
      "epoch": 8.142688679245284,
      "grad_norm": 1.4296875,
      "learning_rate": 8.786677321250757e-05,
      "loss": 4.79,
      "mean_token_accuracy": 0.23222029358148574,
      "num_tokens": 186268510.0,
      "step": 89765
    },
    {
      "entropy": 5.7391290187835695,
      "epoch": 8.143142235123367,
      "grad_norm": 1.71875,
      "learning_rate": 8.784881020191186e-05,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22987106442451477,
      "num_tokens": 186279575.0,
      "step": 89770
    },
    {
      "entropy": 5.6321735858917235,
      "epoch": 8.143595791001452,
      "grad_norm": 1.625,
      "learning_rate": 8.783085106165044e-05,
      "loss": 4.6902,
      "mean_token_accuracy": 0.24514278173446655,
      "num_tokens": 186289521.0,
      "step": 89775
    },
    {
      "entropy": 5.676647758483886,
      "epoch": 8.144049346879536,
      "grad_norm": 1.3828125,
      "learning_rate": 8.781289579209472e-05,
      "loss": 4.7351,
      "mean_token_accuracy": 0.24024679511785507,
      "num_tokens": 186300146.0,
      "step": 89780
    },
    {
      "entropy": 5.682681083679199,
      "epoch": 8.144502902757619,
      "grad_norm": 1.3828125,
      "learning_rate": 8.779494439361601e-05,
      "loss": 4.751,
      "mean_token_accuracy": 0.24325695633888245,
      "num_tokens": 186310656.0,
      "step": 89785
    },
    {
      "entropy": 5.734512424468994,
      "epoch": 8.144956458635704,
      "grad_norm": 1.46875,
      "learning_rate": 8.777699686658556e-05,
      "loss": 4.8887,
      "mean_token_accuracy": 0.2335243418812752,
      "num_tokens": 186321916.0,
      "step": 89790
    },
    {
      "entropy": 5.681565141677856,
      "epoch": 8.145410014513788,
      "grad_norm": 1.4765625,
      "learning_rate": 8.775905321137454e-05,
      "loss": 4.77,
      "mean_token_accuracy": 0.2390258401632309,
      "num_tokens": 186331897.0,
      "step": 89795
    },
    {
      "entropy": 5.69018521308899,
      "epoch": 8.145863570391873,
      "grad_norm": 1.4921875,
      "learning_rate": 8.774111342835395e-05,
      "loss": 4.7299,
      "mean_token_accuracy": 0.23763490915298463,
      "num_tokens": 186341466.0,
      "step": 89800
    },
    {
      "entropy": 5.550119781494141,
      "epoch": 8.146317126269956,
      "grad_norm": 1.578125,
      "learning_rate": 8.77231775178949e-05,
      "loss": 4.6764,
      "mean_token_accuracy": 0.2449374705553055,
      "num_tokens": 186351014.0,
      "step": 89805
    },
    {
      "entropy": 5.66903076171875,
      "epoch": 8.146770682148041,
      "grad_norm": 1.5,
      "learning_rate": 8.770524548036825e-05,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2289290636777878,
      "num_tokens": 186361484.0,
      "step": 89810
    },
    {
      "entropy": 5.69759840965271,
      "epoch": 8.147224238026125,
      "grad_norm": 1.6484375,
      "learning_rate": 8.768731731614488e-05,
      "loss": 4.8581,
      "mean_token_accuracy": 0.23467255085706712,
      "num_tokens": 186371584.0,
      "step": 89815
    },
    {
      "entropy": 5.715124034881592,
      "epoch": 8.14767779390421,
      "grad_norm": 1.515625,
      "learning_rate": 8.76693930255955e-05,
      "loss": 4.884,
      "mean_token_accuracy": 0.23042188286781312,
      "num_tokens": 186383082.0,
      "step": 89820
    },
    {
      "entropy": 5.714970827102661,
      "epoch": 8.148131349782293,
      "grad_norm": 1.5,
      "learning_rate": 8.765147260909081e-05,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23879874646663665,
      "num_tokens": 186393795.0,
      "step": 89825
    },
    {
      "entropy": 5.691426992416382,
      "epoch": 8.148584905660377,
      "grad_norm": 1.4296875,
      "learning_rate": 8.763355606700141e-05,
      "loss": 4.7637,
      "mean_token_accuracy": 0.2455595016479492,
      "num_tokens": 186403546.0,
      "step": 89830
    },
    {
      "entropy": 5.688782596588135,
      "epoch": 8.149038461538462,
      "grad_norm": 1.546875,
      "learning_rate": 8.76156433996978e-05,
      "loss": 4.735,
      "mean_token_accuracy": 0.23904742151498795,
      "num_tokens": 186413972.0,
      "step": 89835
    },
    {
      "entropy": 5.710566616058349,
      "epoch": 8.149492017416545,
      "grad_norm": 1.59375,
      "learning_rate": 8.759773460755044e-05,
      "loss": 4.7997,
      "mean_token_accuracy": 0.2380626156926155,
      "num_tokens": 186425189.0,
      "step": 89840
    },
    {
      "entropy": 5.739260053634643,
      "epoch": 8.14994557329463,
      "grad_norm": 1.5546875,
      "learning_rate": 8.75798296909297e-05,
      "loss": 4.8754,
      "mean_token_accuracy": 0.23716063797473907,
      "num_tokens": 186435726.0,
      "step": 89845
    },
    {
      "entropy": 5.711713361740112,
      "epoch": 8.150399129172714,
      "grad_norm": 1.609375,
      "learning_rate": 8.756192865020584e-05,
      "loss": 4.8047,
      "mean_token_accuracy": 0.24134715050458908,
      "num_tokens": 186445665.0,
      "step": 89850
    },
    {
      "entropy": 5.741569662094117,
      "epoch": 8.150852685050799,
      "grad_norm": 1.5078125,
      "learning_rate": 8.754403148574911e-05,
      "loss": 4.8751,
      "mean_token_accuracy": 0.23268360495567322,
      "num_tokens": 186456217.0,
      "step": 89855
    },
    {
      "entropy": 5.558558940887451,
      "epoch": 8.151306240928882,
      "grad_norm": 1.515625,
      "learning_rate": 8.752613819792952e-05,
      "loss": 4.6473,
      "mean_token_accuracy": 0.25434180945158,
      "num_tokens": 186466499.0,
      "step": 89860
    },
    {
      "entropy": 5.769989490509033,
      "epoch": 8.151759796806967,
      "grad_norm": 1.4296875,
      "learning_rate": 8.750824878711716e-05,
      "loss": 4.9285,
      "mean_token_accuracy": 0.23221694082021713,
      "num_tokens": 186477386.0,
      "step": 89865
    },
    {
      "entropy": 5.688897180557251,
      "epoch": 8.15221335268505,
      "grad_norm": 1.6171875,
      "learning_rate": 8.749036325368197e-05,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23738688081502915,
      "num_tokens": 186487256.0,
      "step": 89870
    },
    {
      "entropy": 5.741700267791748,
      "epoch": 8.152666908563136,
      "grad_norm": 1.4140625,
      "learning_rate": 8.74724815979939e-05,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23063725084066392,
      "num_tokens": 186497497.0,
      "step": 89875
    },
    {
      "entropy": 5.681329298019409,
      "epoch": 8.15312046444122,
      "grad_norm": 1.5703125,
      "learning_rate": 8.745460382042268e-05,
      "loss": 4.7369,
      "mean_token_accuracy": 0.24451829940080644,
      "num_tokens": 186507142.0,
      "step": 89880
    },
    {
      "entropy": 5.677608060836792,
      "epoch": 8.153574020319303,
      "grad_norm": 1.5703125,
      "learning_rate": 8.743672992133809e-05,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23875257670879363,
      "num_tokens": 186517367.0,
      "step": 89885
    },
    {
      "entropy": 5.685663557052612,
      "epoch": 8.154027576197388,
      "grad_norm": 1.3828125,
      "learning_rate": 8.741885990110968e-05,
      "loss": 4.8256,
      "mean_token_accuracy": 0.23723581582307815,
      "num_tokens": 186528556.0,
      "step": 89890
    },
    {
      "entropy": 5.750435829162598,
      "epoch": 8.154481132075471,
      "grad_norm": 1.484375,
      "learning_rate": 8.740099376010705e-05,
      "loss": 4.8527,
      "mean_token_accuracy": 0.23744388967752456,
      "num_tokens": 186539136.0,
      "step": 89895
    },
    {
      "entropy": 5.731344366073609,
      "epoch": 8.154934687953556,
      "grad_norm": 1.515625,
      "learning_rate": 8.738313149869964e-05,
      "loss": 4.814,
      "mean_token_accuracy": 0.23516904562711716,
      "num_tokens": 186548801.0,
      "step": 89900
    },
    {
      "entropy": 5.808780002593994,
      "epoch": 8.15538824383164,
      "grad_norm": 1.6640625,
      "learning_rate": 8.736527311725692e-05,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22750266045331954,
      "num_tokens": 186560702.0,
      "step": 89905
    },
    {
      "entropy": 5.7102607727050785,
      "epoch": 8.155841799709725,
      "grad_norm": 1.359375,
      "learning_rate": 8.734741861614818e-05,
      "loss": 4.8,
      "mean_token_accuracy": 0.2344384014606476,
      "num_tokens": 186571679.0,
      "step": 89910
    },
    {
      "entropy": 5.68328742980957,
      "epoch": 8.156295355587808,
      "grad_norm": 1.6171875,
      "learning_rate": 8.732956799574262e-05,
      "loss": 4.7497,
      "mean_token_accuracy": 0.24353578686714172,
      "num_tokens": 186581032.0,
      "step": 89915
    },
    {
      "entropy": 5.703147268295288,
      "epoch": 8.156748911465893,
      "grad_norm": 1.53125,
      "learning_rate": 8.731172125640943e-05,
      "loss": 4.766,
      "mean_token_accuracy": 0.24492934197187424,
      "num_tokens": 186591077.0,
      "step": 89920
    },
    {
      "entropy": 5.661616468429566,
      "epoch": 8.157202467343977,
      "grad_norm": 1.4765625,
      "learning_rate": 8.729387839851768e-05,
      "loss": 4.7613,
      "mean_token_accuracy": 0.23493383675813675,
      "num_tokens": 186602040.0,
      "step": 89925
    },
    {
      "entropy": 5.73028531074524,
      "epoch": 8.15765602322206,
      "grad_norm": 1.4296875,
      "learning_rate": 8.727603942243637e-05,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23564544767141343,
      "num_tokens": 186614295.0,
      "step": 89930
    },
    {
      "entropy": 5.736703634262085,
      "epoch": 8.158109579100145,
      "grad_norm": 1.578125,
      "learning_rate": 8.725820432853431e-05,
      "loss": 4.838,
      "mean_token_accuracy": 0.2336052030324936,
      "num_tokens": 186624299.0,
      "step": 89935
    },
    {
      "entropy": 5.62271089553833,
      "epoch": 8.158563134978229,
      "grad_norm": 1.53125,
      "learning_rate": 8.724037311718053e-05,
      "loss": 4.7495,
      "mean_token_accuracy": 0.24129231423139572,
      "num_tokens": 186634351.0,
      "step": 89940
    },
    {
      "entropy": 5.697498416900634,
      "epoch": 8.159016690856314,
      "grad_norm": 1.640625,
      "learning_rate": 8.722254578874364e-05,
      "loss": 4.7763,
      "mean_token_accuracy": 0.2447281777858734,
      "num_tokens": 186644656.0,
      "step": 89945
    },
    {
      "entropy": 5.661510753631592,
      "epoch": 8.159470246734397,
      "grad_norm": 1.5546875,
      "learning_rate": 8.720472234359236e-05,
      "loss": 4.7869,
      "mean_token_accuracy": 0.232794126868248,
      "num_tokens": 186654375.0,
      "step": 89950
    },
    {
      "entropy": 5.6936835765838625,
      "epoch": 8.159923802612482,
      "grad_norm": 1.5,
      "learning_rate": 8.718690278209528e-05,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24887388944625854,
      "num_tokens": 186665169.0,
      "step": 89955
    },
    {
      "entropy": 5.597626161575318,
      "epoch": 8.160377358490566,
      "grad_norm": 1.5234375,
      "learning_rate": 8.716908710462094e-05,
      "loss": 4.6615,
      "mean_token_accuracy": 0.24634479135274887,
      "num_tokens": 186675880.0,
      "step": 89960
    },
    {
      "entropy": 5.6912696838378904,
      "epoch": 8.16083091436865,
      "grad_norm": 1.4765625,
      "learning_rate": 8.715127531153773e-05,
      "loss": 4.7956,
      "mean_token_accuracy": 0.2373419627547264,
      "num_tokens": 186686157.0,
      "step": 89965
    },
    {
      "entropy": 5.665458393096924,
      "epoch": 8.161284470246734,
      "grad_norm": 1.5,
      "learning_rate": 8.713346740321396e-05,
      "loss": 4.7244,
      "mean_token_accuracy": 0.25081589370965957,
      "num_tokens": 186696250.0,
      "step": 89970
    },
    {
      "entropy": 5.673393583297729,
      "epoch": 8.16173802612482,
      "grad_norm": 1.5078125,
      "learning_rate": 8.7115663380018e-05,
      "loss": 4.7887,
      "mean_token_accuracy": 0.2426490604877472,
      "num_tokens": 186706616.0,
      "step": 89975
    },
    {
      "entropy": 5.698648452758789,
      "epoch": 8.162191582002903,
      "grad_norm": 1.46875,
      "learning_rate": 8.709786324231798e-05,
      "loss": 4.8473,
      "mean_token_accuracy": 0.23409517258405685,
      "num_tokens": 186717832.0,
      "step": 89980
    },
    {
      "entropy": 5.696640396118164,
      "epoch": 8.162645137880986,
      "grad_norm": 1.6015625,
      "learning_rate": 8.708006699048208e-05,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23183239251375198,
      "num_tokens": 186727550.0,
      "step": 89985
    },
    {
      "entropy": 5.691786766052246,
      "epoch": 8.163098693759071,
      "grad_norm": 1.5234375,
      "learning_rate": 8.706227462487823e-05,
      "loss": 4.7533,
      "mean_token_accuracy": 0.2353050783276558,
      "num_tokens": 186737636.0,
      "step": 89990
    },
    {
      "entropy": 5.74793062210083,
      "epoch": 8.163552249637155,
      "grad_norm": 1.4921875,
      "learning_rate": 8.704448614587448e-05,
      "loss": 4.8653,
      "mean_token_accuracy": 0.23664041459560395,
      "num_tokens": 186748206.0,
      "step": 89995
    },
    {
      "entropy": 5.6199202060699465,
      "epoch": 8.16400580551524,
      "grad_norm": 1.5234375,
      "learning_rate": 8.702670155383863e-05,
      "loss": 4.7585,
      "mean_token_accuracy": 0.24290578812360764,
      "num_tokens": 186759051.0,
      "step": 90000
    },
    {
      "epoch": 8.16400580551524,
      "eval_entropy": 5.400828419875236,
      "eval_loss": 4.912232398986816,
      "eval_mean_token_accuracy": 0.23810991887087465,
      "eval_num_tokens": 186759051.0,
      "eval_runtime": 20.6529,
      "eval_samples_per_second": 1712.111,
      "eval_steps_per_second": 214.014,
      "step": 90000
    },
    {
      "entropy": 5.710803127288818,
      "epoch": 8.164459361393323,
      "grad_norm": 1.6875,
      "learning_rate": 8.700892084913847e-05,
      "loss": 4.8544,
      "mean_token_accuracy": 0.2290244996547699,
      "num_tokens": 186769300.0,
      "step": 90005
    },
    {
      "entropy": 5.7644270896911625,
      "epoch": 8.164912917271408,
      "grad_norm": 1.4765625,
      "learning_rate": 8.699114403214176e-05,
      "loss": 4.9611,
      "mean_token_accuracy": 0.22344272285699845,
      "num_tokens": 186779131.0,
      "step": 90010
    },
    {
      "entropy": 5.7453779697418215,
      "epoch": 8.165366473149492,
      "grad_norm": 1.484375,
      "learning_rate": 8.697337110321609e-05,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22605712562799454,
      "num_tokens": 186789557.0,
      "step": 90015
    },
    {
      "entropy": 5.701339054107666,
      "epoch": 8.165820029027577,
      "grad_norm": 1.4140625,
      "learning_rate": 8.695560206272901e-05,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2336163893342018,
      "num_tokens": 186800802.0,
      "step": 90020
    },
    {
      "entropy": 5.725722074508667,
      "epoch": 8.16627358490566,
      "grad_norm": 1.4609375,
      "learning_rate": 8.693783691104803e-05,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23080952763557433,
      "num_tokens": 186812130.0,
      "step": 90025
    },
    {
      "entropy": 5.670778226852417,
      "epoch": 8.166727140783745,
      "grad_norm": 1.609375,
      "learning_rate": 8.692007564854053e-05,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2346806839108467,
      "num_tokens": 186822835.0,
      "step": 90030
    },
    {
      "entropy": 5.684840059280395,
      "epoch": 8.167180696661829,
      "grad_norm": 1.5390625,
      "learning_rate": 8.690231827557373e-05,
      "loss": 4.7634,
      "mean_token_accuracy": 0.23412077128887177,
      "num_tokens": 186833012.0,
      "step": 90035
    },
    {
      "entropy": 5.7121399402618405,
      "epoch": 8.167634252539912,
      "grad_norm": 1.6171875,
      "learning_rate": 8.688456479251491e-05,
      "loss": 4.7669,
      "mean_token_accuracy": 0.24481943547725676,
      "num_tokens": 186843273.0,
      "step": 90040
    },
    {
      "entropy": 5.688477993011475,
      "epoch": 8.168087808417997,
      "grad_norm": 1.4921875,
      "learning_rate": 8.686681519973128e-05,
      "loss": 4.7977,
      "mean_token_accuracy": 0.23650419861078262,
      "num_tokens": 186854982.0,
      "step": 90045
    },
    {
      "entropy": 5.720480966567993,
      "epoch": 8.16854136429608,
      "grad_norm": 1.40625,
      "learning_rate": 8.684906949758983e-05,
      "loss": 4.8172,
      "mean_token_accuracy": 0.2356798380613327,
      "num_tokens": 186865350.0,
      "step": 90050
    },
    {
      "entropy": 5.713089799880981,
      "epoch": 8.168994920174166,
      "grad_norm": 1.5234375,
      "learning_rate": 8.68313276864576e-05,
      "loss": 4.8576,
      "mean_token_accuracy": 0.23731960952281952,
      "num_tokens": 186875693.0,
      "step": 90055
    },
    {
      "entropy": 5.683885812759399,
      "epoch": 8.16944847605225,
      "grad_norm": 1.5703125,
      "learning_rate": 8.681358976670146e-05,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2437424287199974,
      "num_tokens": 186885780.0,
      "step": 90060
    },
    {
      "entropy": 5.743968057632446,
      "epoch": 8.169902031930334,
      "grad_norm": 1.6328125,
      "learning_rate": 8.67958557386882e-05,
      "loss": 4.8299,
      "mean_token_accuracy": 0.23632963746786118,
      "num_tokens": 186895260.0,
      "step": 90065
    },
    {
      "entropy": 5.665383386611938,
      "epoch": 8.170355587808418,
      "grad_norm": 1.5625,
      "learning_rate": 8.677812560278467e-05,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24059765338897704,
      "num_tokens": 186905043.0,
      "step": 90070
    },
    {
      "entropy": 5.700939083099366,
      "epoch": 8.170809143686503,
      "grad_norm": 1.4765625,
      "learning_rate": 8.676039935935735e-05,
      "loss": 4.8015,
      "mean_token_accuracy": 0.24022364169359206,
      "num_tokens": 186915141.0,
      "step": 90075
    },
    {
      "entropy": 5.670888614654541,
      "epoch": 8.171262699564586,
      "grad_norm": 1.359375,
      "learning_rate": 8.674267700877303e-05,
      "loss": 4.7944,
      "mean_token_accuracy": 0.2412732869386673,
      "num_tokens": 186925246.0,
      "step": 90080
    },
    {
      "entropy": 5.706196069717407,
      "epoch": 8.17171625544267,
      "grad_norm": 1.5703125,
      "learning_rate": 8.672495855139812e-05,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2402112901210785,
      "num_tokens": 186936449.0,
      "step": 90085
    },
    {
      "entropy": 5.686678075790406,
      "epoch": 8.172169811320755,
      "grad_norm": 1.65625,
      "learning_rate": 8.670724398759903e-05,
      "loss": 4.7886,
      "mean_token_accuracy": 0.23209165185689926,
      "num_tokens": 186946529.0,
      "step": 90090
    },
    {
      "entropy": 5.633563613891601,
      "epoch": 8.172623367198838,
      "grad_norm": 1.4765625,
      "learning_rate": 8.668953331774212e-05,
      "loss": 4.718,
      "mean_token_accuracy": 0.24169537127017976,
      "num_tokens": 186957185.0,
      "step": 90095
    },
    {
      "entropy": 5.65764365196228,
      "epoch": 8.173076923076923,
      "grad_norm": 1.4453125,
      "learning_rate": 8.667182654219365e-05,
      "loss": 4.8163,
      "mean_token_accuracy": 0.2363406553864479,
      "num_tokens": 186967307.0,
      "step": 90100
    },
    {
      "entropy": 5.689286899566651,
      "epoch": 8.173530478955007,
      "grad_norm": 1.5390625,
      "learning_rate": 8.665412366131975e-05,
      "loss": 4.7725,
      "mean_token_accuracy": 0.24029868245124816,
      "num_tokens": 186976967.0,
      "step": 90105
    },
    {
      "entropy": 5.6288550853729244,
      "epoch": 8.173984034833092,
      "grad_norm": 1.625,
      "learning_rate": 8.663642467548663e-05,
      "loss": 4.7581,
      "mean_token_accuracy": 0.2440082013607025,
      "num_tokens": 186987320.0,
      "step": 90110
    },
    {
      "entropy": 5.668457126617431,
      "epoch": 8.174437590711175,
      "grad_norm": 1.5546875,
      "learning_rate": 8.661872958506023e-05,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23607144206762315,
      "num_tokens": 186998495.0,
      "step": 90115
    },
    {
      "entropy": 5.73978853225708,
      "epoch": 8.17489114658926,
      "grad_norm": 1.4921875,
      "learning_rate": 8.660103839040653e-05,
      "loss": 4.8516,
      "mean_token_accuracy": 0.2318258285522461,
      "num_tokens": 187009275.0,
      "step": 90120
    },
    {
      "entropy": 5.7566670894622805,
      "epoch": 8.175344702467344,
      "grad_norm": 1.59375,
      "learning_rate": 8.658335109189135e-05,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22871264219284057,
      "num_tokens": 187019231.0,
      "step": 90125
    },
    {
      "entropy": 5.709703588485718,
      "epoch": 8.175798258345429,
      "grad_norm": 1.5,
      "learning_rate": 8.656566768988047e-05,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23782145231962204,
      "num_tokens": 187028609.0,
      "step": 90130
    },
    {
      "entropy": 5.737650442123413,
      "epoch": 8.176251814223512,
      "grad_norm": 1.546875,
      "learning_rate": 8.654798818473964e-05,
      "loss": 4.8807,
      "mean_token_accuracy": 0.2301066905260086,
      "num_tokens": 187037495.0,
      "step": 90135
    },
    {
      "entropy": 5.731504011154175,
      "epoch": 8.176705370101596,
      "grad_norm": 1.4609375,
      "learning_rate": 8.653031257683441e-05,
      "loss": 4.7818,
      "mean_token_accuracy": 0.2416789099574089,
      "num_tokens": 187047709.0,
      "step": 90140
    },
    {
      "entropy": 5.68897852897644,
      "epoch": 8.17715892597968,
      "grad_norm": 1.609375,
      "learning_rate": 8.651264086653034e-05,
      "loss": 4.7806,
      "mean_token_accuracy": 0.23591960817575455,
      "num_tokens": 187058551.0,
      "step": 90145
    },
    {
      "entropy": 5.693097162246704,
      "epoch": 8.177612481857764,
      "grad_norm": 1.5703125,
      "learning_rate": 8.64949730541929e-05,
      "loss": 4.7876,
      "mean_token_accuracy": 0.23703640550374985,
      "num_tokens": 187069850.0,
      "step": 90150
    },
    {
      "entropy": 5.6398993015289305,
      "epoch": 8.17806603773585,
      "grad_norm": 1.5546875,
      "learning_rate": 8.647730914018747e-05,
      "loss": 4.7072,
      "mean_token_accuracy": 0.24271036237478255,
      "num_tokens": 187080028.0,
      "step": 90155
    },
    {
      "entropy": 5.746600341796875,
      "epoch": 8.178519593613933,
      "grad_norm": 1.6640625,
      "learning_rate": 8.64596491248793e-05,
      "loss": 4.9103,
      "mean_token_accuracy": 0.23247306048870087,
      "num_tokens": 187089930.0,
      "step": 90160
    },
    {
      "entropy": 5.736628437042237,
      "epoch": 8.178973149492018,
      "grad_norm": 1.5390625,
      "learning_rate": 8.644199300863365e-05,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22605329751968384,
      "num_tokens": 187100920.0,
      "step": 90165
    },
    {
      "entropy": 5.68293309211731,
      "epoch": 8.179426705370101,
      "grad_norm": 1.53125,
      "learning_rate": 8.642434079181556e-05,
      "loss": 4.7788,
      "mean_token_accuracy": 0.24437706023454667,
      "num_tokens": 187110454.0,
      "step": 90170
    },
    {
      "entropy": 5.678610992431641,
      "epoch": 8.179880261248186,
      "grad_norm": 1.4921875,
      "learning_rate": 8.640669247479025e-05,
      "loss": 4.728,
      "mean_token_accuracy": 0.24301980286836625,
      "num_tokens": 187120724.0,
      "step": 90175
    },
    {
      "entropy": 5.721657609939575,
      "epoch": 8.18033381712627,
      "grad_norm": 1.359375,
      "learning_rate": 8.638904805792255e-05,
      "loss": 4.8185,
      "mean_token_accuracy": 0.23371924012899398,
      "num_tokens": 187131081.0,
      "step": 90180
    },
    {
      "entropy": 5.788612318038941,
      "epoch": 8.180787373004355,
      "grad_norm": 1.5546875,
      "learning_rate": 8.637140754157743e-05,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22076697647571564,
      "num_tokens": 187141211.0,
      "step": 90185
    },
    {
      "entropy": 5.6934346675872805,
      "epoch": 8.181240928882438,
      "grad_norm": 1.53125,
      "learning_rate": 8.635377092611967e-05,
      "loss": 4.6916,
      "mean_token_accuracy": 0.25073114186525347,
      "num_tokens": 187151036.0,
      "step": 90190
    },
    {
      "entropy": 5.711217641830444,
      "epoch": 8.181694484760522,
      "grad_norm": 1.5390625,
      "learning_rate": 8.633613821191398e-05,
      "loss": 4.7729,
      "mean_token_accuracy": 0.23621845692396165,
      "num_tokens": 187161934.0,
      "step": 90195
    },
    {
      "entropy": 5.738018465042114,
      "epoch": 8.182148040638607,
      "grad_norm": 1.4609375,
      "learning_rate": 8.631850939932499e-05,
      "loss": 4.8753,
      "mean_token_accuracy": 0.2316300690174103,
      "num_tokens": 187172999.0,
      "step": 90200
    },
    {
      "entropy": 5.643386173248291,
      "epoch": 8.18260159651669,
      "grad_norm": 1.5390625,
      "learning_rate": 8.630088448871737e-05,
      "loss": 4.7607,
      "mean_token_accuracy": 0.23848002403974533,
      "num_tokens": 187183041.0,
      "step": 90205
    },
    {
      "entropy": 5.709938287734985,
      "epoch": 8.183055152394775,
      "grad_norm": 1.5859375,
      "learning_rate": 8.628326348045553e-05,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23239787369966508,
      "num_tokens": 187192428.0,
      "step": 90210
    },
    {
      "entropy": 5.7409533023834225,
      "epoch": 8.183508708272859,
      "grad_norm": 1.5390625,
      "learning_rate": 8.62656463749039e-05,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22995375394821166,
      "num_tokens": 187203217.0,
      "step": 90215
    },
    {
      "entropy": 5.643398141860962,
      "epoch": 8.183962264150944,
      "grad_norm": 1.375,
      "learning_rate": 8.624803317242679e-05,
      "loss": 4.7299,
      "mean_token_accuracy": 0.24493344128131866,
      "num_tokens": 187213589.0,
      "step": 90220
    },
    {
      "entropy": 5.691924333572388,
      "epoch": 8.184415820029027,
      "grad_norm": 1.4765625,
      "learning_rate": 8.623042387338846e-05,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2405783399939537,
      "num_tokens": 187223910.0,
      "step": 90225
    },
    {
      "entropy": 5.6422281742095945,
      "epoch": 8.184869375907112,
      "grad_norm": 1.6328125,
      "learning_rate": 8.621281847815307e-05,
      "loss": 4.7463,
      "mean_token_accuracy": 0.24079228490591048,
      "num_tokens": 187233678.0,
      "step": 90230
    },
    {
      "entropy": 5.663429832458496,
      "epoch": 8.185322931785196,
      "grad_norm": 1.5625,
      "learning_rate": 8.619521698708464e-05,
      "loss": 4.7699,
      "mean_token_accuracy": 0.24352909326553346,
      "num_tokens": 187243512.0,
      "step": 90235
    },
    {
      "entropy": 5.6230274677276615,
      "epoch": 8.18577648766328,
      "grad_norm": 1.4375,
      "learning_rate": 8.61776194005473e-05,
      "loss": 4.7114,
      "mean_token_accuracy": 0.2434825822710991,
      "num_tokens": 187253644.0,
      "step": 90240
    },
    {
      "entropy": 5.739083766937256,
      "epoch": 8.186230043541364,
      "grad_norm": 1.578125,
      "learning_rate": 8.616002571890491e-05,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2394650772213936,
      "num_tokens": 187263496.0,
      "step": 90245
    },
    {
      "entropy": 5.719364786148072,
      "epoch": 8.186683599419448,
      "grad_norm": 1.546875,
      "learning_rate": 8.61424359425213e-05,
      "loss": 4.8239,
      "mean_token_accuracy": 0.23557454496622085,
      "num_tokens": 187273935.0,
      "step": 90250
    },
    {
      "entropy": 5.666815996170044,
      "epoch": 8.187137155297533,
      "grad_norm": 1.4765625,
      "learning_rate": 8.612485007176027e-05,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23794151544570924,
      "num_tokens": 187283359.0,
      "step": 90255
    },
    {
      "entropy": 5.652565145492554,
      "epoch": 8.187590711175616,
      "grad_norm": 1.515625,
      "learning_rate": 8.610726810698543e-05,
      "loss": 4.7466,
      "mean_token_accuracy": 0.23259689956903457,
      "num_tokens": 187294258.0,
      "step": 90260
    },
    {
      "entropy": 5.726837587356568,
      "epoch": 8.188044267053701,
      "grad_norm": 1.5625,
      "learning_rate": 8.608969004856043e-05,
      "loss": 4.8547,
      "mean_token_accuracy": 0.23367569148540496,
      "num_tokens": 187304715.0,
      "step": 90265
    },
    {
      "entropy": 5.710773849487305,
      "epoch": 8.188497822931785,
      "grad_norm": 1.5,
      "learning_rate": 8.607211589684876e-05,
      "loss": 4.7848,
      "mean_token_accuracy": 0.23991395384073258,
      "num_tokens": 187316063.0,
      "step": 90270
    },
    {
      "entropy": 5.7391773700714115,
      "epoch": 8.18895137880987,
      "grad_norm": 1.4140625,
      "learning_rate": 8.605454565221388e-05,
      "loss": 4.743,
      "mean_token_accuracy": 0.2357721284031868,
      "num_tokens": 187325598.0,
      "step": 90275
    },
    {
      "entropy": 5.66893253326416,
      "epoch": 8.189404934687953,
      "grad_norm": 1.4375,
      "learning_rate": 8.603697931501918e-05,
      "loss": 4.7373,
      "mean_token_accuracy": 0.24935772269964218,
      "num_tokens": 187336285.0,
      "step": 90280
    },
    {
      "entropy": 5.735224056243896,
      "epoch": 8.189858490566039,
      "grad_norm": 1.4609375,
      "learning_rate": 8.601941688562791e-05,
      "loss": 4.8343,
      "mean_token_accuracy": 0.23889282941818238,
      "num_tokens": 187348583.0,
      "step": 90285
    },
    {
      "entropy": 5.698703479766846,
      "epoch": 8.190312046444122,
      "grad_norm": 1.4453125,
      "learning_rate": 8.600185836440322e-05,
      "loss": 4.7997,
      "mean_token_accuracy": 0.24076491594314575,
      "num_tokens": 187358150.0,
      "step": 90290
    },
    {
      "entropy": 5.652228593826294,
      "epoch": 8.190765602322205,
      "grad_norm": 1.4453125,
      "learning_rate": 8.59843037517083e-05,
      "loss": 4.6683,
      "mean_token_accuracy": 0.25299172550439836,
      "num_tokens": 187367590.0,
      "step": 90295
    },
    {
      "entropy": 5.6826667308807375,
      "epoch": 8.19121915820029,
      "grad_norm": 1.6484375,
      "learning_rate": 8.596675304790613e-05,
      "loss": 4.7978,
      "mean_token_accuracy": 0.24498320966959,
      "num_tokens": 187378149.0,
      "step": 90300
    },
    {
      "entropy": 5.655890035629272,
      "epoch": 8.191672714078374,
      "grad_norm": 1.796875,
      "learning_rate": 8.594920625335958e-05,
      "loss": 4.8154,
      "mean_token_accuracy": 0.22678477615118026,
      "num_tokens": 187387840.0,
      "step": 90305
    },
    {
      "entropy": 5.676466608047486,
      "epoch": 8.192126269956459,
      "grad_norm": 1.4609375,
      "learning_rate": 8.593166336843175e-05,
      "loss": 4.7998,
      "mean_token_accuracy": 0.2354747772216797,
      "num_tokens": 187398257.0,
      "step": 90310
    },
    {
      "entropy": 5.706955003738403,
      "epoch": 8.192579825834542,
      "grad_norm": 1.6640625,
      "learning_rate": 8.591412439348525e-05,
      "loss": 4.7565,
      "mean_token_accuracy": 0.24547193795442582,
      "num_tokens": 187407364.0,
      "step": 90315
    },
    {
      "entropy": 5.671437978744507,
      "epoch": 8.193033381712628,
      "grad_norm": 1.625,
      "learning_rate": 8.589658932888287e-05,
      "loss": 4.7903,
      "mean_token_accuracy": 0.23888723105192183,
      "num_tokens": 187417973.0,
      "step": 90320
    },
    {
      "entropy": 5.691803455352783,
      "epoch": 8.193486937590711,
      "grad_norm": 1.515625,
      "learning_rate": 8.58790581749872e-05,
      "loss": 4.8644,
      "mean_token_accuracy": 0.22751857042312623,
      "num_tokens": 187427829.0,
      "step": 90325
    },
    {
      "entropy": 5.647767353057861,
      "epoch": 8.193940493468796,
      "grad_norm": 1.390625,
      "learning_rate": 8.586153093216081e-05,
      "loss": 4.7612,
      "mean_token_accuracy": 0.2418011322617531,
      "num_tokens": 187438842.0,
      "step": 90330
    },
    {
      "entropy": 5.657397222518921,
      "epoch": 8.19439404934688,
      "grad_norm": 1.53125,
      "learning_rate": 8.584400760076613e-05,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23534366339445115,
      "num_tokens": 187448927.0,
      "step": 90335
    },
    {
      "entropy": 5.70513424873352,
      "epoch": 8.194847605224965,
      "grad_norm": 1.53125,
      "learning_rate": 8.582648818116556e-05,
      "loss": 4.844,
      "mean_token_accuracy": 0.22968676686286926,
      "num_tokens": 187458472.0,
      "step": 90340
    },
    {
      "entropy": 5.613517618179321,
      "epoch": 8.195301161103048,
      "grad_norm": 1.5234375,
      "learning_rate": 8.580897267372145e-05,
      "loss": 4.6885,
      "mean_token_accuracy": 0.24258210510015488,
      "num_tokens": 187467911.0,
      "step": 90345
    },
    {
      "entropy": 5.723050212860107,
      "epoch": 8.195754716981131,
      "grad_norm": 1.4765625,
      "learning_rate": 8.579146107879601e-05,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2322138085961342,
      "num_tokens": 187478060.0,
      "step": 90350
    },
    {
      "entropy": 5.653348779678344,
      "epoch": 8.196208272859216,
      "grad_norm": 1.4921875,
      "learning_rate": 8.577395339675136e-05,
      "loss": 4.7101,
      "mean_token_accuracy": 0.24146189242601396,
      "num_tokens": 187488060.0,
      "step": 90355
    },
    {
      "entropy": 5.765227651596069,
      "epoch": 8.1966618287373,
      "grad_norm": 1.4453125,
      "learning_rate": 8.57564496279496e-05,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22861869335174562,
      "num_tokens": 187497972.0,
      "step": 90360
    },
    {
      "entropy": 5.692259788513184,
      "epoch": 8.197115384615385,
      "grad_norm": 1.4453125,
      "learning_rate": 8.573894977275264e-05,
      "loss": 4.7899,
      "mean_token_accuracy": 0.23651032000780106,
      "num_tokens": 187509254.0,
      "step": 90365
    },
    {
      "entropy": 5.719960069656372,
      "epoch": 8.197568940493468,
      "grad_norm": 1.421875,
      "learning_rate": 8.572145383152247e-05,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23140916228294373,
      "num_tokens": 187519330.0,
      "step": 90370
    },
    {
      "entropy": 5.702863550186157,
      "epoch": 8.198022496371554,
      "grad_norm": 1.5234375,
      "learning_rate": 8.570396180462079e-05,
      "loss": 4.8586,
      "mean_token_accuracy": 0.2332441359758377,
      "num_tokens": 187529378.0,
      "step": 90375
    },
    {
      "entropy": 5.637386512756348,
      "epoch": 8.198476052249637,
      "grad_norm": 1.453125,
      "learning_rate": 8.568647369240948e-05,
      "loss": 4.6616,
      "mean_token_accuracy": 0.2469491794705391,
      "num_tokens": 187539446.0,
      "step": 90380
    },
    {
      "entropy": 5.762317705154419,
      "epoch": 8.198929608127722,
      "grad_norm": 1.4921875,
      "learning_rate": 8.566898949525015e-05,
      "loss": 4.8362,
      "mean_token_accuracy": 0.2374938413500786,
      "num_tokens": 187549414.0,
      "step": 90385
    },
    {
      "entropy": 5.66041898727417,
      "epoch": 8.199383164005805,
      "grad_norm": 1.4375,
      "learning_rate": 8.565150921350437e-05,
      "loss": 4.7811,
      "mean_token_accuracy": 0.2352210983633995,
      "num_tokens": 187559595.0,
      "step": 90390
    },
    {
      "entropy": 5.669903230667114,
      "epoch": 8.199836719883889,
      "grad_norm": 1.421875,
      "learning_rate": 8.56340328475336e-05,
      "loss": 4.7595,
      "mean_token_accuracy": 0.2402428701519966,
      "num_tokens": 187569863.0,
      "step": 90395
    },
    {
      "entropy": 5.644478750228882,
      "epoch": 8.200290275761974,
      "grad_norm": 1.5078125,
      "learning_rate": 8.56165603976993e-05,
      "loss": 4.7212,
      "mean_token_accuracy": 0.24067577123641967,
      "num_tokens": 187579704.0,
      "step": 90400
    },
    {
      "entropy": 5.709128761291504,
      "epoch": 8.200743831640057,
      "grad_norm": 1.4296875,
      "learning_rate": 8.559909186436272e-05,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2306541159749031,
      "num_tokens": 187590217.0,
      "step": 90405
    },
    {
      "entropy": 5.704120540618897,
      "epoch": 8.201197387518143,
      "grad_norm": 1.5234375,
      "learning_rate": 8.558162724788526e-05,
      "loss": 4.8209,
      "mean_token_accuracy": 0.23770902454853057,
      "num_tokens": 187600226.0,
      "step": 90410
    },
    {
      "entropy": 5.74227123260498,
      "epoch": 8.201650943396226,
      "grad_norm": 1.40625,
      "learning_rate": 8.556416654862802e-05,
      "loss": 4.8513,
      "mean_token_accuracy": 0.23603795021772384,
      "num_tokens": 187611034.0,
      "step": 90415
    },
    {
      "entropy": 5.722235250473022,
      "epoch": 8.202104499274311,
      "grad_norm": 1.453125,
      "learning_rate": 8.55467097669521e-05,
      "loss": 4.8302,
      "mean_token_accuracy": 0.237490676343441,
      "num_tokens": 187621750.0,
      "step": 90420
    },
    {
      "entropy": 5.626920366287232,
      "epoch": 8.202558055152394,
      "grad_norm": 1.5078125,
      "learning_rate": 8.552925690321847e-05,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24548953175544738,
      "num_tokens": 187631364.0,
      "step": 90425
    },
    {
      "entropy": 5.715682744979858,
      "epoch": 8.20301161103048,
      "grad_norm": 1.3671875,
      "learning_rate": 8.551180795778812e-05,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22224301397800444,
      "num_tokens": 187642989.0,
      "step": 90430
    },
    {
      "entropy": 5.61925892829895,
      "epoch": 8.203465166908563,
      "grad_norm": 1.5390625,
      "learning_rate": 8.549436293102186e-05,
      "loss": 4.6845,
      "mean_token_accuracy": 0.24785236120224,
      "num_tokens": 187653909.0,
      "step": 90435
    },
    {
      "entropy": 5.655036354064942,
      "epoch": 8.203918722786648,
      "grad_norm": 1.5625,
      "learning_rate": 8.547692182328041e-05,
      "loss": 4.7275,
      "mean_token_accuracy": 0.23962656706571578,
      "num_tokens": 187664845.0,
      "step": 90440
    },
    {
      "entropy": 5.660981273651123,
      "epoch": 8.204372278664732,
      "grad_norm": 1.3671875,
      "learning_rate": 8.545948463492457e-05,
      "loss": 4.7628,
      "mean_token_accuracy": 0.23330452144145966,
      "num_tokens": 187675512.0,
      "step": 90445
    },
    {
      "entropy": 5.658112382888794,
      "epoch": 8.204825834542815,
      "grad_norm": 1.5078125,
      "learning_rate": 8.544205136631489e-05,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2372206911444664,
      "num_tokens": 187686104.0,
      "step": 90450
    },
    {
      "entropy": 5.694719505310059,
      "epoch": 8.2052793904209,
      "grad_norm": 1.640625,
      "learning_rate": 8.542462201781187e-05,
      "loss": 4.8241,
      "mean_token_accuracy": 0.23521770238876344,
      "num_tokens": 187696637.0,
      "step": 90455
    },
    {
      "entropy": 5.665052986145019,
      "epoch": 8.205732946298983,
      "grad_norm": 1.53125,
      "learning_rate": 8.540719658977598e-05,
      "loss": 4.823,
      "mean_token_accuracy": 0.23644671738147735,
      "num_tokens": 187707261.0,
      "step": 90460
    },
    {
      "entropy": 5.679069852828979,
      "epoch": 8.206186502177069,
      "grad_norm": 1.609375,
      "learning_rate": 8.53897750825676e-05,
      "loss": 4.7129,
      "mean_token_accuracy": 0.24846579283475875,
      "num_tokens": 187716939.0,
      "step": 90465
    },
    {
      "entropy": 5.726431751251221,
      "epoch": 8.206640058055152,
      "grad_norm": 1.6328125,
      "learning_rate": 8.537235749654698e-05,
      "loss": 4.8665,
      "mean_token_accuracy": 0.23288217931985855,
      "num_tokens": 187727705.0,
      "step": 90470
    },
    {
      "entropy": 5.660924768447876,
      "epoch": 8.207093613933237,
      "grad_norm": 1.5546875,
      "learning_rate": 8.535494383207427e-05,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2418479084968567,
      "num_tokens": 187738111.0,
      "step": 90475
    },
    {
      "entropy": 5.747943449020386,
      "epoch": 8.20754716981132,
      "grad_norm": 1.6328125,
      "learning_rate": 8.53375340895097e-05,
      "loss": 4.9262,
      "mean_token_accuracy": 0.2310672491788864,
      "num_tokens": 187748465.0,
      "step": 90480
    },
    {
      "entropy": 5.684850597381592,
      "epoch": 8.208000725689406,
      "grad_norm": 1.4765625,
      "learning_rate": 8.532012826921329e-05,
      "loss": 4.7735,
      "mean_token_accuracy": 0.24976114332675933,
      "num_tokens": 187760569.0,
      "step": 90485
    },
    {
      "entropy": 5.703832006454467,
      "epoch": 8.208454281567489,
      "grad_norm": 1.375,
      "learning_rate": 8.530272637154493e-05,
      "loss": 4.7635,
      "mean_token_accuracy": 0.2352438360452652,
      "num_tokens": 187770629.0,
      "step": 90490
    },
    {
      "entropy": 5.690430498123169,
      "epoch": 8.208907837445572,
      "grad_norm": 1.4375,
      "learning_rate": 8.528532839686455e-05,
      "loss": 4.8157,
      "mean_token_accuracy": 0.24082791805267334,
      "num_tokens": 187782022.0,
      "step": 90495
    },
    {
      "entropy": 5.683732414245606,
      "epoch": 8.209361393323658,
      "grad_norm": 1.609375,
      "learning_rate": 8.52679343455319e-05,
      "loss": 4.8027,
      "mean_token_accuracy": 0.23736656606197357,
      "num_tokens": 187791648.0,
      "step": 90500
    },
    {
      "entropy": 5.673066568374634,
      "epoch": 8.209814949201741,
      "grad_norm": 1.5,
      "learning_rate": 8.525054421790673e-05,
      "loss": 4.7551,
      "mean_token_accuracy": 0.24031132310628892,
      "num_tokens": 187802026.0,
      "step": 90505
    },
    {
      "entropy": 5.667764043807983,
      "epoch": 8.210268505079826,
      "grad_norm": 1.453125,
      "learning_rate": 8.523315801434861e-05,
      "loss": 4.7461,
      "mean_token_accuracy": 0.23605651557445526,
      "num_tokens": 187812570.0,
      "step": 90510
    },
    {
      "entropy": 5.759746742248535,
      "epoch": 8.21072206095791,
      "grad_norm": 1.515625,
      "learning_rate": 8.521577573521721e-05,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21973448395729064,
      "num_tokens": 187823547.0,
      "step": 90515
    },
    {
      "entropy": 5.670630216598511,
      "epoch": 8.211175616835995,
      "grad_norm": 1.5390625,
      "learning_rate": 8.519839738087188e-05,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23472113013267518,
      "num_tokens": 187833906.0,
      "step": 90520
    },
    {
      "entropy": 5.663003396987915,
      "epoch": 8.211629172714078,
      "grad_norm": 1.609375,
      "learning_rate": 8.51810229516721e-05,
      "loss": 4.7465,
      "mean_token_accuracy": 0.24615169167518616,
      "num_tokens": 187844052.0,
      "step": 90525
    },
    {
      "entropy": 5.644863748550415,
      "epoch": 8.212082728592163,
      "grad_norm": 1.5,
      "learning_rate": 8.516365244797713e-05,
      "loss": 4.7134,
      "mean_token_accuracy": 0.256116783618927,
      "num_tokens": 187854092.0,
      "step": 90530
    },
    {
      "entropy": 5.688611221313477,
      "epoch": 8.212536284470247,
      "grad_norm": 1.578125,
      "learning_rate": 8.51462858701462e-05,
      "loss": 4.7704,
      "mean_token_accuracy": 0.23909449726343154,
      "num_tokens": 187863116.0,
      "step": 90535
    },
    {
      "entropy": 5.6753502368927,
      "epoch": 8.212989840348332,
      "grad_norm": 1.6171875,
      "learning_rate": 8.512892321853846e-05,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23528288751840593,
      "num_tokens": 187873426.0,
      "step": 90540
    },
    {
      "entropy": 5.657914781570435,
      "epoch": 8.213443396226415,
      "grad_norm": 1.53125,
      "learning_rate": 8.511156449351293e-05,
      "loss": 4.765,
      "mean_token_accuracy": 0.23743897825479507,
      "num_tokens": 187882079.0,
      "step": 90545
    },
    {
      "entropy": 5.721566867828369,
      "epoch": 8.213896952104498,
      "grad_norm": 1.5546875,
      "learning_rate": 8.509420969542866e-05,
      "loss": 4.8364,
      "mean_token_accuracy": 0.2531215578317642,
      "num_tokens": 187891829.0,
      "step": 90550
    },
    {
      "entropy": 5.702699279785156,
      "epoch": 8.214350507982584,
      "grad_norm": 1.3671875,
      "learning_rate": 8.507685882464455e-05,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23602920025587082,
      "num_tokens": 187902565.0,
      "step": 90555
    },
    {
      "entropy": 5.724644136428833,
      "epoch": 8.214804063860667,
      "grad_norm": 1.4609375,
      "learning_rate": 8.505951188151939e-05,
      "loss": 4.7584,
      "mean_token_accuracy": 0.23661698549985885,
      "num_tokens": 187912536.0,
      "step": 90560
    },
    {
      "entropy": 5.6670365810394285,
      "epoch": 8.215257619738752,
      "grad_norm": 1.5546875,
      "learning_rate": 8.504216886641199e-05,
      "loss": 4.7565,
      "mean_token_accuracy": 0.23723191916942596,
      "num_tokens": 187922212.0,
      "step": 90565
    },
    {
      "entropy": 5.728718376159668,
      "epoch": 8.215711175616836,
      "grad_norm": 1.4453125,
      "learning_rate": 8.502482977968083e-05,
      "loss": 4.8006,
      "mean_token_accuracy": 0.23957709670066835,
      "num_tokens": 187932922.0,
      "step": 90570
    },
    {
      "entropy": 5.592223358154297,
      "epoch": 8.21616473149492,
      "grad_norm": 1.5625,
      "learning_rate": 8.500749462168467e-05,
      "loss": 4.7005,
      "mean_token_accuracy": 0.2417064368724823,
      "num_tokens": 187942726.0,
      "step": 90575
    },
    {
      "entropy": 5.637595844268799,
      "epoch": 8.216618287373004,
      "grad_norm": 1.640625,
      "learning_rate": 8.49901633927819e-05,
      "loss": 4.7314,
      "mean_token_accuracy": 0.24796562194824218,
      "num_tokens": 187953315.0,
      "step": 90580
    },
    {
      "entropy": 5.657938861846924,
      "epoch": 8.21707184325109,
      "grad_norm": 1.5078125,
      "learning_rate": 8.497283609333099e-05,
      "loss": 4.8105,
      "mean_token_accuracy": 0.2334200918674469,
      "num_tokens": 187964211.0,
      "step": 90585
    },
    {
      "entropy": 5.725555276870727,
      "epoch": 8.217525399129173,
      "grad_norm": 1.6328125,
      "learning_rate": 8.495551272369026e-05,
      "loss": 4.8836,
      "mean_token_accuracy": 0.23539797067642212,
      "num_tokens": 187974719.0,
      "step": 90590
    },
    {
      "entropy": 5.7341221332550045,
      "epoch": 8.217978955007258,
      "grad_norm": 1.5,
      "learning_rate": 8.493819328421796e-05,
      "loss": 4.864,
      "mean_token_accuracy": 0.23280755579471588,
      "num_tokens": 187984109.0,
      "step": 90595
    },
    {
      "entropy": 5.6885425567626955,
      "epoch": 8.218432510885341,
      "grad_norm": 1.4296875,
      "learning_rate": 8.492087777527225e-05,
      "loss": 4.8059,
      "mean_token_accuracy": 0.23600959926843643,
      "num_tokens": 187995621.0,
      "step": 90600
    },
    {
      "entropy": 5.728236961364746,
      "epoch": 8.218886066763424,
      "grad_norm": 1.5078125,
      "learning_rate": 8.490356619721118e-05,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2285924345254898,
      "num_tokens": 188005296.0,
      "step": 90605
    },
    {
      "entropy": 5.711103439331055,
      "epoch": 8.21933962264151,
      "grad_norm": 1.4375,
      "learning_rate": 8.488625855039286e-05,
      "loss": 4.8325,
      "mean_token_accuracy": 0.2319883242249489,
      "num_tokens": 188016944.0,
      "step": 90610
    },
    {
      "entropy": 5.728000211715698,
      "epoch": 8.219793178519593,
      "grad_norm": 1.515625,
      "learning_rate": 8.486895483517515e-05,
      "loss": 4.8011,
      "mean_token_accuracy": 0.23587778806686402,
      "num_tokens": 188027518.0,
      "step": 90615
    },
    {
      "entropy": 5.663213205337525,
      "epoch": 8.220246734397678,
      "grad_norm": 1.4296875,
      "learning_rate": 8.485165505191589e-05,
      "loss": 4.8008,
      "mean_token_accuracy": 0.23983440548181534,
      "num_tokens": 188038178.0,
      "step": 90620
    },
    {
      "entropy": 5.669768238067627,
      "epoch": 8.220700290275762,
      "grad_norm": 1.2421875,
      "learning_rate": 8.483435920097287e-05,
      "loss": 4.7458,
      "mean_token_accuracy": 0.24799128770828247,
      "num_tokens": 188050092.0,
      "step": 90625
    },
    {
      "entropy": 5.659755563735962,
      "epoch": 8.221153846153847,
      "grad_norm": 1.6796875,
      "learning_rate": 8.481706728270378e-05,
      "loss": 4.7304,
      "mean_token_accuracy": 0.24207293689250947,
      "num_tokens": 188060211.0,
      "step": 90630
    },
    {
      "entropy": 5.733222818374633,
      "epoch": 8.22160740203193,
      "grad_norm": 1.375,
      "learning_rate": 8.479977929746617e-05,
      "loss": 4.8998,
      "mean_token_accuracy": 0.23408899307250977,
      "num_tokens": 188071765.0,
      "step": 90635
    },
    {
      "entropy": 5.663028764724731,
      "epoch": 8.222060957910015,
      "grad_norm": 1.4921875,
      "learning_rate": 8.478249524561758e-05,
      "loss": 4.7894,
      "mean_token_accuracy": 0.24704430997371674,
      "num_tokens": 188082440.0,
      "step": 90640
    },
    {
      "entropy": 5.697838640213012,
      "epoch": 8.222514513788099,
      "grad_norm": 1.671875,
      "learning_rate": 8.476521512751548e-05,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2378432810306549,
      "num_tokens": 188091315.0,
      "step": 90645
    },
    {
      "entropy": 5.76552381515503,
      "epoch": 8.222968069666184,
      "grad_norm": 1.59375,
      "learning_rate": 8.474793894351724e-05,
      "loss": 4.8332,
      "mean_token_accuracy": 0.23845240771770476,
      "num_tokens": 188100906.0,
      "step": 90650
    },
    {
      "entropy": 5.657237243652344,
      "epoch": 8.223421625544267,
      "grad_norm": 1.421875,
      "learning_rate": 8.473066669398005e-05,
      "loss": 4.7409,
      "mean_token_accuracy": 0.24897727817296983,
      "num_tokens": 188112062.0,
      "step": 90655
    },
    {
      "entropy": 5.674151372909546,
      "epoch": 8.22387518142235,
      "grad_norm": 1.4296875,
      "learning_rate": 8.47133983792612e-05,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23070373386144638,
      "num_tokens": 188122677.0,
      "step": 90660
    },
    {
      "entropy": 5.718736410140991,
      "epoch": 8.224328737300436,
      "grad_norm": 1.3828125,
      "learning_rate": 8.469613399971779e-05,
      "loss": 4.8755,
      "mean_token_accuracy": 0.23415968865156173,
      "num_tokens": 188133479.0,
      "step": 90665
    },
    {
      "entropy": 5.702302932739258,
      "epoch": 8.224782293178519,
      "grad_norm": 1.4921875,
      "learning_rate": 8.467887355570678e-05,
      "loss": 4.8538,
      "mean_token_accuracy": 0.2276550352573395,
      "num_tokens": 188144347.0,
      "step": 90670
    },
    {
      "entropy": 5.764096784591675,
      "epoch": 8.225235849056604,
      "grad_norm": 1.5390625,
      "learning_rate": 8.466161704758511e-05,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2311725750565529,
      "num_tokens": 188155773.0,
      "step": 90675
    },
    {
      "entropy": 5.675661945343018,
      "epoch": 8.225689404934688,
      "grad_norm": 1.578125,
      "learning_rate": 8.464436447570978e-05,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2390719845890999,
      "num_tokens": 188166211.0,
      "step": 90680
    },
    {
      "entropy": 5.687388181686401,
      "epoch": 8.226142960812773,
      "grad_norm": 1.4140625,
      "learning_rate": 8.46271158404375e-05,
      "loss": 4.779,
      "mean_token_accuracy": 0.24124813377857207,
      "num_tokens": 188176156.0,
      "step": 90685
    },
    {
      "entropy": 5.695997047424316,
      "epoch": 8.226596516690856,
      "grad_norm": 1.5,
      "learning_rate": 8.460987114212495e-05,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22932445108890534,
      "num_tokens": 188186614.0,
      "step": 90690
    },
    {
      "entropy": 5.723404884338379,
      "epoch": 8.227050072568941,
      "grad_norm": 1.453125,
      "learning_rate": 8.45926303811288e-05,
      "loss": 4.7865,
      "mean_token_accuracy": 0.2434905543923378,
      "num_tokens": 188196901.0,
      "step": 90695
    },
    {
      "entropy": 5.654957628250122,
      "epoch": 8.227503628447025,
      "grad_norm": 1.6328125,
      "learning_rate": 8.457539355780558e-05,
      "loss": 4.6795,
      "mean_token_accuracy": 0.23796603679656983,
      "num_tokens": 188206491.0,
      "step": 90700
    },
    {
      "entropy": 5.722637367248535,
      "epoch": 8.227957184325108,
      "grad_norm": 1.5390625,
      "learning_rate": 8.455816067251171e-05,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22381583005189895,
      "num_tokens": 188216095.0,
      "step": 90705
    },
    {
      "entropy": 5.72762713432312,
      "epoch": 8.228410740203193,
      "grad_norm": 1.3515625,
      "learning_rate": 8.454093172560364e-05,
      "loss": 4.8269,
      "mean_token_accuracy": 0.23034408390522004,
      "num_tokens": 188227743.0,
      "step": 90710
    },
    {
      "entropy": 5.612669038772583,
      "epoch": 8.228864296081277,
      "grad_norm": 1.4375,
      "learning_rate": 8.452370671743761e-05,
      "loss": 4.7063,
      "mean_token_accuracy": 0.2481731131672859,
      "num_tokens": 188238061.0,
      "step": 90715
    },
    {
      "entropy": 5.681066036224365,
      "epoch": 8.229317851959362,
      "grad_norm": 1.5703125,
      "learning_rate": 8.45064856483699e-05,
      "loss": 4.7455,
      "mean_token_accuracy": 0.24704018533229827,
      "num_tokens": 188247320.0,
      "step": 90720
    },
    {
      "entropy": 5.682439994812012,
      "epoch": 8.229771407837445,
      "grad_norm": 1.640625,
      "learning_rate": 8.44892685187566e-05,
      "loss": 4.747,
      "mean_token_accuracy": 0.24201081246137618,
      "num_tokens": 188257675.0,
      "step": 90725
    },
    {
      "entropy": 5.606034183502198,
      "epoch": 8.23022496371553,
      "grad_norm": 1.46875,
      "learning_rate": 8.447205532895378e-05,
      "loss": 4.6998,
      "mean_token_accuracy": 0.23944831192493438,
      "num_tokens": 188268191.0,
      "step": 90730
    },
    {
      "entropy": 5.67895474433899,
      "epoch": 8.230678519593614,
      "grad_norm": 1.5546875,
      "learning_rate": 8.445484607931742e-05,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2319626271724701,
      "num_tokens": 188277972.0,
      "step": 90735
    },
    {
      "entropy": 5.634367179870606,
      "epoch": 8.231132075471699,
      "grad_norm": 1.4765625,
      "learning_rate": 8.443764077020331e-05,
      "loss": 4.7572,
      "mean_token_accuracy": 0.2410231277346611,
      "num_tokens": 188288750.0,
      "step": 90740
    },
    {
      "entropy": 5.700469493865967,
      "epoch": 8.231585631349782,
      "grad_norm": 1.6796875,
      "learning_rate": 8.442043940196744e-05,
      "loss": 4.8165,
      "mean_token_accuracy": 0.2311058148741722,
      "num_tokens": 188299848.0,
      "step": 90745
    },
    {
      "entropy": 5.719078588485718,
      "epoch": 8.232039187227867,
      "grad_norm": 1.5390625,
      "learning_rate": 8.440324197496544e-05,
      "loss": 4.8483,
      "mean_token_accuracy": 0.2360844060778618,
      "num_tokens": 188309770.0,
      "step": 90750
    },
    {
      "entropy": 5.640816879272461,
      "epoch": 8.23249274310595,
      "grad_norm": 1.578125,
      "learning_rate": 8.438604848955295e-05,
      "loss": 4.7382,
      "mean_token_accuracy": 0.24104428589344024,
      "num_tokens": 188320855.0,
      "step": 90755
    },
    {
      "entropy": 5.674601888656616,
      "epoch": 8.232946298984034,
      "grad_norm": 1.671875,
      "learning_rate": 8.436885894608554e-05,
      "loss": 4.8186,
      "mean_token_accuracy": 0.23779881596565247,
      "num_tokens": 188330861.0,
      "step": 90760
    },
    {
      "entropy": 5.62109603881836,
      "epoch": 8.23339985486212,
      "grad_norm": 1.5,
      "learning_rate": 8.435167334491876e-05,
      "loss": 4.7093,
      "mean_token_accuracy": 0.23880542665719987,
      "num_tokens": 188340793.0,
      "step": 90765
    },
    {
      "entropy": 5.735506248474121,
      "epoch": 8.233853410740203,
      "grad_norm": 1.5859375,
      "learning_rate": 8.43344916864079e-05,
      "loss": 4.8708,
      "mean_token_accuracy": 0.225460222363472,
      "num_tokens": 188350881.0,
      "step": 90770
    },
    {
      "entropy": 5.7191814422607425,
      "epoch": 8.234306966618288,
      "grad_norm": 1.6015625,
      "learning_rate": 8.43173139709083e-05,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2381407767534256,
      "num_tokens": 188361386.0,
      "step": 90775
    },
    {
      "entropy": 5.656911468505859,
      "epoch": 8.234760522496371,
      "grad_norm": 1.4296875,
      "learning_rate": 8.43001401987753e-05,
      "loss": 4.7817,
      "mean_token_accuracy": 0.2324432522058487,
      "num_tokens": 188371585.0,
      "step": 90780
    },
    {
      "entropy": 5.732047033309937,
      "epoch": 8.235214078374456,
      "grad_norm": 1.515625,
      "learning_rate": 8.428297037036401e-05,
      "loss": 4.89,
      "mean_token_accuracy": 0.2290794476866722,
      "num_tokens": 188382798.0,
      "step": 90785
    },
    {
      "entropy": 5.682984685897827,
      "epoch": 8.23566763425254,
      "grad_norm": 1.546875,
      "learning_rate": 8.426580448602948e-05,
      "loss": 4.7422,
      "mean_token_accuracy": 0.23907275199890138,
      "num_tokens": 188393376.0,
      "step": 90790
    },
    {
      "entropy": 5.617674732208252,
      "epoch": 8.236121190130625,
      "grad_norm": 1.5,
      "learning_rate": 8.424864254612674e-05,
      "loss": 4.7337,
      "mean_token_accuracy": 0.24467982202768326,
      "num_tokens": 188404805.0,
      "step": 90795
    },
    {
      "entropy": 5.684377908706665,
      "epoch": 8.236574746008708,
      "grad_norm": 1.546875,
      "learning_rate": 8.423148455101066e-05,
      "loss": 4.7545,
      "mean_token_accuracy": 0.23891610354185105,
      "num_tokens": 188414821.0,
      "step": 90800
    },
    {
      "entropy": 5.646154356002808,
      "epoch": 8.237028301886792,
      "grad_norm": 1.4140625,
      "learning_rate": 8.421433050103609e-05,
      "loss": 4.7373,
      "mean_token_accuracy": 0.23824537694454193,
      "num_tokens": 188425226.0,
      "step": 90805
    },
    {
      "entropy": 5.647101974487304,
      "epoch": 8.237481857764877,
      "grad_norm": 1.5625,
      "learning_rate": 8.419718039655774e-05,
      "loss": 4.7749,
      "mean_token_accuracy": 0.2339727059006691,
      "num_tokens": 188435215.0,
      "step": 90810
    },
    {
      "entropy": 5.646821308135986,
      "epoch": 8.23793541364296,
      "grad_norm": 1.484375,
      "learning_rate": 8.418003423793035e-05,
      "loss": 4.7758,
      "mean_token_accuracy": 0.23676653504371642,
      "num_tokens": 188446289.0,
      "step": 90815
    },
    {
      "entropy": 5.6946385383605955,
      "epoch": 8.238388969521045,
      "grad_norm": 1.59375,
      "learning_rate": 8.41628920255085e-05,
      "loss": 4.836,
      "mean_token_accuracy": 0.23635853826999664,
      "num_tokens": 188456010.0,
      "step": 90820
    },
    {
      "entropy": 5.65877628326416,
      "epoch": 8.238842525399129,
      "grad_norm": 1.453125,
      "learning_rate": 8.414575375964667e-05,
      "loss": 4.8509,
      "mean_token_accuracy": 0.24171533584594726,
      "num_tokens": 188467944.0,
      "step": 90825
    },
    {
      "entropy": 5.691052198410034,
      "epoch": 8.239296081277214,
      "grad_norm": 1.625,
      "learning_rate": 8.41286194406993e-05,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22991663366556167,
      "num_tokens": 188477302.0,
      "step": 90830
    },
    {
      "entropy": 5.702075672149658,
      "epoch": 8.239749637155297,
      "grad_norm": 1.4453125,
      "learning_rate": 8.411148906902072e-05,
      "loss": 4.7724,
      "mean_token_accuracy": 0.2407807245850563,
      "num_tokens": 188487536.0,
      "step": 90835
    },
    {
      "entropy": 5.707656288146973,
      "epoch": 8.240203193033382,
      "grad_norm": 1.6328125,
      "learning_rate": 8.409436264496518e-05,
      "loss": 4.8018,
      "mean_token_accuracy": 0.2361203223466873,
      "num_tokens": 188497226.0,
      "step": 90840
    },
    {
      "entropy": 5.730699348449707,
      "epoch": 8.240656748911466,
      "grad_norm": 1.453125,
      "learning_rate": 8.40772401688868e-05,
      "loss": 4.8629,
      "mean_token_accuracy": 0.2298883393406868,
      "num_tokens": 188507524.0,
      "step": 90845
    },
    {
      "entropy": 5.733546447753906,
      "epoch": 8.24111030478955,
      "grad_norm": 1.5859375,
      "learning_rate": 8.406012164113982e-05,
      "loss": 4.8069,
      "mean_token_accuracy": 0.23315664380788803,
      "num_tokens": 188517229.0,
      "step": 90850
    },
    {
      "entropy": 5.613424444198609,
      "epoch": 8.241563860667634,
      "grad_norm": 1.46875,
      "learning_rate": 8.404300706207818e-05,
      "loss": 4.6344,
      "mean_token_accuracy": 0.24877965897321702,
      "num_tokens": 188527627.0,
      "step": 90855
    },
    {
      "entropy": 5.703477048873902,
      "epoch": 8.242017416545718,
      "grad_norm": 1.6484375,
      "learning_rate": 8.402589643205579e-05,
      "loss": 4.7614,
      "mean_token_accuracy": 0.23944674730300902,
      "num_tokens": 188537753.0,
      "step": 90860
    },
    {
      "entropy": 5.730878114700317,
      "epoch": 8.242470972423803,
      "grad_norm": 1.625,
      "learning_rate": 8.400878975142657e-05,
      "loss": 4.8624,
      "mean_token_accuracy": 0.2261812672019005,
      "num_tokens": 188549136.0,
      "step": 90865
    },
    {
      "entropy": 5.69645619392395,
      "epoch": 8.242924528301886,
      "grad_norm": 1.546875,
      "learning_rate": 8.399168702054422e-05,
      "loss": 4.839,
      "mean_token_accuracy": 0.23746226727962494,
      "num_tokens": 188559849.0,
      "step": 90870
    },
    {
      "entropy": 5.648223209381103,
      "epoch": 8.243378084179971,
      "grad_norm": 1.46875,
      "learning_rate": 8.397458823976248e-05,
      "loss": 4.7717,
      "mean_token_accuracy": 0.23996011912822723,
      "num_tokens": 188570141.0,
      "step": 90875
    },
    {
      "entropy": 5.7418945789337155,
      "epoch": 8.243831640058055,
      "grad_norm": 1.5,
      "learning_rate": 8.395749340943485e-05,
      "loss": 4.8496,
      "mean_token_accuracy": 0.23362127989530562,
      "num_tokens": 188579373.0,
      "step": 90880
    },
    {
      "entropy": 5.675340700149536,
      "epoch": 8.24428519593614,
      "grad_norm": 1.3984375,
      "learning_rate": 8.394040252991503e-05,
      "loss": 4.7974,
      "mean_token_accuracy": 0.237145134806633,
      "num_tokens": 188590120.0,
      "step": 90885
    },
    {
      "entropy": 5.715720510482788,
      "epoch": 8.244738751814223,
      "grad_norm": 1.484375,
      "learning_rate": 8.392331560155634e-05,
      "loss": 4.8717,
      "mean_token_accuracy": 0.23605694025754928,
      "num_tokens": 188601420.0,
      "step": 90890
    },
    {
      "entropy": 5.638871192932129,
      "epoch": 8.245192307692308,
      "grad_norm": 1.515625,
      "learning_rate": 8.39062326247122e-05,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2494267150759697,
      "num_tokens": 188611806.0,
      "step": 90895
    },
    {
      "entropy": 5.699211835861206,
      "epoch": 8.245645863570392,
      "grad_norm": 1.578125,
      "learning_rate": 8.388915359973586e-05,
      "loss": 4.8212,
      "mean_token_accuracy": 0.23421671241521835,
      "num_tokens": 188622816.0,
      "step": 90900
    },
    {
      "entropy": 5.73767204284668,
      "epoch": 8.246099419448477,
      "grad_norm": 1.53125,
      "learning_rate": 8.38720785269805e-05,
      "loss": 4.8398,
      "mean_token_accuracy": 0.23513120859861375,
      "num_tokens": 188633193.0,
      "step": 90905
    },
    {
      "entropy": 5.694986295700073,
      "epoch": 8.24655297532656,
      "grad_norm": 1.4140625,
      "learning_rate": 8.385500740679922e-05,
      "loss": 4.8101,
      "mean_token_accuracy": 0.2276940628886223,
      "num_tokens": 188644197.0,
      "step": 90910
    },
    {
      "entropy": 5.708482265472412,
      "epoch": 8.247006531204644,
      "grad_norm": 1.4140625,
      "learning_rate": 8.383794023954517e-05,
      "loss": 4.8184,
      "mean_token_accuracy": 0.23473510444164275,
      "num_tokens": 188654420.0,
      "step": 90915
    },
    {
      "entropy": 5.661064004898071,
      "epoch": 8.247460087082729,
      "grad_norm": 1.53125,
      "learning_rate": 8.38208770255712e-05,
      "loss": 4.7786,
      "mean_token_accuracy": 0.2343516990542412,
      "num_tokens": 188664690.0,
      "step": 90920
    },
    {
      "entropy": 5.7116907119750975,
      "epoch": 8.247913642960812,
      "grad_norm": 1.5703125,
      "learning_rate": 8.380381776523024e-05,
      "loss": 4.7977,
      "mean_token_accuracy": 0.23192149102687837,
      "num_tokens": 188675677.0,
      "step": 90925
    },
    {
      "entropy": 5.671830034255981,
      "epoch": 8.248367198838897,
      "grad_norm": 1.59375,
      "learning_rate": 8.378676245887504e-05,
      "loss": 4.7131,
      "mean_token_accuracy": 0.24885273873806,
      "num_tokens": 188685152.0,
      "step": 90930
    },
    {
      "entropy": 5.717337799072266,
      "epoch": 8.24882075471698,
      "grad_norm": 1.4921875,
      "learning_rate": 8.37697111068583e-05,
      "loss": 4.7567,
      "mean_token_accuracy": 0.2413076266646385,
      "num_tokens": 188695974.0,
      "step": 90935
    },
    {
      "entropy": 5.663115406036377,
      "epoch": 8.249274310595066,
      "grad_norm": 1.5,
      "learning_rate": 8.375266370953272e-05,
      "loss": 4.7535,
      "mean_token_accuracy": 0.23847136348485948,
      "num_tokens": 188706379.0,
      "step": 90940
    },
    {
      "entropy": 5.761938953399659,
      "epoch": 8.24972786647315,
      "grad_norm": 1.46875,
      "learning_rate": 8.373562026725067e-05,
      "loss": 4.8426,
      "mean_token_accuracy": 0.22768578827381133,
      "num_tokens": 188716559.0,
      "step": 90945
    },
    {
      "entropy": 5.667680644989014,
      "epoch": 8.250181422351234,
      "grad_norm": 1.515625,
      "learning_rate": 8.371858078036481e-05,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2350558638572693,
      "num_tokens": 188726561.0,
      "step": 90950
    },
    {
      "entropy": 5.699882221221924,
      "epoch": 8.250634978229318,
      "grad_norm": 1.5078125,
      "learning_rate": 8.370154524922749e-05,
      "loss": 4.7661,
      "mean_token_accuracy": 0.24269693046808244,
      "num_tokens": 188736009.0,
      "step": 90955
    },
    {
      "entropy": 5.623003816604614,
      "epoch": 8.251088534107401,
      "grad_norm": 1.4921875,
      "learning_rate": 8.368451367419092e-05,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2397315576672554,
      "num_tokens": 188746608.0,
      "step": 90960
    },
    {
      "entropy": 5.65512638092041,
      "epoch": 8.251542089985486,
      "grad_norm": 1.484375,
      "learning_rate": 8.366748605560735e-05,
      "loss": 4.7154,
      "mean_token_accuracy": 0.24203592985868455,
      "num_tokens": 188757750.0,
      "step": 90965
    },
    {
      "entropy": 5.667241811752319,
      "epoch": 8.25199564586357,
      "grad_norm": 1.609375,
      "learning_rate": 8.365046239382893e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.23083252608776092,
      "num_tokens": 188767407.0,
      "step": 90970
    },
    {
      "entropy": 5.669921159744263,
      "epoch": 8.252449201741655,
      "grad_norm": 1.3984375,
      "learning_rate": 8.363344268920774e-05,
      "loss": 4.8046,
      "mean_token_accuracy": 0.22944841235876084,
      "num_tokens": 188779056.0,
      "step": 90975
    },
    {
      "entropy": 5.581432819366455,
      "epoch": 8.252902757619738,
      "grad_norm": 1.609375,
      "learning_rate": 8.361642694209563e-05,
      "loss": 4.6432,
      "mean_token_accuracy": 0.2550755888223648,
      "num_tokens": 188788659.0,
      "step": 90980
    },
    {
      "entropy": 5.675892639160156,
      "epoch": 8.253356313497823,
      "grad_norm": 1.6171875,
      "learning_rate": 8.359941515284463e-05,
      "loss": 4.7921,
      "mean_token_accuracy": 0.24188755601644515,
      "num_tokens": 188798603.0,
      "step": 90985
    },
    {
      "entropy": 5.707578516006469,
      "epoch": 8.253809869375907,
      "grad_norm": 1.546875,
      "learning_rate": 8.35824073218066e-05,
      "loss": 4.8143,
      "mean_token_accuracy": 0.2408279910683632,
      "num_tokens": 188809431.0,
      "step": 90990
    },
    {
      "entropy": 5.647011947631836,
      "epoch": 8.254263425253992,
      "grad_norm": 1.7109375,
      "learning_rate": 8.356540344933308e-05,
      "loss": 4.7477,
      "mean_token_accuracy": 0.23518364131450653,
      "num_tokens": 188819643.0,
      "step": 90995
    },
    {
      "entropy": 5.76546802520752,
      "epoch": 8.254716981132075,
      "grad_norm": 1.5625,
      "learning_rate": 8.35484035357758e-05,
      "loss": 4.8347,
      "mean_token_accuracy": 0.23089670836925508,
      "num_tokens": 188829699.0,
      "step": 91000
    },
    {
      "entropy": 5.7088910102844235,
      "epoch": 8.25517053701016,
      "grad_norm": 1.53125,
      "learning_rate": 8.353140758148629e-05,
      "loss": 4.8079,
      "mean_token_accuracy": 0.23575983941555023,
      "num_tokens": 188840632.0,
      "step": 91005
    },
    {
      "entropy": 5.73811354637146,
      "epoch": 8.255624092888244,
      "grad_norm": 1.5703125,
      "learning_rate": 8.35144155868161e-05,
      "loss": 4.7802,
      "mean_token_accuracy": 0.2354110911488533,
      "num_tokens": 188851036.0,
      "step": 91010
    },
    {
      "entropy": 5.71325454711914,
      "epoch": 8.256077648766327,
      "grad_norm": 1.546875,
      "learning_rate": 8.349742755211659e-05,
      "loss": 4.773,
      "mean_token_accuracy": 0.23925278186798096,
      "num_tokens": 188861696.0,
      "step": 91015
    },
    {
      "entropy": 5.66933856010437,
      "epoch": 8.256531204644412,
      "grad_norm": 1.453125,
      "learning_rate": 8.348044347773908e-05,
      "loss": 4.7776,
      "mean_token_accuracy": 0.23229749947786332,
      "num_tokens": 188872014.0,
      "step": 91020
    },
    {
      "entropy": 5.6787138938903805,
      "epoch": 8.256984760522496,
      "grad_norm": 1.4921875,
      "learning_rate": 8.346346336403481e-05,
      "loss": 4.7808,
      "mean_token_accuracy": 0.23812047243118287,
      "num_tokens": 188883301.0,
      "step": 91025
    },
    {
      "entropy": 5.683319807052612,
      "epoch": 8.257438316400581,
      "grad_norm": 1.4140625,
      "learning_rate": 8.344648721135492e-05,
      "loss": 4.777,
      "mean_token_accuracy": 0.23910706788301467,
      "num_tokens": 188893613.0,
      "step": 91030
    },
    {
      "entropy": 5.6317604064941404,
      "epoch": 8.257891872278664,
      "grad_norm": 1.484375,
      "learning_rate": 8.342951502005048e-05,
      "loss": 4.7421,
      "mean_token_accuracy": 0.24230569303035737,
      "num_tokens": 188903820.0,
      "step": 91035
    },
    {
      "entropy": 5.633017063140869,
      "epoch": 8.25834542815675,
      "grad_norm": 1.59375,
      "learning_rate": 8.341254679047247e-05,
      "loss": 4.7167,
      "mean_token_accuracy": 0.24456077963113784,
      "num_tokens": 188914117.0,
      "step": 91040
    },
    {
      "entropy": 5.672479963302612,
      "epoch": 8.258798984034833,
      "grad_norm": 1.5625,
      "learning_rate": 8.339558252297185e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.235787533223629,
      "num_tokens": 188922950.0,
      "step": 91045
    },
    {
      "entropy": 5.6558003425598145,
      "epoch": 8.259252539912918,
      "grad_norm": 1.484375,
      "learning_rate": 8.337862221789941e-05,
      "loss": 4.752,
      "mean_token_accuracy": 0.23898226618766785,
      "num_tokens": 188933458.0,
      "step": 91050
    },
    {
      "entropy": 5.694613409042359,
      "epoch": 8.259706095791001,
      "grad_norm": 1.4609375,
      "learning_rate": 8.336166587560586e-05,
      "loss": 4.8592,
      "mean_token_accuracy": 0.23445525169372558,
      "num_tokens": 188943758.0,
      "step": 91055
    },
    {
      "entropy": 5.641618394851685,
      "epoch": 8.260159651669085,
      "grad_norm": 1.5,
      "learning_rate": 8.334471349644194e-05,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2495507001876831,
      "num_tokens": 188953957.0,
      "step": 91060
    },
    {
      "entropy": 5.667987298965454,
      "epoch": 8.26061320754717,
      "grad_norm": 1.609375,
      "learning_rate": 8.332776508075814e-05,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23443879783153534,
      "num_tokens": 188963663.0,
      "step": 91065
    },
    {
      "entropy": 5.648628664016724,
      "epoch": 8.261066763425253,
      "grad_norm": 1.5390625,
      "learning_rate": 8.331082062890505e-05,
      "loss": 4.7421,
      "mean_token_accuracy": 0.23244993686676024,
      "num_tokens": 188973208.0,
      "step": 91070
    },
    {
      "entropy": 5.6652419090271,
      "epoch": 8.261520319303338,
      "grad_norm": 1.59375,
      "learning_rate": 8.329388014123296e-05,
      "loss": 4.7715,
      "mean_token_accuracy": 0.24369381964206696,
      "num_tokens": 188983914.0,
      "step": 91075
    },
    {
      "entropy": 5.690786457061767,
      "epoch": 8.261973875181422,
      "grad_norm": 1.515625,
      "learning_rate": 8.327694361809235e-05,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2423340141773224,
      "num_tokens": 188993701.0,
      "step": 91080
    },
    {
      "entropy": 5.6744856357574465,
      "epoch": 8.262427431059507,
      "grad_norm": 1.625,
      "learning_rate": 8.326001105983337e-05,
      "loss": 4.7576,
      "mean_token_accuracy": 0.24924541264772415,
      "num_tokens": 189004055.0,
      "step": 91085
    },
    {
      "entropy": 5.7078940868377686,
      "epoch": 8.26288098693759,
      "grad_norm": 1.4296875,
      "learning_rate": 8.324308246680624e-05,
      "loss": 4.8947,
      "mean_token_accuracy": 0.23371261060237886,
      "num_tokens": 189014350.0,
      "step": 91090
    },
    {
      "entropy": 5.698531007766723,
      "epoch": 8.263334542815675,
      "grad_norm": 1.65625,
      "learning_rate": 8.322615783936103e-05,
      "loss": 4.7905,
      "mean_token_accuracy": 0.24108764082193374,
      "num_tokens": 189024162.0,
      "step": 91095
    },
    {
      "entropy": 5.700304269790649,
      "epoch": 8.263788098693759,
      "grad_norm": 1.609375,
      "learning_rate": 8.320923717784775e-05,
      "loss": 4.776,
      "mean_token_accuracy": 0.2365760773420334,
      "num_tokens": 189033668.0,
      "step": 91100
    },
    {
      "entropy": 5.676487016677856,
      "epoch": 8.264241654571844,
      "grad_norm": 1.3984375,
      "learning_rate": 8.319232048261633e-05,
      "loss": 4.803,
      "mean_token_accuracy": 0.23442634791135789,
      "num_tokens": 189044360.0,
      "step": 91105
    },
    {
      "entropy": 5.638352394104004,
      "epoch": 8.264695210449927,
      "grad_norm": 1.625,
      "learning_rate": 8.317540775401655e-05,
      "loss": 4.7212,
      "mean_token_accuracy": 0.24250977039337157,
      "num_tokens": 189054505.0,
      "step": 91110
    },
    {
      "entropy": 5.709730195999145,
      "epoch": 8.26514876632801,
      "grad_norm": 1.59375,
      "learning_rate": 8.315849899239823e-05,
      "loss": 4.7691,
      "mean_token_accuracy": 0.23037175983190536,
      "num_tokens": 189064909.0,
      "step": 91115
    },
    {
      "entropy": 5.675620269775391,
      "epoch": 8.265602322206096,
      "grad_norm": 1.53125,
      "learning_rate": 8.314159419811108e-05,
      "loss": 4.7342,
      "mean_token_accuracy": 0.24408797174692154,
      "num_tokens": 189075564.0,
      "step": 91120
    },
    {
      "entropy": 5.63213529586792,
      "epoch": 8.26605587808418,
      "grad_norm": 1.4921875,
      "learning_rate": 8.312469337150463e-05,
      "loss": 4.7283,
      "mean_token_accuracy": 0.237675803899765,
      "num_tokens": 189086100.0,
      "step": 91125
    },
    {
      "entropy": 5.680723047256469,
      "epoch": 8.266509433962264,
      "grad_norm": 1.578125,
      "learning_rate": 8.310779651292844e-05,
      "loss": 4.8196,
      "mean_token_accuracy": 0.2267441600561142,
      "num_tokens": 189096313.0,
      "step": 91130
    },
    {
      "entropy": 5.650304222106934,
      "epoch": 8.266962989840348,
      "grad_norm": 1.453125,
      "learning_rate": 8.309090362273189e-05,
      "loss": 4.7474,
      "mean_token_accuracy": 0.24272837042808532,
      "num_tokens": 189106682.0,
      "step": 91135
    },
    {
      "entropy": 5.699005937576294,
      "epoch": 8.267416545718433,
      "grad_norm": 1.5625,
      "learning_rate": 8.307401470126439e-05,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23474827408790588,
      "num_tokens": 189116294.0,
      "step": 91140
    },
    {
      "entropy": 5.660797643661499,
      "epoch": 8.267870101596516,
      "grad_norm": 1.515625,
      "learning_rate": 8.305712974887508e-05,
      "loss": 4.738,
      "mean_token_accuracy": 0.23638831228017806,
      "num_tokens": 189126528.0,
      "step": 91145
    },
    {
      "entropy": 5.6780167579650875,
      "epoch": 8.268323657474602,
      "grad_norm": 1.5625,
      "learning_rate": 8.304024876591332e-05,
      "loss": 4.8012,
      "mean_token_accuracy": 0.23904983848333358,
      "num_tokens": 189136823.0,
      "step": 91150
    },
    {
      "entropy": 5.665111446380616,
      "epoch": 8.268777213352685,
      "grad_norm": 1.6015625,
      "learning_rate": 8.302337175272815e-05,
      "loss": 4.7292,
      "mean_token_accuracy": 0.24108168482780457,
      "num_tokens": 189147427.0,
      "step": 91155
    },
    {
      "entropy": 5.633035373687744,
      "epoch": 8.26923076923077,
      "grad_norm": 1.4765625,
      "learning_rate": 8.300649870966854e-05,
      "loss": 4.7001,
      "mean_token_accuracy": 0.243505822122097,
      "num_tokens": 189157273.0,
      "step": 91160
    },
    {
      "entropy": 5.6729286193847654,
      "epoch": 8.269684325108853,
      "grad_norm": 1.5859375,
      "learning_rate": 8.298962963708347e-05,
      "loss": 4.7969,
      "mean_token_accuracy": 0.23705631494522095,
      "num_tokens": 189167588.0,
      "step": 91165
    },
    {
      "entropy": 5.664297580718994,
      "epoch": 8.270137880986937,
      "grad_norm": 1.5,
      "learning_rate": 8.297276453532182e-05,
      "loss": 4.7888,
      "mean_token_accuracy": 0.23419599533081054,
      "num_tokens": 189179007.0,
      "step": 91170
    },
    {
      "entropy": 5.666149377822876,
      "epoch": 8.270591436865022,
      "grad_norm": 1.5390625,
      "learning_rate": 8.29559034047323e-05,
      "loss": 4.7657,
      "mean_token_accuracy": 0.24655920565128325,
      "num_tokens": 189189985.0,
      "step": 91175
    },
    {
      "entropy": 5.711884498596191,
      "epoch": 8.271044992743105,
      "grad_norm": 1.578125,
      "learning_rate": 8.293904624566359e-05,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23051103353500366,
      "num_tokens": 189200691.0,
      "step": 91180
    },
    {
      "entropy": 5.661417627334595,
      "epoch": 8.27149854862119,
      "grad_norm": 1.375,
      "learning_rate": 8.29221930584644e-05,
      "loss": 4.8269,
      "mean_token_accuracy": 0.23124615997076034,
      "num_tokens": 189212746.0,
      "step": 91185
    },
    {
      "entropy": 5.638424253463745,
      "epoch": 8.271952104499274,
      "grad_norm": 1.5234375,
      "learning_rate": 8.29053438434832e-05,
      "loss": 4.7484,
      "mean_token_accuracy": 0.2541544303297997,
      "num_tokens": 189222934.0,
      "step": 91190
    },
    {
      "entropy": 5.653606700897217,
      "epoch": 8.272405660377359,
      "grad_norm": 1.4609375,
      "learning_rate": 8.288849860106846e-05,
      "loss": 4.7149,
      "mean_token_accuracy": 0.24383406341075897,
      "num_tokens": 189233445.0,
      "step": 91195
    },
    {
      "entropy": 5.61715612411499,
      "epoch": 8.272859216255442,
      "grad_norm": 1.453125,
      "learning_rate": 8.28716573315685e-05,
      "loss": 4.7666,
      "mean_token_accuracy": 0.24111021608114241,
      "num_tokens": 189244979.0,
      "step": 91200
    },
    {
      "entropy": 5.682700872421265,
      "epoch": 8.273312772133528,
      "grad_norm": 1.6796875,
      "learning_rate": 8.285482003533162e-05,
      "loss": 4.8237,
      "mean_token_accuracy": 0.24117216765880584,
      "num_tokens": 189255122.0,
      "step": 91205
    },
    {
      "entropy": 5.701036691665649,
      "epoch": 8.273766328011611,
      "grad_norm": 1.5859375,
      "learning_rate": 8.283798671270601e-05,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23829482346773148,
      "num_tokens": 189265512.0,
      "step": 91210
    },
    {
      "entropy": 5.652448272705078,
      "epoch": 8.274219883889696,
      "grad_norm": 1.484375,
      "learning_rate": 8.282115736403983e-05,
      "loss": 4.7723,
      "mean_token_accuracy": 0.24222447574138642,
      "num_tokens": 189276373.0,
      "step": 91215
    },
    {
      "entropy": 5.55056939125061,
      "epoch": 8.27467343976778,
      "grad_norm": 1.421875,
      "learning_rate": 8.28043319896811e-05,
      "loss": 4.7022,
      "mean_token_accuracy": 0.24007759988307953,
      "num_tokens": 189286989.0,
      "step": 91220
    },
    {
      "entropy": 5.662331199645996,
      "epoch": 8.275126995645863,
      "grad_norm": 1.4453125,
      "learning_rate": 8.278751058997777e-05,
      "loss": 4.7852,
      "mean_token_accuracy": 0.23369783759117127,
      "num_tokens": 189296825.0,
      "step": 91225
    },
    {
      "entropy": 5.720290851593018,
      "epoch": 8.275580551523948,
      "grad_norm": 1.5625,
      "learning_rate": 8.27706931652777e-05,
      "loss": 4.774,
      "mean_token_accuracy": 0.24735423773527146,
      "num_tokens": 189307683.0,
      "step": 91230
    },
    {
      "entropy": 5.709004211425781,
      "epoch": 8.276034107402031,
      "grad_norm": 1.5703125,
      "learning_rate": 8.275387971592866e-05,
      "loss": 4.7918,
      "mean_token_accuracy": 0.24456779956817626,
      "num_tokens": 189318356.0,
      "step": 91235
    },
    {
      "entropy": 5.7114012241363525,
      "epoch": 8.276487663280117,
      "grad_norm": 1.6015625,
      "learning_rate": 8.273707024227839e-05,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22712286114692687,
      "num_tokens": 189328646.0,
      "step": 91240
    },
    {
      "entropy": 5.709577941894532,
      "epoch": 8.2769412191582,
      "grad_norm": 1.53125,
      "learning_rate": 8.272026474467445e-05,
      "loss": 4.8254,
      "mean_token_accuracy": 0.233471816778183,
      "num_tokens": 189338616.0,
      "step": 91245
    },
    {
      "entropy": 5.6288166522979735,
      "epoch": 8.277394775036285,
      "grad_norm": 1.4921875,
      "learning_rate": 8.270346322346448e-05,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2426729068160057,
      "num_tokens": 189349151.0,
      "step": 91250
    },
    {
      "entropy": 5.677606725692749,
      "epoch": 8.277848330914368,
      "grad_norm": 1.375,
      "learning_rate": 8.268666567899588e-05,
      "loss": 4.8081,
      "mean_token_accuracy": 0.23195482045412064,
      "num_tokens": 189360305.0,
      "step": 91255
    },
    {
      "entropy": 5.61046199798584,
      "epoch": 8.278301886792454,
      "grad_norm": 1.5234375,
      "learning_rate": 8.266987211161607e-05,
      "loss": 4.6944,
      "mean_token_accuracy": 0.24619982689619063,
      "num_tokens": 189370575.0,
      "step": 91260
    },
    {
      "entropy": 5.672236108779908,
      "epoch": 8.278755442670537,
      "grad_norm": 1.515625,
      "learning_rate": 8.265308252167231e-05,
      "loss": 4.7924,
      "mean_token_accuracy": 0.23242373615503312,
      "num_tokens": 189381006.0,
      "step": 91265
    },
    {
      "entropy": 5.748619604110718,
      "epoch": 8.27920899854862,
      "grad_norm": 1.8125,
      "learning_rate": 8.26362969095118e-05,
      "loss": 4.8767,
      "mean_token_accuracy": 0.23340865075588227,
      "num_tokens": 189392567.0,
      "step": 91270
    },
    {
      "entropy": 5.702742195129394,
      "epoch": 8.279662554426706,
      "grad_norm": 1.6328125,
      "learning_rate": 8.261951527548172e-05,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2371993511915207,
      "num_tokens": 189402265.0,
      "step": 91275
    },
    {
      "entropy": 5.691094255447387,
      "epoch": 8.280116110304789,
      "grad_norm": 1.4765625,
      "learning_rate": 8.260273761992902e-05,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23076874017715454,
      "num_tokens": 189412790.0,
      "step": 91280
    },
    {
      "entropy": 5.6367696762084964,
      "epoch": 8.280569666182874,
      "grad_norm": 1.4765625,
      "learning_rate": 8.258596394320078e-05,
      "loss": 4.7521,
      "mean_token_accuracy": 0.23383352011442185,
      "num_tokens": 189422752.0,
      "step": 91285
    },
    {
      "entropy": 5.60863847732544,
      "epoch": 8.281023222060957,
      "grad_norm": 1.546875,
      "learning_rate": 8.256919424564385e-05,
      "loss": 4.7112,
      "mean_token_accuracy": 0.2404153048992157,
      "num_tokens": 189433254.0,
      "step": 91290
    },
    {
      "entropy": 5.688035678863526,
      "epoch": 8.281476777939043,
      "grad_norm": 1.4296875,
      "learning_rate": 8.255242852760503e-05,
      "loss": 4.7991,
      "mean_token_accuracy": 0.24310170263051986,
      "num_tokens": 189444000.0,
      "step": 91295
    },
    {
      "entropy": 5.6335193634033205,
      "epoch": 8.281930333817126,
      "grad_norm": 1.390625,
      "learning_rate": 8.253566678943101e-05,
      "loss": 4.789,
      "mean_token_accuracy": 0.23475051671266556,
      "num_tokens": 189455923.0,
      "step": 91300
    },
    {
      "entropy": 5.718120336532593,
      "epoch": 8.282383889695211,
      "grad_norm": 1.4921875,
      "learning_rate": 8.251890903146845e-05,
      "loss": 4.87,
      "mean_token_accuracy": 0.2396745502948761,
      "num_tokens": 189466634.0,
      "step": 91305
    },
    {
      "entropy": 5.729834842681885,
      "epoch": 8.282837445573294,
      "grad_norm": 1.4375,
      "learning_rate": 8.250215525406391e-05,
      "loss": 4.8296,
      "mean_token_accuracy": 0.2366866648197174,
      "num_tokens": 189478320.0,
      "step": 91310
    },
    {
      "entropy": 5.689100217819214,
      "epoch": 8.28329100145138,
      "grad_norm": 1.4765625,
      "learning_rate": 8.24854054575638e-05,
      "loss": 4.7465,
      "mean_token_accuracy": 0.23936735689640046,
      "num_tokens": 189488466.0,
      "step": 91315
    },
    {
      "entropy": 5.712624549865723,
      "epoch": 8.283744557329463,
      "grad_norm": 1.4375,
      "learning_rate": 8.246865964231459e-05,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23833743184804917,
      "num_tokens": 189499226.0,
      "step": 91320
    },
    {
      "entropy": 5.702564239501953,
      "epoch": 8.284198113207546,
      "grad_norm": 1.453125,
      "learning_rate": 8.24519178086626e-05,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2364235669374466,
      "num_tokens": 189510531.0,
      "step": 91325
    },
    {
      "entropy": 5.721005487442016,
      "epoch": 8.284651669085632,
      "grad_norm": 1.484375,
      "learning_rate": 8.243517995695401e-05,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23581395149230958,
      "num_tokens": 189521789.0,
      "step": 91330
    },
    {
      "entropy": 5.708850383758545,
      "epoch": 8.285105224963715,
      "grad_norm": 1.5703125,
      "learning_rate": 8.241844608753494e-05,
      "loss": 4.8303,
      "mean_token_accuracy": 0.23872508406639098,
      "num_tokens": 189531716.0,
      "step": 91335
    },
    {
      "entropy": 5.670679235458374,
      "epoch": 8.2855587808418,
      "grad_norm": 1.4921875,
      "learning_rate": 8.240171620075148e-05,
      "loss": 4.8058,
      "mean_token_accuracy": 0.23411405086517334,
      "num_tokens": 189542678.0,
      "step": 91340
    },
    {
      "entropy": 5.7177512645721436,
      "epoch": 8.286012336719883,
      "grad_norm": 1.515625,
      "learning_rate": 8.238499029694966e-05,
      "loss": 4.8391,
      "mean_token_accuracy": 0.24208205044269562,
      "num_tokens": 189552219.0,
      "step": 91345
    },
    {
      "entropy": 5.611488151550293,
      "epoch": 8.286465892597969,
      "grad_norm": 1.6015625,
      "learning_rate": 8.236826837647524e-05,
      "loss": 4.7163,
      "mean_token_accuracy": 0.23780472576618195,
      "num_tokens": 189562373.0,
      "step": 91350
    },
    {
      "entropy": 5.672631502151489,
      "epoch": 8.286919448476052,
      "grad_norm": 1.671875,
      "learning_rate": 8.235155043967419e-05,
      "loss": 4.7882,
      "mean_token_accuracy": 0.24001587182283401,
      "num_tokens": 189571767.0,
      "step": 91355
    },
    {
      "entropy": 5.6875648021698,
      "epoch": 8.287373004354137,
      "grad_norm": 1.6640625,
      "learning_rate": 8.233483648689214e-05,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23812215477228166,
      "num_tokens": 189581282.0,
      "step": 91360
    },
    {
      "entropy": 5.633417463302612,
      "epoch": 8.28782656023222,
      "grad_norm": 1.5625,
      "learning_rate": 8.231812651847477e-05,
      "loss": 4.7664,
      "mean_token_accuracy": 0.23806109577417373,
      "num_tokens": 189591438.0,
      "step": 91365
    },
    {
      "entropy": 5.772697877883911,
      "epoch": 8.288280116110304,
      "grad_norm": 1.609375,
      "learning_rate": 8.230142053476766e-05,
      "loss": 4.9088,
      "mean_token_accuracy": 0.22837339490652084,
      "num_tokens": 189602496.0,
      "step": 91370
    },
    {
      "entropy": 5.742119646072387,
      "epoch": 8.288733671988389,
      "grad_norm": 1.4453125,
      "learning_rate": 8.228471853611627e-05,
      "loss": 4.8385,
      "mean_token_accuracy": 0.2406722366809845,
      "num_tokens": 189613198.0,
      "step": 91375
    },
    {
      "entropy": 5.7164387702941895,
      "epoch": 8.289187227866472,
      "grad_norm": 1.65625,
      "learning_rate": 8.226802052286603e-05,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23862916678190232,
      "num_tokens": 189623713.0,
      "step": 91380
    },
    {
      "entropy": 5.7487226009368895,
      "epoch": 8.289640783744558,
      "grad_norm": 1.5078125,
      "learning_rate": 8.225132649536215e-05,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23983938843011857,
      "num_tokens": 189634933.0,
      "step": 91385
    },
    {
      "entropy": 5.733004426956176,
      "epoch": 8.290094339622641,
      "grad_norm": 1.625,
      "learning_rate": 8.223463645395005e-05,
      "loss": 4.7881,
      "mean_token_accuracy": 0.2411866620182991,
      "num_tokens": 189645381.0,
      "step": 91390
    },
    {
      "entropy": 5.731591606140137,
      "epoch": 8.290547895500726,
      "grad_norm": 1.4375,
      "learning_rate": 8.221795039897476e-05,
      "loss": 4.828,
      "mean_token_accuracy": 0.23020764589309692,
      "num_tokens": 189656782.0,
      "step": 91395
    },
    {
      "entropy": 5.64011435508728,
      "epoch": 8.29100145137881,
      "grad_norm": 1.4375,
      "learning_rate": 8.22012683307814e-05,
      "loss": 4.7439,
      "mean_token_accuracy": 0.23596824258565902,
      "num_tokens": 189667183.0,
      "step": 91400
    },
    {
      "entropy": 5.679515361785889,
      "epoch": 8.291455007256895,
      "grad_norm": 1.515625,
      "learning_rate": 8.218459024971492e-05,
      "loss": 4.7258,
      "mean_token_accuracy": 0.2493066519498825,
      "num_tokens": 189677336.0,
      "step": 91405
    },
    {
      "entropy": 5.684364700317383,
      "epoch": 8.291908563134978,
      "grad_norm": 1.5078125,
      "learning_rate": 8.216791615612026e-05,
      "loss": 4.7874,
      "mean_token_accuracy": 0.24481741338968277,
      "num_tokens": 189687753.0,
      "step": 91410
    },
    {
      "entropy": 5.721015310287475,
      "epoch": 8.292362119013063,
      "grad_norm": 1.453125,
      "learning_rate": 8.215124605034224e-05,
      "loss": 4.7933,
      "mean_token_accuracy": 0.23677874803543092,
      "num_tokens": 189697599.0,
      "step": 91415
    },
    {
      "entropy": 5.670612382888794,
      "epoch": 8.292815674891147,
      "grad_norm": 1.53125,
      "learning_rate": 8.21345799327256e-05,
      "loss": 4.7939,
      "mean_token_accuracy": 0.23864287585020066,
      "num_tokens": 189708811.0,
      "step": 91420
    },
    {
      "entropy": 5.6615478515625,
      "epoch": 8.29326923076923,
      "grad_norm": 1.5859375,
      "learning_rate": 8.2117917803615e-05,
      "loss": 4.7551,
      "mean_token_accuracy": 0.2433062493801117,
      "num_tokens": 189719467.0,
      "step": 91425
    },
    {
      "entropy": 5.7621828556060795,
      "epoch": 8.293722786647315,
      "grad_norm": 1.4375,
      "learning_rate": 8.210125966335499e-05,
      "loss": 4.857,
      "mean_token_accuracy": 0.22767778038978576,
      "num_tokens": 189730249.0,
      "step": 91430
    },
    {
      "entropy": 5.64457335472107,
      "epoch": 8.294176342525398,
      "grad_norm": 1.53125,
      "learning_rate": 8.208460551229011e-05,
      "loss": 4.7155,
      "mean_token_accuracy": 0.2451528415083885,
      "num_tokens": 189739766.0,
      "step": 91435
    },
    {
      "entropy": 5.6550205707550045,
      "epoch": 8.294629898403484,
      "grad_norm": 1.4765625,
      "learning_rate": 8.206795535076472e-05,
      "loss": 4.7692,
      "mean_token_accuracy": 0.24460023045539855,
      "num_tokens": 189750220.0,
      "step": 91440
    },
    {
      "entropy": 5.708153581619262,
      "epoch": 8.295083454281567,
      "grad_norm": 1.5859375,
      "learning_rate": 8.205130917912316e-05,
      "loss": 4.7793,
      "mean_token_accuracy": 0.24181172996759415,
      "num_tokens": 189760371.0,
      "step": 91445
    },
    {
      "entropy": 5.675099325180054,
      "epoch": 8.295537010159652,
      "grad_norm": 1.484375,
      "learning_rate": 8.20346669977097e-05,
      "loss": 4.8006,
      "mean_token_accuracy": 0.24025055319070815,
      "num_tokens": 189770326.0,
      "step": 91450
    },
    {
      "entropy": 5.683142805099488,
      "epoch": 8.295990566037736,
      "grad_norm": 1.5390625,
      "learning_rate": 8.201802880686854e-05,
      "loss": 4.7654,
      "mean_token_accuracy": 0.23959133625030518,
      "num_tokens": 189780733.0,
      "step": 91455
    },
    {
      "entropy": 5.705726528167725,
      "epoch": 8.29644412191582,
      "grad_norm": 1.5859375,
      "learning_rate": 8.200139460694372e-05,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23478049039840698,
      "num_tokens": 189792003.0,
      "step": 91460
    },
    {
      "entropy": 5.779649496078491,
      "epoch": 8.296897677793904,
      "grad_norm": 1.5546875,
      "learning_rate": 8.198476439827923e-05,
      "loss": 4.8734,
      "mean_token_accuracy": 0.23393121510744094,
      "num_tokens": 189802402.0,
      "step": 91465
    },
    {
      "entropy": 5.702347183227539,
      "epoch": 8.29735123367199,
      "grad_norm": 1.5703125,
      "learning_rate": 8.1968138181219e-05,
      "loss": 4.7842,
      "mean_token_accuracy": 0.23702313899993896,
      "num_tokens": 189811761.0,
      "step": 91470
    },
    {
      "entropy": 5.6976659297943115,
      "epoch": 8.297804789550073,
      "grad_norm": 1.578125,
      "learning_rate": 8.195151595610686e-05,
      "loss": 4.8527,
      "mean_token_accuracy": 0.22957748174667358,
      "num_tokens": 189822061.0,
      "step": 91475
    },
    {
      "entropy": 5.672809839248657,
      "epoch": 8.298258345428156,
      "grad_norm": 1.5234375,
      "learning_rate": 8.193489772328653e-05,
      "loss": 4.8345,
      "mean_token_accuracy": 0.23928043991327286,
      "num_tokens": 189831627.0,
      "step": 91480
    },
    {
      "entropy": 5.693524599075317,
      "epoch": 8.298711901306241,
      "grad_norm": 1.4140625,
      "learning_rate": 8.191828348310174e-05,
      "loss": 4.8033,
      "mean_token_accuracy": 0.23612256348133087,
      "num_tokens": 189843036.0,
      "step": 91485
    },
    {
      "entropy": 5.765378856658936,
      "epoch": 8.299165457184325,
      "grad_norm": 1.5859375,
      "learning_rate": 8.190167323589604e-05,
      "loss": 4.8759,
      "mean_token_accuracy": 0.2377047762274742,
      "num_tokens": 189853691.0,
      "step": 91490
    },
    {
      "entropy": 5.717033004760742,
      "epoch": 8.29961901306241,
      "grad_norm": 1.484375,
      "learning_rate": 8.188506698201297e-05,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23923903852701187,
      "num_tokens": 189864392.0,
      "step": 91495
    },
    {
      "entropy": 5.729044580459595,
      "epoch": 8.300072568940493,
      "grad_norm": 1.5625,
      "learning_rate": 8.186846472179593e-05,
      "loss": 4.8671,
      "mean_token_accuracy": 0.2322054997086525,
      "num_tokens": 189874965.0,
      "step": 91500
    },
    {
      "entropy": 5.674850225448608,
      "epoch": 8.300526124818578,
      "grad_norm": 1.5625,
      "learning_rate": 8.185186645558825e-05,
      "loss": 4.7515,
      "mean_token_accuracy": 0.23550060242414475,
      "num_tokens": 189884790.0,
      "step": 91505
    },
    {
      "entropy": 5.703609752655029,
      "epoch": 8.300979680696662,
      "grad_norm": 1.703125,
      "learning_rate": 8.183527218373319e-05,
      "loss": 4.7768,
      "mean_token_accuracy": 0.23781219571828843,
      "num_tokens": 189893980.0,
      "step": 91510
    },
    {
      "entropy": 5.653223180770874,
      "epoch": 8.301433236574747,
      "grad_norm": 1.4609375,
      "learning_rate": 8.181868190657387e-05,
      "loss": 4.7859,
      "mean_token_accuracy": 0.2282302126288414,
      "num_tokens": 189905088.0,
      "step": 91515
    },
    {
      "entropy": 5.698764848709106,
      "epoch": 8.30188679245283,
      "grad_norm": 1.40625,
      "learning_rate": 8.180209562445349e-05,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23361641615629197,
      "num_tokens": 189915134.0,
      "step": 91520
    },
    {
      "entropy": 5.641634511947632,
      "epoch": 8.302340348330915,
      "grad_norm": 1.375,
      "learning_rate": 8.178551333771502e-05,
      "loss": 4.7933,
      "mean_token_accuracy": 0.23234456777572632,
      "num_tokens": 189925480.0,
      "step": 91525
    },
    {
      "entropy": 5.678431463241577,
      "epoch": 8.302793904208999,
      "grad_norm": 1.5,
      "learning_rate": 8.176893504670136e-05,
      "loss": 4.7835,
      "mean_token_accuracy": 0.24756639897823335,
      "num_tokens": 189936056.0,
      "step": 91530
    },
    {
      "entropy": 5.66237244606018,
      "epoch": 8.303247460087082,
      "grad_norm": 1.6015625,
      "learning_rate": 8.175236075175534e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.2382289469242096,
      "num_tokens": 189946379.0,
      "step": 91535
    },
    {
      "entropy": 5.684008646011352,
      "epoch": 8.303701015965167,
      "grad_norm": 1.5625,
      "learning_rate": 8.173579045321977e-05,
      "loss": 4.8541,
      "mean_token_accuracy": 0.2296686977148056,
      "num_tokens": 189956788.0,
      "step": 91540
    },
    {
      "entropy": 5.72836856842041,
      "epoch": 8.30415457184325,
      "grad_norm": 1.5625,
      "learning_rate": 8.171922415143726e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.23313085734844208,
      "num_tokens": 189967450.0,
      "step": 91545
    },
    {
      "entropy": 5.6760108947753904,
      "epoch": 8.304608127721336,
      "grad_norm": 1.515625,
      "learning_rate": 8.170266184675049e-05,
      "loss": 4.7615,
      "mean_token_accuracy": 0.23439770936965942,
      "num_tokens": 189978885.0,
      "step": 91550
    },
    {
      "entropy": 5.638380146026611,
      "epoch": 8.305061683599419,
      "grad_norm": 1.5078125,
      "learning_rate": 8.168610353950195e-05,
      "loss": 4.7118,
      "mean_token_accuracy": 0.24343233108520507,
      "num_tokens": 189989467.0,
      "step": 91555
    },
    {
      "entropy": 5.718387126922607,
      "epoch": 8.305515239477504,
      "grad_norm": 1.6796875,
      "learning_rate": 8.166954923003403e-05,
      "loss": 4.7964,
      "mean_token_accuracy": 0.23331850916147232,
      "num_tokens": 190000007.0,
      "step": 91560
    },
    {
      "entropy": 5.678527688980102,
      "epoch": 8.305968795355588,
      "grad_norm": 1.53125,
      "learning_rate": 8.165299891868911e-05,
      "loss": 4.7563,
      "mean_token_accuracy": 0.2420475661754608,
      "num_tokens": 190009865.0,
      "step": 91565
    },
    {
      "entropy": 5.736617946624756,
      "epoch": 8.306422351233673,
      "grad_norm": 1.4140625,
      "learning_rate": 8.16364526058094e-05,
      "loss": 4.8288,
      "mean_token_accuracy": 0.2385522410273552,
      "num_tokens": 190022442.0,
      "step": 91570
    },
    {
      "entropy": 5.755638647079468,
      "epoch": 8.306875907111756,
      "grad_norm": 1.625,
      "learning_rate": 8.161991029173718e-05,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22584497630596162,
      "num_tokens": 190033395.0,
      "step": 91575
    },
    {
      "entropy": 5.740290641784668,
      "epoch": 8.30732946298984,
      "grad_norm": 1.453125,
      "learning_rate": 8.160337197681442e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2399817019701004,
      "num_tokens": 190043207.0,
      "step": 91580
    },
    {
      "entropy": 5.706753730773926,
      "epoch": 8.307783018867925,
      "grad_norm": 1.5703125,
      "learning_rate": 8.158683766138325e-05,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22962946742773055,
      "num_tokens": 190053320.0,
      "step": 91585
    },
    {
      "entropy": 5.728376245498657,
      "epoch": 8.308236574746008,
      "grad_norm": 1.609375,
      "learning_rate": 8.157030734578559e-05,
      "loss": 4.8344,
      "mean_token_accuracy": 0.2294747292995453,
      "num_tokens": 190063261.0,
      "step": 91590
    },
    {
      "entropy": 5.689309978485108,
      "epoch": 8.308690130624093,
      "grad_norm": 1.3984375,
      "learning_rate": 8.155378103036324e-05,
      "loss": 4.6742,
      "mean_token_accuracy": 0.24593827426433562,
      "num_tokens": 190074016.0,
      "step": 91595
    },
    {
      "entropy": 5.670113229751587,
      "epoch": 8.309143686502177,
      "grad_norm": 1.625,
      "learning_rate": 8.153725871545801e-05,
      "loss": 4.7749,
      "mean_token_accuracy": 0.2472894459962845,
      "num_tokens": 190084448.0,
      "step": 91600
    },
    {
      "entropy": 5.6930989742279055,
      "epoch": 8.309597242380262,
      "grad_norm": 1.484375,
      "learning_rate": 8.152074040141159e-05,
      "loss": 4.854,
      "mean_token_accuracy": 0.23591738045215607,
      "num_tokens": 190094465.0,
      "step": 91605
    },
    {
      "entropy": 5.691711139678955,
      "epoch": 8.310050798258345,
      "grad_norm": 1.6875,
      "learning_rate": 8.150422608856554e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.23558350205421447,
      "num_tokens": 190105806.0,
      "step": 91610
    },
    {
      "entropy": 5.731919956207276,
      "epoch": 8.31050435413643,
      "grad_norm": 1.53125,
      "learning_rate": 8.148771577726133e-05,
      "loss": 4.7841,
      "mean_token_accuracy": 0.23956916779279708,
      "num_tokens": 190115927.0,
      "step": 91615
    },
    {
      "entropy": 5.623065376281739,
      "epoch": 8.310957910014514,
      "grad_norm": 1.578125,
      "learning_rate": 8.147120946784054e-05,
      "loss": 4.7829,
      "mean_token_accuracy": 0.22882957607507706,
      "num_tokens": 190126042.0,
      "step": 91620
    },
    {
      "entropy": 5.724824333190918,
      "epoch": 8.311411465892599,
      "grad_norm": 1.5625,
      "learning_rate": 8.145470716064448e-05,
      "loss": 4.8356,
      "mean_token_accuracy": 0.2310678854584694,
      "num_tokens": 190135604.0,
      "step": 91625
    },
    {
      "entropy": 5.6413815975189205,
      "epoch": 8.311865021770682,
      "grad_norm": 1.546875,
      "learning_rate": 8.143820885601435e-05,
      "loss": 4.7551,
      "mean_token_accuracy": 0.24400605857372284,
      "num_tokens": 190145874.0,
      "step": 91630
    },
    {
      "entropy": 5.729534578323364,
      "epoch": 8.312318577648766,
      "grad_norm": 1.4765625,
      "learning_rate": 8.142171455429142e-05,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2444741353392601,
      "num_tokens": 190155673.0,
      "step": 91635
    },
    {
      "entropy": 5.703633403778076,
      "epoch": 8.31277213352685,
      "grad_norm": 1.53125,
      "learning_rate": 8.140522425581678e-05,
      "loss": 4.7986,
      "mean_token_accuracy": 0.23875530511140824,
      "num_tokens": 190165972.0,
      "step": 91640
    },
    {
      "entropy": 5.765063905715943,
      "epoch": 8.313225689404934,
      "grad_norm": 1.71875,
      "learning_rate": 8.138873796093138e-05,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23679169565439223,
      "num_tokens": 190176716.0,
      "step": 91645
    },
    {
      "entropy": 5.639843273162842,
      "epoch": 8.31367924528302,
      "grad_norm": 1.6328125,
      "learning_rate": 8.137225566997622e-05,
      "loss": 4.7167,
      "mean_token_accuracy": 0.2398155927658081,
      "num_tokens": 190187418.0,
      "step": 91650
    },
    {
      "entropy": 5.754892349243164,
      "epoch": 8.314132801161103,
      "grad_norm": 1.4140625,
      "learning_rate": 8.135577738329216e-05,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22912233471870422,
      "num_tokens": 190198552.0,
      "step": 91655
    },
    {
      "entropy": 5.730822563171387,
      "epoch": 8.314586357039188,
      "grad_norm": 1.8203125,
      "learning_rate": 8.133930310122001e-05,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23914394229650499,
      "num_tokens": 190208790.0,
      "step": 91660
    },
    {
      "entropy": 5.6848742961883545,
      "epoch": 8.315039912917271,
      "grad_norm": 1.65625,
      "learning_rate": 8.132283282410039e-05,
      "loss": 4.7517,
      "mean_token_accuracy": 0.2493545189499855,
      "num_tokens": 190218039.0,
      "step": 91665
    },
    {
      "entropy": 5.75002760887146,
      "epoch": 8.315493468795356,
      "grad_norm": 1.59375,
      "learning_rate": 8.130636655227396e-05,
      "loss": 4.8155,
      "mean_token_accuracy": 0.22982093095779418,
      "num_tokens": 190229024.0,
      "step": 91670
    },
    {
      "entropy": 5.6409564971923825,
      "epoch": 8.31594702467344,
      "grad_norm": 1.53125,
      "learning_rate": 8.128990428608124e-05,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2344442218542099,
      "num_tokens": 190238593.0,
      "step": 91675
    },
    {
      "entropy": 5.664513874053955,
      "epoch": 8.316400580551523,
      "grad_norm": 1.640625,
      "learning_rate": 8.127344602586268e-05,
      "loss": 4.8288,
      "mean_token_accuracy": 0.2382392942905426,
      "num_tokens": 190249813.0,
      "step": 91680
    },
    {
      "entropy": 5.772431707382202,
      "epoch": 8.316854136429608,
      "grad_norm": 1.625,
      "learning_rate": 8.125699177195853e-05,
      "loss": 4.8714,
      "mean_token_accuracy": 0.2296572893857956,
      "num_tokens": 190259548.0,
      "step": 91685
    },
    {
      "entropy": 5.686837387084961,
      "epoch": 8.317307692307692,
      "grad_norm": 1.46875,
      "learning_rate": 8.12405415247092e-05,
      "loss": 4.8019,
      "mean_token_accuracy": 0.23731733709573746,
      "num_tokens": 190270125.0,
      "step": 91690
    },
    {
      "entropy": 5.671729898452758,
      "epoch": 8.317761248185777,
      "grad_norm": 1.5859375,
      "learning_rate": 8.122409528445488e-05,
      "loss": 4.7216,
      "mean_token_accuracy": 0.24675049930810927,
      "num_tokens": 190280818.0,
      "step": 91695
    },
    {
      "entropy": 5.681562805175782,
      "epoch": 8.31821480406386,
      "grad_norm": 1.6015625,
      "learning_rate": 8.120765305153563e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.23885756880044937,
      "num_tokens": 190290418.0,
      "step": 91700
    },
    {
      "entropy": 5.639808416366577,
      "epoch": 8.318668359941945,
      "grad_norm": 1.59375,
      "learning_rate": 8.119121482629154e-05,
      "loss": 4.7134,
      "mean_token_accuracy": 0.23810600042343139,
      "num_tokens": 190300957.0,
      "step": 91705
    },
    {
      "entropy": 5.678966903686524,
      "epoch": 8.319121915820029,
      "grad_norm": 1.7265625,
      "learning_rate": 8.117478060906246e-05,
      "loss": 4.81,
      "mean_token_accuracy": 0.23634969145059587,
      "num_tokens": 190310662.0,
      "step": 91710
    },
    {
      "entropy": 5.633798837661743,
      "epoch": 8.319575471698114,
      "grad_norm": 1.5859375,
      "learning_rate": 8.115835040018835e-05,
      "loss": 4.7322,
      "mean_token_accuracy": 0.23844640851020812,
      "num_tokens": 190320768.0,
      "step": 91715
    },
    {
      "entropy": 5.747520160675049,
      "epoch": 8.320029027576197,
      "grad_norm": 1.484375,
      "learning_rate": 8.114192420000888e-05,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2289664030075073,
      "num_tokens": 190332296.0,
      "step": 91720
    },
    {
      "entropy": 5.7300656795501705,
      "epoch": 8.320482583454282,
      "grad_norm": 1.53125,
      "learning_rate": 8.112550200886387e-05,
      "loss": 4.8271,
      "mean_token_accuracy": 0.237321574985981,
      "num_tokens": 190341911.0,
      "step": 91725
    },
    {
      "entropy": 5.648892736434936,
      "epoch": 8.320936139332366,
      "grad_norm": 1.4375,
      "learning_rate": 8.110908382709289e-05,
      "loss": 4.812,
      "mean_token_accuracy": 0.2343614086508751,
      "num_tokens": 190352764.0,
      "step": 91730
    },
    {
      "entropy": 5.755800104141235,
      "epoch": 8.32138969521045,
      "grad_norm": 1.59375,
      "learning_rate": 8.109266965503547e-05,
      "loss": 4.854,
      "mean_token_accuracy": 0.2376081019639969,
      "num_tokens": 190362544.0,
      "step": 91735
    },
    {
      "entropy": 5.64747314453125,
      "epoch": 8.321843251088534,
      "grad_norm": 1.484375,
      "learning_rate": 8.107625949303105e-05,
      "loss": 4.7453,
      "mean_token_accuracy": 0.23849257826805115,
      "num_tokens": 190372295.0,
      "step": 91740
    },
    {
      "entropy": 5.632679414749146,
      "epoch": 8.322296806966618,
      "grad_norm": 1.484375,
      "learning_rate": 8.1059853341419e-05,
      "loss": 4.733,
      "mean_token_accuracy": 0.24440272897481918,
      "num_tokens": 190382241.0,
      "step": 91745
    },
    {
      "entropy": 5.668793773651123,
      "epoch": 8.322750362844703,
      "grad_norm": 1.4921875,
      "learning_rate": 8.104345120053858e-05,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23117756545543672,
      "num_tokens": 190392064.0,
      "step": 91750
    },
    {
      "entropy": 5.739709711074829,
      "epoch": 8.323203918722786,
      "grad_norm": 1.4921875,
      "learning_rate": 8.102705307072903e-05,
      "loss": 4.7962,
      "mean_token_accuracy": 0.24141203761100768,
      "num_tokens": 190402297.0,
      "step": 91755
    },
    {
      "entropy": 5.7139544010162355,
      "epoch": 8.323657474600871,
      "grad_norm": 1.4140625,
      "learning_rate": 8.101065895232945e-05,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23189535439014436,
      "num_tokens": 190412579.0,
      "step": 91760
    },
    {
      "entropy": 5.68811821937561,
      "epoch": 8.324111030478955,
      "grad_norm": 1.578125,
      "learning_rate": 8.099426884567888e-05,
      "loss": 4.7947,
      "mean_token_accuracy": 0.2460833266377449,
      "num_tokens": 190422676.0,
      "step": 91765
    },
    {
      "entropy": 5.687349653244018,
      "epoch": 8.32456458635704,
      "grad_norm": 1.5859375,
      "learning_rate": 8.097788275111628e-05,
      "loss": 4.7945,
      "mean_token_accuracy": 0.23795802295207977,
      "num_tokens": 190433288.0,
      "step": 91770
    },
    {
      "entropy": 5.754226493835449,
      "epoch": 8.325018142235123,
      "grad_norm": 1.5859375,
      "learning_rate": 8.096150066898053e-05,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2296549692749977,
      "num_tokens": 190444412.0,
      "step": 91775
    },
    {
      "entropy": 5.720509195327759,
      "epoch": 8.325471698113208,
      "grad_norm": 1.71875,
      "learning_rate": 8.094512259961036e-05,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23832205981016158,
      "num_tokens": 190455074.0,
      "step": 91780
    },
    {
      "entropy": 5.688826227188111,
      "epoch": 8.325925253991292,
      "grad_norm": 1.546875,
      "learning_rate": 8.092874854334446e-05,
      "loss": 4.7803,
      "mean_token_accuracy": 0.2355889141559601,
      "num_tokens": 190464591.0,
      "step": 91785
    },
    {
      "entropy": 5.574213314056396,
      "epoch": 8.326378809869375,
      "grad_norm": 1.5546875,
      "learning_rate": 8.091237850052155e-05,
      "loss": 4.694,
      "mean_token_accuracy": 0.2447711259126663,
      "num_tokens": 190474442.0,
      "step": 91790
    },
    {
      "entropy": 5.665397119522095,
      "epoch": 8.32683236574746,
      "grad_norm": 1.515625,
      "learning_rate": 8.089601247148013e-05,
      "loss": 4.8228,
      "mean_token_accuracy": 0.233237923681736,
      "num_tokens": 190485791.0,
      "step": 91795
    },
    {
      "entropy": 5.747836971282959,
      "epoch": 8.327285921625544,
      "grad_norm": 1.65625,
      "learning_rate": 8.087965045655859e-05,
      "loss": 4.8721,
      "mean_token_accuracy": 0.2331380397081375,
      "num_tokens": 190496675.0,
      "step": 91800
    },
    {
      "entropy": 5.618617248535156,
      "epoch": 8.327739477503629,
      "grad_norm": 1.5703125,
      "learning_rate": 8.086329245609541e-05,
      "loss": 4.7194,
      "mean_token_accuracy": 0.24047164916992186,
      "num_tokens": 190506722.0,
      "step": 91805
    },
    {
      "entropy": 5.706476211547852,
      "epoch": 8.328193033381712,
      "grad_norm": 1.453125,
      "learning_rate": 8.084693847042876e-05,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23580015301704407,
      "num_tokens": 190517274.0,
      "step": 91810
    },
    {
      "entropy": 5.7589888095855715,
      "epoch": 8.328646589259797,
      "grad_norm": 1.5390625,
      "learning_rate": 8.083058849989691e-05,
      "loss": 4.8905,
      "mean_token_accuracy": 0.23750253319740294,
      "num_tokens": 190527448.0,
      "step": 91815
    },
    {
      "entropy": 5.663724327087403,
      "epoch": 8.32910014513788,
      "grad_norm": 1.5390625,
      "learning_rate": 8.081424254483792e-05,
      "loss": 4.7433,
      "mean_token_accuracy": 0.24051753133535386,
      "num_tokens": 190537211.0,
      "step": 91820
    },
    {
      "entropy": 5.725901699066162,
      "epoch": 8.329553701015966,
      "grad_norm": 1.546875,
      "learning_rate": 8.079790060558995e-05,
      "loss": 4.8755,
      "mean_token_accuracy": 0.24262404143810273,
      "num_tokens": 190547238.0,
      "step": 91825
    },
    {
      "entropy": 5.698735666275025,
      "epoch": 8.33000725689405,
      "grad_norm": 1.4921875,
      "learning_rate": 8.078156268249088e-05,
      "loss": 4.865,
      "mean_token_accuracy": 0.2350073993206024,
      "num_tokens": 190557425.0,
      "step": 91830
    },
    {
      "entropy": 5.658678197860718,
      "epoch": 8.330460812772133,
      "grad_norm": 1.6171875,
      "learning_rate": 8.076522877587855e-05,
      "loss": 4.739,
      "mean_token_accuracy": 0.2476708173751831,
      "num_tokens": 190567503.0,
      "step": 91835
    },
    {
      "entropy": 5.687074661254883,
      "epoch": 8.330914368650218,
      "grad_norm": 1.515625,
      "learning_rate": 8.074889888609081e-05,
      "loss": 4.8354,
      "mean_token_accuracy": 0.2351404219865799,
      "num_tokens": 190577557.0,
      "step": 91840
    },
    {
      "entropy": 5.657991027832031,
      "epoch": 8.331367924528301,
      "grad_norm": 1.484375,
      "learning_rate": 8.073257301346534e-05,
      "loss": 4.7547,
      "mean_token_accuracy": 0.2365456759929657,
      "num_tokens": 190587663.0,
      "step": 91845
    },
    {
      "entropy": 5.655710506439209,
      "epoch": 8.331821480406386,
      "grad_norm": 1.484375,
      "learning_rate": 8.071625115833979e-05,
      "loss": 4.7208,
      "mean_token_accuracy": 0.24522019922733307,
      "num_tokens": 190597263.0,
      "step": 91850
    },
    {
      "entropy": 5.67295093536377,
      "epoch": 8.33227503628447,
      "grad_norm": 1.53125,
      "learning_rate": 8.069993332105165e-05,
      "loss": 4.7338,
      "mean_token_accuracy": 0.23884033113718034,
      "num_tokens": 190607745.0,
      "step": 91855
    },
    {
      "entropy": 5.73189902305603,
      "epoch": 8.332728592162555,
      "grad_norm": 1.6875,
      "learning_rate": 8.06836195019384e-05,
      "loss": 4.8543,
      "mean_token_accuracy": 0.23332078456878663,
      "num_tokens": 190618143.0,
      "step": 91860
    },
    {
      "entropy": 5.672039985656738,
      "epoch": 8.333182148040638,
      "grad_norm": 1.6015625,
      "learning_rate": 8.066730970133746e-05,
      "loss": 4.7354,
      "mean_token_accuracy": 0.24520112574100494,
      "num_tokens": 190627713.0,
      "step": 91865
    },
    {
      "entropy": 5.653444433212281,
      "epoch": 8.333635703918723,
      "grad_norm": 1.4453125,
      "learning_rate": 8.065100391958602e-05,
      "loss": 4.7189,
      "mean_token_accuracy": 0.23763144314289092,
      "num_tokens": 190638351.0,
      "step": 91870
    },
    {
      "entropy": 5.696833276748658,
      "epoch": 8.334089259796807,
      "grad_norm": 1.5078125,
      "learning_rate": 8.063470215702136e-05,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23128566592931749,
      "num_tokens": 190647719.0,
      "step": 91875
    },
    {
      "entropy": 5.7203521728515625,
      "epoch": 8.334542815674892,
      "grad_norm": 1.53125,
      "learning_rate": 8.061840441398055e-05,
      "loss": 4.8951,
      "mean_token_accuracy": 0.23077562004327773,
      "num_tokens": 190659430.0,
      "step": 91880
    },
    {
      "entropy": 5.7118877410888675,
      "epoch": 8.334996371552975,
      "grad_norm": 1.59375,
      "learning_rate": 8.060211069080073e-05,
      "loss": 4.7488,
      "mean_token_accuracy": 0.2401622414588928,
      "num_tokens": 190669587.0,
      "step": 91885
    },
    {
      "entropy": 5.699384021759033,
      "epoch": 8.335449927431059,
      "grad_norm": 1.5078125,
      "learning_rate": 8.058582098781876e-05,
      "loss": 4.8484,
      "mean_token_accuracy": 0.2339882045984268,
      "num_tokens": 190680256.0,
      "step": 91890
    },
    {
      "entropy": 5.698443222045898,
      "epoch": 8.335903483309144,
      "grad_norm": 1.5625,
      "learning_rate": 8.056953530537158e-05,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2353268101811409,
      "num_tokens": 190691304.0,
      "step": 91895
    },
    {
      "entropy": 5.702356815338135,
      "epoch": 8.336357039187227,
      "grad_norm": 1.4453125,
      "learning_rate": 8.055325364379596e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.24100791066884994,
      "num_tokens": 190702520.0,
      "step": 91900
    },
    {
      "entropy": 5.619276666641236,
      "epoch": 8.336810595065312,
      "grad_norm": 1.53125,
      "learning_rate": 8.053697600342862e-05,
      "loss": 4.7123,
      "mean_token_accuracy": 0.2424662873148918,
      "num_tokens": 190712534.0,
      "step": 91905
    },
    {
      "entropy": 5.690115213394165,
      "epoch": 8.337264150943396,
      "grad_norm": 1.53125,
      "learning_rate": 8.052070238460613e-05,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23329519629478454,
      "num_tokens": 190722796.0,
      "step": 91910
    },
    {
      "entropy": 5.687467288970947,
      "epoch": 8.337717706821481,
      "grad_norm": 1.828125,
      "learning_rate": 8.050443278766505e-05,
      "loss": 4.8377,
      "mean_token_accuracy": 0.22925582826137542,
      "num_tokens": 190731855.0,
      "step": 91915
    },
    {
      "entropy": 5.698151350021362,
      "epoch": 8.338171262699564,
      "grad_norm": 1.6640625,
      "learning_rate": 8.048816721294192e-05,
      "loss": 4.7788,
      "mean_token_accuracy": 0.2394220471382141,
      "num_tokens": 190742280.0,
      "step": 91920
    },
    {
      "entropy": 5.687841844558716,
      "epoch": 8.33862481857765,
      "grad_norm": 1.4921875,
      "learning_rate": 8.047190566077306e-05,
      "loss": 4.7998,
      "mean_token_accuracy": 0.2383695736527443,
      "num_tokens": 190753120.0,
      "step": 91925
    },
    {
      "entropy": 5.653875541687012,
      "epoch": 8.339078374455733,
      "grad_norm": 1.7265625,
      "learning_rate": 8.045564813149475e-05,
      "loss": 4.7489,
      "mean_token_accuracy": 0.238041952252388,
      "num_tokens": 190763961.0,
      "step": 91930
    },
    {
      "entropy": 5.701025295257568,
      "epoch": 8.339531930333816,
      "grad_norm": 1.6484375,
      "learning_rate": 8.043939462544322e-05,
      "loss": 4.7619,
      "mean_token_accuracy": 0.23870974630117417,
      "num_tokens": 190773549.0,
      "step": 91935
    },
    {
      "entropy": 5.660439586639404,
      "epoch": 8.339985486211901,
      "grad_norm": 1.4453125,
      "learning_rate": 8.042314514295458e-05,
      "loss": 4.7295,
      "mean_token_accuracy": 0.23999673277139663,
      "num_tokens": 190784186.0,
      "step": 91940
    },
    {
      "entropy": 5.708781576156616,
      "epoch": 8.340439042089985,
      "grad_norm": 1.59375,
      "learning_rate": 8.040689968436489e-05,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23362500220537186,
      "num_tokens": 190795378.0,
      "step": 91945
    },
    {
      "entropy": 5.623921632766724,
      "epoch": 8.34089259796807,
      "grad_norm": 1.546875,
      "learning_rate": 8.039065825001006e-05,
      "loss": 4.7256,
      "mean_token_accuracy": 0.24049397110939025,
      "num_tokens": 190804573.0,
      "step": 91950
    },
    {
      "entropy": 5.747682189941406,
      "epoch": 8.341346153846153,
      "grad_norm": 1.546875,
      "learning_rate": 8.037442084022604e-05,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22737568318843843,
      "num_tokens": 190814063.0,
      "step": 91955
    },
    {
      "entropy": 5.60358304977417,
      "epoch": 8.341799709724238,
      "grad_norm": 1.5,
      "learning_rate": 8.035818745534861e-05,
      "loss": 4.6563,
      "mean_token_accuracy": 0.2491367280483246,
      "num_tokens": 190823911.0,
      "step": 91960
    },
    {
      "entropy": 5.700808906555176,
      "epoch": 8.342253265602322,
      "grad_norm": 1.40625,
      "learning_rate": 8.034195809571345e-05,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22846725583076477,
      "num_tokens": 190836174.0,
      "step": 91965
    },
    {
      "entropy": 5.696722078323364,
      "epoch": 8.342706821480407,
      "grad_norm": 1.484375,
      "learning_rate": 8.03257327616562e-05,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23368950933218002,
      "num_tokens": 190847680.0,
      "step": 91970
    },
    {
      "entropy": 5.612892580032349,
      "epoch": 8.34316037735849,
      "grad_norm": 1.5390625,
      "learning_rate": 8.030951145351243e-05,
      "loss": 4.6356,
      "mean_token_accuracy": 0.24728035032749177,
      "num_tokens": 190857287.0,
      "step": 91975
    },
    {
      "entropy": 5.649921703338623,
      "epoch": 8.343613933236576,
      "grad_norm": 1.5234375,
      "learning_rate": 8.029329417161755e-05,
      "loss": 4.7609,
      "mean_token_accuracy": 0.23638941049575807,
      "num_tokens": 190867243.0,
      "step": 91980
    },
    {
      "entropy": 5.680807781219483,
      "epoch": 8.344067489114659,
      "grad_norm": 1.546875,
      "learning_rate": 8.027708091630694e-05,
      "loss": 4.7737,
      "mean_token_accuracy": 0.24095385223627092,
      "num_tokens": 190876653.0,
      "step": 91985
    },
    {
      "entropy": 5.613863563537597,
      "epoch": 8.344521044992742,
      "grad_norm": 1.5859375,
      "learning_rate": 8.026087168791596e-05,
      "loss": 4.7225,
      "mean_token_accuracy": 0.24401870220899582,
      "num_tokens": 190886783.0,
      "step": 91990
    },
    {
      "entropy": 5.641596794128418,
      "epoch": 8.344974600870827,
      "grad_norm": 1.4140625,
      "learning_rate": 8.024466648677976e-05,
      "loss": 4.6931,
      "mean_token_accuracy": 0.24409301280975343,
      "num_tokens": 190897376.0,
      "step": 91995
    },
    {
      "entropy": 5.639432430267334,
      "epoch": 8.34542815674891,
      "grad_norm": 1.4609375,
      "learning_rate": 8.022846531323349e-05,
      "loss": 4.7736,
      "mean_token_accuracy": 0.24106374084949495,
      "num_tokens": 190908320.0,
      "step": 92000
    },
    {
      "entropy": 5.6920699119567875,
      "epoch": 8.345881712626996,
      "grad_norm": 1.4375,
      "learning_rate": 8.02122681676122e-05,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2252932757139206,
      "num_tokens": 190918952.0,
      "step": 92005
    },
    {
      "entropy": 5.71896595954895,
      "epoch": 8.34633526850508,
      "grad_norm": 1.453125,
      "learning_rate": 8.019607505025085e-05,
      "loss": 4.8837,
      "mean_token_accuracy": 0.23055660128593444,
      "num_tokens": 190928780.0,
      "step": 92010
    },
    {
      "entropy": 5.654683017730713,
      "epoch": 8.346788824383164,
      "grad_norm": 1.4609375,
      "learning_rate": 8.01798859614843e-05,
      "loss": 4.8341,
      "mean_token_accuracy": 0.2418380469083786,
      "num_tokens": 190939243.0,
      "step": 92015
    },
    {
      "entropy": 5.707570934295655,
      "epoch": 8.347242380261248,
      "grad_norm": 1.40625,
      "learning_rate": 8.016370090164732e-05,
      "loss": 4.7683,
      "mean_token_accuracy": 0.24595048725605012,
      "num_tokens": 190950069.0,
      "step": 92020
    },
    {
      "entropy": 5.695918893814087,
      "epoch": 8.347695936139333,
      "grad_norm": 1.4921875,
      "learning_rate": 8.014751987107467e-05,
      "loss": 4.7688,
      "mean_token_accuracy": 0.23776875883340837,
      "num_tokens": 190960732.0,
      "step": 92025
    },
    {
      "entropy": 5.722405004501343,
      "epoch": 8.348149492017416,
      "grad_norm": 1.515625,
      "learning_rate": 8.013134287010099e-05,
      "loss": 4.8466,
      "mean_token_accuracy": 0.23795915991067887,
      "num_tokens": 190971274.0,
      "step": 92030
    },
    {
      "entropy": 5.686155319213867,
      "epoch": 8.348603047895502,
      "grad_norm": 1.4453125,
      "learning_rate": 8.01151698990608e-05,
      "loss": 4.799,
      "mean_token_accuracy": 0.23796162903308868,
      "num_tokens": 190982125.0,
      "step": 92035
    },
    {
      "entropy": 5.725762605667114,
      "epoch": 8.349056603773585,
      "grad_norm": 1.6640625,
      "learning_rate": 8.009900095828855e-05,
      "loss": 4.7434,
      "mean_token_accuracy": 0.24323491156101226,
      "num_tokens": 190991599.0,
      "step": 92040
    },
    {
      "entropy": 5.727030944824219,
      "epoch": 8.349510159651668,
      "grad_norm": 1.4453125,
      "learning_rate": 8.008283604811863e-05,
      "loss": 4.8657,
      "mean_token_accuracy": 0.23508499413728715,
      "num_tokens": 191001149.0,
      "step": 92045
    },
    {
      "entropy": 5.674269008636474,
      "epoch": 8.349963715529753,
      "grad_norm": 1.515625,
      "learning_rate": 8.006667516888528e-05,
      "loss": 4.8485,
      "mean_token_accuracy": 0.24036488682031631,
      "num_tokens": 191011674.0,
      "step": 92050
    },
    {
      "entropy": 5.727038145065308,
      "epoch": 8.350417271407837,
      "grad_norm": 1.5703125,
      "learning_rate": 8.005051832092281e-05,
      "loss": 4.8703,
      "mean_token_accuracy": 0.24516990929841995,
      "num_tokens": 191021893.0,
      "step": 92055
    },
    {
      "entropy": 5.616587209701538,
      "epoch": 8.350870827285922,
      "grad_norm": 1.609375,
      "learning_rate": 8.00343655045653e-05,
      "loss": 4.7212,
      "mean_token_accuracy": 0.24609310030937195,
      "num_tokens": 191032419.0,
      "step": 92060
    },
    {
      "entropy": 5.691227626800537,
      "epoch": 8.351324383164005,
      "grad_norm": 1.4609375,
      "learning_rate": 8.00182167201468e-05,
      "loss": 4.7517,
      "mean_token_accuracy": 0.23814696073532104,
      "num_tokens": 191043569.0,
      "step": 92065
    },
    {
      "entropy": 5.7088405132293705,
      "epoch": 8.35177793904209,
      "grad_norm": 1.5625,
      "learning_rate": 8.000207196800124e-05,
      "loss": 4.7816,
      "mean_token_accuracy": 0.2441389739513397,
      "num_tokens": 191052820.0,
      "step": 92070
    },
    {
      "entropy": 5.781645345687866,
      "epoch": 8.352231494920174,
      "grad_norm": 1.59375,
      "learning_rate": 7.998593124846257e-05,
      "loss": 4.8603,
      "mean_token_accuracy": 0.23515585511922837,
      "num_tokens": 191063718.0,
      "step": 92075
    },
    {
      "entropy": 5.6596955299377445,
      "epoch": 8.352685050798259,
      "grad_norm": 1.5,
      "learning_rate": 7.99697945618645e-05,
      "loss": 4.7487,
      "mean_token_accuracy": 0.23881043791770934,
      "num_tokens": 191073611.0,
      "step": 92080
    },
    {
      "entropy": 5.6644891738891605,
      "epoch": 8.353138606676342,
      "grad_norm": 1.4296875,
      "learning_rate": 7.995366190854074e-05,
      "loss": 4.8068,
      "mean_token_accuracy": 0.24041538685560226,
      "num_tokens": 191083785.0,
      "step": 92085
    },
    {
      "entropy": 5.651388359069824,
      "epoch": 8.353592162554428,
      "grad_norm": 1.53125,
      "learning_rate": 7.993753328882501e-05,
      "loss": 4.7794,
      "mean_token_accuracy": 0.2392260029911995,
      "num_tokens": 191095035.0,
      "step": 92090
    },
    {
      "entropy": 5.672310924530029,
      "epoch": 8.354045718432511,
      "grad_norm": 1.4921875,
      "learning_rate": 7.992140870305078e-05,
      "loss": 4.7655,
      "mean_token_accuracy": 0.23113106489181517,
      "num_tokens": 191105730.0,
      "step": 92095
    },
    {
      "entropy": 5.736989831924438,
      "epoch": 8.354499274310594,
      "grad_norm": 1.8125,
      "learning_rate": 7.990528815155157e-05,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23179846704006196,
      "num_tokens": 191115177.0,
      "step": 92100
    },
    {
      "entropy": 5.639191770553589,
      "epoch": 8.35495283018868,
      "grad_norm": 1.53125,
      "learning_rate": 7.988917163466069e-05,
      "loss": 4.6926,
      "mean_token_accuracy": 0.24855578541755677,
      "num_tokens": 191125715.0,
      "step": 92105
    },
    {
      "entropy": 5.760159301757812,
      "epoch": 8.355406386066763,
      "grad_norm": 1.4921875,
      "learning_rate": 7.987305915271146e-05,
      "loss": 4.889,
      "mean_token_accuracy": 0.2310703217983246,
      "num_tokens": 191136316.0,
      "step": 92110
    },
    {
      "entropy": 5.662558317184448,
      "epoch": 8.355859941944848,
      "grad_norm": 1.609375,
      "learning_rate": 7.98569507060371e-05,
      "loss": 4.7475,
      "mean_token_accuracy": 0.24164448529481888,
      "num_tokens": 191146265.0,
      "step": 92115
    },
    {
      "entropy": 5.728250932693482,
      "epoch": 8.356313497822931,
      "grad_norm": 1.515625,
      "learning_rate": 7.984084629497065e-05,
      "loss": 4.8479,
      "mean_token_accuracy": 0.23122627586126326,
      "num_tokens": 191156433.0,
      "step": 92120
    },
    {
      "entropy": 5.613675498962403,
      "epoch": 8.356767053701017,
      "grad_norm": 1.5078125,
      "learning_rate": 7.982474591984531e-05,
      "loss": 4.7864,
      "mean_token_accuracy": 0.24308521747589112,
      "num_tokens": 191167065.0,
      "step": 92125
    },
    {
      "entropy": 5.758504009246826,
      "epoch": 8.3572206095791,
      "grad_norm": 1.4140625,
      "learning_rate": 7.980864958099395e-05,
      "loss": 4.8904,
      "mean_token_accuracy": 0.23197929859161376,
      "num_tokens": 191177672.0,
      "step": 92130
    },
    {
      "entropy": 5.692176675796508,
      "epoch": 8.357674165457185,
      "grad_norm": 1.6484375,
      "learning_rate": 7.979255727874948e-05,
      "loss": 4.8179,
      "mean_token_accuracy": 0.23384953737258912,
      "num_tokens": 191187044.0,
      "step": 92135
    },
    {
      "entropy": 5.665296936035157,
      "epoch": 8.358127721335268,
      "grad_norm": 1.4296875,
      "learning_rate": 7.977646901344463e-05,
      "loss": 4.7791,
      "mean_token_accuracy": 0.240150648355484,
      "num_tokens": 191197922.0,
      "step": 92140
    },
    {
      "entropy": 5.715367174148559,
      "epoch": 8.358581277213352,
      "grad_norm": 1.5625,
      "learning_rate": 7.976038478541219e-05,
      "loss": 4.8566,
      "mean_token_accuracy": 0.23229830712080002,
      "num_tokens": 191208961.0,
      "step": 92145
    },
    {
      "entropy": 5.701867580413818,
      "epoch": 8.359034833091437,
      "grad_norm": 1.359375,
      "learning_rate": 7.974430459498473e-05,
      "loss": 4.8673,
      "mean_token_accuracy": 0.23060673773288726,
      "num_tokens": 191218664.0,
      "step": 92150
    },
    {
      "entropy": 5.666635131835937,
      "epoch": 8.35948838896952,
      "grad_norm": 1.5703125,
      "learning_rate": 7.972822844249473e-05,
      "loss": 4.6981,
      "mean_token_accuracy": 0.24991913884878159,
      "num_tokens": 191228283.0,
      "step": 92155
    },
    {
      "entropy": 5.745589017868042,
      "epoch": 8.359941944847606,
      "grad_norm": 1.3359375,
      "learning_rate": 7.97121563282748e-05,
      "loss": 4.8433,
      "mean_token_accuracy": 0.22986778765916824,
      "num_tokens": 191238555.0,
      "step": 92160
    },
    {
      "entropy": 5.666846847534179,
      "epoch": 8.360395500725689,
      "grad_norm": 1.46875,
      "learning_rate": 7.969608825265726e-05,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2435883551836014,
      "num_tokens": 191250028.0,
      "step": 92165
    },
    {
      "entropy": 5.67900710105896,
      "epoch": 8.360849056603774,
      "grad_norm": 1.4296875,
      "learning_rate": 7.968002421597438e-05,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23915527164936065,
      "num_tokens": 191261285.0,
      "step": 92170
    },
    {
      "entropy": 5.617982053756714,
      "epoch": 8.361302612481857,
      "grad_norm": 1.46875,
      "learning_rate": 7.966396421855836e-05,
      "loss": 4.7467,
      "mean_token_accuracy": 0.24303310960531235,
      "num_tokens": 191271973.0,
      "step": 92175
    },
    {
      "entropy": 5.719082450866699,
      "epoch": 8.361756168359943,
      "grad_norm": 1.515625,
      "learning_rate": 7.964790826074132e-05,
      "loss": 4.8185,
      "mean_token_accuracy": 0.23973968923091887,
      "num_tokens": 191281250.0,
      "step": 92180
    },
    {
      "entropy": 5.753865385055542,
      "epoch": 8.362209724238026,
      "grad_norm": 1.5,
      "learning_rate": 7.963185634285536e-05,
      "loss": 4.8737,
      "mean_token_accuracy": 0.23301859945058823,
      "num_tokens": 191291743.0,
      "step": 92185
    },
    {
      "entropy": 5.726372575759887,
      "epoch": 8.362663280116111,
      "grad_norm": 1.515625,
      "learning_rate": 7.96158084652323e-05,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23993410915136337,
      "num_tokens": 191302865.0,
      "step": 92190
    },
    {
      "entropy": 5.646310949325562,
      "epoch": 8.363116835994195,
      "grad_norm": 1.6328125,
      "learning_rate": 7.95997646282042e-05,
      "loss": 4.7459,
      "mean_token_accuracy": 0.24220197945833205,
      "num_tokens": 191312748.0,
      "step": 92195
    },
    {
      "entropy": 5.707769441604614,
      "epoch": 8.363570391872278,
      "grad_norm": 1.7109375,
      "learning_rate": 7.958372483210276e-05,
      "loss": 4.8345,
      "mean_token_accuracy": 0.23956738859415055,
      "num_tokens": 191323004.0,
      "step": 92200
    },
    {
      "entropy": 5.710848617553711,
      "epoch": 8.364023947750363,
      "grad_norm": 1.5703125,
      "learning_rate": 7.956768907725967e-05,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2333690732717514,
      "num_tokens": 191333687.0,
      "step": 92205
    },
    {
      "entropy": 5.693916320800781,
      "epoch": 8.364477503628446,
      "grad_norm": 1.5078125,
      "learning_rate": 7.955165736400656e-05,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23018960803747177,
      "num_tokens": 191343981.0,
      "step": 92210
    },
    {
      "entropy": 5.685241317749023,
      "epoch": 8.364931059506532,
      "grad_norm": 1.6015625,
      "learning_rate": 7.9535629692675e-05,
      "loss": 4.7643,
      "mean_token_accuracy": 0.24342720359563827,
      "num_tokens": 191353726.0,
      "step": 92215
    },
    {
      "entropy": 5.666869640350342,
      "epoch": 8.365384615384615,
      "grad_norm": 1.4296875,
      "learning_rate": 7.951960606359643e-05,
      "loss": 4.7322,
      "mean_token_accuracy": 0.23771337270736695,
      "num_tokens": 191363582.0,
      "step": 92220
    },
    {
      "entropy": 5.703732967376709,
      "epoch": 8.3658381712627,
      "grad_norm": 1.7421875,
      "learning_rate": 7.950358647710218e-05,
      "loss": 4.8302,
      "mean_token_accuracy": 0.23328164517879485,
      "num_tokens": 191373619.0,
      "step": 92225
    },
    {
      "entropy": 5.732791137695313,
      "epoch": 8.366291727140784,
      "grad_norm": 1.5078125,
      "learning_rate": 7.948757093352361e-05,
      "loss": 4.8786,
      "mean_token_accuracy": 0.23402535170316696,
      "num_tokens": 191383992.0,
      "step": 92230
    },
    {
      "entropy": 5.778372287750244,
      "epoch": 8.366745283018869,
      "grad_norm": 1.375,
      "learning_rate": 7.94715594331919e-05,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22793344259262086,
      "num_tokens": 191394970.0,
      "step": 92235
    },
    {
      "entropy": 5.635058641433716,
      "epoch": 8.367198838896952,
      "grad_norm": 1.4921875,
      "learning_rate": 7.945555197643818e-05,
      "loss": 4.747,
      "mean_token_accuracy": 0.23768343329429625,
      "num_tokens": 191404901.0,
      "step": 92240
    },
    {
      "entropy": 5.707110643386841,
      "epoch": 8.367652394775035,
      "grad_norm": 1.546875,
      "learning_rate": 7.943954856359347e-05,
      "loss": 4.826,
      "mean_token_accuracy": 0.24156933277845383,
      "num_tokens": 191414515.0,
      "step": 92245
    },
    {
      "entropy": 5.624000072479248,
      "epoch": 8.36810595065312,
      "grad_norm": 1.453125,
      "learning_rate": 7.942354919498874e-05,
      "loss": 4.7048,
      "mean_token_accuracy": 0.2423827365040779,
      "num_tokens": 191425116.0,
      "step": 92250
    },
    {
      "entropy": 5.715579700469971,
      "epoch": 8.368559506531204,
      "grad_norm": 1.5078125,
      "learning_rate": 7.940755387095476e-05,
      "loss": 4.862,
      "mean_token_accuracy": 0.22564934194087982,
      "num_tokens": 191435706.0,
      "step": 92255
    },
    {
      "entropy": 5.745800638198853,
      "epoch": 8.369013062409289,
      "grad_norm": 1.4296875,
      "learning_rate": 7.93915625918225e-05,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2292199432849884,
      "num_tokens": 191446560.0,
      "step": 92260
    },
    {
      "entropy": 5.72046012878418,
      "epoch": 8.369466618287372,
      "grad_norm": 1.5703125,
      "learning_rate": 7.937557535792255e-05,
      "loss": 4.7866,
      "mean_token_accuracy": 0.2380295068025589,
      "num_tokens": 191456121.0,
      "step": 92265
    },
    {
      "entropy": 5.652900838851929,
      "epoch": 8.369920174165458,
      "grad_norm": 1.453125,
      "learning_rate": 7.935959216958562e-05,
      "loss": 4.7672,
      "mean_token_accuracy": 0.2381893903017044,
      "num_tokens": 191466954.0,
      "step": 92270
    },
    {
      "entropy": 5.732302570343018,
      "epoch": 8.370373730043541,
      "grad_norm": 1.328125,
      "learning_rate": 7.934361302714211e-05,
      "loss": 4.7874,
      "mean_token_accuracy": 0.23434510529041291,
      "num_tokens": 191476786.0,
      "step": 92275
    },
    {
      "entropy": 5.721403121948242,
      "epoch": 8.370827285921626,
      "grad_norm": 1.65625,
      "learning_rate": 7.932763793092257e-05,
      "loss": 4.7923,
      "mean_token_accuracy": 0.23915897607803344,
      "num_tokens": 191486288.0,
      "step": 92280
    },
    {
      "entropy": 5.717854261398315,
      "epoch": 8.37128084179971,
      "grad_norm": 1.4375,
      "learning_rate": 7.931166688125726e-05,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23984139710664748,
      "num_tokens": 191496497.0,
      "step": 92285
    },
    {
      "entropy": 5.757662725448609,
      "epoch": 8.371734397677795,
      "grad_norm": 1.640625,
      "learning_rate": 7.929569987847662e-05,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22860673516988755,
      "num_tokens": 191505727.0,
      "step": 92290
    },
    {
      "entropy": 5.648762941360474,
      "epoch": 8.372187953555878,
      "grad_norm": 1.4375,
      "learning_rate": 7.927973692291077e-05,
      "loss": 4.7494,
      "mean_token_accuracy": 0.2408456400036812,
      "num_tokens": 191516245.0,
      "step": 92295
    },
    {
      "entropy": 5.7279798030853275,
      "epoch": 8.372641509433961,
      "grad_norm": 1.6484375,
      "learning_rate": 7.926377801488983e-05,
      "loss": 4.8798,
      "mean_token_accuracy": 0.23471296727657318,
      "num_tokens": 191528367.0,
      "step": 92300
    },
    {
      "entropy": 5.641056394577026,
      "epoch": 8.373095065312047,
      "grad_norm": 1.59375,
      "learning_rate": 7.924782315474385e-05,
      "loss": 4.6733,
      "mean_token_accuracy": 0.24163510799407958,
      "num_tokens": 191539113.0,
      "step": 92305
    },
    {
      "entropy": 5.720043087005616,
      "epoch": 8.37354862119013,
      "grad_norm": 1.59375,
      "learning_rate": 7.923187234280277e-05,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23237097710371019,
      "num_tokens": 191549002.0,
      "step": 92310
    },
    {
      "entropy": 5.681570434570313,
      "epoch": 8.374002177068215,
      "grad_norm": 1.453125,
      "learning_rate": 7.921592557939646e-05,
      "loss": 4.7907,
      "mean_token_accuracy": 0.23962552100419998,
      "num_tokens": 191560143.0,
      "step": 92315
    },
    {
      "entropy": 5.6742723941802975,
      "epoch": 8.374455732946299,
      "grad_norm": 1.546875,
      "learning_rate": 7.919998286485462e-05,
      "loss": 4.766,
      "mean_token_accuracy": 0.2371736705303192,
      "num_tokens": 191569439.0,
      "step": 92320
    },
    {
      "entropy": 5.7178443431854244,
      "epoch": 8.374909288824384,
      "grad_norm": 1.515625,
      "learning_rate": 7.91840441995071e-05,
      "loss": 4.8151,
      "mean_token_accuracy": 0.234354467689991,
      "num_tokens": 191579522.0,
      "step": 92325
    },
    {
      "entropy": 5.598841142654419,
      "epoch": 8.375362844702467,
      "grad_norm": 1.6328125,
      "learning_rate": 7.916810958368347e-05,
      "loss": 4.6839,
      "mean_token_accuracy": 0.2505872040987015,
      "num_tokens": 191589731.0,
      "step": 92330
    },
    {
      "entropy": 5.638553810119629,
      "epoch": 8.375816400580552,
      "grad_norm": 1.5078125,
      "learning_rate": 7.915217901771321e-05,
      "loss": 4.695,
      "mean_token_accuracy": 0.25020917505025864,
      "num_tokens": 191599522.0,
      "step": 92335
    },
    {
      "entropy": 5.6313426971435545,
      "epoch": 8.376269956458636,
      "grad_norm": 1.5078125,
      "learning_rate": 7.91362525019258e-05,
      "loss": 4.7151,
      "mean_token_accuracy": 0.2496527314186096,
      "num_tokens": 191609996.0,
      "step": 92340
    },
    {
      "entropy": 5.725091743469238,
      "epoch": 8.37672351233672,
      "grad_norm": 1.3984375,
      "learning_rate": 7.912033003665061e-05,
      "loss": 4.7825,
      "mean_token_accuracy": 0.24229619950056075,
      "num_tokens": 191620353.0,
      "step": 92345
    },
    {
      "entropy": 5.5742490768432615,
      "epoch": 8.377177068214804,
      "grad_norm": 1.4921875,
      "learning_rate": 7.910441162221687e-05,
      "loss": 4.6265,
      "mean_token_accuracy": 0.25167723298072814,
      "num_tokens": 191629909.0,
      "step": 92350
    },
    {
      "entropy": 5.750825309753418,
      "epoch": 8.377630624092888,
      "grad_norm": 1.5546875,
      "learning_rate": 7.908849725895383e-05,
      "loss": 4.8637,
      "mean_token_accuracy": 0.23867860585451126,
      "num_tokens": 191640348.0,
      "step": 92355
    },
    {
      "entropy": 5.617571115493774,
      "epoch": 8.378084179970973,
      "grad_norm": 1.4765625,
      "learning_rate": 7.907258694719062e-05,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2347486659884453,
      "num_tokens": 191650807.0,
      "step": 92360
    },
    {
      "entropy": 5.692871570587158,
      "epoch": 8.378537735849056,
      "grad_norm": 1.4921875,
      "learning_rate": 7.905668068725625e-05,
      "loss": 4.8604,
      "mean_token_accuracy": 0.24378167539834977,
      "num_tokens": 191660811.0,
      "step": 92365
    },
    {
      "entropy": 5.689439487457276,
      "epoch": 8.378991291727141,
      "grad_norm": 1.546875,
      "learning_rate": 7.904077847947963e-05,
      "loss": 4.7983,
      "mean_token_accuracy": 0.23646894544363023,
      "num_tokens": 191670249.0,
      "step": 92370
    },
    {
      "entropy": 5.749256229400634,
      "epoch": 8.379444847605225,
      "grad_norm": 1.546875,
      "learning_rate": 7.902488032418964e-05,
      "loss": 4.8471,
      "mean_token_accuracy": 0.23197481781244278,
      "num_tokens": 191681038.0,
      "step": 92375
    },
    {
      "entropy": 5.716463708877564,
      "epoch": 8.37989840348331,
      "grad_norm": 1.4921875,
      "learning_rate": 7.900898622171509e-05,
      "loss": 4.8527,
      "mean_token_accuracy": 0.231599260866642,
      "num_tokens": 191691145.0,
      "step": 92380
    },
    {
      "entropy": 5.720598459243774,
      "epoch": 8.380351959361393,
      "grad_norm": 1.6171875,
      "learning_rate": 7.899309617238461e-05,
      "loss": 4.856,
      "mean_token_accuracy": 0.23282465785741807,
      "num_tokens": 191701576.0,
      "step": 92385
    },
    {
      "entropy": 5.679355525970459,
      "epoch": 8.380805515239478,
      "grad_norm": 1.609375,
      "learning_rate": 7.897721017652687e-05,
      "loss": 4.8681,
      "mean_token_accuracy": 0.23570276647806168,
      "num_tokens": 191711865.0,
      "step": 92390
    },
    {
      "entropy": 5.685268878936768,
      "epoch": 8.381259071117562,
      "grad_norm": 1.421875,
      "learning_rate": 7.896132823447038e-05,
      "loss": 4.7192,
      "mean_token_accuracy": 0.2426590844988823,
      "num_tokens": 191722230.0,
      "step": 92395
    },
    {
      "entropy": 5.713074350357056,
      "epoch": 8.381712626995645,
      "grad_norm": 1.3046875,
      "learning_rate": 7.894545034654358e-05,
      "loss": 4.7896,
      "mean_token_accuracy": 0.2379963517189026,
      "num_tokens": 191732461.0,
      "step": 92400
    },
    {
      "entropy": 5.600406551361084,
      "epoch": 8.38216618287373,
      "grad_norm": 1.6328125,
      "learning_rate": 7.892957651307483e-05,
      "loss": 4.6494,
      "mean_token_accuracy": 0.24536392390727996,
      "num_tokens": 191741885.0,
      "step": 92405
    },
    {
      "entropy": 5.739694309234619,
      "epoch": 8.382619738751814,
      "grad_norm": 1.59375,
      "learning_rate": 7.891370673439241e-05,
      "loss": 4.8582,
      "mean_token_accuracy": 0.22875114679336547,
      "num_tokens": 191752413.0,
      "step": 92410
    },
    {
      "entropy": 5.684728193283081,
      "epoch": 8.383073294629899,
      "grad_norm": 1.5234375,
      "learning_rate": 7.88978410108245e-05,
      "loss": 4.8061,
      "mean_token_accuracy": 0.23917364329099655,
      "num_tokens": 191762566.0,
      "step": 92415
    },
    {
      "entropy": 5.610015201568603,
      "epoch": 8.383526850507982,
      "grad_norm": 1.546875,
      "learning_rate": 7.888197934269917e-05,
      "loss": 4.6781,
      "mean_token_accuracy": 0.25000960677862166,
      "num_tokens": 191773380.0,
      "step": 92420
    },
    {
      "entropy": 5.682780075073242,
      "epoch": 8.383980406386067,
      "grad_norm": 1.6640625,
      "learning_rate": 7.886612173034451e-05,
      "loss": 4.7794,
      "mean_token_accuracy": 0.24034024626016617,
      "num_tokens": 191783913.0,
      "step": 92425
    },
    {
      "entropy": 5.7239683151245115,
      "epoch": 8.38443396226415,
      "grad_norm": 1.375,
      "learning_rate": 7.885026817408844e-05,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23561981916427613,
      "num_tokens": 191794293.0,
      "step": 92430
    },
    {
      "entropy": 5.7120317935943605,
      "epoch": 8.384887518142236,
      "grad_norm": 1.5703125,
      "learning_rate": 7.883441867425883e-05,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23593526482582092,
      "num_tokens": 191804932.0,
      "step": 92435
    },
    {
      "entropy": 5.691127109527588,
      "epoch": 8.38534107402032,
      "grad_norm": 1.5546875,
      "learning_rate": 7.881857323118343e-05,
      "loss": 4.802,
      "mean_token_accuracy": 0.2408740133047104,
      "num_tokens": 191814912.0,
      "step": 92440
    },
    {
      "entropy": 5.698393154144287,
      "epoch": 8.385794629898404,
      "grad_norm": 1.375,
      "learning_rate": 7.880273184518989e-05,
      "loss": 4.7202,
      "mean_token_accuracy": 0.24694366306066512,
      "num_tokens": 191824822.0,
      "step": 92445
    },
    {
      "entropy": 5.672655296325684,
      "epoch": 8.386248185776488,
      "grad_norm": 1.578125,
      "learning_rate": 7.878689451660589e-05,
      "loss": 4.7529,
      "mean_token_accuracy": 0.24432074278593063,
      "num_tokens": 191835892.0,
      "step": 92450
    },
    {
      "entropy": 5.708679437637329,
      "epoch": 8.386701741654571,
      "grad_norm": 1.578125,
      "learning_rate": 7.877106124575887e-05,
      "loss": 4.7457,
      "mean_token_accuracy": 0.24729938060045242,
      "num_tokens": 191845204.0,
      "step": 92455
    },
    {
      "entropy": 5.63621335029602,
      "epoch": 8.387155297532656,
      "grad_norm": 1.59375,
      "learning_rate": 7.875523203297633e-05,
      "loss": 4.7694,
      "mean_token_accuracy": 0.24395667016506195,
      "num_tokens": 191855456.0,
      "step": 92460
    },
    {
      "entropy": 5.637161493301392,
      "epoch": 8.38760885341074,
      "grad_norm": 1.5859375,
      "learning_rate": 7.873940687858561e-05,
      "loss": 4.7785,
      "mean_token_accuracy": 0.2403674080967903,
      "num_tokens": 191865073.0,
      "step": 92465
    },
    {
      "entropy": 5.734465169906616,
      "epoch": 8.388062409288825,
      "grad_norm": 1.515625,
      "learning_rate": 7.872358578291395e-05,
      "loss": 4.8269,
      "mean_token_accuracy": 0.24063560366630554,
      "num_tokens": 191875235.0,
      "step": 92470
    },
    {
      "entropy": 5.689170789718628,
      "epoch": 8.388515965166908,
      "grad_norm": 1.515625,
      "learning_rate": 7.870776874628857e-05,
      "loss": 4.8558,
      "mean_token_accuracy": 0.23054732978343964,
      "num_tokens": 191885712.0,
      "step": 92475
    },
    {
      "entropy": 5.626595211029053,
      "epoch": 8.388969521044993,
      "grad_norm": 1.46875,
      "learning_rate": 7.869195576903656e-05,
      "loss": 4.6948,
      "mean_token_accuracy": 0.24259495437145234,
      "num_tokens": 191896727.0,
      "step": 92480
    },
    {
      "entropy": 5.657906723022461,
      "epoch": 8.389423076923077,
      "grad_norm": 1.5078125,
      "learning_rate": 7.867614685148494e-05,
      "loss": 4.7118,
      "mean_token_accuracy": 0.24634173810482024,
      "num_tokens": 191906515.0,
      "step": 92485
    },
    {
      "entropy": 5.681627893447876,
      "epoch": 8.389876632801162,
      "grad_norm": 1.3984375,
      "learning_rate": 7.866034199396056e-05,
      "loss": 4.7938,
      "mean_token_accuracy": 0.23900715708732606,
      "num_tokens": 191916281.0,
      "step": 92490
    },
    {
      "entropy": 5.668304824829102,
      "epoch": 8.390330188679245,
      "grad_norm": 1.625,
      "learning_rate": 7.864454119679039e-05,
      "loss": 4.7773,
      "mean_token_accuracy": 0.2348772406578064,
      "num_tokens": 191926049.0,
      "step": 92495
    },
    {
      "entropy": 5.692406463623047,
      "epoch": 8.390783744557329,
      "grad_norm": 1.6171875,
      "learning_rate": 7.862874446030116e-05,
      "loss": 4.7875,
      "mean_token_accuracy": 0.2380746364593506,
      "num_tokens": 191935318.0,
      "step": 92500
    },
    {
      "entropy": 5.685004329681396,
      "epoch": 8.391237300435414,
      "grad_norm": 1.421875,
      "learning_rate": 7.861295178481954e-05,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23519809991121293,
      "num_tokens": 191946279.0,
      "step": 92505
    },
    {
      "entropy": 5.731167364120483,
      "epoch": 8.391690856313497,
      "grad_norm": 1.546875,
      "learning_rate": 7.85971631706721e-05,
      "loss": 4.8075,
      "mean_token_accuracy": 0.24146389961242676,
      "num_tokens": 191956385.0,
      "step": 92510
    },
    {
      "entropy": 5.724434852600098,
      "epoch": 8.392144412191582,
      "grad_norm": 1.5625,
      "learning_rate": 7.85813786181854e-05,
      "loss": 4.8098,
      "mean_token_accuracy": 0.23290004581212997,
      "num_tokens": 191966010.0,
      "step": 92515
    },
    {
      "entropy": 5.681364250183106,
      "epoch": 8.392597968069666,
      "grad_norm": 1.5078125,
      "learning_rate": 7.856559812768583e-05,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23815384209156037,
      "num_tokens": 191976680.0,
      "step": 92520
    },
    {
      "entropy": 5.620184278488159,
      "epoch": 8.39305152394775,
      "grad_norm": 1.3671875,
      "learning_rate": 7.85498216994997e-05,
      "loss": 4.6952,
      "mean_token_accuracy": 0.23867657035589218,
      "num_tokens": 191987227.0,
      "step": 92525
    },
    {
      "entropy": 5.696701717376709,
      "epoch": 8.393505079825834,
      "grad_norm": 1.546875,
      "learning_rate": 7.853404933395336e-05,
      "loss": 4.7993,
      "mean_token_accuracy": 0.23608187586069107,
      "num_tokens": 191997701.0,
      "step": 92530
    },
    {
      "entropy": 5.703978824615478,
      "epoch": 8.39395863570392,
      "grad_norm": 1.5625,
      "learning_rate": 7.851828103137296e-05,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23313675075769424,
      "num_tokens": 192007617.0,
      "step": 92535
    },
    {
      "entropy": 5.733535385131836,
      "epoch": 8.394412191582003,
      "grad_norm": 1.4765625,
      "learning_rate": 7.850251679208458e-05,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23726292997598647,
      "num_tokens": 192017223.0,
      "step": 92540
    },
    {
      "entropy": 5.6070859909057615,
      "epoch": 8.394865747460088,
      "grad_norm": 1.34375,
      "learning_rate": 7.84867566164142e-05,
      "loss": 4.725,
      "mean_token_accuracy": 0.22960295230150224,
      "num_tokens": 192028023.0,
      "step": 92545
    },
    {
      "entropy": 5.641626977920533,
      "epoch": 8.395319303338171,
      "grad_norm": 1.4453125,
      "learning_rate": 7.847100050468779e-05,
      "loss": 4.7487,
      "mean_token_accuracy": 0.23682444244623185,
      "num_tokens": 192039579.0,
      "step": 92550
    },
    {
      "entropy": 5.663353490829468,
      "epoch": 8.395772859216255,
      "grad_norm": 1.671875,
      "learning_rate": 7.845524845723114e-05,
      "loss": 4.775,
      "mean_token_accuracy": 0.23682355284690856,
      "num_tokens": 192049383.0,
      "step": 92555
    },
    {
      "entropy": 5.788677406311035,
      "epoch": 8.39622641509434,
      "grad_norm": 1.4609375,
      "learning_rate": 7.843950047437006e-05,
      "loss": 4.8467,
      "mean_token_accuracy": 0.24058905988931656,
      "num_tokens": 192060094.0,
      "step": 92560
    },
    {
      "entropy": 5.785530710220337,
      "epoch": 8.396679970972423,
      "grad_norm": 1.71875,
      "learning_rate": 7.842375655643019e-05,
      "loss": 4.8797,
      "mean_token_accuracy": 0.23488228321075438,
      "num_tokens": 192069068.0,
      "step": 92565
    },
    {
      "entropy": 5.6925252914428714,
      "epoch": 8.397133526850508,
      "grad_norm": 1.5546875,
      "learning_rate": 7.840801670373715e-05,
      "loss": 4.7833,
      "mean_token_accuracy": 0.23629907965660096,
      "num_tokens": 192078901.0,
      "step": 92570
    },
    {
      "entropy": 5.671352863311768,
      "epoch": 8.397587082728592,
      "grad_norm": 1.4609375,
      "learning_rate": 7.839228091661643e-05,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23532569408416748,
      "num_tokens": 192088878.0,
      "step": 92575
    },
    {
      "entropy": 5.753569936752319,
      "epoch": 8.398040638606677,
      "grad_norm": 1.3984375,
      "learning_rate": 7.837654919539341e-05,
      "loss": 4.8394,
      "mean_token_accuracy": 0.23064774572849273,
      "num_tokens": 192098662.0,
      "step": 92580
    },
    {
      "entropy": 5.659621000289917,
      "epoch": 8.39849419448476,
      "grad_norm": 1.578125,
      "learning_rate": 7.836082154039348e-05,
      "loss": 4.7177,
      "mean_token_accuracy": 0.24764469116926194,
      "num_tokens": 192108240.0,
      "step": 92585
    },
    {
      "entropy": 5.688381147384644,
      "epoch": 8.398947750362845,
      "grad_norm": 1.6015625,
      "learning_rate": 7.834509795194183e-05,
      "loss": 4.7692,
      "mean_token_accuracy": 0.23767871409654617,
      "num_tokens": 192118735.0,
      "step": 92590
    },
    {
      "entropy": 5.644028234481811,
      "epoch": 8.399401306240929,
      "grad_norm": 1.5859375,
      "learning_rate": 7.83293784303637e-05,
      "loss": 4.7717,
      "mean_token_accuracy": 0.23651232123374938,
      "num_tokens": 192128624.0,
      "step": 92595
    },
    {
      "entropy": 5.730756092071533,
      "epoch": 8.399854862119014,
      "grad_norm": 1.6328125,
      "learning_rate": 7.831366297598415e-05,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22752784788608552,
      "num_tokens": 192139799.0,
      "step": 92600
    },
    {
      "entropy": 5.711366319656372,
      "epoch": 8.400308417997097,
      "grad_norm": 1.546875,
      "learning_rate": 7.829795158912816e-05,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23331552147865295,
      "num_tokens": 192149699.0,
      "step": 92605
    },
    {
      "entropy": 5.678016233444214,
      "epoch": 8.40076197387518,
      "grad_norm": 1.46875,
      "learning_rate": 7.828224427012066e-05,
      "loss": 4.7619,
      "mean_token_accuracy": 0.24033265858888625,
      "num_tokens": 192159583.0,
      "step": 92610
    },
    {
      "entropy": 5.637264537811279,
      "epoch": 8.401215529753266,
      "grad_norm": 1.5078125,
      "learning_rate": 7.826654101928647e-05,
      "loss": 4.7648,
      "mean_token_accuracy": 0.2307988315820694,
      "num_tokens": 192169926.0,
      "step": 92615
    },
    {
      "entropy": 5.701725912094116,
      "epoch": 8.40166908563135,
      "grad_norm": 1.4296875,
      "learning_rate": 7.825084183695037e-05,
      "loss": 4.8526,
      "mean_token_accuracy": 0.23454838395118713,
      "num_tokens": 192180140.0,
      "step": 92620
    },
    {
      "entropy": 5.7013989925384525,
      "epoch": 8.402122641509434,
      "grad_norm": 1.4609375,
      "learning_rate": 7.823514672343696e-05,
      "loss": 4.7241,
      "mean_token_accuracy": 0.24041315615177156,
      "num_tokens": 192190169.0,
      "step": 92625
    },
    {
      "entropy": 5.682688760757446,
      "epoch": 8.402576197387518,
      "grad_norm": 1.5234375,
      "learning_rate": 7.821945567907088e-05,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23338676542043685,
      "num_tokens": 192200886.0,
      "step": 92630
    },
    {
      "entropy": 5.723102617263794,
      "epoch": 8.403029753265603,
      "grad_norm": 1.4609375,
      "learning_rate": 7.82037687041766e-05,
      "loss": 4.792,
      "mean_token_accuracy": 0.24823664277791976,
      "num_tokens": 192210452.0,
      "step": 92635
    },
    {
      "entropy": 5.705462121963501,
      "epoch": 8.403483309143686,
      "grad_norm": 1.6796875,
      "learning_rate": 7.818808579907853e-05,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22836311161518097,
      "num_tokens": 192220612.0,
      "step": 92640
    },
    {
      "entropy": 5.700375652313232,
      "epoch": 8.403936865021771,
      "grad_norm": 1.4140625,
      "learning_rate": 7.8172406964101e-05,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23493196964263915,
      "num_tokens": 192230864.0,
      "step": 92645
    },
    {
      "entropy": 5.661522674560547,
      "epoch": 8.404390420899855,
      "grad_norm": 1.484375,
      "learning_rate": 7.815673219956826e-05,
      "loss": 4.7908,
      "mean_token_accuracy": 0.23637471944093705,
      "num_tokens": 192241823.0,
      "step": 92650
    },
    {
      "entropy": 5.633738613128662,
      "epoch": 8.40484397677794,
      "grad_norm": 1.5546875,
      "learning_rate": 7.814106150580447e-05,
      "loss": 4.8009,
      "mean_token_accuracy": 0.24063934087753297,
      "num_tokens": 192252429.0,
      "step": 92655
    },
    {
      "entropy": 5.691025590896606,
      "epoch": 8.405297532656023,
      "grad_norm": 1.5,
      "learning_rate": 7.812539488313365e-05,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23361166417598725,
      "num_tokens": 192262578.0,
      "step": 92660
    },
    {
      "entropy": 5.743885803222656,
      "epoch": 8.405751088534107,
      "grad_norm": 1.515625,
      "learning_rate": 7.810973233187986e-05,
      "loss": 4.8769,
      "mean_token_accuracy": 0.23368813395500182,
      "num_tokens": 192272737.0,
      "step": 92665
    },
    {
      "entropy": 5.694544887542724,
      "epoch": 8.406204644412192,
      "grad_norm": 1.4375,
      "learning_rate": 7.809407385236697e-05,
      "loss": 4.7835,
      "mean_token_accuracy": 0.23519905656576157,
      "num_tokens": 192283765.0,
      "step": 92670
    },
    {
      "entropy": 5.609095811843872,
      "epoch": 8.406658200290275,
      "grad_norm": 1.4453125,
      "learning_rate": 7.807841944491885e-05,
      "loss": 4.757,
      "mean_token_accuracy": 0.24238947331905364,
      "num_tokens": 192293341.0,
      "step": 92675
    },
    {
      "entropy": 5.651995754241943,
      "epoch": 8.40711175616836,
      "grad_norm": 1.5390625,
      "learning_rate": 7.806276910985917e-05,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2485881805419922,
      "num_tokens": 192304052.0,
      "step": 92680
    },
    {
      "entropy": 5.71121129989624,
      "epoch": 8.407565312046444,
      "grad_norm": 1.34375,
      "learning_rate": 7.804712284751162e-05,
      "loss": 4.8644,
      "mean_token_accuracy": 0.2236899822950363,
      "num_tokens": 192315671.0,
      "step": 92685
    },
    {
      "entropy": 5.691309404373169,
      "epoch": 8.408018867924529,
      "grad_norm": 1.484375,
      "learning_rate": 7.803148065819974e-05,
      "loss": 4.7921,
      "mean_token_accuracy": 0.23447708189487457,
      "num_tokens": 192326898.0,
      "step": 92690
    },
    {
      "entropy": 5.69489016532898,
      "epoch": 8.408472423802612,
      "grad_norm": 1.5703125,
      "learning_rate": 7.801584254224706e-05,
      "loss": 4.8037,
      "mean_token_accuracy": 0.2349283665418625,
      "num_tokens": 192336582.0,
      "step": 92695
    },
    {
      "entropy": 5.710627555847168,
      "epoch": 8.408925979680697,
      "grad_norm": 1.6015625,
      "learning_rate": 7.800020849997693e-05,
      "loss": 4.7509,
      "mean_token_accuracy": 0.23743321746587753,
      "num_tokens": 192346781.0,
      "step": 92700
    },
    {
      "entropy": 5.7254109382629395,
      "epoch": 8.40937953555878,
      "grad_norm": 1.359375,
      "learning_rate": 7.79845785317127e-05,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23052965998649597,
      "num_tokens": 192357395.0,
      "step": 92705
    },
    {
      "entropy": 5.730925178527832,
      "epoch": 8.409833091436864,
      "grad_norm": 1.671875,
      "learning_rate": 7.796895263777759e-05,
      "loss": 4.8029,
      "mean_token_accuracy": 0.24254192113876344,
      "num_tokens": 192368942.0,
      "step": 92710
    },
    {
      "entropy": 5.713073062896728,
      "epoch": 8.41028664731495,
      "grad_norm": 1.484375,
      "learning_rate": 7.795333081849475e-05,
      "loss": 4.8603,
      "mean_token_accuracy": 0.23328596651554107,
      "num_tokens": 192379724.0,
      "step": 92715
    },
    {
      "entropy": 5.723507308959961,
      "epoch": 8.410740203193033,
      "grad_norm": 1.4921875,
      "learning_rate": 7.793771307418719e-05,
      "loss": 4.8676,
      "mean_token_accuracy": 0.23518994003534316,
      "num_tokens": 192389927.0,
      "step": 92720
    },
    {
      "entropy": 5.642443609237671,
      "epoch": 8.411193759071118,
      "grad_norm": 1.5390625,
      "learning_rate": 7.792209940517798e-05,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23840092569589616,
      "num_tokens": 192400890.0,
      "step": 92725
    },
    {
      "entropy": 5.710469675064087,
      "epoch": 8.411647314949201,
      "grad_norm": 1.3515625,
      "learning_rate": 7.790648981178999e-05,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23944758474826813,
      "num_tokens": 192411588.0,
      "step": 92730
    },
    {
      "entropy": 5.646076965332031,
      "epoch": 8.412100870827286,
      "grad_norm": 1.5234375,
      "learning_rate": 7.789088429434604e-05,
      "loss": 4.7674,
      "mean_token_accuracy": 0.24412407726049423,
      "num_tokens": 192422631.0,
      "step": 92735
    },
    {
      "entropy": 5.746943044662475,
      "epoch": 8.41255442670537,
      "grad_norm": 1.6328125,
      "learning_rate": 7.787528285316881e-05,
      "loss": 4.9368,
      "mean_token_accuracy": 0.22209490239620208,
      "num_tokens": 192435496.0,
      "step": 92740
    },
    {
      "entropy": 5.739799165725708,
      "epoch": 8.413007982583455,
      "grad_norm": 1.453125,
      "learning_rate": 7.785968548858097e-05,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22808157056570053,
      "num_tokens": 192446012.0,
      "step": 92745
    },
    {
      "entropy": 5.644110918045044,
      "epoch": 8.413461538461538,
      "grad_norm": 1.4609375,
      "learning_rate": 7.784409220090504e-05,
      "loss": 4.7585,
      "mean_token_accuracy": 0.2365832269191742,
      "num_tokens": 192456248.0,
      "step": 92750
    },
    {
      "entropy": 5.662847566604614,
      "epoch": 8.413915094339623,
      "grad_norm": 1.484375,
      "learning_rate": 7.782850299046352e-05,
      "loss": 4.8055,
      "mean_token_accuracy": 0.23829707950353624,
      "num_tokens": 192466195.0,
      "step": 92755
    },
    {
      "entropy": 5.7540483474731445,
      "epoch": 8.414368650217707,
      "grad_norm": 1.453125,
      "learning_rate": 7.781291785757882e-05,
      "loss": 4.7815,
      "mean_token_accuracy": 0.2370308145880699,
      "num_tokens": 192476581.0,
      "step": 92760
    },
    {
      "entropy": 5.680645370483399,
      "epoch": 8.41482220609579,
      "grad_norm": 1.5625,
      "learning_rate": 7.779733680257324e-05,
      "loss": 4.7806,
      "mean_token_accuracy": 0.23028838485479355,
      "num_tokens": 192486168.0,
      "step": 92765
    },
    {
      "entropy": 5.727762985229492,
      "epoch": 8.415275761973875,
      "grad_norm": 1.4296875,
      "learning_rate": 7.778175982576897e-05,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2311079904437065,
      "num_tokens": 192496973.0,
      "step": 92770
    },
    {
      "entropy": 5.6752852439880375,
      "epoch": 8.415729317851959,
      "grad_norm": 1.4140625,
      "learning_rate": 7.776618692748819e-05,
      "loss": 4.8022,
      "mean_token_accuracy": 0.24153702557086945,
      "num_tokens": 192507613.0,
      "step": 92775
    },
    {
      "entropy": 5.652023124694824,
      "epoch": 8.416182873730044,
      "grad_norm": 1.625,
      "learning_rate": 7.77506181080529e-05,
      "loss": 4.7012,
      "mean_token_accuracy": 0.245028255879879,
      "num_tokens": 192517491.0,
      "step": 92780
    },
    {
      "entropy": 5.586600732803345,
      "epoch": 8.416636429608127,
      "grad_norm": 1.5390625,
      "learning_rate": 7.773505336778514e-05,
      "loss": 4.696,
      "mean_token_accuracy": 0.244073686003685,
      "num_tokens": 192527605.0,
      "step": 92785
    },
    {
      "entropy": 5.715696191787719,
      "epoch": 8.417089985486212,
      "grad_norm": 1.390625,
      "learning_rate": 7.771949270700667e-05,
      "loss": 4.7954,
      "mean_token_accuracy": 0.24661228358745574,
      "num_tokens": 192538017.0,
      "step": 92790
    },
    {
      "entropy": 5.718160104751587,
      "epoch": 8.417543541364296,
      "grad_norm": 1.4453125,
      "learning_rate": 7.770393612603941e-05,
      "loss": 4.8319,
      "mean_token_accuracy": 0.24074102193117142,
      "num_tokens": 192548102.0,
      "step": 92795
    },
    {
      "entropy": 5.642234516143799,
      "epoch": 8.417997097242381,
      "grad_norm": 1.453125,
      "learning_rate": 7.768838362520501e-05,
      "loss": 4.7666,
      "mean_token_accuracy": 0.23418055325746537,
      "num_tokens": 192558645.0,
      "step": 92800
    },
    {
      "entropy": 5.720122671127319,
      "epoch": 8.418450653120464,
      "grad_norm": 1.453125,
      "learning_rate": 7.767283520482514e-05,
      "loss": 4.7993,
      "mean_token_accuracy": 0.23783656507730483,
      "num_tokens": 192568436.0,
      "step": 92805
    },
    {
      "entropy": 5.661160182952881,
      "epoch": 8.418904208998548,
      "grad_norm": 1.46875,
      "learning_rate": 7.765729086522132e-05,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24182106107473372,
      "num_tokens": 192579798.0,
      "step": 92810
    },
    {
      "entropy": 5.747420072555542,
      "epoch": 8.419357764876633,
      "grad_norm": 1.375,
      "learning_rate": 7.764175060671502e-05,
      "loss": 4.936,
      "mean_token_accuracy": 0.22691156417131425,
      "num_tokens": 192590100.0,
      "step": 92815
    },
    {
      "entropy": 5.688242673873901,
      "epoch": 8.419811320754716,
      "grad_norm": 1.515625,
      "learning_rate": 7.762621442962758e-05,
      "loss": 4.791,
      "mean_token_accuracy": 0.23885277956724166,
      "num_tokens": 192600290.0,
      "step": 92820
    },
    {
      "entropy": 5.698468446731567,
      "epoch": 8.420264876632801,
      "grad_norm": 1.4765625,
      "learning_rate": 7.761068233428032e-05,
      "loss": 4.8027,
      "mean_token_accuracy": 0.23331386893987655,
      "num_tokens": 192610738.0,
      "step": 92825
    },
    {
      "entropy": 5.709515237808228,
      "epoch": 8.420718432510885,
      "grad_norm": 1.59375,
      "learning_rate": 7.759515432099446e-05,
      "loss": 4.8021,
      "mean_token_accuracy": 0.23598660826683043,
      "num_tokens": 192620939.0,
      "step": 92830
    },
    {
      "entropy": 5.649885272979736,
      "epoch": 8.42117198838897,
      "grad_norm": 1.5625,
      "learning_rate": 7.757963039009113e-05,
      "loss": 4.699,
      "mean_token_accuracy": 0.23990747332572937,
      "num_tokens": 192630646.0,
      "step": 92835
    },
    {
      "entropy": 5.6851777076721195,
      "epoch": 8.421625544267053,
      "grad_norm": 1.5625,
      "learning_rate": 7.756411054189135e-05,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23859914690256118,
      "num_tokens": 192641543.0,
      "step": 92840
    },
    {
      "entropy": 5.696933174133301,
      "epoch": 8.422079100145138,
      "grad_norm": 1.4296875,
      "learning_rate": 7.754859477671608e-05,
      "loss": 4.7725,
      "mean_token_accuracy": 0.23852260708808898,
      "num_tokens": 192652165.0,
      "step": 92845
    },
    {
      "entropy": 5.642933702468872,
      "epoch": 8.422532656023222,
      "grad_norm": 1.46875,
      "learning_rate": 7.753308309488617e-05,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2358841747045517,
      "num_tokens": 192661734.0,
      "step": 92850
    },
    {
      "entropy": 5.761564779281616,
      "epoch": 8.422986211901307,
      "grad_norm": 1.453125,
      "learning_rate": 7.751757549672239e-05,
      "loss": 4.8911,
      "mean_token_accuracy": 0.22477843165397643,
      "num_tokens": 192671893.0,
      "step": 92855
    },
    {
      "entropy": 5.702061796188355,
      "epoch": 8.42343976777939,
      "grad_norm": 1.3828125,
      "learning_rate": 7.750207198254553e-05,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23289241790771484,
      "num_tokens": 192682411.0,
      "step": 92860
    },
    {
      "entropy": 5.6743875503540036,
      "epoch": 8.423893323657474,
      "grad_norm": 1.421875,
      "learning_rate": 7.748657255267614e-05,
      "loss": 4.8402,
      "mean_token_accuracy": 0.2316809341311455,
      "num_tokens": 192692854.0,
      "step": 92865
    },
    {
      "entropy": 5.646349096298218,
      "epoch": 8.424346879535559,
      "grad_norm": 1.421875,
      "learning_rate": 7.747107720743473e-05,
      "loss": 4.7119,
      "mean_token_accuracy": 0.24391277134418488,
      "num_tokens": 192704567.0,
      "step": 92870
    },
    {
      "entropy": 5.649553442001343,
      "epoch": 8.424800435413642,
      "grad_norm": 1.4765625,
      "learning_rate": 7.745558594714182e-05,
      "loss": 4.7606,
      "mean_token_accuracy": 0.23837122321128845,
      "num_tokens": 192715554.0,
      "step": 92875
    },
    {
      "entropy": 5.673343706130981,
      "epoch": 8.425253991291727,
      "grad_norm": 1.59375,
      "learning_rate": 7.744009877211771e-05,
      "loss": 4.7786,
      "mean_token_accuracy": 0.23355868011713027,
      "num_tokens": 192724932.0,
      "step": 92880
    },
    {
      "entropy": 5.658164548873901,
      "epoch": 8.42570754716981,
      "grad_norm": 1.40625,
      "learning_rate": 7.742461568268274e-05,
      "loss": 4.7829,
      "mean_token_accuracy": 0.24119779467582703,
      "num_tokens": 192735545.0,
      "step": 92885
    },
    {
      "entropy": 5.668352031707764,
      "epoch": 8.426161103047896,
      "grad_norm": 1.4609375,
      "learning_rate": 7.740913667915697e-05,
      "loss": 4.7991,
      "mean_token_accuracy": 0.2328184053301811,
      "num_tokens": 192745313.0,
      "step": 92890
    },
    {
      "entropy": 5.666409587860107,
      "epoch": 8.42661465892598,
      "grad_norm": 1.5859375,
      "learning_rate": 7.739366176186069e-05,
      "loss": 4.7813,
      "mean_token_accuracy": 0.23563982546329498,
      "num_tokens": 192756423.0,
      "step": 92895
    },
    {
      "entropy": 5.719196844100952,
      "epoch": 8.427068214804065,
      "grad_norm": 1.515625,
      "learning_rate": 7.737819093111383e-05,
      "loss": 4.8701,
      "mean_token_accuracy": 0.24225496649742126,
      "num_tokens": 192766977.0,
      "step": 92900
    },
    {
      "entropy": 5.631129217147827,
      "epoch": 8.427521770682148,
      "grad_norm": 1.546875,
      "learning_rate": 7.736272418723634e-05,
      "loss": 4.7089,
      "mean_token_accuracy": 0.24925229400396348,
      "num_tokens": 192778160.0,
      "step": 92905
    },
    {
      "entropy": 5.597680568695068,
      "epoch": 8.427975326560233,
      "grad_norm": 1.484375,
      "learning_rate": 7.734726153054808e-05,
      "loss": 4.6942,
      "mean_token_accuracy": 0.246396104991436,
      "num_tokens": 192788529.0,
      "step": 92910
    },
    {
      "entropy": 5.704134511947632,
      "epoch": 8.428428882438316,
      "grad_norm": 1.4921875,
      "learning_rate": 7.733180296136882e-05,
      "loss": 4.842,
      "mean_token_accuracy": 0.22955500185489655,
      "num_tokens": 192798208.0,
      "step": 92915
    },
    {
      "entropy": 5.6678861618042,
      "epoch": 8.4288824383164,
      "grad_norm": 1.5625,
      "learning_rate": 7.731634848001824e-05,
      "loss": 4.7548,
      "mean_token_accuracy": 0.24232445210218428,
      "num_tokens": 192808180.0,
      "step": 92920
    },
    {
      "entropy": 5.623684024810791,
      "epoch": 8.429335994194485,
      "grad_norm": 1.3203125,
      "learning_rate": 7.730089808681592e-05,
      "loss": 4.7595,
      "mean_token_accuracy": 0.23542524576187135,
      "num_tokens": 192819091.0,
      "step": 92925
    },
    {
      "entropy": 5.690916299819946,
      "epoch": 8.429789550072568,
      "grad_norm": 1.515625,
      "learning_rate": 7.728545178208142e-05,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23816716223955153,
      "num_tokens": 192829600.0,
      "step": 92930
    },
    {
      "entropy": 5.627197360992431,
      "epoch": 8.430243105950654,
      "grad_norm": 1.6171875,
      "learning_rate": 7.727000956613418e-05,
      "loss": 4.6829,
      "mean_token_accuracy": 0.25077303647995,
      "num_tokens": 192839932.0,
      "step": 92935
    },
    {
      "entropy": 5.7550066947937015,
      "epoch": 8.430696661828737,
      "grad_norm": 1.4296875,
      "learning_rate": 7.725457143929352e-05,
      "loss": 4.9223,
      "mean_token_accuracy": 0.22930645793676377,
      "num_tokens": 192850109.0,
      "step": 92940
    },
    {
      "entropy": 5.618574380874634,
      "epoch": 8.431150217706822,
      "grad_norm": 1.578125,
      "learning_rate": 7.723913740187873e-05,
      "loss": 4.7061,
      "mean_token_accuracy": 0.24309777915477754,
      "num_tokens": 192860015.0,
      "step": 92945
    },
    {
      "entropy": 5.714534425735474,
      "epoch": 8.431603773584905,
      "grad_norm": 1.6015625,
      "learning_rate": 7.722370745420895e-05,
      "loss": 4.8217,
      "mean_token_accuracy": 0.23244935721158982,
      "num_tokens": 192870450.0,
      "step": 92950
    },
    {
      "entropy": 5.6867759227752686,
      "epoch": 8.43205732946299,
      "grad_norm": 1.578125,
      "learning_rate": 7.720828159660333e-05,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24136497378349303,
      "num_tokens": 192880785.0,
      "step": 92955
    },
    {
      "entropy": 5.710818529129028,
      "epoch": 8.432510885341074,
      "grad_norm": 1.4921875,
      "learning_rate": 7.719285982938079e-05,
      "loss": 4.8677,
      "mean_token_accuracy": 0.23210861086845397,
      "num_tokens": 192892342.0,
      "step": 92960
    },
    {
      "entropy": 5.726980066299438,
      "epoch": 8.43296444121916,
      "grad_norm": 1.5390625,
      "learning_rate": 7.717744215286037e-05,
      "loss": 4.8181,
      "mean_token_accuracy": 0.2305077061057091,
      "num_tokens": 192901373.0,
      "step": 92965
    },
    {
      "entropy": 5.645923852920532,
      "epoch": 8.433417997097242,
      "grad_norm": 1.3671875,
      "learning_rate": 7.716202856736084e-05,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24221621006727218,
      "num_tokens": 192912142.0,
      "step": 92970
    },
    {
      "entropy": 5.660545873641968,
      "epoch": 8.433871552975326,
      "grad_norm": 1.40625,
      "learning_rate": 7.714661907320098e-05,
      "loss": 4.7536,
      "mean_token_accuracy": 0.24153545796871184,
      "num_tokens": 192923098.0,
      "step": 92975
    },
    {
      "entropy": 5.7509543895721436,
      "epoch": 8.434325108853411,
      "grad_norm": 1.453125,
      "learning_rate": 7.713121367069948e-05,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23120716959238052,
      "num_tokens": 192934085.0,
      "step": 92980
    },
    {
      "entropy": 5.708499813079834,
      "epoch": 8.434778664731494,
      "grad_norm": 1.5234375,
      "learning_rate": 7.711581236017488e-05,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23999734818935395,
      "num_tokens": 192944519.0,
      "step": 92985
    },
    {
      "entropy": 5.702391147613525,
      "epoch": 8.43523222060958,
      "grad_norm": 1.453125,
      "learning_rate": 7.710041514194571e-05,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2331417202949524,
      "num_tokens": 192955900.0,
      "step": 92990
    },
    {
      "entropy": 5.782951688766479,
      "epoch": 8.435685776487663,
      "grad_norm": 1.5,
      "learning_rate": 7.708502201633034e-05,
      "loss": 4.978,
      "mean_token_accuracy": 0.2229061856865883,
      "num_tokens": 192965732.0,
      "step": 92995
    },
    {
      "entropy": 5.684912395477295,
      "epoch": 8.436139332365748,
      "grad_norm": 1.515625,
      "learning_rate": 7.70696329836472e-05,
      "loss": 4.7685,
      "mean_token_accuracy": 0.23828232139348984,
      "num_tokens": 192974985.0,
      "step": 93000
    },
    {
      "epoch": 8.436139332365748,
      "eval_entropy": 5.410148927528934,
      "eval_loss": 4.91062593460083,
      "eval_mean_token_accuracy": 0.23847595102141075,
      "eval_num_tokens": 192974985.0,
      "eval_runtime": 20.707,
      "eval_samples_per_second": 1707.633,
      "eval_steps_per_second": 213.454,
      "step": 93000
    },
    {
      "entropy": 5.686800479888916,
      "epoch": 8.436592888243831,
      "grad_norm": 1.609375,
      "learning_rate": 7.705424804421448e-05,
      "loss": 4.7444,
      "mean_token_accuracy": 0.24464187026023865,
      "num_tokens": 192985692.0,
      "step": 93005
    },
    {
      "entropy": 5.6981041431427,
      "epoch": 8.437046444121917,
      "grad_norm": 1.65625,
      "learning_rate": 7.703886719835036e-05,
      "loss": 4.7864,
      "mean_token_accuracy": 0.24306486696004867,
      "num_tokens": 192995668.0,
      "step": 93010
    },
    {
      "entropy": 5.711143636703492,
      "epoch": 8.4375,
      "grad_norm": 1.6484375,
      "learning_rate": 7.702349044637291e-05,
      "loss": 4.7646,
      "mean_token_accuracy": 0.2510621383786201,
      "num_tokens": 193006508.0,
      "step": 93015
    },
    {
      "entropy": 5.708534002304077,
      "epoch": 8.437953555878083,
      "grad_norm": 1.5703125,
      "learning_rate": 7.700811778860013e-05,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23911079168319702,
      "num_tokens": 193016021.0,
      "step": 93020
    },
    {
      "entropy": 5.6774519920349125,
      "epoch": 8.438407111756169,
      "grad_norm": 1.5,
      "learning_rate": 7.699274922534993e-05,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23678857684135438,
      "num_tokens": 193026591.0,
      "step": 93025
    },
    {
      "entropy": 5.6498815536499025,
      "epoch": 8.438860667634252,
      "grad_norm": 1.640625,
      "learning_rate": 7.697738475694009e-05,
      "loss": 4.738,
      "mean_token_accuracy": 0.2460998475551605,
      "num_tokens": 193036003.0,
      "step": 93030
    },
    {
      "entropy": 5.6169696807861325,
      "epoch": 8.439314223512337,
      "grad_norm": 1.3984375,
      "learning_rate": 7.696202438368845e-05,
      "loss": 4.7413,
      "mean_token_accuracy": 0.23596541881561278,
      "num_tokens": 193047242.0,
      "step": 93035
    },
    {
      "entropy": 5.712754917144776,
      "epoch": 8.43976777939042,
      "grad_norm": 1.578125,
      "learning_rate": 7.69466681059126e-05,
      "loss": 4.7778,
      "mean_token_accuracy": 0.24360256195068358,
      "num_tokens": 193056874.0,
      "step": 93040
    },
    {
      "entropy": 5.677986478805542,
      "epoch": 8.440221335268506,
      "grad_norm": 1.5859375,
      "learning_rate": 7.693131592393011e-05,
      "loss": 4.7732,
      "mean_token_accuracy": 0.23797551542520523,
      "num_tokens": 193065896.0,
      "step": 93045
    },
    {
      "entropy": 5.653041553497315,
      "epoch": 8.440674891146589,
      "grad_norm": 1.3125,
      "learning_rate": 7.691596783805848e-05,
      "loss": 4.7819,
      "mean_token_accuracy": 0.24023456871509552,
      "num_tokens": 193076815.0,
      "step": 93050
    },
    {
      "entropy": 5.692468452453613,
      "epoch": 8.441128447024674,
      "grad_norm": 1.6015625,
      "learning_rate": 7.690062384861514e-05,
      "loss": 4.7726,
      "mean_token_accuracy": 0.23440420627593994,
      "num_tokens": 193086914.0,
      "step": 93055
    },
    {
      "entropy": 5.7060174465179445,
      "epoch": 8.441582002902758,
      "grad_norm": 1.53125,
      "learning_rate": 7.688528395591733e-05,
      "loss": 4.8414,
      "mean_token_accuracy": 0.24065957814455033,
      "num_tokens": 193097942.0,
      "step": 93060
    },
    {
      "entropy": 5.716087007522583,
      "epoch": 8.44203555878084,
      "grad_norm": 1.5625,
      "learning_rate": 7.686994816028237e-05,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23487140387296676,
      "num_tokens": 193107894.0,
      "step": 93065
    },
    {
      "entropy": 5.714432334899902,
      "epoch": 8.442489114658926,
      "grad_norm": 1.625,
      "learning_rate": 7.685461646202737e-05,
      "loss": 4.8104,
      "mean_token_accuracy": 0.2347624868154526,
      "num_tokens": 193118471.0,
      "step": 93070
    },
    {
      "entropy": 5.615514612197876,
      "epoch": 8.44294267053701,
      "grad_norm": 1.4921875,
      "learning_rate": 7.683928886146939e-05,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2443877711892128,
      "num_tokens": 193128693.0,
      "step": 93075
    },
    {
      "entropy": 5.729672145843506,
      "epoch": 8.443396226415095,
      "grad_norm": 1.765625,
      "learning_rate": 7.682396535892544e-05,
      "loss": 4.822,
      "mean_token_accuracy": 0.23406241238117217,
      "num_tokens": 193138938.0,
      "step": 93080
    },
    {
      "entropy": 5.703195333480835,
      "epoch": 8.443849782293178,
      "grad_norm": 1.546875,
      "learning_rate": 7.680864595471235e-05,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23878685981035233,
      "num_tokens": 193148337.0,
      "step": 93085
    },
    {
      "entropy": 5.734991025924683,
      "epoch": 8.444303338171263,
      "grad_norm": 1.5859375,
      "learning_rate": 7.679333064914699e-05,
      "loss": 4.9268,
      "mean_token_accuracy": 0.223747119307518,
      "num_tokens": 193158538.0,
      "step": 93090
    },
    {
      "entropy": 5.61878023147583,
      "epoch": 8.444756894049346,
      "grad_norm": 1.5234375,
      "learning_rate": 7.677801944254602e-05,
      "loss": 4.7147,
      "mean_token_accuracy": 0.2467123970389366,
      "num_tokens": 193168557.0,
      "step": 93095
    },
    {
      "entropy": 5.707261276245117,
      "epoch": 8.445210449927432,
      "grad_norm": 1.59375,
      "learning_rate": 7.676271233522615e-05,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23540110439062117,
      "num_tokens": 193178988.0,
      "step": 93100
    },
    {
      "entropy": 5.674047756195068,
      "epoch": 8.445664005805515,
      "grad_norm": 1.5234375,
      "learning_rate": 7.674740932750391e-05,
      "loss": 4.7328,
      "mean_token_accuracy": 0.24221277981996536,
      "num_tokens": 193190161.0,
      "step": 93105
    },
    {
      "entropy": 5.734266901016236,
      "epoch": 8.4461175616836,
      "grad_norm": 1.6328125,
      "learning_rate": 7.673211041969579e-05,
      "loss": 4.8736,
      "mean_token_accuracy": 0.23978820890188218,
      "num_tokens": 193199464.0,
      "step": 93110
    },
    {
      "entropy": 5.704270458221435,
      "epoch": 8.446571117561684,
      "grad_norm": 1.546875,
      "learning_rate": 7.671681561211812e-05,
      "loss": 4.7771,
      "mean_token_accuracy": 0.24160640835762023,
      "num_tokens": 193209680.0,
      "step": 93115
    },
    {
      "entropy": 5.807614374160766,
      "epoch": 8.447024673439767,
      "grad_norm": 1.6171875,
      "learning_rate": 7.670152490508723e-05,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2385701298713684,
      "num_tokens": 193220758.0,
      "step": 93120
    },
    {
      "entropy": 5.656894826889038,
      "epoch": 8.447478229317852,
      "grad_norm": 1.53125,
      "learning_rate": 7.668623829891937e-05,
      "loss": 4.7145,
      "mean_token_accuracy": 0.24254560321569443,
      "num_tokens": 193230983.0,
      "step": 93125
    },
    {
      "entropy": 5.666531133651733,
      "epoch": 8.447931785195935,
      "grad_norm": 1.3984375,
      "learning_rate": 7.667095579393061e-05,
      "loss": 4.7526,
      "mean_token_accuracy": 0.239847095310688,
      "num_tokens": 193242179.0,
      "step": 93130
    },
    {
      "entropy": 5.708649778366089,
      "epoch": 8.44838534107402,
      "grad_norm": 1.359375,
      "learning_rate": 7.665567739043704e-05,
      "loss": 4.823,
      "mean_token_accuracy": 0.2316316172480583,
      "num_tokens": 193253009.0,
      "step": 93135
    },
    {
      "entropy": 5.648462438583374,
      "epoch": 8.448838896952104,
      "grad_norm": 1.4609375,
      "learning_rate": 7.664040308875459e-05,
      "loss": 4.6935,
      "mean_token_accuracy": 0.25114252120256425,
      "num_tokens": 193263764.0,
      "step": 93140
    },
    {
      "entropy": 5.66619086265564,
      "epoch": 8.44929245283019,
      "grad_norm": 1.671875,
      "learning_rate": 7.662513288919918e-05,
      "loss": 4.7525,
      "mean_token_accuracy": 0.2430346578359604,
      "num_tokens": 193272459.0,
      "step": 93145
    },
    {
      "entropy": 5.670949125289917,
      "epoch": 8.449746008708273,
      "grad_norm": 1.484375,
      "learning_rate": 7.660986679208655e-05,
      "loss": 4.8009,
      "mean_token_accuracy": 0.2372065454721451,
      "num_tokens": 193283534.0,
      "step": 93150
    },
    {
      "entropy": 5.660113334655762,
      "epoch": 8.450199564586358,
      "grad_norm": 1.4921875,
      "learning_rate": 7.65946047977324e-05,
      "loss": 4.8196,
      "mean_token_accuracy": 0.23852505832910537,
      "num_tokens": 193293791.0,
      "step": 93155
    },
    {
      "entropy": 5.72044415473938,
      "epoch": 8.450653120464441,
      "grad_norm": 1.59375,
      "learning_rate": 7.657934690645238e-05,
      "loss": 4.9255,
      "mean_token_accuracy": 0.2316076248884201,
      "num_tokens": 193304249.0,
      "step": 93160
    },
    {
      "entropy": 5.623410224914551,
      "epoch": 8.451106676342526,
      "grad_norm": 1.5234375,
      "learning_rate": 7.656409311856208e-05,
      "loss": 4.7118,
      "mean_token_accuracy": 0.24786169826984406,
      "num_tokens": 193314981.0,
      "step": 93165
    },
    {
      "entropy": 5.690154981613159,
      "epoch": 8.45156023222061,
      "grad_norm": 1.4609375,
      "learning_rate": 7.654884343437688e-05,
      "loss": 4.8266,
      "mean_token_accuracy": 0.233661949634552,
      "num_tokens": 193326409.0,
      "step": 93170
    },
    {
      "entropy": 5.67969913482666,
      "epoch": 8.452013788098693,
      "grad_norm": 1.4921875,
      "learning_rate": 7.653359785421219e-05,
      "loss": 4.8058,
      "mean_token_accuracy": 0.24020648896694183,
      "num_tokens": 193336091.0,
      "step": 93175
    },
    {
      "entropy": 5.675713205337525,
      "epoch": 8.452467343976778,
      "grad_norm": 1.515625,
      "learning_rate": 7.651835637838325e-05,
      "loss": 4.7432,
      "mean_token_accuracy": 0.2484314560890198,
      "num_tokens": 193346338.0,
      "step": 93180
    },
    {
      "entropy": 5.743917608261109,
      "epoch": 8.452920899854862,
      "grad_norm": 1.59375,
      "learning_rate": 7.650311900720532e-05,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22728272825479506,
      "num_tokens": 193356730.0,
      "step": 93185
    },
    {
      "entropy": 5.615808629989624,
      "epoch": 8.453374455732947,
      "grad_norm": 1.4296875,
      "learning_rate": 7.648788574099337e-05,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22828672975301742,
      "num_tokens": 193367760.0,
      "step": 93190
    },
    {
      "entropy": 5.66716365814209,
      "epoch": 8.45382801161103,
      "grad_norm": 1.65625,
      "learning_rate": 7.647265658006261e-05,
      "loss": 4.8142,
      "mean_token_accuracy": 0.2399092823266983,
      "num_tokens": 193378138.0,
      "step": 93195
    },
    {
      "entropy": 5.746873044967652,
      "epoch": 8.454281567489115,
      "grad_norm": 1.5859375,
      "learning_rate": 7.64574315247279e-05,
      "loss": 4.858,
      "mean_token_accuracy": 0.2375028282403946,
      "num_tokens": 193388598.0,
      "step": 93200
    },
    {
      "entropy": 5.703601694107055,
      "epoch": 8.454735123367199,
      "grad_norm": 1.5625,
      "learning_rate": 7.64422105753041e-05,
      "loss": 4.7897,
      "mean_token_accuracy": 0.23451476097106932,
      "num_tokens": 193397600.0,
      "step": 93205
    },
    {
      "entropy": 5.6958507061004635,
      "epoch": 8.455188679245284,
      "grad_norm": 1.6328125,
      "learning_rate": 7.642699373210598e-05,
      "loss": 4.8514,
      "mean_token_accuracy": 0.23666499853134154,
      "num_tokens": 193408183.0,
      "step": 93210
    },
    {
      "entropy": 5.711565542221069,
      "epoch": 8.455642235123367,
      "grad_norm": 1.6171875,
      "learning_rate": 7.641178099544821e-05,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2312620088458061,
      "num_tokens": 193418194.0,
      "step": 93215
    },
    {
      "entropy": 5.639948034286499,
      "epoch": 8.456095791001452,
      "grad_norm": 1.6484375,
      "learning_rate": 7.639657236564546e-05,
      "loss": 4.7297,
      "mean_token_accuracy": 0.23954293876886368,
      "num_tokens": 193426954.0,
      "step": 93220
    },
    {
      "entropy": 5.700588226318359,
      "epoch": 8.456549346879536,
      "grad_norm": 1.5390625,
      "learning_rate": 7.638136784301213e-05,
      "loss": 4.7535,
      "mean_token_accuracy": 0.24056895822286606,
      "num_tokens": 193437440.0,
      "step": 93225
    },
    {
      "entropy": 5.69061918258667,
      "epoch": 8.457002902757619,
      "grad_norm": 1.3671875,
      "learning_rate": 7.636616742786275e-05,
      "loss": 4.8638,
      "mean_token_accuracy": 0.23394751846790313,
      "num_tokens": 193448138.0,
      "step": 93230
    },
    {
      "entropy": 5.705939102172851,
      "epoch": 8.457456458635704,
      "grad_norm": 1.5703125,
      "learning_rate": 7.635097112051167e-05,
      "loss": 4.8083,
      "mean_token_accuracy": 0.24339917153120041,
      "num_tokens": 193459187.0,
      "step": 93235
    },
    {
      "entropy": 5.667025518417359,
      "epoch": 8.457910014513788,
      "grad_norm": 1.515625,
      "learning_rate": 7.633577892127312e-05,
      "loss": 4.8006,
      "mean_token_accuracy": 0.23591413497924804,
      "num_tokens": 193469568.0,
      "step": 93240
    },
    {
      "entropy": 5.645644426345825,
      "epoch": 8.458363570391873,
      "grad_norm": 1.4453125,
      "learning_rate": 7.632059083046134e-05,
      "loss": 4.7215,
      "mean_token_accuracy": 0.25009095668792725,
      "num_tokens": 193480260.0,
      "step": 93245
    },
    {
      "entropy": 5.639613962173462,
      "epoch": 8.458817126269956,
      "grad_norm": 1.4921875,
      "learning_rate": 7.63054068483903e-05,
      "loss": 4.7174,
      "mean_token_accuracy": 0.24412593245506287,
      "num_tokens": 193490598.0,
      "step": 93250
    },
    {
      "entropy": 5.606719160079956,
      "epoch": 8.459270682148041,
      "grad_norm": 1.421875,
      "learning_rate": 7.629022697537411e-05,
      "loss": 4.6796,
      "mean_token_accuracy": 0.25012718290090563,
      "num_tokens": 193500350.0,
      "step": 93255
    },
    {
      "entropy": 5.6820097923278805,
      "epoch": 8.459724238026125,
      "grad_norm": 1.46875,
      "learning_rate": 7.62750512117266e-05,
      "loss": 4.8624,
      "mean_token_accuracy": 0.23375002443790435,
      "num_tokens": 193511628.0,
      "step": 93260
    },
    {
      "entropy": 5.642878818511963,
      "epoch": 8.46017779390421,
      "grad_norm": 1.625,
      "learning_rate": 7.625987955776175e-05,
      "loss": 4.7298,
      "mean_token_accuracy": 0.24668803215026855,
      "num_tokens": 193521156.0,
      "step": 93265
    },
    {
      "entropy": 5.672129917144775,
      "epoch": 8.460631349782293,
      "grad_norm": 1.4765625,
      "learning_rate": 7.624471201379322e-05,
      "loss": 4.7828,
      "mean_token_accuracy": 0.23482609093189238,
      "num_tokens": 193531514.0,
      "step": 93270
    },
    {
      "entropy": 5.747758150100708,
      "epoch": 8.461084905660377,
      "grad_norm": 1.4921875,
      "learning_rate": 7.62295485801347e-05,
      "loss": 4.8631,
      "mean_token_accuracy": 0.23191613852977752,
      "num_tokens": 193541903.0,
      "step": 93275
    },
    {
      "entropy": 5.779682540893555,
      "epoch": 8.461538461538462,
      "grad_norm": 1.5390625,
      "learning_rate": 7.621438925709975e-05,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23286849856376649,
      "num_tokens": 193551678.0,
      "step": 93280
    },
    {
      "entropy": 5.601801061630249,
      "epoch": 8.461992017416545,
      "grad_norm": 1.5390625,
      "learning_rate": 7.61992340450019e-05,
      "loss": 4.6972,
      "mean_token_accuracy": 0.24042701870203018,
      "num_tokens": 193561717.0,
      "step": 93285
    },
    {
      "entropy": 5.655725622177124,
      "epoch": 8.46244557329463,
      "grad_norm": 1.4609375,
      "learning_rate": 7.618408294415457e-05,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2369446262717247,
      "num_tokens": 193572143.0,
      "step": 93290
    },
    {
      "entropy": 5.677638673782349,
      "epoch": 8.462899129172714,
      "grad_norm": 1.4453125,
      "learning_rate": 7.616893595487101e-05,
      "loss": 4.7581,
      "mean_token_accuracy": 0.2449764996767044,
      "num_tokens": 193582679.0,
      "step": 93295
    },
    {
      "entropy": 5.770569944381714,
      "epoch": 8.463352685050799,
      "grad_norm": 1.5625,
      "learning_rate": 7.615379307746454e-05,
      "loss": 4.8472,
      "mean_token_accuracy": 0.23510292321443557,
      "num_tokens": 193592324.0,
      "step": 93300
    },
    {
      "entropy": 5.630569791793823,
      "epoch": 8.463806240928882,
      "grad_norm": 1.5625,
      "learning_rate": 7.613865431224834e-05,
      "loss": 4.6779,
      "mean_token_accuracy": 0.25182937681674955,
      "num_tokens": 193603416.0,
      "step": 93305
    },
    {
      "entropy": 5.740942859649659,
      "epoch": 8.464259796806967,
      "grad_norm": 1.34375,
      "learning_rate": 7.612351965953543e-05,
      "loss": 4.8611,
      "mean_token_accuracy": 0.23543600291013717,
      "num_tokens": 193614711.0,
      "step": 93310
    },
    {
      "entropy": 5.707372617721558,
      "epoch": 8.46471335268505,
      "grad_norm": 1.515625,
      "learning_rate": 7.61083891196388e-05,
      "loss": 4.9441,
      "mean_token_accuracy": 0.23077948540449142,
      "num_tokens": 193624969.0,
      "step": 93315
    },
    {
      "entropy": 5.753130722045898,
      "epoch": 8.465166908563136,
      "grad_norm": 1.6328125,
      "learning_rate": 7.60932626928714e-05,
      "loss": 4.91,
      "mean_token_accuracy": 0.2337425097823143,
      "num_tokens": 193636000.0,
      "step": 93320
    },
    {
      "entropy": 5.658727407455444,
      "epoch": 8.46562046444122,
      "grad_norm": 1.4609375,
      "learning_rate": 7.607814037954597e-05,
      "loss": 4.7641,
      "mean_token_accuracy": 0.24058697372674942,
      "num_tokens": 193647207.0,
      "step": 93325
    },
    {
      "entropy": 5.688840007781982,
      "epoch": 8.466074020319303,
      "grad_norm": 1.578125,
      "learning_rate": 7.606302217997527e-05,
      "loss": 4.7879,
      "mean_token_accuracy": 0.2391711100935936,
      "num_tokens": 193657032.0,
      "step": 93330
    },
    {
      "entropy": 5.668672180175781,
      "epoch": 8.466527576197388,
      "grad_norm": 1.4921875,
      "learning_rate": 7.6047908094472e-05,
      "loss": 4.783,
      "mean_token_accuracy": 0.2357383742928505,
      "num_tokens": 193667447.0,
      "step": 93335
    },
    {
      "entropy": 5.75125150680542,
      "epoch": 8.466981132075471,
      "grad_norm": 1.421875,
      "learning_rate": 7.60327981233487e-05,
      "loss": 4.8368,
      "mean_token_accuracy": 0.23645258843898773,
      "num_tokens": 193678393.0,
      "step": 93340
    },
    {
      "entropy": 5.747914838790893,
      "epoch": 8.467434687953556,
      "grad_norm": 1.4375,
      "learning_rate": 7.601769226691782e-05,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2378832906484604,
      "num_tokens": 193689818.0,
      "step": 93345
    },
    {
      "entropy": 5.722551774978638,
      "epoch": 8.46788824383164,
      "grad_norm": 1.5859375,
      "learning_rate": 7.60025905254918e-05,
      "loss": 4.882,
      "mean_token_accuracy": 0.23275430351495743,
      "num_tokens": 193700504.0,
      "step": 93350
    },
    {
      "entropy": 5.724836969375611,
      "epoch": 8.468341799709725,
      "grad_norm": 1.53125,
      "learning_rate": 7.598749289938289e-05,
      "loss": 4.8266,
      "mean_token_accuracy": 0.23130162805318832,
      "num_tokens": 193711063.0,
      "step": 93355
    },
    {
      "entropy": 5.654139280319214,
      "epoch": 8.468795355587808,
      "grad_norm": 1.5625,
      "learning_rate": 7.597239938890331e-05,
      "loss": 4.7687,
      "mean_token_accuracy": 0.23260634541511535,
      "num_tokens": 193720420.0,
      "step": 93360
    },
    {
      "entropy": 5.651057338714599,
      "epoch": 8.469248911465893,
      "grad_norm": 1.59375,
      "learning_rate": 7.595730999436529e-05,
      "loss": 4.7441,
      "mean_token_accuracy": 0.2512804135680199,
      "num_tokens": 193730904.0,
      "step": 93365
    },
    {
      "entropy": 5.638110113143921,
      "epoch": 8.469702467343977,
      "grad_norm": 1.484375,
      "learning_rate": 7.594222471608081e-05,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2404665991663933,
      "num_tokens": 193741547.0,
      "step": 93370
    },
    {
      "entropy": 5.798304748535156,
      "epoch": 8.47015602322206,
      "grad_norm": 1.6328125,
      "learning_rate": 7.592714355436183e-05,
      "loss": 4.9675,
      "mean_token_accuracy": 0.22390456199645997,
      "num_tokens": 193751456.0,
      "step": 93375
    },
    {
      "entropy": 5.720625209808349,
      "epoch": 8.470609579100145,
      "grad_norm": 1.671875,
      "learning_rate": 7.591206650952029e-05,
      "loss": 4.7987,
      "mean_token_accuracy": 0.23205334991216658,
      "num_tokens": 193760751.0,
      "step": 93380
    },
    {
      "entropy": 5.746635770797729,
      "epoch": 8.471063134978229,
      "grad_norm": 1.5234375,
      "learning_rate": 7.58969935818679e-05,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23684865683317186,
      "num_tokens": 193770743.0,
      "step": 93385
    },
    {
      "entropy": 5.60735239982605,
      "epoch": 8.471516690856314,
      "grad_norm": 1.40625,
      "learning_rate": 7.588192477171644e-05,
      "loss": 4.661,
      "mean_token_accuracy": 0.2517309859395027,
      "num_tokens": 193781036.0,
      "step": 93390
    },
    {
      "entropy": 5.670959949493408,
      "epoch": 8.471970246734397,
      "grad_norm": 1.5859375,
      "learning_rate": 7.586686007937749e-05,
      "loss": 4.8086,
      "mean_token_accuracy": 0.23705879747867584,
      "num_tokens": 193792213.0,
      "step": 93395
    },
    {
      "entropy": 5.741263055801392,
      "epoch": 8.472423802612482,
      "grad_norm": 1.5390625,
      "learning_rate": 7.585179950516265e-05,
      "loss": 4.8325,
      "mean_token_accuracy": 0.23127986043691634,
      "num_tokens": 193802466.0,
      "step": 93400
    },
    {
      "entropy": 5.692383718490601,
      "epoch": 8.472877358490566,
      "grad_norm": 1.59375,
      "learning_rate": 7.583674304938335e-05,
      "loss": 4.769,
      "mean_token_accuracy": 0.22789095491170883,
      "num_tokens": 193812674.0,
      "step": 93405
    },
    {
      "entropy": 5.681033277511597,
      "epoch": 8.47333091436865,
      "grad_norm": 1.5546875,
      "learning_rate": 7.582169071235094e-05,
      "loss": 4.8276,
      "mean_token_accuracy": 0.2355302557349205,
      "num_tokens": 193822928.0,
      "step": 93410
    },
    {
      "entropy": 5.713939428329468,
      "epoch": 8.473784470246734,
      "grad_norm": 1.484375,
      "learning_rate": 7.580664249437672e-05,
      "loss": 4.8518,
      "mean_token_accuracy": 0.23632689267396928,
      "num_tokens": 193833435.0,
      "step": 93415
    },
    {
      "entropy": 5.683598470687866,
      "epoch": 8.47423802612482,
      "grad_norm": 1.5546875,
      "learning_rate": 7.579159839577193e-05,
      "loss": 4.7886,
      "mean_token_accuracy": 0.2411939173936844,
      "num_tokens": 193843094.0,
      "step": 93420
    },
    {
      "entropy": 5.694723749160767,
      "epoch": 8.474691582002903,
      "grad_norm": 1.5625,
      "learning_rate": 7.577655841684762e-05,
      "loss": 4.8785,
      "mean_token_accuracy": 0.22985480427742006,
      "num_tokens": 193852643.0,
      "step": 93425
    },
    {
      "entropy": 5.71142258644104,
      "epoch": 8.475145137880986,
      "grad_norm": 1.453125,
      "learning_rate": 7.576152255791482e-05,
      "loss": 4.8695,
      "mean_token_accuracy": 0.22931859493255616,
      "num_tokens": 193863381.0,
      "step": 93430
    },
    {
      "entropy": 5.638113260269165,
      "epoch": 8.475598693759071,
      "grad_norm": 1.4921875,
      "learning_rate": 7.574649081928455e-05,
      "loss": 4.7354,
      "mean_token_accuracy": 0.25275971442461015,
      "num_tokens": 193874756.0,
      "step": 93435
    },
    {
      "entropy": 5.672728395462036,
      "epoch": 8.476052249637155,
      "grad_norm": 1.5625,
      "learning_rate": 7.57314632012676e-05,
      "loss": 4.7504,
      "mean_token_accuracy": 0.23844944536685944,
      "num_tokens": 193884738.0,
      "step": 93440
    },
    {
      "entropy": 5.722782325744629,
      "epoch": 8.47650580551524,
      "grad_norm": 1.4375,
      "learning_rate": 7.57164397041748e-05,
      "loss": 4.7665,
      "mean_token_accuracy": 0.24104480743408202,
      "num_tokens": 193894313.0,
      "step": 93445
    },
    {
      "entropy": 5.662847089767456,
      "epoch": 8.476959361393323,
      "grad_norm": 1.421875,
      "learning_rate": 7.570142032831679e-05,
      "loss": 4.7408,
      "mean_token_accuracy": 0.2389300510287285,
      "num_tokens": 193905443.0,
      "step": 93450
    },
    {
      "entropy": 5.68264536857605,
      "epoch": 8.477412917271408,
      "grad_norm": 1.5703125,
      "learning_rate": 7.56864050740042e-05,
      "loss": 4.7827,
      "mean_token_accuracy": 0.24232831597328186,
      "num_tokens": 193916322.0,
      "step": 93455
    },
    {
      "entropy": 5.677496337890625,
      "epoch": 8.477866473149492,
      "grad_norm": 1.453125,
      "learning_rate": 7.567139394154756e-05,
      "loss": 4.7951,
      "mean_token_accuracy": 0.23405563682317734,
      "num_tokens": 193926578.0,
      "step": 93460
    },
    {
      "entropy": 5.617102909088135,
      "epoch": 8.478320029027577,
      "grad_norm": 1.46875,
      "learning_rate": 7.565638693125723e-05,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23930806666612625,
      "num_tokens": 193936612.0,
      "step": 93465
    },
    {
      "entropy": 5.627625131607056,
      "epoch": 8.47877358490566,
      "grad_norm": 1.484375,
      "learning_rate": 7.564138404344367e-05,
      "loss": 4.7765,
      "mean_token_accuracy": 0.23702559024095535,
      "num_tokens": 193946378.0,
      "step": 93470
    },
    {
      "entropy": 5.734845733642578,
      "epoch": 8.479227140783745,
      "grad_norm": 1.5703125,
      "learning_rate": 7.562638527841709e-05,
      "loss": 4.876,
      "mean_token_accuracy": 0.2378571704030037,
      "num_tokens": 193956048.0,
      "step": 93475
    },
    {
      "entropy": 5.689678764343261,
      "epoch": 8.479680696661829,
      "grad_norm": 1.421875,
      "learning_rate": 7.561139063648765e-05,
      "loss": 4.7972,
      "mean_token_accuracy": 0.23850373476743697,
      "num_tokens": 193966213.0,
      "step": 93480
    },
    {
      "entropy": 5.788786697387695,
      "epoch": 8.480134252539912,
      "grad_norm": 1.3671875,
      "learning_rate": 7.559640011796548e-05,
      "loss": 4.9095,
      "mean_token_accuracy": 0.23369574993848802,
      "num_tokens": 193975893.0,
      "step": 93485
    },
    {
      "entropy": 5.7167542457580565,
      "epoch": 8.480587808417997,
      "grad_norm": 1.5234375,
      "learning_rate": 7.558141372316056e-05,
      "loss": 4.7933,
      "mean_token_accuracy": 0.23693958371877671,
      "num_tokens": 193986979.0,
      "step": 93490
    },
    {
      "entropy": 5.777075338363647,
      "epoch": 8.48104136429608,
      "grad_norm": 1.703125,
      "learning_rate": 7.556643145238281e-05,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23732443600893022,
      "num_tokens": 193995959.0,
      "step": 93495
    },
    {
      "entropy": 5.684382009506225,
      "epoch": 8.481494920174166,
      "grad_norm": 1.4765625,
      "learning_rate": 7.555145330594208e-05,
      "loss": 4.7801,
      "mean_token_accuracy": 0.24538350552320481,
      "num_tokens": 194007382.0,
      "step": 93500
    },
    {
      "entropy": 5.738353157043457,
      "epoch": 8.48194847605225,
      "grad_norm": 1.5,
      "learning_rate": 7.553647928414813e-05,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2307626336812973,
      "num_tokens": 194017973.0,
      "step": 93505
    },
    {
      "entropy": 5.6638024806976315,
      "epoch": 8.482402031930334,
      "grad_norm": 1.5703125,
      "learning_rate": 7.552150938731065e-05,
      "loss": 4.7309,
      "mean_token_accuracy": 0.2432885855436325,
      "num_tokens": 194027876.0,
      "step": 93510
    },
    {
      "entropy": 5.6905694007873535,
      "epoch": 8.482855587808418,
      "grad_norm": 1.5234375,
      "learning_rate": 7.550654361573914e-05,
      "loss": 4.7685,
      "mean_token_accuracy": 0.23816954642534255,
      "num_tokens": 194038828.0,
      "step": 93515
    },
    {
      "entropy": 5.672300529479981,
      "epoch": 8.483309143686503,
      "grad_norm": 1.65625,
      "learning_rate": 7.549158196974317e-05,
      "loss": 4.7855,
      "mean_token_accuracy": 0.23057586550712586,
      "num_tokens": 194048639.0,
      "step": 93520
    },
    {
      "entropy": 5.667972707748413,
      "epoch": 8.483762699564586,
      "grad_norm": 1.6015625,
      "learning_rate": 7.547662444963215e-05,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22838054746389388,
      "num_tokens": 194059180.0,
      "step": 93525
    },
    {
      "entropy": 5.670329523086548,
      "epoch": 8.484216255442671,
      "grad_norm": 1.546875,
      "learning_rate": 7.546167105571534e-05,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23142619132995607,
      "num_tokens": 194069446.0,
      "step": 93530
    },
    {
      "entropy": 5.691270780563355,
      "epoch": 8.484669811320755,
      "grad_norm": 1.5234375,
      "learning_rate": 7.544672178830199e-05,
      "loss": 4.7552,
      "mean_token_accuracy": 0.24129331558942796,
      "num_tokens": 194080212.0,
      "step": 93535
    },
    {
      "entropy": 5.689349174499512,
      "epoch": 8.485123367198838,
      "grad_norm": 1.5078125,
      "learning_rate": 7.543177664770134e-05,
      "loss": 4.7901,
      "mean_token_accuracy": 0.2354146048426628,
      "num_tokens": 194091329.0,
      "step": 93540
    },
    {
      "entropy": 5.672479963302612,
      "epoch": 8.485576923076923,
      "grad_norm": 1.6875,
      "learning_rate": 7.541683563422239e-05,
      "loss": 4.764,
      "mean_token_accuracy": 0.23918985873460769,
      "num_tokens": 194101036.0,
      "step": 93545
    },
    {
      "entropy": 5.644822502136231,
      "epoch": 8.486030478955007,
      "grad_norm": 1.7109375,
      "learning_rate": 7.54018987481742e-05,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2370796486735344,
      "num_tokens": 194112295.0,
      "step": 93550
    },
    {
      "entropy": 5.709794807434082,
      "epoch": 8.486484034833092,
      "grad_norm": 1.6171875,
      "learning_rate": 7.538696598986556e-05,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23097901046276093,
      "num_tokens": 194123245.0,
      "step": 93555
    },
    {
      "entropy": 5.666625118255615,
      "epoch": 8.486937590711175,
      "grad_norm": 1.4140625,
      "learning_rate": 7.537203735960526e-05,
      "loss": 4.8084,
      "mean_token_accuracy": 0.24039072096347808,
      "num_tokens": 194133788.0,
      "step": 93560
    },
    {
      "entropy": 5.707431221008301,
      "epoch": 8.48739114658926,
      "grad_norm": 1.4609375,
      "learning_rate": 7.535711285770216e-05,
      "loss": 4.7844,
      "mean_token_accuracy": 0.2396704986691475,
      "num_tokens": 194145759.0,
      "step": 93565
    },
    {
      "entropy": 5.703523731231689,
      "epoch": 8.487844702467344,
      "grad_norm": 1.484375,
      "learning_rate": 7.534219248446485e-05,
      "loss": 4.7492,
      "mean_token_accuracy": 0.23787622451782225,
      "num_tokens": 194155328.0,
      "step": 93570
    },
    {
      "entropy": 5.751754331588745,
      "epoch": 8.488298258345429,
      "grad_norm": 1.6328125,
      "learning_rate": 7.532727624020187e-05,
      "loss": 4.8547,
      "mean_token_accuracy": 0.23400946855545043,
      "num_tokens": 194165211.0,
      "step": 93575
    },
    {
      "entropy": 5.665249872207641,
      "epoch": 8.488751814223512,
      "grad_norm": 1.4765625,
      "learning_rate": 7.531236412522171e-05,
      "loss": 4.7901,
      "mean_token_accuracy": 0.24283961504697799,
      "num_tokens": 194176346.0,
      "step": 93580
    },
    {
      "entropy": 5.640337657928467,
      "epoch": 8.489205370101596,
      "grad_norm": 1.5859375,
      "learning_rate": 7.529745613983274e-05,
      "loss": 4.7324,
      "mean_token_accuracy": 0.2358388602733612,
      "num_tokens": 194185504.0,
      "step": 93585
    },
    {
      "entropy": 5.746907329559326,
      "epoch": 8.48965892597968,
      "grad_norm": 1.6171875,
      "learning_rate": 7.528255228434327e-05,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23230403363704683,
      "num_tokens": 194195160.0,
      "step": 93590
    },
    {
      "entropy": 5.710511589050293,
      "epoch": 8.490112481857764,
      "grad_norm": 1.5,
      "learning_rate": 7.526765255906146e-05,
      "loss": 4.9298,
      "mean_token_accuracy": 0.2346382200717926,
      "num_tokens": 194206290.0,
      "step": 93595
    },
    {
      "entropy": 5.671810579299927,
      "epoch": 8.49056603773585,
      "grad_norm": 1.7421875,
      "learning_rate": 7.525275696429554e-05,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23810648769140244,
      "num_tokens": 194215705.0,
      "step": 93600
    },
    {
      "entropy": 5.713294315338135,
      "epoch": 8.491019593613933,
      "grad_norm": 1.5625,
      "learning_rate": 7.523786550035351e-05,
      "loss": 4.829,
      "mean_token_accuracy": 0.23398037850856782,
      "num_tokens": 194226237.0,
      "step": 93605
    },
    {
      "entropy": 5.644697999954223,
      "epoch": 8.491473149492018,
      "grad_norm": 1.6875,
      "learning_rate": 7.522297816754333e-05,
      "loss": 4.7356,
      "mean_token_accuracy": 0.23708377480506898,
      "num_tokens": 194236278.0,
      "step": 93610
    },
    {
      "entropy": 5.723438501358032,
      "epoch": 8.491926705370101,
      "grad_norm": 1.6484375,
      "learning_rate": 7.520809496617288e-05,
      "loss": 4.7766,
      "mean_token_accuracy": 0.23483167588710785,
      "num_tokens": 194245289.0,
      "step": 93615
    },
    {
      "entropy": 5.6955023288726805,
      "epoch": 8.492380261248186,
      "grad_norm": 1.546875,
      "learning_rate": 7.519321589654993e-05,
      "loss": 4.7487,
      "mean_token_accuracy": 0.24258464872837066,
      "num_tokens": 194255463.0,
      "step": 93620
    },
    {
      "entropy": 5.6439004898071286,
      "epoch": 8.49283381712627,
      "grad_norm": 1.421875,
      "learning_rate": 7.517834095898219e-05,
      "loss": 4.7675,
      "mean_token_accuracy": 0.23848024904727935,
      "num_tokens": 194266332.0,
      "step": 93625
    },
    {
      "entropy": 5.71943473815918,
      "epoch": 8.493287373004355,
      "grad_norm": 1.6171875,
      "learning_rate": 7.516347015377722e-05,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23615852296352385,
      "num_tokens": 194276167.0,
      "step": 93630
    },
    {
      "entropy": 5.711739110946655,
      "epoch": 8.493740928882438,
      "grad_norm": 1.640625,
      "learning_rate": 7.514860348124267e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.24891038239002228,
      "num_tokens": 194286627.0,
      "step": 93635
    },
    {
      "entropy": 5.678492689132691,
      "epoch": 8.494194484760522,
      "grad_norm": 1.546875,
      "learning_rate": 7.513374094168595e-05,
      "loss": 4.8252,
      "mean_token_accuracy": 0.23322683274745942,
      "num_tokens": 194297551.0,
      "step": 93640
    },
    {
      "entropy": 5.777190780639648,
      "epoch": 8.494648040638607,
      "grad_norm": 1.40625,
      "learning_rate": 7.511888253541437e-05,
      "loss": 4.9081,
      "mean_token_accuracy": 0.2306770548224449,
      "num_tokens": 194309246.0,
      "step": 93645
    },
    {
      "entropy": 5.676389551162719,
      "epoch": 8.49510159651669,
      "grad_norm": 1.609375,
      "learning_rate": 7.510402826273523e-05,
      "loss": 4.7686,
      "mean_token_accuracy": 0.24452030062675476,
      "num_tokens": 194320337.0,
      "step": 93650
    },
    {
      "entropy": 5.752158498764038,
      "epoch": 8.495555152394775,
      "grad_norm": 1.453125,
      "learning_rate": 7.508917812395574e-05,
      "loss": 4.8365,
      "mean_token_accuracy": 0.2394584447145462,
      "num_tokens": 194329457.0,
      "step": 93655
    },
    {
      "entropy": 5.642902278900147,
      "epoch": 8.496008708272859,
      "grad_norm": 1.546875,
      "learning_rate": 7.5074332119383e-05,
      "loss": 4.6937,
      "mean_token_accuracy": 0.24509461224079132,
      "num_tokens": 194339560.0,
      "step": 93660
    },
    {
      "entropy": 5.688702535629273,
      "epoch": 8.496462264150944,
      "grad_norm": 1.5625,
      "learning_rate": 7.505949024932394e-05,
      "loss": 4.8656,
      "mean_token_accuracy": 0.23427194207906724,
      "num_tokens": 194349814.0,
      "step": 93665
    },
    {
      "entropy": 5.709981918334961,
      "epoch": 8.496915820029027,
      "grad_norm": 1.640625,
      "learning_rate": 7.504465251408565e-05,
      "loss": 4.8732,
      "mean_token_accuracy": 0.23438937962055206,
      "num_tokens": 194360128.0,
      "step": 93670
    },
    {
      "entropy": 5.652327585220337,
      "epoch": 8.497369375907112,
      "grad_norm": 1.5234375,
      "learning_rate": 7.50298189139749e-05,
      "loss": 4.7866,
      "mean_token_accuracy": 0.2386720210313797,
      "num_tokens": 194369604.0,
      "step": 93675
    },
    {
      "entropy": 5.731446647644043,
      "epoch": 8.497822931785196,
      "grad_norm": 1.546875,
      "learning_rate": 7.501498944929842e-05,
      "loss": 4.7205,
      "mean_token_accuracy": 0.23897896558046341,
      "num_tokens": 194378998.0,
      "step": 93680
    },
    {
      "entropy": 5.72858247756958,
      "epoch": 8.49827648766328,
      "grad_norm": 1.546875,
      "learning_rate": 7.500016412036296e-05,
      "loss": 4.8521,
      "mean_token_accuracy": 0.23420632928609847,
      "num_tokens": 194389967.0,
      "step": 93685
    },
    {
      "entropy": 5.625341415405273,
      "epoch": 8.498730043541364,
      "grad_norm": 1.5859375,
      "learning_rate": 7.498534292747506e-05,
      "loss": 4.7565,
      "mean_token_accuracy": 0.23809790313243867,
      "num_tokens": 194400308.0,
      "step": 93690
    },
    {
      "entropy": 5.654445648193359,
      "epoch": 8.499183599419448,
      "grad_norm": 1.3203125,
      "learning_rate": 7.497052587094119e-05,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2430286943912506,
      "num_tokens": 194410565.0,
      "step": 93695
    },
    {
      "entropy": 5.688559818267822,
      "epoch": 8.499637155297533,
      "grad_norm": 1.328125,
      "learning_rate": 7.495571295106786e-05,
      "loss": 4.8155,
      "mean_token_accuracy": 0.24082447290420533,
      "num_tokens": 194421799.0,
      "step": 93700
    },
    {
      "entropy": 5.665348672866822,
      "epoch": 8.500090711175616,
      "grad_norm": 1.5546875,
      "learning_rate": 7.494090416816136e-05,
      "loss": 4.7345,
      "mean_token_accuracy": 0.25088845640420915,
      "num_tokens": 194431625.0,
      "step": 93705
    },
    {
      "entropy": 5.722607707977295,
      "epoch": 8.500544267053701,
      "grad_norm": 1.4921875,
      "learning_rate": 7.492609952252795e-05,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23634153455495835,
      "num_tokens": 194441945.0,
      "step": 93710
    },
    {
      "entropy": 5.686397647857666,
      "epoch": 8.500997822931785,
      "grad_norm": 1.484375,
      "learning_rate": 7.491129901447379e-05,
      "loss": 4.7898,
      "mean_token_accuracy": 0.24567523449659348,
      "num_tokens": 194452583.0,
      "step": 93715
    },
    {
      "entropy": 5.723160743713379,
      "epoch": 8.50145137880987,
      "grad_norm": 1.59375,
      "learning_rate": 7.489650264430493e-05,
      "loss": 4.8404,
      "mean_token_accuracy": 0.22502860873937608,
      "num_tokens": 194462901.0,
      "step": 93720
    },
    {
      "entropy": 5.768000507354737,
      "epoch": 8.501904934687953,
      "grad_norm": 1.4921875,
      "learning_rate": 7.488171041232742e-05,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2310188189148903,
      "num_tokens": 194473615.0,
      "step": 93725
    },
    {
      "entropy": 5.743111228942871,
      "epoch": 8.502358490566039,
      "grad_norm": 1.6953125,
      "learning_rate": 7.48669223188471e-05,
      "loss": 4.8435,
      "mean_token_accuracy": 0.2334517851471901,
      "num_tokens": 194483732.0,
      "step": 93730
    },
    {
      "entropy": 5.701708793640137,
      "epoch": 8.502812046444122,
      "grad_norm": 1.59375,
      "learning_rate": 7.485213836416985e-05,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23417342007160186,
      "num_tokens": 194493606.0,
      "step": 93735
    },
    {
      "entropy": 5.681463956832886,
      "epoch": 8.503265602322205,
      "grad_norm": 1.484375,
      "learning_rate": 7.483735854860135e-05,
      "loss": 4.7854,
      "mean_token_accuracy": 0.24034926742315293,
      "num_tokens": 194503397.0,
      "step": 93740
    },
    {
      "entropy": 5.681715869903565,
      "epoch": 8.50371915820029,
      "grad_norm": 1.4921875,
      "learning_rate": 7.482258287244734e-05,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22993445694446563,
      "num_tokens": 194513860.0,
      "step": 93745
    },
    {
      "entropy": 5.657442474365235,
      "epoch": 8.504172714078374,
      "grad_norm": 1.4375,
      "learning_rate": 7.480781133601331e-05,
      "loss": 4.7487,
      "mean_token_accuracy": 0.23797498941421508,
      "num_tokens": 194523841.0,
      "step": 93750
    },
    {
      "entropy": 5.73021969795227,
      "epoch": 8.504626269956459,
      "grad_norm": 1.578125,
      "learning_rate": 7.479304393960474e-05,
      "loss": 4.8381,
      "mean_token_accuracy": 0.229271537065506,
      "num_tokens": 194533856.0,
      "step": 93755
    },
    {
      "entropy": 5.64417085647583,
      "epoch": 8.505079825834542,
      "grad_norm": 1.4453125,
      "learning_rate": 7.477828068352703e-05,
      "loss": 4.7353,
      "mean_token_accuracy": 0.2461184784770012,
      "num_tokens": 194545505.0,
      "step": 93760
    },
    {
      "entropy": 5.68509955406189,
      "epoch": 8.505533381712628,
      "grad_norm": 1.5546875,
      "learning_rate": 7.476352156808548e-05,
      "loss": 4.8128,
      "mean_token_accuracy": 0.2345714285969734,
      "num_tokens": 194555700.0,
      "step": 93765
    },
    {
      "entropy": 5.683234310150146,
      "epoch": 8.505986937590711,
      "grad_norm": 1.5859375,
      "learning_rate": 7.474876659358537e-05,
      "loss": 4.7741,
      "mean_token_accuracy": 0.23276560604572297,
      "num_tokens": 194566304.0,
      "step": 93770
    },
    {
      "entropy": 5.639089822769165,
      "epoch": 8.506440493468796,
      "grad_norm": 1.484375,
      "learning_rate": 7.473401576033178e-05,
      "loss": 4.7226,
      "mean_token_accuracy": 0.23492567986249924,
      "num_tokens": 194576588.0,
      "step": 93775
    },
    {
      "entropy": 5.6470483303070065,
      "epoch": 8.50689404934688,
      "grad_norm": 1.5390625,
      "learning_rate": 7.47192690686298e-05,
      "loss": 4.6301,
      "mean_token_accuracy": 0.2506174400448799,
      "num_tokens": 194586858.0,
      "step": 93780
    },
    {
      "entropy": 5.7215046882629395,
      "epoch": 8.507347605224965,
      "grad_norm": 1.5,
      "learning_rate": 7.470452651878438e-05,
      "loss": 4.8567,
      "mean_token_accuracy": 0.24071243405342102,
      "num_tokens": 194597374.0,
      "step": 93785
    },
    {
      "entropy": 5.692077779769898,
      "epoch": 8.507801161103048,
      "grad_norm": 1.3828125,
      "learning_rate": 7.468978811110039e-05,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23707726001739501,
      "num_tokens": 194607222.0,
      "step": 93790
    },
    {
      "entropy": 5.682783079147339,
      "epoch": 8.508254716981131,
      "grad_norm": 1.4296875,
      "learning_rate": 7.46750538458826e-05,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2421549767255783,
      "num_tokens": 194617940.0,
      "step": 93795
    },
    {
      "entropy": 5.660887622833252,
      "epoch": 8.508708272859216,
      "grad_norm": 1.3984375,
      "learning_rate": 7.466032372343572e-05,
      "loss": 4.7207,
      "mean_token_accuracy": 0.24176487028598787,
      "num_tokens": 194627974.0,
      "step": 93800
    },
    {
      "entropy": 5.696955013275146,
      "epoch": 8.5091618287373,
      "grad_norm": 1.59375,
      "learning_rate": 7.464559774406442e-05,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2285324051976204,
      "num_tokens": 194638071.0,
      "step": 93805
    },
    {
      "entropy": 5.705987024307251,
      "epoch": 8.509615384615385,
      "grad_norm": 1.4765625,
      "learning_rate": 7.463087590807324e-05,
      "loss": 4.7724,
      "mean_token_accuracy": 0.24416628032922744,
      "num_tokens": 194648548.0,
      "step": 93810
    },
    {
      "entropy": 5.687578248977661,
      "epoch": 8.510068940493468,
      "grad_norm": 1.4765625,
      "learning_rate": 7.461615821576661e-05,
      "loss": 4.7854,
      "mean_token_accuracy": 0.2437348946928978,
      "num_tokens": 194659794.0,
      "step": 93815
    },
    {
      "entropy": 5.661159610748291,
      "epoch": 8.510522496371554,
      "grad_norm": 1.3203125,
      "learning_rate": 7.46014446674489e-05,
      "loss": 4.7722,
      "mean_token_accuracy": 0.24172955751419067,
      "num_tokens": 194670694.0,
      "step": 93820
    },
    {
      "entropy": 5.733769226074219,
      "epoch": 8.510976052249637,
      "grad_norm": 1.484375,
      "learning_rate": 7.458673526342436e-05,
      "loss": 4.8728,
      "mean_token_accuracy": 0.23263511955738067,
      "num_tokens": 194680762.0,
      "step": 93825
    },
    {
      "entropy": 5.680635118484497,
      "epoch": 8.511429608127722,
      "grad_norm": 1.5625,
      "learning_rate": 7.45720300039972e-05,
      "loss": 4.7987,
      "mean_token_accuracy": 0.24010874330997467,
      "num_tokens": 194691178.0,
      "step": 93830
    },
    {
      "entropy": 5.7112163543701175,
      "epoch": 8.511883164005805,
      "grad_norm": 1.5625,
      "learning_rate": 7.45573288894715e-05,
      "loss": 4.8544,
      "mean_token_accuracy": 0.23455893248319626,
      "num_tokens": 194700985.0,
      "step": 93835
    },
    {
      "entropy": 5.6854864120483395,
      "epoch": 8.51233671988389,
      "grad_norm": 1.4609375,
      "learning_rate": 7.454263192015134e-05,
      "loss": 4.8442,
      "mean_token_accuracy": 0.2338208556175232,
      "num_tokens": 194711540.0,
      "step": 93840
    },
    {
      "entropy": 5.6987025260925295,
      "epoch": 8.512790275761974,
      "grad_norm": 1.5078125,
      "learning_rate": 7.452793909634068e-05,
      "loss": 4.7947,
      "mean_token_accuracy": 0.2366634875535965,
      "num_tokens": 194720803.0,
      "step": 93845
    },
    {
      "entropy": 5.620936727523803,
      "epoch": 8.513243831640057,
      "grad_norm": 1.5,
      "learning_rate": 7.451325041834328e-05,
      "loss": 4.7187,
      "mean_token_accuracy": 0.24061695784330367,
      "num_tokens": 194732312.0,
      "step": 93850
    },
    {
      "entropy": 5.66216139793396,
      "epoch": 8.513697387518143,
      "grad_norm": 1.46875,
      "learning_rate": 7.449856588646299e-05,
      "loss": 4.8274,
      "mean_token_accuracy": 0.22902048379182816,
      "num_tokens": 194743212.0,
      "step": 93855
    },
    {
      "entropy": 5.635796070098877,
      "epoch": 8.514150943396226,
      "grad_norm": 1.546875,
      "learning_rate": 7.448388550100343e-05,
      "loss": 4.721,
      "mean_token_accuracy": 0.24024134874343872,
      "num_tokens": 194752847.0,
      "step": 93860
    },
    {
      "entropy": 5.658990955352783,
      "epoch": 8.514604499274311,
      "grad_norm": 1.4609375,
      "learning_rate": 7.446920926226822e-05,
      "loss": 4.7671,
      "mean_token_accuracy": 0.24240746945142747,
      "num_tokens": 194764121.0,
      "step": 93865
    },
    {
      "entropy": 5.6856043338775635,
      "epoch": 8.515058055152394,
      "grad_norm": 1.4453125,
      "learning_rate": 7.445453717056082e-05,
      "loss": 4.8043,
      "mean_token_accuracy": 0.2401679664850235,
      "num_tokens": 194774138.0,
      "step": 93870
    },
    {
      "entropy": 5.750414085388184,
      "epoch": 8.51551161103048,
      "grad_norm": 1.4765625,
      "learning_rate": 7.443986922618475e-05,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2246122255921364,
      "num_tokens": 194784915.0,
      "step": 93875
    },
    {
      "entropy": 5.612877321243286,
      "epoch": 8.515965166908563,
      "grad_norm": 1.578125,
      "learning_rate": 7.442520542944328e-05,
      "loss": 4.7331,
      "mean_token_accuracy": 0.24199249297380448,
      "num_tokens": 194794579.0,
      "step": 93880
    },
    {
      "entropy": 5.728250360488891,
      "epoch": 8.516418722786648,
      "grad_norm": 1.671875,
      "learning_rate": 7.441054578063969e-05,
      "loss": 4.8401,
      "mean_token_accuracy": 0.23458222001791001,
      "num_tokens": 194804881.0,
      "step": 93885
    },
    {
      "entropy": 5.7163121700286865,
      "epoch": 8.516872278664732,
      "grad_norm": 1.578125,
      "learning_rate": 7.439589028007711e-05,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23251252770423889,
      "num_tokens": 194816183.0,
      "step": 93890
    },
    {
      "entropy": 5.645302677154541,
      "epoch": 8.517325834542815,
      "grad_norm": 1.546875,
      "learning_rate": 7.438123892805864e-05,
      "loss": 4.7483,
      "mean_token_accuracy": 0.24330573827028273,
      "num_tokens": 194826613.0,
      "step": 93895
    },
    {
      "entropy": 5.670778083801269,
      "epoch": 8.5177793904209,
      "grad_norm": 1.4765625,
      "learning_rate": 7.436659172488724e-05,
      "loss": 4.7534,
      "mean_token_accuracy": 0.24062785804271697,
      "num_tokens": 194837077.0,
      "step": 93900
    },
    {
      "entropy": 5.750493144989013,
      "epoch": 8.518232946298983,
      "grad_norm": 1.6328125,
      "learning_rate": 7.43519486708659e-05,
      "loss": 4.8385,
      "mean_token_accuracy": 0.236009319126606,
      "num_tokens": 194847250.0,
      "step": 93905
    },
    {
      "entropy": 5.659757280349732,
      "epoch": 8.518686502177069,
      "grad_norm": 1.421875,
      "learning_rate": 7.433730976629737e-05,
      "loss": 4.7734,
      "mean_token_accuracy": 0.23466332852840424,
      "num_tokens": 194858095.0,
      "step": 93910
    },
    {
      "entropy": 5.702312326431274,
      "epoch": 8.519140058055152,
      "grad_norm": 1.703125,
      "learning_rate": 7.432267501148444e-05,
      "loss": 4.8106,
      "mean_token_accuracy": 0.2372767597436905,
      "num_tokens": 194867437.0,
      "step": 93915
    },
    {
      "entropy": 5.70022144317627,
      "epoch": 8.519593613933237,
      "grad_norm": 1.5,
      "learning_rate": 7.43080444067297e-05,
      "loss": 4.7931,
      "mean_token_accuracy": 0.24161761552095412,
      "num_tokens": 194877225.0,
      "step": 93920
    },
    {
      "entropy": 5.690079689025879,
      "epoch": 8.52004716981132,
      "grad_norm": 1.65625,
      "learning_rate": 7.429341795233573e-05,
      "loss": 4.8106,
      "mean_token_accuracy": 0.23397720605134964,
      "num_tokens": 194886886.0,
      "step": 93925
    },
    {
      "entropy": 5.663588285446167,
      "epoch": 8.520500725689406,
      "grad_norm": 1.5625,
      "learning_rate": 7.427879564860505e-05,
      "loss": 4.8046,
      "mean_token_accuracy": 0.2353224501013756,
      "num_tokens": 194896536.0,
      "step": 93930
    },
    {
      "entropy": 5.740483522415161,
      "epoch": 8.520954281567489,
      "grad_norm": 1.5546875,
      "learning_rate": 7.426417749584e-05,
      "loss": 4.8891,
      "mean_token_accuracy": 0.23010599613189697,
      "num_tokens": 194907429.0,
      "step": 93935
    },
    {
      "entropy": 5.603059148788452,
      "epoch": 8.521407837445572,
      "grad_norm": 1.6015625,
      "learning_rate": 7.424956349434293e-05,
      "loss": 4.6846,
      "mean_token_accuracy": 0.2443855345249176,
      "num_tokens": 194917412.0,
      "step": 93940
    },
    {
      "entropy": 5.618256616592407,
      "epoch": 8.521861393323658,
      "grad_norm": 1.65625,
      "learning_rate": 7.423495364441606e-05,
      "loss": 4.719,
      "mean_token_accuracy": 0.24244046807289124,
      "num_tokens": 194928262.0,
      "step": 93945
    },
    {
      "entropy": 5.72190294265747,
      "epoch": 8.522314949201741,
      "grad_norm": 1.4765625,
      "learning_rate": 7.422034794636149e-05,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2344077482819557,
      "num_tokens": 194940478.0,
      "step": 93950
    },
    {
      "entropy": 5.668279838562012,
      "epoch": 8.522768505079826,
      "grad_norm": 1.6953125,
      "learning_rate": 7.420574640048131e-05,
      "loss": 4.7809,
      "mean_token_accuracy": 0.23416607826948166,
      "num_tokens": 194949735.0,
      "step": 93955
    },
    {
      "entropy": 5.687791395187378,
      "epoch": 8.52322206095791,
      "grad_norm": 1.5546875,
      "learning_rate": 7.419114900707746e-05,
      "loss": 4.7674,
      "mean_token_accuracy": 0.23650048673152924,
      "num_tokens": 194959696.0,
      "step": 93960
    },
    {
      "entropy": 5.694263172149658,
      "epoch": 8.523675616835995,
      "grad_norm": 1.5625,
      "learning_rate": 7.41765557664518e-05,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23107799589633943,
      "num_tokens": 194970600.0,
      "step": 93965
    },
    {
      "entropy": 5.628978490829468,
      "epoch": 8.524129172714078,
      "grad_norm": 1.546875,
      "learning_rate": 7.416196667890611e-05,
      "loss": 4.7261,
      "mean_token_accuracy": 0.24162825345993041,
      "num_tokens": 194981837.0,
      "step": 93970
    },
    {
      "entropy": 5.664199590682983,
      "epoch": 8.524582728592163,
      "grad_norm": 1.453125,
      "learning_rate": 7.414738174474224e-05,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23597351014614104,
      "num_tokens": 194991574.0,
      "step": 93975
    },
    {
      "entropy": 5.669850158691406,
      "epoch": 8.525036284470247,
      "grad_norm": 1.65625,
      "learning_rate": 7.413280096426162e-05,
      "loss": 4.7395,
      "mean_token_accuracy": 0.2370482861995697,
      "num_tokens": 195001426.0,
      "step": 93980
    },
    {
      "entropy": 5.685875940322876,
      "epoch": 8.525489840348332,
      "grad_norm": 1.4453125,
      "learning_rate": 7.411822433776586e-05,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23326359391212464,
      "num_tokens": 195011992.0,
      "step": 93985
    },
    {
      "entropy": 5.711050319671631,
      "epoch": 8.525943396226415,
      "grad_norm": 1.4609375,
      "learning_rate": 7.410365186555644e-05,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23564700335264205,
      "num_tokens": 195022192.0,
      "step": 93990
    },
    {
      "entropy": 5.705986881256104,
      "epoch": 8.526396952104498,
      "grad_norm": 1.4296875,
      "learning_rate": 7.408908354793464e-05,
      "loss": 4.7496,
      "mean_token_accuracy": 0.239516182243824,
      "num_tokens": 195032682.0,
      "step": 93995
    },
    {
      "entropy": 5.655035018920898,
      "epoch": 8.526850507982584,
      "grad_norm": 1.4453125,
      "learning_rate": 7.407451938520185e-05,
      "loss": 4.757,
      "mean_token_accuracy": 0.23572385609149932,
      "num_tokens": 195043376.0,
      "step": 94000
    },
    {
      "entropy": 5.720546627044678,
      "epoch": 8.527304063860667,
      "grad_norm": 1.390625,
      "learning_rate": 7.405995937765919e-05,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2452506884932518,
      "num_tokens": 195053997.0,
      "step": 94005
    },
    {
      "entropy": 5.676597166061401,
      "epoch": 8.527757619738752,
      "grad_norm": 1.5703125,
      "learning_rate": 7.404540352560776e-05,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23944885432720184,
      "num_tokens": 195063031.0,
      "step": 94010
    },
    {
      "entropy": 5.74200472831726,
      "epoch": 8.528211175616836,
      "grad_norm": 1.5625,
      "learning_rate": 7.403085182934859e-05,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22729749083518982,
      "num_tokens": 195073206.0,
      "step": 94015
    },
    {
      "entropy": 5.659286308288574,
      "epoch": 8.52866473149492,
      "grad_norm": 1.40625,
      "learning_rate": 7.401630428918264e-05,
      "loss": 4.7318,
      "mean_token_accuracy": 0.2441360682249069,
      "num_tokens": 195083819.0,
      "step": 94020
    },
    {
      "entropy": 5.655841827392578,
      "epoch": 8.529118287373004,
      "grad_norm": 1.5390625,
      "learning_rate": 7.400176090541073e-05,
      "loss": 4.7708,
      "mean_token_accuracy": 0.24328860640525818,
      "num_tokens": 195093937.0,
      "step": 94025
    },
    {
      "entropy": 5.675010824203492,
      "epoch": 8.52957184325109,
      "grad_norm": 1.5859375,
      "learning_rate": 7.398722167833356e-05,
      "loss": 4.7867,
      "mean_token_accuracy": 0.23476916402578354,
      "num_tokens": 195104033.0,
      "step": 94030
    },
    {
      "entropy": 5.768413305282593,
      "epoch": 8.530025399129173,
      "grad_norm": 1.390625,
      "learning_rate": 7.397268660825196e-05,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2312764272093773,
      "num_tokens": 195115381.0,
      "step": 94035
    },
    {
      "entropy": 5.6836179256439205,
      "epoch": 8.530478955007258,
      "grad_norm": 1.421875,
      "learning_rate": 7.395815569546638e-05,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23757259249687196,
      "num_tokens": 195127071.0,
      "step": 94040
    },
    {
      "entropy": 5.702208518981934,
      "epoch": 8.530932510885341,
      "grad_norm": 1.4921875,
      "learning_rate": 7.39436289402774e-05,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22833265364170074,
      "num_tokens": 195137671.0,
      "step": 94045
    },
    {
      "entropy": 5.6328857898712155,
      "epoch": 8.531386066763424,
      "grad_norm": 1.578125,
      "learning_rate": 7.392910634298539e-05,
      "loss": 4.7406,
      "mean_token_accuracy": 0.24570234417915343,
      "num_tokens": 195148171.0,
      "step": 94050
    },
    {
      "entropy": 5.637293004989624,
      "epoch": 8.53183962264151,
      "grad_norm": 1.6328125,
      "learning_rate": 7.391458790389071e-05,
      "loss": 4.7694,
      "mean_token_accuracy": 0.234410460293293,
      "num_tokens": 195157693.0,
      "step": 94055
    },
    {
      "entropy": 5.723610544204712,
      "epoch": 8.532293178519593,
      "grad_norm": 1.3984375,
      "learning_rate": 7.39000736232936e-05,
      "loss": 4.7922,
      "mean_token_accuracy": 0.2387584000825882,
      "num_tokens": 195167951.0,
      "step": 94060
    },
    {
      "entropy": 5.617111253738403,
      "epoch": 8.532746734397678,
      "grad_norm": 1.5390625,
      "learning_rate": 7.388556350149417e-05,
      "loss": 4.7395,
      "mean_token_accuracy": 0.23967081010341645,
      "num_tokens": 195179138.0,
      "step": 94065
    },
    {
      "entropy": 5.689618396759033,
      "epoch": 8.533200290275762,
      "grad_norm": 1.5234375,
      "learning_rate": 7.387105753879258e-05,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23239668607711791,
      "num_tokens": 195189716.0,
      "step": 94070
    },
    {
      "entropy": 5.644873332977295,
      "epoch": 8.533653846153847,
      "grad_norm": 1.5703125,
      "learning_rate": 7.385655573548875e-05,
      "loss": 4.6768,
      "mean_token_accuracy": 0.24712323099374772,
      "num_tokens": 195199961.0,
      "step": 94075
    },
    {
      "entropy": 5.69436206817627,
      "epoch": 8.53410740203193,
      "grad_norm": 1.609375,
      "learning_rate": 7.384205809188262e-05,
      "loss": 4.7771,
      "mean_token_accuracy": 0.24402301013469696,
      "num_tokens": 195210577.0,
      "step": 94080
    },
    {
      "entropy": 5.714770793914795,
      "epoch": 8.534560957910015,
      "grad_norm": 1.6328125,
      "learning_rate": 7.382756460827402e-05,
      "loss": 4.82,
      "mean_token_accuracy": 0.23405176550149917,
      "num_tokens": 195220325.0,
      "step": 94085
    },
    {
      "entropy": 5.650350189208984,
      "epoch": 8.535014513788099,
      "grad_norm": 1.6015625,
      "learning_rate": 7.38130752849626e-05,
      "loss": 4.7434,
      "mean_token_accuracy": 0.2452193856239319,
      "num_tokens": 195229745.0,
      "step": 94090
    },
    {
      "entropy": 5.700692081451416,
      "epoch": 8.535468069666184,
      "grad_norm": 1.546875,
      "learning_rate": 7.379859012224807e-05,
      "loss": 4.8019,
      "mean_token_accuracy": 0.2382549077272415,
      "num_tokens": 195239822.0,
      "step": 94095
    },
    {
      "entropy": 5.668958854675293,
      "epoch": 8.535921625544267,
      "grad_norm": 1.5234375,
      "learning_rate": 7.378410912042993e-05,
      "loss": 4.7894,
      "mean_token_accuracy": 0.24201864153146743,
      "num_tokens": 195250247.0,
      "step": 94100
    },
    {
      "entropy": 5.608560514450073,
      "epoch": 8.53637518142235,
      "grad_norm": 1.4765625,
      "learning_rate": 7.376963227980774e-05,
      "loss": 4.7315,
      "mean_token_accuracy": 0.23886346817016602,
      "num_tokens": 195262383.0,
      "step": 94105
    },
    {
      "entropy": 5.625151205062866,
      "epoch": 8.536828737300436,
      "grad_norm": 1.359375,
      "learning_rate": 7.375515960068083e-05,
      "loss": 4.7029,
      "mean_token_accuracy": 0.2439226284623146,
      "num_tokens": 195273895.0,
      "step": 94110
    },
    {
      "entropy": 5.709638690948486,
      "epoch": 8.537282293178519,
      "grad_norm": 1.640625,
      "learning_rate": 7.37406910833485e-05,
      "loss": 4.8073,
      "mean_token_accuracy": 0.240726500749588,
      "num_tokens": 195283640.0,
      "step": 94115
    },
    {
      "entropy": 5.657980918884277,
      "epoch": 8.537735849056604,
      "grad_norm": 1.7109375,
      "learning_rate": 7.372622672810995e-05,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2480802372097969,
      "num_tokens": 195293585.0,
      "step": 94120
    },
    {
      "entropy": 5.6294358253479,
      "epoch": 8.538189404934688,
      "grad_norm": 1.46875,
      "learning_rate": 7.371176653526435e-05,
      "loss": 4.6793,
      "mean_token_accuracy": 0.24644236415624618,
      "num_tokens": 195304287.0,
      "step": 94125
    },
    {
      "entropy": 5.612281227111817,
      "epoch": 8.538642960812773,
      "grad_norm": 1.5625,
      "learning_rate": 7.369731050511069e-05,
      "loss": 4.712,
      "mean_token_accuracy": 0.2435792401432991,
      "num_tokens": 195314159.0,
      "step": 94130
    },
    {
      "entropy": 5.722635698318482,
      "epoch": 8.539096516690856,
      "grad_norm": 1.515625,
      "learning_rate": 7.368285863794792e-05,
      "loss": 4.8621,
      "mean_token_accuracy": 0.23010614663362502,
      "num_tokens": 195325428.0,
      "step": 94135
    },
    {
      "entropy": 5.714094114303589,
      "epoch": 8.539550072568941,
      "grad_norm": 1.5625,
      "learning_rate": 7.366841093407495e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.23804748803377151,
      "num_tokens": 195336176.0,
      "step": 94140
    },
    {
      "entropy": 5.679920053482055,
      "epoch": 8.540003628447025,
      "grad_norm": 1.5234375,
      "learning_rate": 7.365396739379058e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.24184437841176987,
      "num_tokens": 195347129.0,
      "step": 94145
    },
    {
      "entropy": 5.668152904510498,
      "epoch": 8.540457184325108,
      "grad_norm": 1.5703125,
      "learning_rate": 7.363952801739346e-05,
      "loss": 4.7537,
      "mean_token_accuracy": 0.23977120816707612,
      "num_tokens": 195357338.0,
      "step": 94150
    },
    {
      "entropy": 5.766324520111084,
      "epoch": 8.540910740203193,
      "grad_norm": 1.625,
      "learning_rate": 7.362509280518221e-05,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2297292485833168,
      "num_tokens": 195367650.0,
      "step": 94155
    },
    {
      "entropy": 5.760770702362061,
      "epoch": 8.541364296081277,
      "grad_norm": 1.578125,
      "learning_rate": 7.361066175745536e-05,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2416550874710083,
      "num_tokens": 195378289.0,
      "step": 94160
    },
    {
      "entropy": 5.67740912437439,
      "epoch": 8.541817851959362,
      "grad_norm": 1.3671875,
      "learning_rate": 7.359623487451135e-05,
      "loss": 4.8082,
      "mean_token_accuracy": 0.23872574865818025,
      "num_tokens": 195389165.0,
      "step": 94165
    },
    {
      "entropy": 5.668388795852661,
      "epoch": 8.542271407837445,
      "grad_norm": 1.734375,
      "learning_rate": 7.358181215664846e-05,
      "loss": 4.716,
      "mean_token_accuracy": 0.249479441344738,
      "num_tokens": 195399879.0,
      "step": 94170
    },
    {
      "entropy": 5.7310902118682865,
      "epoch": 8.54272496371553,
      "grad_norm": 1.609375,
      "learning_rate": 7.35673936041651e-05,
      "loss": 4.8553,
      "mean_token_accuracy": 0.23467669188976287,
      "num_tokens": 195410421.0,
      "step": 94175
    },
    {
      "entropy": 5.724639463424682,
      "epoch": 8.543178519593614,
      "grad_norm": 1.59375,
      "learning_rate": 7.355297921735937e-05,
      "loss": 4.7579,
      "mean_token_accuracy": 0.23506741225719452,
      "num_tokens": 195419805.0,
      "step": 94180
    },
    {
      "entropy": 5.69322099685669,
      "epoch": 8.543632075471699,
      "grad_norm": 1.421875,
      "learning_rate": 7.353856899652935e-05,
      "loss": 4.7513,
      "mean_token_accuracy": 0.23476245254278183,
      "num_tokens": 195430339.0,
      "step": 94185
    },
    {
      "entropy": 5.636162567138672,
      "epoch": 8.544085631349782,
      "grad_norm": 1.46875,
      "learning_rate": 7.352416294197306e-05,
      "loss": 4.7487,
      "mean_token_accuracy": 0.24125289618968965,
      "num_tokens": 195439955.0,
      "step": 94190
    },
    {
      "entropy": 5.7033086776733395,
      "epoch": 8.544539187227866,
      "grad_norm": 1.6171875,
      "learning_rate": 7.350976105398842e-05,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2376231610774994,
      "num_tokens": 195449151.0,
      "step": 94195
    },
    {
      "entropy": 5.725208759307861,
      "epoch": 8.54499274310595,
      "grad_norm": 1.359375,
      "learning_rate": 7.349536333287324e-05,
      "loss": 4.8599,
      "mean_token_accuracy": 0.2319241389632225,
      "num_tokens": 195460067.0,
      "step": 94200
    },
    {
      "entropy": 5.655210828781128,
      "epoch": 8.545446298984034,
      "grad_norm": 1.359375,
      "learning_rate": 7.348096977892532e-05,
      "loss": 4.7277,
      "mean_token_accuracy": 0.2350369319319725,
      "num_tokens": 195470377.0,
      "step": 94205
    },
    {
      "entropy": 5.683418798446655,
      "epoch": 8.54589985486212,
      "grad_norm": 1.46875,
      "learning_rate": 7.34665803924423e-05,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23749839067459105,
      "num_tokens": 195481062.0,
      "step": 94210
    },
    {
      "entropy": 5.641357135772705,
      "epoch": 8.546353410740203,
      "grad_norm": 1.6328125,
      "learning_rate": 7.345219517372176e-05,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2300418049097061,
      "num_tokens": 195492279.0,
      "step": 94215
    },
    {
      "entropy": 5.744388437271118,
      "epoch": 8.546806966618288,
      "grad_norm": 1.7109375,
      "learning_rate": 7.34378141230612e-05,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2359330549836159,
      "num_tokens": 195501637.0,
      "step": 94220
    },
    {
      "entropy": 5.627642822265625,
      "epoch": 8.547260522496371,
      "grad_norm": 1.46875,
      "learning_rate": 7.342343724075798e-05,
      "loss": 4.7338,
      "mean_token_accuracy": 0.2473328649997711,
      "num_tokens": 195511874.0,
      "step": 94225
    },
    {
      "entropy": 5.6389518737792965,
      "epoch": 8.547714078374456,
      "grad_norm": 1.5390625,
      "learning_rate": 7.340906452710946e-05,
      "loss": 4.6693,
      "mean_token_accuracy": 0.2474084824323654,
      "num_tokens": 195521387.0,
      "step": 94230
    },
    {
      "entropy": 5.603719902038574,
      "epoch": 8.54816763425254,
      "grad_norm": 1.5390625,
      "learning_rate": 7.339469598241281e-05,
      "loss": 4.6795,
      "mean_token_accuracy": 0.24668262749910355,
      "num_tokens": 195531339.0,
      "step": 94235
    },
    {
      "entropy": 5.660577917098999,
      "epoch": 8.548621190130625,
      "grad_norm": 1.53125,
      "learning_rate": 7.338033160696525e-05,
      "loss": 4.765,
      "mean_token_accuracy": 0.2425484612584114,
      "num_tokens": 195541519.0,
      "step": 94240
    },
    {
      "entropy": 5.626969671249389,
      "epoch": 8.549074746008708,
      "grad_norm": 1.5625,
      "learning_rate": 7.336597140106384e-05,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2332247644662857,
      "num_tokens": 195552213.0,
      "step": 94245
    },
    {
      "entropy": 5.675989151000977,
      "epoch": 8.549528301886792,
      "grad_norm": 1.546875,
      "learning_rate": 7.335161536500553e-05,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23355694264173507,
      "num_tokens": 195561468.0,
      "step": 94250
    },
    {
      "entropy": 5.6671562671661375,
      "epoch": 8.549981857764877,
      "grad_norm": 1.5,
      "learning_rate": 7.333726349908717e-05,
      "loss": 4.7359,
      "mean_token_accuracy": 0.24166358411312103,
      "num_tokens": 195571629.0,
      "step": 94255
    },
    {
      "entropy": 5.819631481170655,
      "epoch": 8.55043541364296,
      "grad_norm": 1.5546875,
      "learning_rate": 7.332291580360561e-05,
      "loss": 4.9823,
      "mean_token_accuracy": 0.22282091975212098,
      "num_tokens": 195582801.0,
      "step": 94260
    },
    {
      "entropy": 5.67402138710022,
      "epoch": 8.550888969521045,
      "grad_norm": 1.4921875,
      "learning_rate": 7.330857227885752e-05,
      "loss": 4.7514,
      "mean_token_accuracy": 0.24114301353693007,
      "num_tokens": 195595234.0,
      "step": 94265
    },
    {
      "entropy": 5.693351316452026,
      "epoch": 8.551342525399129,
      "grad_norm": 1.5390625,
      "learning_rate": 7.329423292513954e-05,
      "loss": 4.8172,
      "mean_token_accuracy": 0.2291499614715576,
      "num_tokens": 195605847.0,
      "step": 94270
    },
    {
      "entropy": 5.6541553974151615,
      "epoch": 8.551796081277214,
      "grad_norm": 1.546875,
      "learning_rate": 7.327989774274825e-05,
      "loss": 4.8026,
      "mean_token_accuracy": 0.23711526840925218,
      "num_tokens": 195616186.0,
      "step": 94275
    },
    {
      "entropy": 5.807525777816773,
      "epoch": 8.552249637155297,
      "grad_norm": 1.5546875,
      "learning_rate": 7.32655667319801e-05,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22861477732658386,
      "num_tokens": 195627463.0,
      "step": 94280
    },
    {
      "entropy": 5.639752197265625,
      "epoch": 8.552703193033382,
      "grad_norm": 1.453125,
      "learning_rate": 7.325123989313141e-05,
      "loss": 4.6677,
      "mean_token_accuracy": 0.24762416630983353,
      "num_tokens": 195637034.0,
      "step": 94285
    },
    {
      "entropy": 5.701354312896728,
      "epoch": 8.553156748911466,
      "grad_norm": 1.4765625,
      "learning_rate": 7.32369172264985e-05,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23562699407339097,
      "num_tokens": 195648644.0,
      "step": 94290
    },
    {
      "entropy": 5.725522708892822,
      "epoch": 8.55361030478955,
      "grad_norm": 1.578125,
      "learning_rate": 7.322259873237756e-05,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23981697410345076,
      "num_tokens": 195657832.0,
      "step": 94295
    },
    {
      "entropy": 5.674843740463257,
      "epoch": 8.554063860667634,
      "grad_norm": 1.546875,
      "learning_rate": 7.32082844110647e-05,
      "loss": 4.7697,
      "mean_token_accuracy": 0.2393873006105423,
      "num_tokens": 195668405.0,
      "step": 94300
    },
    {
      "entropy": 5.6688862323760985,
      "epoch": 8.554517416545718,
      "grad_norm": 1.4921875,
      "learning_rate": 7.319397426285589e-05,
      "loss": 4.7575,
      "mean_token_accuracy": 0.24280677884817123,
      "num_tokens": 195679396.0,
      "step": 94305
    },
    {
      "entropy": 5.677715921401978,
      "epoch": 8.554970972423803,
      "grad_norm": 1.53125,
      "learning_rate": 7.317966828804717e-05,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23169533908367157,
      "num_tokens": 195689645.0,
      "step": 94310
    },
    {
      "entropy": 5.670571231842041,
      "epoch": 8.555424528301886,
      "grad_norm": 1.59375,
      "learning_rate": 7.316536648693434e-05,
      "loss": 4.8083,
      "mean_token_accuracy": 0.23344988226890565,
      "num_tokens": 195699566.0,
      "step": 94315
    },
    {
      "entropy": 5.704926443099976,
      "epoch": 8.555878084179971,
      "grad_norm": 1.421875,
      "learning_rate": 7.315106885981316e-05,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22960792928934098,
      "num_tokens": 195710289.0,
      "step": 94320
    },
    {
      "entropy": 5.69570779800415,
      "epoch": 8.556331640058055,
      "grad_norm": 1.7109375,
      "learning_rate": 7.313677540697933e-05,
      "loss": 4.7775,
      "mean_token_accuracy": 0.241356760263443,
      "num_tokens": 195720079.0,
      "step": 94325
    },
    {
      "entropy": 5.667790651321411,
      "epoch": 8.55678519593614,
      "grad_norm": 1.5,
      "learning_rate": 7.31224861287284e-05,
      "loss": 4.8033,
      "mean_token_accuracy": 0.23674533516168594,
      "num_tokens": 195730943.0,
      "step": 94330
    },
    {
      "entropy": 5.69333438873291,
      "epoch": 8.557238751814223,
      "grad_norm": 1.5625,
      "learning_rate": 7.31082010253559e-05,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23865926265716553,
      "num_tokens": 195741267.0,
      "step": 94335
    },
    {
      "entropy": 5.68373384475708,
      "epoch": 8.557692307692308,
      "grad_norm": 1.5390625,
      "learning_rate": 7.309392009715724e-05,
      "loss": 4.8578,
      "mean_token_accuracy": 0.23334995657205582,
      "num_tokens": 195751541.0,
      "step": 94340
    },
    {
      "entropy": 5.656467056274414,
      "epoch": 8.558145863570392,
      "grad_norm": 1.5859375,
      "learning_rate": 7.307964334442778e-05,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23778942227363586,
      "num_tokens": 195761429.0,
      "step": 94345
    },
    {
      "entropy": 5.678757286071777,
      "epoch": 8.558599419448477,
      "grad_norm": 1.4765625,
      "learning_rate": 7.306537076746274e-05,
      "loss": 4.7832,
      "mean_token_accuracy": 0.23660916537046434,
      "num_tokens": 195771754.0,
      "step": 94350
    },
    {
      "entropy": 5.7558674812316895,
      "epoch": 8.55905297532656,
      "grad_norm": 1.6484375,
      "learning_rate": 7.305110236655732e-05,
      "loss": 4.9801,
      "mean_token_accuracy": 0.22032182514667512,
      "num_tokens": 195782448.0,
      "step": 94355
    },
    {
      "entropy": 5.696052742004395,
      "epoch": 8.559506531204644,
      "grad_norm": 1.6875,
      "learning_rate": 7.303683814200655e-05,
      "loss": 4.7565,
      "mean_token_accuracy": 0.24218276739120484,
      "num_tokens": 195791634.0,
      "step": 94360
    },
    {
      "entropy": 5.5764467239379885,
      "epoch": 8.559960087082729,
      "grad_norm": 1.5234375,
      "learning_rate": 7.302257809410544e-05,
      "loss": 4.6359,
      "mean_token_accuracy": 0.25293537229299545,
      "num_tokens": 195801735.0,
      "step": 94365
    },
    {
      "entropy": 5.692608594894409,
      "epoch": 8.560413642960812,
      "grad_norm": 1.4375,
      "learning_rate": 7.300832222314888e-05,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2413598507642746,
      "num_tokens": 195811883.0,
      "step": 94370
    },
    {
      "entropy": 5.670693588256836,
      "epoch": 8.560867198838897,
      "grad_norm": 1.6328125,
      "learning_rate": 7.299407052943166e-05,
      "loss": 4.7663,
      "mean_token_accuracy": 0.2400139883160591,
      "num_tokens": 195822981.0,
      "step": 94375
    },
    {
      "entropy": 5.66664981842041,
      "epoch": 8.56132075471698,
      "grad_norm": 1.609375,
      "learning_rate": 7.297982301324857e-05,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2392853245139122,
      "num_tokens": 195832785.0,
      "step": 94380
    },
    {
      "entropy": 5.655135250091552,
      "epoch": 8.561774310595066,
      "grad_norm": 1.546875,
      "learning_rate": 7.296557967489422e-05,
      "loss": 4.8207,
      "mean_token_accuracy": 0.2308182090520859,
      "num_tokens": 195843176.0,
      "step": 94385
    },
    {
      "entropy": 5.733922910690308,
      "epoch": 8.56222786647315,
      "grad_norm": 1.4609375,
      "learning_rate": 7.295134051466318e-05,
      "loss": 4.8283,
      "mean_token_accuracy": 0.2353613406419754,
      "num_tokens": 195854085.0,
      "step": 94390
    },
    {
      "entropy": 5.565691137313843,
      "epoch": 8.562681422351234,
      "grad_norm": 1.5078125,
      "learning_rate": 7.293710553284988e-05,
      "loss": 4.7293,
      "mean_token_accuracy": 0.23832617551088334,
      "num_tokens": 195864579.0,
      "step": 94395
    },
    {
      "entropy": 5.609237241744995,
      "epoch": 8.563134978229318,
      "grad_norm": 1.3828125,
      "learning_rate": 7.292287472974875e-05,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2480481669306755,
      "num_tokens": 195874966.0,
      "step": 94400
    },
    {
      "entropy": 5.63356647491455,
      "epoch": 8.563588534107403,
      "grad_norm": 1.6015625,
      "learning_rate": 7.290864810565405e-05,
      "loss": 4.7213,
      "mean_token_accuracy": 0.25247475504875183,
      "num_tokens": 195885228.0,
      "step": 94405
    },
    {
      "entropy": 5.686657524108886,
      "epoch": 8.564042089985486,
      "grad_norm": 1.4609375,
      "learning_rate": 7.289442566086002e-05,
      "loss": 4.8452,
      "mean_token_accuracy": 0.23447750210762025,
      "num_tokens": 195896813.0,
      "step": 94410
    },
    {
      "entropy": 5.595412254333496,
      "epoch": 8.56449564586357,
      "grad_norm": 1.4296875,
      "learning_rate": 7.288020739566075e-05,
      "loss": 4.6965,
      "mean_token_accuracy": 0.248620143532753,
      "num_tokens": 195907218.0,
      "step": 94415
    },
    {
      "entropy": 5.711633396148682,
      "epoch": 8.564949201741655,
      "grad_norm": 1.53125,
      "learning_rate": 7.286599331035034e-05,
      "loss": 4.8349,
      "mean_token_accuracy": 0.23132581859827042,
      "num_tokens": 195917708.0,
      "step": 94420
    },
    {
      "entropy": 5.673502969741821,
      "epoch": 8.565402757619738,
      "grad_norm": 1.546875,
      "learning_rate": 7.285178340522266e-05,
      "loss": 4.7558,
      "mean_token_accuracy": 0.24354860931634903,
      "num_tokens": 195928304.0,
      "step": 94425
    },
    {
      "entropy": 5.703619337081909,
      "epoch": 8.565856313497823,
      "grad_norm": 1.5078125,
      "learning_rate": 7.283757768057164e-05,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23409816175699233,
      "num_tokens": 195937455.0,
      "step": 94430
    },
    {
      "entropy": 5.670676422119141,
      "epoch": 8.566309869375907,
      "grad_norm": 1.40625,
      "learning_rate": 7.282337613669096e-05,
      "loss": 4.7434,
      "mean_token_accuracy": 0.249437615275383,
      "num_tokens": 195948397.0,
      "step": 94435
    },
    {
      "entropy": 5.6910031795501705,
      "epoch": 8.566763425253992,
      "grad_norm": 1.5625,
      "learning_rate": 7.280917877387445e-05,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23202915042638778,
      "num_tokens": 195958324.0,
      "step": 94440
    },
    {
      "entropy": 5.67702693939209,
      "epoch": 8.567216981132075,
      "grad_norm": 1.515625,
      "learning_rate": 7.279498559241563e-05,
      "loss": 4.7657,
      "mean_token_accuracy": 0.2360619455575943,
      "num_tokens": 195968812.0,
      "step": 94445
    },
    {
      "entropy": 5.679183006286621,
      "epoch": 8.56767053701016,
      "grad_norm": 1.4375,
      "learning_rate": 7.278079659260804e-05,
      "loss": 4.7768,
      "mean_token_accuracy": 0.23678742200136185,
      "num_tokens": 195979993.0,
      "step": 94450
    },
    {
      "entropy": 5.681818056106567,
      "epoch": 8.568124092888244,
      "grad_norm": 1.40625,
      "learning_rate": 7.276661177474511e-05,
      "loss": 4.7788,
      "mean_token_accuracy": 0.23662405610084533,
      "num_tokens": 195990212.0,
      "step": 94455
    },
    {
      "entropy": 5.662840890884399,
      "epoch": 8.568577648766327,
      "grad_norm": 1.453125,
      "learning_rate": 7.275243113912018e-05,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23792511224746704,
      "num_tokens": 196001148.0,
      "step": 94460
    },
    {
      "entropy": 5.695492792129516,
      "epoch": 8.569031204644412,
      "grad_norm": 1.53125,
      "learning_rate": 7.273825468602653e-05,
      "loss": 4.8536,
      "mean_token_accuracy": 0.24012472629547119,
      "num_tokens": 196011898.0,
      "step": 94465
    },
    {
      "entropy": 5.7153356075286865,
      "epoch": 8.569484760522496,
      "grad_norm": 1.453125,
      "learning_rate": 7.272408241575723e-05,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23932512402534484,
      "num_tokens": 196022572.0,
      "step": 94470
    },
    {
      "entropy": 5.754334259033203,
      "epoch": 8.569938316400581,
      "grad_norm": 1.5546875,
      "learning_rate": 7.270991432860554e-05,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2352418676018715,
      "num_tokens": 196033797.0,
      "step": 94475
    },
    {
      "entropy": 5.700373888015747,
      "epoch": 8.570391872278664,
      "grad_norm": 1.6875,
      "learning_rate": 7.269575042486436e-05,
      "loss": 4.8232,
      "mean_token_accuracy": 0.22631948590278625,
      "num_tokens": 196043389.0,
      "step": 94480
    },
    {
      "entropy": 5.650119733810425,
      "epoch": 8.57084542815675,
      "grad_norm": 1.5390625,
      "learning_rate": 7.268159070482662e-05,
      "loss": 4.74,
      "mean_token_accuracy": 0.24306602030992508,
      "num_tokens": 196054323.0,
      "step": 94485
    },
    {
      "entropy": 5.73320517539978,
      "epoch": 8.571298984034833,
      "grad_norm": 1.5234375,
      "learning_rate": 7.266743516878512e-05,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23571187257766724,
      "num_tokens": 196064874.0,
      "step": 94490
    },
    {
      "entropy": 5.638005256652832,
      "epoch": 8.571752539912918,
      "grad_norm": 1.59375,
      "learning_rate": 7.265328381703262e-05,
      "loss": 4.7266,
      "mean_token_accuracy": 0.24880505502223968,
      "num_tokens": 196075212.0,
      "step": 94495
    },
    {
      "entropy": 5.680828619003296,
      "epoch": 8.572206095791001,
      "grad_norm": 1.5,
      "learning_rate": 7.263913664986178e-05,
      "loss": 4.7628,
      "mean_token_accuracy": 0.24456705152988434,
      "num_tokens": 196085891.0,
      "step": 94500
    },
    {
      "entropy": 5.7024740219116214,
      "epoch": 8.572659651669085,
      "grad_norm": 1.53125,
      "learning_rate": 7.262499366756514e-05,
      "loss": 4.7882,
      "mean_token_accuracy": 0.2401447042822838,
      "num_tokens": 196095687.0,
      "step": 94505
    },
    {
      "entropy": 5.7137692928314205,
      "epoch": 8.57311320754717,
      "grad_norm": 1.609375,
      "learning_rate": 7.261085487043525e-05,
      "loss": 4.7865,
      "mean_token_accuracy": 0.23809174001216887,
      "num_tokens": 196105482.0,
      "step": 94510
    },
    {
      "entropy": 5.679796600341797,
      "epoch": 8.573566763425253,
      "grad_norm": 1.515625,
      "learning_rate": 7.259672025876441e-05,
      "loss": 4.7605,
      "mean_token_accuracy": 0.24407099932432175,
      "num_tokens": 196117232.0,
      "step": 94515
    },
    {
      "entropy": 5.663522148132325,
      "epoch": 8.574020319303338,
      "grad_norm": 1.484375,
      "learning_rate": 7.2582589832845e-05,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23675646483898163,
      "num_tokens": 196127944.0,
      "step": 94520
    },
    {
      "entropy": 5.629887437820434,
      "epoch": 8.574473875181422,
      "grad_norm": 1.53125,
      "learning_rate": 7.256846359296923e-05,
      "loss": 4.805,
      "mean_token_accuracy": 0.23390933573246003,
      "num_tokens": 196139270.0,
      "step": 94525
    },
    {
      "entropy": 5.7859807968139645,
      "epoch": 8.574927431059507,
      "grad_norm": 1.6328125,
      "learning_rate": 7.25543415394292e-05,
      "loss": 4.9158,
      "mean_token_accuracy": 0.23189949244260788,
      "num_tokens": 196148464.0,
      "step": 94530
    },
    {
      "entropy": 5.675659036636352,
      "epoch": 8.57538098693759,
      "grad_norm": 1.578125,
      "learning_rate": 7.254022367251692e-05,
      "loss": 4.7803,
      "mean_token_accuracy": 0.23718903809785843,
      "num_tokens": 196159482.0,
      "step": 94535
    },
    {
      "entropy": 5.678581047058105,
      "epoch": 8.575834542815675,
      "grad_norm": 1.5859375,
      "learning_rate": 7.252610999252449e-05,
      "loss": 4.7649,
      "mean_token_accuracy": 0.24647595435380937,
      "num_tokens": 196169621.0,
      "step": 94540
    },
    {
      "entropy": 5.6911022663116455,
      "epoch": 8.576288098693759,
      "grad_norm": 1.5,
      "learning_rate": 7.251200049974367e-05,
      "loss": 4.847,
      "mean_token_accuracy": 0.23341186940670014,
      "num_tokens": 196179439.0,
      "step": 94545
    },
    {
      "entropy": 5.688672590255737,
      "epoch": 8.576741654571844,
      "grad_norm": 1.5703125,
      "learning_rate": 7.249789519446628e-05,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23815277069807053,
      "num_tokens": 196190184.0,
      "step": 94550
    },
    {
      "entropy": 5.639182710647583,
      "epoch": 8.577195210449927,
      "grad_norm": 1.59375,
      "learning_rate": 7.248379407698404e-05,
      "loss": 4.7475,
      "mean_token_accuracy": 0.23852334022521973,
      "num_tokens": 196200772.0,
      "step": 94555
    },
    {
      "entropy": 5.693669939041138,
      "epoch": 8.57764876632801,
      "grad_norm": 1.421875,
      "learning_rate": 7.246969714758853e-05,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22334963828325272,
      "num_tokens": 196212233.0,
      "step": 94560
    },
    {
      "entropy": 5.752459096908569,
      "epoch": 8.578102322206096,
      "grad_norm": 1.4609375,
      "learning_rate": 7.24556044065713e-05,
      "loss": 4.922,
      "mean_token_accuracy": 0.22973606139421462,
      "num_tokens": 196222308.0,
      "step": 94565
    },
    {
      "entropy": 5.746529150009155,
      "epoch": 8.57855587808418,
      "grad_norm": 1.5234375,
      "learning_rate": 7.244151585422374e-05,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2327503576874733,
      "num_tokens": 196232267.0,
      "step": 94570
    },
    {
      "entropy": 5.65711555480957,
      "epoch": 8.579009433962264,
      "grad_norm": 1.546875,
      "learning_rate": 7.242743149083727e-05,
      "loss": 4.7128,
      "mean_token_accuracy": 0.24721868932247162,
      "num_tokens": 196242617.0,
      "step": 94575
    },
    {
      "entropy": 5.6597391128540036,
      "epoch": 8.579462989840348,
      "grad_norm": 1.546875,
      "learning_rate": 7.241335131670319e-05,
      "loss": 4.7533,
      "mean_token_accuracy": 0.24245658814907073,
      "num_tokens": 196253079.0,
      "step": 94580
    },
    {
      "entropy": 5.700562906265259,
      "epoch": 8.579916545718433,
      "grad_norm": 1.5625,
      "learning_rate": 7.23992753321126e-05,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23892745971679688,
      "num_tokens": 196263641.0,
      "step": 94585
    },
    {
      "entropy": 5.691742324829102,
      "epoch": 8.580370101596516,
      "grad_norm": 1.515625,
      "learning_rate": 7.238520353735661e-05,
      "loss": 4.7709,
      "mean_token_accuracy": 0.2354100838303566,
      "num_tokens": 196273908.0,
      "step": 94590
    },
    {
      "entropy": 5.638352251052856,
      "epoch": 8.580823657474602,
      "grad_norm": 1.625,
      "learning_rate": 7.237113593272627e-05,
      "loss": 4.7056,
      "mean_token_accuracy": 0.24435534924268723,
      "num_tokens": 196282923.0,
      "step": 94595
    },
    {
      "entropy": 5.7443708896636965,
      "epoch": 8.581277213352685,
      "grad_norm": 1.5625,
      "learning_rate": 7.235707251851247e-05,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2403291329741478,
      "num_tokens": 196295104.0,
      "step": 94600
    },
    {
      "entropy": 5.719149780273438,
      "epoch": 8.58173076923077,
      "grad_norm": 1.46875,
      "learning_rate": 7.234301329500601e-05,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22967143803834916,
      "num_tokens": 196305270.0,
      "step": 94605
    },
    {
      "entropy": 5.7358299732208256,
      "epoch": 8.582184325108853,
      "grad_norm": 1.640625,
      "learning_rate": 7.23289582624977e-05,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2342930942773819,
      "num_tokens": 196316264.0,
      "step": 94610
    },
    {
      "entropy": 5.675779199600219,
      "epoch": 8.582637880986937,
      "grad_norm": 1.5546875,
      "learning_rate": 7.231490742127819e-05,
      "loss": 4.7633,
      "mean_token_accuracy": 0.23412229865789413,
      "num_tokens": 196325897.0,
      "step": 94615
    },
    {
      "entropy": 5.704056358337402,
      "epoch": 8.583091436865022,
      "grad_norm": 1.5234375,
      "learning_rate": 7.230086077163806e-05,
      "loss": 4.7672,
      "mean_token_accuracy": 0.24103871434926988,
      "num_tokens": 196336049.0,
      "step": 94620
    },
    {
      "entropy": 5.658065986633301,
      "epoch": 8.583544992743105,
      "grad_norm": 1.4453125,
      "learning_rate": 7.228681831386776e-05,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24973852038383484,
      "num_tokens": 196346782.0,
      "step": 94625
    },
    {
      "entropy": 5.6876262664794925,
      "epoch": 8.58399854862119,
      "grad_norm": 1.40625,
      "learning_rate": 7.227278004825775e-05,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23741000145673752,
      "num_tokens": 196357012.0,
      "step": 94630
    },
    {
      "entropy": 5.68438458442688,
      "epoch": 8.584452104499274,
      "grad_norm": 1.5078125,
      "learning_rate": 7.225874597509827e-05,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2317606747150421,
      "num_tokens": 196368244.0,
      "step": 94635
    },
    {
      "entropy": 5.573481798171997,
      "epoch": 8.584905660377359,
      "grad_norm": 1.546875,
      "learning_rate": 7.224471609467957e-05,
      "loss": 4.6812,
      "mean_token_accuracy": 0.24060437977313995,
      "num_tokens": 196378699.0,
      "step": 94640
    },
    {
      "entropy": 5.711899518966675,
      "epoch": 8.585359216255442,
      "grad_norm": 1.5078125,
      "learning_rate": 7.223069040729183e-05,
      "loss": 4.8284,
      "mean_token_accuracy": 0.24115556180477143,
      "num_tokens": 196388720.0,
      "step": 94645
    },
    {
      "entropy": 5.694104480743408,
      "epoch": 8.585812772133528,
      "grad_norm": 1.4375,
      "learning_rate": 7.221666891322509e-05,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2332816869020462,
      "num_tokens": 196399426.0,
      "step": 94650
    },
    {
      "entropy": 5.653076648712158,
      "epoch": 8.586266328011611,
      "grad_norm": 1.5546875,
      "learning_rate": 7.220265161276934e-05,
      "loss": 4.7384,
      "mean_token_accuracy": 0.24366451650857926,
      "num_tokens": 196409372.0,
      "step": 94655
    },
    {
      "entropy": 5.754181337356568,
      "epoch": 8.586719883889696,
      "grad_norm": 1.5078125,
      "learning_rate": 7.218863850621441e-05,
      "loss": 4.8998,
      "mean_token_accuracy": 0.22952752858400344,
      "num_tokens": 196419962.0,
      "step": 94660
    },
    {
      "entropy": 5.79524040222168,
      "epoch": 8.58717343976778,
      "grad_norm": 1.3828125,
      "learning_rate": 7.217462959385011e-05,
      "loss": 4.8939,
      "mean_token_accuracy": 0.2301830992102623,
      "num_tokens": 196431063.0,
      "step": 94665
    },
    {
      "entropy": 5.697041416168213,
      "epoch": 8.587626995645863,
      "grad_norm": 1.625,
      "learning_rate": 7.21606248759662e-05,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2310629740357399,
      "num_tokens": 196440892.0,
      "step": 94670
    },
    {
      "entropy": 5.75733494758606,
      "epoch": 8.588080551523948,
      "grad_norm": 1.6015625,
      "learning_rate": 7.214662435285217e-05,
      "loss": 4.8562,
      "mean_token_accuracy": 0.2343314602971077,
      "num_tokens": 196450449.0,
      "step": 94675
    },
    {
      "entropy": 5.665273284912109,
      "epoch": 8.588534107402031,
      "grad_norm": 1.5546875,
      "learning_rate": 7.213262802479769e-05,
      "loss": 4.7353,
      "mean_token_accuracy": 0.24354532957077027,
      "num_tokens": 196460100.0,
      "step": 94680
    },
    {
      "entropy": 5.73109803199768,
      "epoch": 8.588987663280117,
      "grad_norm": 1.59375,
      "learning_rate": 7.211863589209215e-05,
      "loss": 4.8606,
      "mean_token_accuracy": 0.23550112992525102,
      "num_tokens": 196470312.0,
      "step": 94685
    },
    {
      "entropy": 5.699900484085083,
      "epoch": 8.5894412191582,
      "grad_norm": 1.5234375,
      "learning_rate": 7.210464795502495e-05,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23620200902223587,
      "num_tokens": 196480359.0,
      "step": 94690
    },
    {
      "entropy": 5.640513563156128,
      "epoch": 8.589894775036285,
      "grad_norm": 1.421875,
      "learning_rate": 7.209066421388533e-05,
      "loss": 4.7199,
      "mean_token_accuracy": 0.2480606347322464,
      "num_tokens": 196490547.0,
      "step": 94695
    },
    {
      "entropy": 5.684369897842407,
      "epoch": 8.590348330914368,
      "grad_norm": 1.453125,
      "learning_rate": 7.207668466896247e-05,
      "loss": 4.7576,
      "mean_token_accuracy": 0.23547848016023637,
      "num_tokens": 196501199.0,
      "step": 94700
    },
    {
      "entropy": 5.644621133804321,
      "epoch": 8.590801886792454,
      "grad_norm": 1.4609375,
      "learning_rate": 7.206270932054543e-05,
      "loss": 4.728,
      "mean_token_accuracy": 0.2414749503135681,
      "num_tokens": 196511601.0,
      "step": 94705
    },
    {
      "entropy": 5.688805055618286,
      "epoch": 8.591255442670537,
      "grad_norm": 1.5703125,
      "learning_rate": 7.204873816892332e-05,
      "loss": 4.7626,
      "mean_token_accuracy": 0.2458677664399147,
      "num_tokens": 196521281.0,
      "step": 94710
    },
    {
      "entropy": 5.698372793197632,
      "epoch": 8.591708998548622,
      "grad_norm": 1.5078125,
      "learning_rate": 7.203477121438505e-05,
      "loss": 4.7642,
      "mean_token_accuracy": 0.23540908247232437,
      "num_tokens": 196531743.0,
      "step": 94715
    },
    {
      "entropy": 5.7792706966400145,
      "epoch": 8.592162554426706,
      "grad_norm": 1.515625,
      "learning_rate": 7.202080845721939e-05,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22887277752161025,
      "num_tokens": 196542685.0,
      "step": 94720
    },
    {
      "entropy": 5.648996305465698,
      "epoch": 8.592616110304789,
      "grad_norm": 1.5,
      "learning_rate": 7.200684989771517e-05,
      "loss": 4.7505,
      "mean_token_accuracy": 0.2457796037197113,
      "num_tokens": 196552746.0,
      "step": 94725
    },
    {
      "entropy": 5.724352407455444,
      "epoch": 8.593069666182874,
      "grad_norm": 1.5859375,
      "learning_rate": 7.199289553616103e-05,
      "loss": 4.89,
      "mean_token_accuracy": 0.23233200907707213,
      "num_tokens": 196563520.0,
      "step": 94730
    },
    {
      "entropy": 5.615899276733399,
      "epoch": 8.593523222060957,
      "grad_norm": 1.5078125,
      "learning_rate": 7.197894537284551e-05,
      "loss": 4.683,
      "mean_token_accuracy": 0.24735037982463837,
      "num_tokens": 196574620.0,
      "step": 94735
    },
    {
      "entropy": 5.672311592102051,
      "epoch": 8.593976777939043,
      "grad_norm": 1.6953125,
      "learning_rate": 7.196499940805712e-05,
      "loss": 4.7402,
      "mean_token_accuracy": 0.2401280388236046,
      "num_tokens": 196584487.0,
      "step": 94740
    },
    {
      "entropy": 5.6812645435333256,
      "epoch": 8.594430333817126,
      "grad_norm": 1.578125,
      "learning_rate": 7.19510576420843e-05,
      "loss": 4.8101,
      "mean_token_accuracy": 0.2345762401819229,
      "num_tokens": 196595244.0,
      "step": 94745
    },
    {
      "entropy": 5.614375638961792,
      "epoch": 8.594883889695211,
      "grad_norm": 1.421875,
      "learning_rate": 7.193712007521538e-05,
      "loss": 4.718,
      "mean_token_accuracy": 0.23916582465171815,
      "num_tokens": 196606614.0,
      "step": 94750
    },
    {
      "entropy": 5.764910507202148,
      "epoch": 8.595337445573294,
      "grad_norm": 1.4453125,
      "learning_rate": 7.192318670773854e-05,
      "loss": 4.8753,
      "mean_token_accuracy": 0.23975420296192168,
      "num_tokens": 196618034.0,
      "step": 94755
    },
    {
      "entropy": 5.682952451705932,
      "epoch": 8.595791001451378,
      "grad_norm": 1.4921875,
      "learning_rate": 7.190925753994197e-05,
      "loss": 4.834,
      "mean_token_accuracy": 0.23435427248477936,
      "num_tokens": 196627982.0,
      "step": 94760
    },
    {
      "entropy": 5.713671445846558,
      "epoch": 8.596244557329463,
      "grad_norm": 1.578125,
      "learning_rate": 7.189533257211369e-05,
      "loss": 4.8174,
      "mean_token_accuracy": 0.22808821946382524,
      "num_tokens": 196638219.0,
      "step": 94765
    },
    {
      "entropy": 5.744552803039551,
      "epoch": 8.596698113207546,
      "grad_norm": 1.515625,
      "learning_rate": 7.18814118045417e-05,
      "loss": 4.8712,
      "mean_token_accuracy": 0.22950932383537292,
      "num_tokens": 196647538.0,
      "step": 94770
    },
    {
      "entropy": 5.66174283027649,
      "epoch": 8.597151669085632,
      "grad_norm": 1.5078125,
      "learning_rate": 7.186749523751382e-05,
      "loss": 4.7597,
      "mean_token_accuracy": 0.24075293242931367,
      "num_tokens": 196656712.0,
      "step": 94775
    },
    {
      "entropy": 5.696488380432129,
      "epoch": 8.597605224963715,
      "grad_norm": 1.6328125,
      "learning_rate": 7.185358287131795e-05,
      "loss": 4.7492,
      "mean_token_accuracy": 0.23832842111587524,
      "num_tokens": 196666236.0,
      "step": 94780
    },
    {
      "entropy": 5.723865747451782,
      "epoch": 8.5980587808418,
      "grad_norm": 1.5,
      "learning_rate": 7.183967470624176e-05,
      "loss": 4.852,
      "mean_token_accuracy": 0.23657076209783554,
      "num_tokens": 196678082.0,
      "step": 94785
    },
    {
      "entropy": 5.688290214538574,
      "epoch": 8.598512336719883,
      "grad_norm": 1.609375,
      "learning_rate": 7.182577074257286e-05,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23676783442497254,
      "num_tokens": 196687234.0,
      "step": 94790
    },
    {
      "entropy": 5.692222213745117,
      "epoch": 8.598965892597969,
      "grad_norm": 1.4375,
      "learning_rate": 7.18118709805988e-05,
      "loss": 4.7968,
      "mean_token_accuracy": 0.23788036853075029,
      "num_tokens": 196698565.0,
      "step": 94795
    },
    {
      "entropy": 5.731159830093384,
      "epoch": 8.599419448476052,
      "grad_norm": 1.453125,
      "learning_rate": 7.179797542060703e-05,
      "loss": 4.8358,
      "mean_token_accuracy": 0.23487270027399063,
      "num_tokens": 196710312.0,
      "step": 94800
    },
    {
      "entropy": 5.641317844390869,
      "epoch": 8.599873004354137,
      "grad_norm": 1.5625,
      "learning_rate": 7.178408406288491e-05,
      "loss": 4.6906,
      "mean_token_accuracy": 0.242224083840847,
      "num_tokens": 196720052.0,
      "step": 94805
    },
    {
      "entropy": 5.652112579345703,
      "epoch": 8.60032656023222,
      "grad_norm": 1.640625,
      "learning_rate": 7.177019690771967e-05,
      "loss": 4.7495,
      "mean_token_accuracy": 0.24145755618810655,
      "num_tokens": 196729953.0,
      "step": 94810
    },
    {
      "entropy": 5.664963579177856,
      "epoch": 8.600780116110304,
      "grad_norm": 1.5078125,
      "learning_rate": 7.17563139553986e-05,
      "loss": 4.7662,
      "mean_token_accuracy": 0.24032330363988877,
      "num_tokens": 196741293.0,
      "step": 94815
    },
    {
      "entropy": 5.5781848430633545,
      "epoch": 8.601233671988389,
      "grad_norm": 1.5859375,
      "learning_rate": 7.174243520620875e-05,
      "loss": 4.7233,
      "mean_token_accuracy": 0.24013564139604568,
      "num_tokens": 196751743.0,
      "step": 94820
    },
    {
      "entropy": 5.669715929031372,
      "epoch": 8.601687227866472,
      "grad_norm": 1.6875,
      "learning_rate": 7.172856066043712e-05,
      "loss": 4.7105,
      "mean_token_accuracy": 0.25009963363409043,
      "num_tokens": 196761443.0,
      "step": 94825
    },
    {
      "entropy": 5.671115159988403,
      "epoch": 8.602140783744558,
      "grad_norm": 1.6171875,
      "learning_rate": 7.171469031837072e-05,
      "loss": 4.7255,
      "mean_token_accuracy": 0.24138682782649995,
      "num_tokens": 196772260.0,
      "step": 94830
    },
    {
      "entropy": 5.660001802444458,
      "epoch": 8.602594339622641,
      "grad_norm": 1.4453125,
      "learning_rate": 7.170082418029622e-05,
      "loss": 4.7387,
      "mean_token_accuracy": 0.24183064997196196,
      "num_tokens": 196782618.0,
      "step": 94835
    },
    {
      "entropy": 5.650151681900025,
      "epoch": 8.603047895500726,
      "grad_norm": 1.4921875,
      "learning_rate": 7.168696224650053e-05,
      "loss": 4.6951,
      "mean_token_accuracy": 0.24997652024030687,
      "num_tokens": 196793008.0,
      "step": 94840
    },
    {
      "entropy": 5.765087080001831,
      "epoch": 8.60350145137881,
      "grad_norm": 1.4296875,
      "learning_rate": 7.167310451727028e-05,
      "loss": 4.9035,
      "mean_token_accuracy": 0.23501957207918167,
      "num_tokens": 196803713.0,
      "step": 94845
    },
    {
      "entropy": 5.693394899368286,
      "epoch": 8.603955007256895,
      "grad_norm": 1.5,
      "learning_rate": 7.165925099289202e-05,
      "loss": 4.7886,
      "mean_token_accuracy": 0.23908661156892777,
      "num_tokens": 196813124.0,
      "step": 94850
    },
    {
      "entropy": 5.669821548461914,
      "epoch": 8.604408563134978,
      "grad_norm": 1.546875,
      "learning_rate": 7.164540167365232e-05,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23356460481882096,
      "num_tokens": 196823466.0,
      "step": 94855
    },
    {
      "entropy": 5.638911008834839,
      "epoch": 8.604862119013063,
      "grad_norm": 1.4921875,
      "learning_rate": 7.163155655983748e-05,
      "loss": 4.7627,
      "mean_token_accuracy": 0.24194009453058243,
      "num_tokens": 196833966.0,
      "step": 94860
    },
    {
      "entropy": 5.652277183532715,
      "epoch": 8.605315674891147,
      "grad_norm": 1.578125,
      "learning_rate": 7.16177156517339e-05,
      "loss": 4.7705,
      "mean_token_accuracy": 0.2394824355840683,
      "num_tokens": 196843684.0,
      "step": 94865
    },
    {
      "entropy": 5.728207397460937,
      "epoch": 8.60576923076923,
      "grad_norm": 1.578125,
      "learning_rate": 7.160387894962772e-05,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23029269129037858,
      "num_tokens": 196853686.0,
      "step": 94870
    },
    {
      "entropy": 5.703083229064942,
      "epoch": 8.606222786647315,
      "grad_norm": 1.71875,
      "learning_rate": 7.159004645380524e-05,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23779848217964172,
      "num_tokens": 196863877.0,
      "step": 94875
    },
    {
      "entropy": 5.673625326156616,
      "epoch": 8.606676342525398,
      "grad_norm": 1.546875,
      "learning_rate": 7.15762181645524e-05,
      "loss": 4.758,
      "mean_token_accuracy": 0.24108802974224092,
      "num_tokens": 196873615.0,
      "step": 94880
    },
    {
      "entropy": 5.650042867660522,
      "epoch": 8.607129898403484,
      "grad_norm": 1.59375,
      "learning_rate": 7.15623940821552e-05,
      "loss": 4.6715,
      "mean_token_accuracy": 0.24489308297634124,
      "num_tokens": 196884174.0,
      "step": 94885
    },
    {
      "entropy": 5.685844135284424,
      "epoch": 8.607583454281567,
      "grad_norm": 1.46875,
      "learning_rate": 7.154857420689956e-05,
      "loss": 4.8966,
      "mean_token_accuracy": 0.23742512315511705,
      "num_tokens": 196896138.0,
      "step": 94890
    },
    {
      "entropy": 5.632209777832031,
      "epoch": 8.608037010159652,
      "grad_norm": 1.671875,
      "learning_rate": 7.153475853907125e-05,
      "loss": 4.6698,
      "mean_token_accuracy": 0.24822169095277785,
      "num_tokens": 196905483.0,
      "step": 94895
    },
    {
      "entropy": 5.682334852218628,
      "epoch": 8.608490566037736,
      "grad_norm": 1.4453125,
      "learning_rate": 7.152094707895596e-05,
      "loss": 4.7788,
      "mean_token_accuracy": 0.24478573650121688,
      "num_tokens": 196915372.0,
      "step": 94900
    },
    {
      "entropy": 5.7545990467071535,
      "epoch": 8.60894412191582,
      "grad_norm": 1.4765625,
      "learning_rate": 7.150713982683932e-05,
      "loss": 4.8886,
      "mean_token_accuracy": 0.22742469906806945,
      "num_tokens": 196925995.0,
      "step": 94905
    },
    {
      "entropy": 5.603221988677978,
      "epoch": 8.609397677793904,
      "grad_norm": 1.453125,
      "learning_rate": 7.14933367830069e-05,
      "loss": 4.6531,
      "mean_token_accuracy": 0.24557895511388778,
      "num_tokens": 196935878.0,
      "step": 94910
    },
    {
      "entropy": 5.702280282974243,
      "epoch": 8.60985123367199,
      "grad_norm": 1.4296875,
      "learning_rate": 7.147953794774416e-05,
      "loss": 4.8285,
      "mean_token_accuracy": 0.2317585051059723,
      "num_tokens": 196947435.0,
      "step": 94915
    },
    {
      "entropy": 5.572736930847168,
      "epoch": 8.610304789550073,
      "grad_norm": 1.5078125,
      "learning_rate": 7.146574332133643e-05,
      "loss": 4.6691,
      "mean_token_accuracy": 0.24795842468738555,
      "num_tokens": 196958243.0,
      "step": 94920
    },
    {
      "entropy": 5.724387550354004,
      "epoch": 8.610758345428156,
      "grad_norm": 1.4765625,
      "learning_rate": 7.1451952904069e-05,
      "loss": 4.8217,
      "mean_token_accuracy": 0.23934231996536254,
      "num_tokens": 196969166.0,
      "step": 94925
    },
    {
      "entropy": 5.730712890625,
      "epoch": 8.611211901306241,
      "grad_norm": 1.578125,
      "learning_rate": 7.143816669622703e-05,
      "loss": 4.8082,
      "mean_token_accuracy": 0.2458129718899727,
      "num_tokens": 196978113.0,
      "step": 94930
    },
    {
      "entropy": 5.655264663696289,
      "epoch": 8.611665457184325,
      "grad_norm": 1.46875,
      "learning_rate": 7.142438469809567e-05,
      "loss": 4.7835,
      "mean_token_accuracy": 0.23800960928201675,
      "num_tokens": 196989034.0,
      "step": 94935
    },
    {
      "entropy": 5.776662540435791,
      "epoch": 8.61211901306241,
      "grad_norm": 1.46875,
      "learning_rate": 7.141060690995988e-05,
      "loss": 4.875,
      "mean_token_accuracy": 0.2289124384522438,
      "num_tokens": 197000865.0,
      "step": 94940
    },
    {
      "entropy": 5.736805200576782,
      "epoch": 8.612572568940493,
      "grad_norm": 1.5859375,
      "learning_rate": 7.139683333210463e-05,
      "loss": 4.8397,
      "mean_token_accuracy": 0.23285259157419205,
      "num_tokens": 197010747.0,
      "step": 94945
    },
    {
      "entropy": 5.688477849960327,
      "epoch": 8.613026124818578,
      "grad_norm": 1.5625,
      "learning_rate": 7.138306396481477e-05,
      "loss": 4.8205,
      "mean_token_accuracy": 0.23395639061927795,
      "num_tokens": 197021117.0,
      "step": 94950
    },
    {
      "entropy": 5.677611589431763,
      "epoch": 8.613479680696662,
      "grad_norm": 1.421875,
      "learning_rate": 7.136929880837503e-05,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24319749921560288,
      "num_tokens": 197030825.0,
      "step": 94955
    },
    {
      "entropy": 5.654498100280762,
      "epoch": 8.613933236574747,
      "grad_norm": 1.453125,
      "learning_rate": 7.13555378630701e-05,
      "loss": 4.7651,
      "mean_token_accuracy": 0.23401979357004166,
      "num_tokens": 197041339.0,
      "step": 94960
    },
    {
      "entropy": 5.734561538696289,
      "epoch": 8.61438679245283,
      "grad_norm": 1.5703125,
      "learning_rate": 7.134178112918452e-05,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2375440165400505,
      "num_tokens": 197051621.0,
      "step": 94965
    },
    {
      "entropy": 5.618127775192261,
      "epoch": 8.614840348330915,
      "grad_norm": 1.5234375,
      "learning_rate": 7.132802860700281e-05,
      "loss": 4.6677,
      "mean_token_accuracy": 0.24781762808561325,
      "num_tokens": 197061531.0,
      "step": 94970
    },
    {
      "entropy": 5.636903381347656,
      "epoch": 8.615293904208999,
      "grad_norm": 1.75,
      "learning_rate": 7.131428029680934e-05,
      "loss": 4.7046,
      "mean_token_accuracy": 0.23445375263690948,
      "num_tokens": 197070206.0,
      "step": 94975
    },
    {
      "entropy": 5.659935140609742,
      "epoch": 8.615747460087082,
      "grad_norm": 1.703125,
      "learning_rate": 7.130053619888847e-05,
      "loss": 4.7837,
      "mean_token_accuracy": 0.24158926755189897,
      "num_tokens": 197079705.0,
      "step": 94980
    },
    {
      "entropy": 5.60902509689331,
      "epoch": 8.616201015965167,
      "grad_norm": 1.421875,
      "learning_rate": 7.128679631352445e-05,
      "loss": 4.7308,
      "mean_token_accuracy": 0.23951330333948134,
      "num_tokens": 197090433.0,
      "step": 94985
    },
    {
      "entropy": 5.6956336975097654,
      "epoch": 8.61665457184325,
      "grad_norm": 1.5390625,
      "learning_rate": 7.127306064100138e-05,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2379427060484886,
      "num_tokens": 197100368.0,
      "step": 94990
    },
    {
      "entropy": 5.638569688796997,
      "epoch": 8.617108127721336,
      "grad_norm": 1.5546875,
      "learning_rate": 7.125932918160334e-05,
      "loss": 4.6967,
      "mean_token_accuracy": 0.24127282500267028,
      "num_tokens": 197110587.0,
      "step": 94995
    },
    {
      "entropy": 5.651538419723511,
      "epoch": 8.617561683599419,
      "grad_norm": 1.421875,
      "learning_rate": 7.124560193561427e-05,
      "loss": 4.7873,
      "mean_token_accuracy": 0.24205703288316727,
      "num_tokens": 197121605.0,
      "step": 95000
    },
    {
      "entropy": 5.625376605987549,
      "epoch": 8.618015239477504,
      "grad_norm": 1.6171875,
      "learning_rate": 7.123187890331805e-05,
      "loss": 4.7014,
      "mean_token_accuracy": 0.24297119230031966,
      "num_tokens": 197131616.0,
      "step": 95005
    },
    {
      "entropy": 5.684009552001953,
      "epoch": 8.618468795355588,
      "grad_norm": 1.4296875,
      "learning_rate": 7.121816008499856e-05,
      "loss": 4.8206,
      "mean_token_accuracy": 0.23718346953392028,
      "num_tokens": 197142595.0,
      "step": 95010
    },
    {
      "entropy": 5.690876197814942,
      "epoch": 8.618922351233673,
      "grad_norm": 1.5625,
      "learning_rate": 7.120444548093941e-05,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23294976353645325,
      "num_tokens": 197152198.0,
      "step": 95015
    },
    {
      "entropy": 5.678248071670533,
      "epoch": 8.619375907111756,
      "grad_norm": 1.5625,
      "learning_rate": 7.11907350914243e-05,
      "loss": 4.7727,
      "mean_token_accuracy": 0.2339020073413849,
      "num_tokens": 197162625.0,
      "step": 95020
    },
    {
      "entropy": 5.692843818664551,
      "epoch": 8.61982946298984,
      "grad_norm": 1.4140625,
      "learning_rate": 7.117702891673671e-05,
      "loss": 4.804,
      "mean_token_accuracy": 0.23119928389787675,
      "num_tokens": 197173258.0,
      "step": 95025
    },
    {
      "entropy": 5.706052207946778,
      "epoch": 8.620283018867925,
      "grad_norm": 1.40625,
      "learning_rate": 7.11633269571601e-05,
      "loss": 4.7865,
      "mean_token_accuracy": 0.24456001073122025,
      "num_tokens": 197184122.0,
      "step": 95030
    },
    {
      "entropy": 5.658166885375977,
      "epoch": 8.620736574746008,
      "grad_norm": 1.6015625,
      "learning_rate": 7.114962921297784e-05,
      "loss": 4.7283,
      "mean_token_accuracy": 0.2430923119187355,
      "num_tokens": 197193978.0,
      "step": 95035
    },
    {
      "entropy": 5.720619058609008,
      "epoch": 8.621190130624093,
      "grad_norm": 1.5546875,
      "learning_rate": 7.113593568447313e-05,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22350432723760605,
      "num_tokens": 197204112.0,
      "step": 95040
    },
    {
      "entropy": 5.6687750816345215,
      "epoch": 8.621643686502177,
      "grad_norm": 1.578125,
      "learning_rate": 7.112224637192926e-05,
      "loss": 4.7557,
      "mean_token_accuracy": 0.238428096473217,
      "num_tokens": 197214055.0,
      "step": 95045
    },
    {
      "entropy": 5.699739646911621,
      "epoch": 8.622097242380262,
      "grad_norm": 1.5390625,
      "learning_rate": 7.110856127562932e-05,
      "loss": 4.7148,
      "mean_token_accuracy": 0.24720858484506608,
      "num_tokens": 197222524.0,
      "step": 95050
    },
    {
      "entropy": 5.735777044296265,
      "epoch": 8.622550798258345,
      "grad_norm": 1.328125,
      "learning_rate": 7.10948803958563e-05,
      "loss": 4.8722,
      "mean_token_accuracy": 0.23189130276441575,
      "num_tokens": 197234160.0,
      "step": 95055
    },
    {
      "entropy": 5.634521436691284,
      "epoch": 8.62300435413643,
      "grad_norm": 1.4296875,
      "learning_rate": 7.108120373289307e-05,
      "loss": 4.7105,
      "mean_token_accuracy": 0.2516647219657898,
      "num_tokens": 197244632.0,
      "step": 95060
    },
    {
      "entropy": 5.670110750198364,
      "epoch": 8.623457910014514,
      "grad_norm": 1.4921875,
      "learning_rate": 7.106753128702256e-05,
      "loss": 4.8189,
      "mean_token_accuracy": 0.23405254483222962,
      "num_tokens": 197255621.0,
      "step": 95065
    },
    {
      "entropy": 5.730222511291504,
      "epoch": 8.623911465892597,
      "grad_norm": 1.421875,
      "learning_rate": 7.105386305852744e-05,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22500621676445007,
      "num_tokens": 197266863.0,
      "step": 95070
    },
    {
      "entropy": 5.6851233959198,
      "epoch": 8.624365021770682,
      "grad_norm": 1.46875,
      "learning_rate": 7.104019904769036e-05,
      "loss": 4.8777,
      "mean_token_accuracy": 0.23248545527458192,
      "num_tokens": 197276664.0,
      "step": 95075
    },
    {
      "entropy": 5.666268348693848,
      "epoch": 8.624818577648766,
      "grad_norm": 1.53125,
      "learning_rate": 7.102653925479398e-05,
      "loss": 4.794,
      "mean_token_accuracy": 0.2392224431037903,
      "num_tokens": 197287200.0,
      "step": 95080
    },
    {
      "entropy": 5.678859901428223,
      "epoch": 8.62527213352685,
      "grad_norm": 1.515625,
      "learning_rate": 7.101288368012075e-05,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23805034309625625,
      "num_tokens": 197297995.0,
      "step": 95085
    },
    {
      "entropy": 5.684841156005859,
      "epoch": 8.625725689404934,
      "grad_norm": 1.4375,
      "learning_rate": 7.099923232395309e-05,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22658603638410568,
      "num_tokens": 197308732.0,
      "step": 95090
    },
    {
      "entropy": 5.721112060546875,
      "epoch": 8.62617924528302,
      "grad_norm": 1.46875,
      "learning_rate": 7.098558518657326e-05,
      "loss": 4.8595,
      "mean_token_accuracy": 0.23420030921697615,
      "num_tokens": 197319206.0,
      "step": 95095
    },
    {
      "entropy": 5.74666018486023,
      "epoch": 8.626632801161103,
      "grad_norm": 1.609375,
      "learning_rate": 7.097194226826353e-05,
      "loss": 4.859,
      "mean_token_accuracy": 0.23269812464714051,
      "num_tokens": 197329753.0,
      "step": 95100
    },
    {
      "entropy": 5.707572317123413,
      "epoch": 8.627086357039188,
      "grad_norm": 1.4375,
      "learning_rate": 7.095830356930604e-05,
      "loss": 4.8366,
      "mean_token_accuracy": 0.2354089617729187,
      "num_tokens": 197339373.0,
      "step": 95105
    },
    {
      "entropy": 5.675622367858887,
      "epoch": 8.627539912917271,
      "grad_norm": 1.578125,
      "learning_rate": 7.094466908998276e-05,
      "loss": 4.746,
      "mean_token_accuracy": 0.24991348534822463,
      "num_tokens": 197350107.0,
      "step": 95110
    },
    {
      "entropy": 5.637261962890625,
      "epoch": 8.627993468795356,
      "grad_norm": 1.4375,
      "learning_rate": 7.093103883057579e-05,
      "loss": 4.7433,
      "mean_token_accuracy": 0.24105162769556046,
      "num_tokens": 197360945.0,
      "step": 95115
    },
    {
      "entropy": 5.672143888473511,
      "epoch": 8.62844702467344,
      "grad_norm": 1.546875,
      "learning_rate": 7.091741279136694e-05,
      "loss": 4.7986,
      "mean_token_accuracy": 0.23424244672060013,
      "num_tokens": 197371509.0,
      "step": 95120
    },
    {
      "entropy": 5.720341587066651,
      "epoch": 8.628900580551523,
      "grad_norm": 1.53125,
      "learning_rate": 7.090379097263797e-05,
      "loss": 4.8198,
      "mean_token_accuracy": 0.24137205332517625,
      "num_tokens": 197382186.0,
      "step": 95125
    },
    {
      "entropy": 5.6264242172241214,
      "epoch": 8.629354136429608,
      "grad_norm": 1.53125,
      "learning_rate": 7.089017337467064e-05,
      "loss": 4.7323,
      "mean_token_accuracy": 0.24271779805421828,
      "num_tokens": 197393129.0,
      "step": 95130
    },
    {
      "entropy": 5.668461275100708,
      "epoch": 8.629807692307692,
      "grad_norm": 1.5546875,
      "learning_rate": 7.087655999774653e-05,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23710295110940932,
      "num_tokens": 197402674.0,
      "step": 95135
    },
    {
      "entropy": 5.719234323501587,
      "epoch": 8.630261248185777,
      "grad_norm": 1.65625,
      "learning_rate": 7.086295084214715e-05,
      "loss": 4.7464,
      "mean_token_accuracy": 0.23859935700893403,
      "num_tokens": 197411827.0,
      "step": 95140
    },
    {
      "entropy": 5.715149307250977,
      "epoch": 8.63071480406386,
      "grad_norm": 1.390625,
      "learning_rate": 7.084934590815398e-05,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22769208252429962,
      "num_tokens": 197422759.0,
      "step": 95145
    },
    {
      "entropy": 5.701634693145752,
      "epoch": 8.631168359941945,
      "grad_norm": 1.6953125,
      "learning_rate": 7.083574519604835e-05,
      "loss": 4.8847,
      "mean_token_accuracy": 0.23433021157979966,
      "num_tokens": 197433636.0,
      "step": 95150
    },
    {
      "entropy": 5.6984124183654785,
      "epoch": 8.631621915820029,
      "grad_norm": 1.3828125,
      "learning_rate": 7.082214870611157e-05,
      "loss": 4.7888,
      "mean_token_accuracy": 0.2322283461689949,
      "num_tokens": 197444246.0,
      "step": 95155
    },
    {
      "entropy": 5.707924938201904,
      "epoch": 8.632075471698114,
      "grad_norm": 1.3828125,
      "learning_rate": 7.080855643862475e-05,
      "loss": 4.8606,
      "mean_token_accuracy": 0.23680632561445236,
      "num_tokens": 197455654.0,
      "step": 95160
    },
    {
      "entropy": 5.7242649555206295,
      "epoch": 8.632529027576197,
      "grad_norm": 1.484375,
      "learning_rate": 7.079496839386907e-05,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23469428718090057,
      "num_tokens": 197466088.0,
      "step": 95165
    },
    {
      "entropy": 5.63460373878479,
      "epoch": 8.632982583454282,
      "grad_norm": 1.5546875,
      "learning_rate": 7.078138457212545e-05,
      "loss": 4.7431,
      "mean_token_accuracy": 0.24898913353681565,
      "num_tokens": 197476609.0,
      "step": 95170
    },
    {
      "entropy": 5.721384429931641,
      "epoch": 8.633436139332366,
      "grad_norm": 1.453125,
      "learning_rate": 7.076780497367483e-05,
      "loss": 4.7919,
      "mean_token_accuracy": 0.23880569338798524,
      "num_tokens": 197489105.0,
      "step": 95175
    },
    {
      "entropy": 5.700876665115357,
      "epoch": 8.63388969521045,
      "grad_norm": 1.4453125,
      "learning_rate": 7.075422959879803e-05,
      "loss": 4.7384,
      "mean_token_accuracy": 0.24162575155496596,
      "num_tokens": 197499779.0,
      "step": 95180
    },
    {
      "entropy": 5.638337802886963,
      "epoch": 8.634343251088534,
      "grad_norm": 1.4453125,
      "learning_rate": 7.074065844777584e-05,
      "loss": 4.7152,
      "mean_token_accuracy": 0.25191932171583176,
      "num_tokens": 197510234.0,
      "step": 95185
    },
    {
      "entropy": 5.675269079208374,
      "epoch": 8.634796806966618,
      "grad_norm": 1.5,
      "learning_rate": 7.072709152088887e-05,
      "loss": 4.7479,
      "mean_token_accuracy": 0.23909763395786285,
      "num_tokens": 197520550.0,
      "step": 95190
    },
    {
      "entropy": 5.75360689163208,
      "epoch": 8.635250362844703,
      "grad_norm": 1.484375,
      "learning_rate": 7.071352881841772e-05,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22927155196666718,
      "num_tokens": 197530874.0,
      "step": 95195
    },
    {
      "entropy": 5.69561619758606,
      "epoch": 8.635703918722786,
      "grad_norm": 1.34375,
      "learning_rate": 7.069997034064285e-05,
      "loss": 4.805,
      "mean_token_accuracy": 0.2385726109147072,
      "num_tokens": 197541072.0,
      "step": 95200
    },
    {
      "entropy": 5.742883634567261,
      "epoch": 8.636157474600871,
      "grad_norm": 1.640625,
      "learning_rate": 7.068641608784463e-05,
      "loss": 4.8937,
      "mean_token_accuracy": 0.230411659181118,
      "num_tokens": 197550941.0,
      "step": 95205
    },
    {
      "entropy": 5.676323890686035,
      "epoch": 8.636611030478955,
      "grad_norm": 1.5078125,
      "learning_rate": 7.067286606030339e-05,
      "loss": 4.7803,
      "mean_token_accuracy": 0.2380809709429741,
      "num_tokens": 197561952.0,
      "step": 95210
    },
    {
      "entropy": 5.72922420501709,
      "epoch": 8.63706458635704,
      "grad_norm": 1.5390625,
      "learning_rate": 7.065932025829935e-05,
      "loss": 4.8759,
      "mean_token_accuracy": 0.23350640386343002,
      "num_tokens": 197572776.0,
      "step": 95215
    },
    {
      "entropy": 5.676231861114502,
      "epoch": 8.637518142235123,
      "grad_norm": 1.453125,
      "learning_rate": 7.064577868211269e-05,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2286875605583191,
      "num_tokens": 197582813.0,
      "step": 95220
    },
    {
      "entropy": 5.732288217544555,
      "epoch": 8.637971698113208,
      "grad_norm": 1.7890625,
      "learning_rate": 7.063224133202337e-05,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22916201800107955,
      "num_tokens": 197592182.0,
      "step": 95225
    },
    {
      "entropy": 5.672436761856079,
      "epoch": 8.638425253991292,
      "grad_norm": 1.4921875,
      "learning_rate": 7.061870820831138e-05,
      "loss": 4.766,
      "mean_token_accuracy": 0.23603868931531907,
      "num_tokens": 197602825.0,
      "step": 95230
    },
    {
      "entropy": 5.754458332061768,
      "epoch": 8.638878809869375,
      "grad_norm": 1.4140625,
      "learning_rate": 7.060517931125659e-05,
      "loss": 4.86,
      "mean_token_accuracy": 0.23397577852010726,
      "num_tokens": 197613867.0,
      "step": 95235
    },
    {
      "entropy": 5.692811775207519,
      "epoch": 8.63933236574746,
      "grad_norm": 1.703125,
      "learning_rate": 7.059165464113879e-05,
      "loss": 4.742,
      "mean_token_accuracy": 0.24165432304143905,
      "num_tokens": 197623867.0,
      "step": 95240
    },
    {
      "entropy": 5.662004184722901,
      "epoch": 8.639785921625544,
      "grad_norm": 1.7890625,
      "learning_rate": 7.057813419823759e-05,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23962380588054658,
      "num_tokens": 197634572.0,
      "step": 95245
    },
    {
      "entropy": 5.7044144630432125,
      "epoch": 8.640239477503629,
      "grad_norm": 1.6171875,
      "learning_rate": 7.056461798283271e-05,
      "loss": 4.7729,
      "mean_token_accuracy": 0.24324689656496049,
      "num_tokens": 197644041.0,
      "step": 95250
    },
    {
      "entropy": 5.677141094207764,
      "epoch": 8.640693033381712,
      "grad_norm": 1.65625,
      "learning_rate": 7.055110599520369e-05,
      "loss": 4.7561,
      "mean_token_accuracy": 0.2506050065159798,
      "num_tokens": 197653109.0,
      "step": 95255
    },
    {
      "entropy": 5.6990844249725345,
      "epoch": 8.641146589259797,
      "grad_norm": 1.4765625,
      "learning_rate": 7.053759823562986e-05,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23772214949131013,
      "num_tokens": 197663864.0,
      "step": 95260
    },
    {
      "entropy": 5.624838590621948,
      "epoch": 8.64160014513788,
      "grad_norm": 1.40625,
      "learning_rate": 7.05240947043906e-05,
      "loss": 4.6849,
      "mean_token_accuracy": 0.2534373924136162,
      "num_tokens": 197673792.0,
      "step": 95265
    },
    {
      "entropy": 5.6983002662658695,
      "epoch": 8.642053701015966,
      "grad_norm": 1.53125,
      "learning_rate": 7.051059540176515e-05,
      "loss": 4.8306,
      "mean_token_accuracy": 0.23000657111406325,
      "num_tokens": 197683721.0,
      "step": 95270
    },
    {
      "entropy": 5.643569707870483,
      "epoch": 8.64250725689405,
      "grad_norm": 1.625,
      "learning_rate": 7.049710032803268e-05,
      "loss": 4.7137,
      "mean_token_accuracy": 0.23975110054016113,
      "num_tokens": 197693594.0,
      "step": 95275
    },
    {
      "entropy": 5.7573343276977536,
      "epoch": 8.642960812772134,
      "grad_norm": 1.6171875,
      "learning_rate": 7.04836094834723e-05,
      "loss": 4.819,
      "mean_token_accuracy": 0.2380548134446144,
      "num_tokens": 197703829.0,
      "step": 95280
    },
    {
      "entropy": 5.712864875793457,
      "epoch": 8.643414368650218,
      "grad_norm": 1.609375,
      "learning_rate": 7.047012286836301e-05,
      "loss": 4.8129,
      "mean_token_accuracy": 0.2356785163283348,
      "num_tokens": 197712847.0,
      "step": 95285
    },
    {
      "entropy": 5.6852270603179935,
      "epoch": 8.643867924528301,
      "grad_norm": 1.3984375,
      "learning_rate": 7.04566404829837e-05,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23344888538122177,
      "num_tokens": 197723464.0,
      "step": 95290
    },
    {
      "entropy": 5.618294668197632,
      "epoch": 8.644321480406386,
      "grad_norm": 1.546875,
      "learning_rate": 7.044316232761319e-05,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2501975238323212,
      "num_tokens": 197733938.0,
      "step": 95295
    },
    {
      "entropy": 5.666362714767456,
      "epoch": 8.64477503628447,
      "grad_norm": 1.484375,
      "learning_rate": 7.04296884025302e-05,
      "loss": 4.75,
      "mean_token_accuracy": 0.23920781761407853,
      "num_tokens": 197744428.0,
      "step": 95300
    },
    {
      "entropy": 5.6364258289337155,
      "epoch": 8.645228592162555,
      "grad_norm": 1.640625,
      "learning_rate": 7.041621870801335e-05,
      "loss": 4.7368,
      "mean_token_accuracy": 0.24926314502954483,
      "num_tokens": 197754369.0,
      "step": 95305
    },
    {
      "entropy": 5.644780492782592,
      "epoch": 8.645682148040638,
      "grad_norm": 1.359375,
      "learning_rate": 7.040275324434127e-05,
      "loss": 4.7453,
      "mean_token_accuracy": 0.24356791824102403,
      "num_tokens": 197764427.0,
      "step": 95310
    },
    {
      "entropy": 5.6703606128692625,
      "epoch": 8.646135703918723,
      "grad_norm": 1.5390625,
      "learning_rate": 7.038929201179241e-05,
      "loss": 4.744,
      "mean_token_accuracy": 0.24787445217370987,
      "num_tokens": 197773622.0,
      "step": 95315
    },
    {
      "entropy": 5.719189548492432,
      "epoch": 8.646589259796807,
      "grad_norm": 1.3984375,
      "learning_rate": 7.037583501064509e-05,
      "loss": 4.8186,
      "mean_token_accuracy": 0.23478558510541916,
      "num_tokens": 197785064.0,
      "step": 95320
    },
    {
      "entropy": 5.690788793563843,
      "epoch": 8.647042815674892,
      "grad_norm": 1.640625,
      "learning_rate": 7.036238224117769e-05,
      "loss": 4.8436,
      "mean_token_accuracy": 0.23364771157503128,
      "num_tokens": 197795406.0,
      "step": 95325
    },
    {
      "entropy": 5.710599899291992,
      "epoch": 8.647496371552975,
      "grad_norm": 1.40625,
      "learning_rate": 7.034893370366835e-05,
      "loss": 4.8054,
      "mean_token_accuracy": 0.2439720317721367,
      "num_tokens": 197806194.0,
      "step": 95330
    },
    {
      "entropy": 5.707115221023559,
      "epoch": 8.647949927431059,
      "grad_norm": 1.6015625,
      "learning_rate": 7.03354893983952e-05,
      "loss": 4.808,
      "mean_token_accuracy": 0.2327781304717064,
      "num_tokens": 197816798.0,
      "step": 95335
    },
    {
      "entropy": 5.719246435165405,
      "epoch": 8.648403483309144,
      "grad_norm": 1.484375,
      "learning_rate": 7.032204932563626e-05,
      "loss": 4.7761,
      "mean_token_accuracy": 0.24258027821779252,
      "num_tokens": 197826947.0,
      "step": 95340
    },
    {
      "entropy": 5.654751396179199,
      "epoch": 8.648857039187227,
      "grad_norm": 1.5703125,
      "learning_rate": 7.030861348566956e-05,
      "loss": 4.8358,
      "mean_token_accuracy": 0.2342439353466034,
      "num_tokens": 197837605.0,
      "step": 95345
    },
    {
      "entropy": 5.6973655223846436,
      "epoch": 8.649310595065312,
      "grad_norm": 1.6171875,
      "learning_rate": 7.029518187877288e-05,
      "loss": 4.7732,
      "mean_token_accuracy": 0.24479202032089234,
      "num_tokens": 197847877.0,
      "step": 95350
    },
    {
      "entropy": 5.67369818687439,
      "epoch": 8.649764150943396,
      "grad_norm": 1.5,
      "learning_rate": 7.0281754505224e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2348834380507469,
      "num_tokens": 197858811.0,
      "step": 95355
    },
    {
      "entropy": 5.606340932846069,
      "epoch": 8.650217706821481,
      "grad_norm": 1.6484375,
      "learning_rate": 7.026833136530058e-05,
      "loss": 4.7302,
      "mean_token_accuracy": 0.23764235377311707,
      "num_tokens": 197870297.0,
      "step": 95360
    },
    {
      "entropy": 5.698521614074707,
      "epoch": 8.650671262699564,
      "grad_norm": 1.3828125,
      "learning_rate": 7.025491245928029e-05,
      "loss": 4.7698,
      "mean_token_accuracy": 0.2360885724425316,
      "num_tokens": 197881358.0,
      "step": 95365
    },
    {
      "entropy": 5.7302323341369625,
      "epoch": 8.65112481857765,
      "grad_norm": 1.5859375,
      "learning_rate": 7.024149778744056e-05,
      "loss": 4.8653,
      "mean_token_accuracy": 0.23707131445407867,
      "num_tokens": 197892350.0,
      "step": 95370
    },
    {
      "entropy": 5.660697889328003,
      "epoch": 8.651578374455733,
      "grad_norm": 1.5703125,
      "learning_rate": 7.022808735005878e-05,
      "loss": 4.7736,
      "mean_token_accuracy": 0.23567523807287216,
      "num_tokens": 197902249.0,
      "step": 95375
    },
    {
      "entropy": 5.69198751449585,
      "epoch": 8.652031930333816,
      "grad_norm": 1.375,
      "learning_rate": 7.021468114741239e-05,
      "loss": 4.7764,
      "mean_token_accuracy": 0.24160308092832566,
      "num_tokens": 197914652.0,
      "step": 95380
    },
    {
      "entropy": 5.699692630767823,
      "epoch": 8.652485486211901,
      "grad_norm": 1.484375,
      "learning_rate": 7.020127917977858e-05,
      "loss": 4.8208,
      "mean_token_accuracy": 0.22992523461580278,
      "num_tokens": 197925219.0,
      "step": 95385
    },
    {
      "entropy": 5.720066690444947,
      "epoch": 8.652939042089985,
      "grad_norm": 1.5625,
      "learning_rate": 7.01878814474345e-05,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22313937693834304,
      "num_tokens": 197936462.0,
      "step": 95390
    },
    {
      "entropy": 5.706318521499634,
      "epoch": 8.65339259796807,
      "grad_norm": 1.53125,
      "learning_rate": 7.017448795065722e-05,
      "loss": 4.8005,
      "mean_token_accuracy": 0.23946398943662645,
      "num_tokens": 197946864.0,
      "step": 95395
    },
    {
      "entropy": 5.640585470199585,
      "epoch": 8.653846153846153,
      "grad_norm": 1.4921875,
      "learning_rate": 7.016109868972373e-05,
      "loss": 4.7203,
      "mean_token_accuracy": 0.24671844244003296,
      "num_tokens": 197957314.0,
      "step": 95400
    },
    {
      "entropy": 5.759293413162231,
      "epoch": 8.654299709724238,
      "grad_norm": 1.484375,
      "learning_rate": 7.014771366491093e-05,
      "loss": 4.9275,
      "mean_token_accuracy": 0.22576853036880493,
      "num_tokens": 197967726.0,
      "step": 95405
    },
    {
      "entropy": 5.77863655090332,
      "epoch": 8.654753265602322,
      "grad_norm": 1.7890625,
      "learning_rate": 7.013433287649555e-05,
      "loss": 4.9139,
      "mean_token_accuracy": 0.2249475061893463,
      "num_tokens": 197978166.0,
      "step": 95410
    },
    {
      "entropy": 5.668653202056885,
      "epoch": 8.655206821480407,
      "grad_norm": 1.515625,
      "learning_rate": 7.012095632475443e-05,
      "loss": 4.7406,
      "mean_token_accuracy": 0.23648591786623002,
      "num_tokens": 197988729.0,
      "step": 95415
    },
    {
      "entropy": 5.706142950057983,
      "epoch": 8.65566037735849,
      "grad_norm": 1.578125,
      "learning_rate": 7.010758400996414e-05,
      "loss": 4.856,
      "mean_token_accuracy": 0.23830369263887405,
      "num_tokens": 197998948.0,
      "step": 95420
    },
    {
      "entropy": 5.752370166778564,
      "epoch": 8.656113933236576,
      "grad_norm": 1.8046875,
      "learning_rate": 7.009421593240123e-05,
      "loss": 4.8268,
      "mean_token_accuracy": 0.2363789498806,
      "num_tokens": 198008630.0,
      "step": 95425
    },
    {
      "entropy": 5.8089361667633055,
      "epoch": 8.656567489114659,
      "grad_norm": 1.34375,
      "learning_rate": 7.008085209234215e-05,
      "loss": 4.9446,
      "mean_token_accuracy": 0.22857287377119065,
      "num_tokens": 198019809.0,
      "step": 95430
    },
    {
      "entropy": 5.623050165176392,
      "epoch": 8.657021044992742,
      "grad_norm": 1.5625,
      "learning_rate": 7.006749249006325e-05,
      "loss": 4.7084,
      "mean_token_accuracy": 0.24597703367471696,
      "num_tokens": 198029647.0,
      "step": 95435
    },
    {
      "entropy": 5.72647647857666,
      "epoch": 8.657474600870827,
      "grad_norm": 1.421875,
      "learning_rate": 7.005413712584084e-05,
      "loss": 4.8571,
      "mean_token_accuracy": 0.23679377883672714,
      "num_tokens": 198040062.0,
      "step": 95440
    },
    {
      "entropy": 5.723336458206177,
      "epoch": 8.65792815674891,
      "grad_norm": 1.59375,
      "learning_rate": 7.004078599995108e-05,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23935515880584718,
      "num_tokens": 198050762.0,
      "step": 95445
    },
    {
      "entropy": 5.647134113311767,
      "epoch": 8.658381712626996,
      "grad_norm": 1.609375,
      "learning_rate": 7.00274391126701e-05,
      "loss": 4.7871,
      "mean_token_accuracy": 0.238438880443573,
      "num_tokens": 198060704.0,
      "step": 95450
    },
    {
      "entropy": 5.715012264251709,
      "epoch": 8.65883526850508,
      "grad_norm": 1.3828125,
      "learning_rate": 7.001409646427396e-05,
      "loss": 4.8352,
      "mean_token_accuracy": 0.22851217538118362,
      "num_tokens": 198072331.0,
      "step": 95455
    },
    {
      "entropy": 5.755070400238037,
      "epoch": 8.659288824383164,
      "grad_norm": 1.5234375,
      "learning_rate": 7.00007580550385e-05,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23865915834903717,
      "num_tokens": 198082702.0,
      "step": 95460
    },
    {
      "entropy": 5.6912517070770265,
      "epoch": 8.659742380261248,
      "grad_norm": 1.5390625,
      "learning_rate": 6.998742388523962e-05,
      "loss": 4.7932,
      "mean_token_accuracy": 0.2358444258570671,
      "num_tokens": 198093996.0,
      "step": 95465
    },
    {
      "entropy": 5.72051215171814,
      "epoch": 8.660195936139333,
      "grad_norm": 1.4765625,
      "learning_rate": 6.997409395515304e-05,
      "loss": 4.8508,
      "mean_token_accuracy": 0.23860757052898407,
      "num_tokens": 198105439.0,
      "step": 95470
    },
    {
      "entropy": 5.649917125701904,
      "epoch": 8.660649492017416,
      "grad_norm": 1.5234375,
      "learning_rate": 6.996076826505447e-05,
      "loss": 4.7998,
      "mean_token_accuracy": 0.23520273566246033,
      "num_tokens": 198116004.0,
      "step": 95475
    },
    {
      "entropy": 5.650346755981445,
      "epoch": 8.661103047895502,
      "grad_norm": 1.46875,
      "learning_rate": 6.994744681521941e-05,
      "loss": 4.7184,
      "mean_token_accuracy": 0.24433062523603438,
      "num_tokens": 198126132.0,
      "step": 95480
    },
    {
      "entropy": 5.6963708877563475,
      "epoch": 8.661556603773585,
      "grad_norm": 1.5234375,
      "learning_rate": 6.993412960592344e-05,
      "loss": 4.8383,
      "mean_token_accuracy": 0.24266148060560228,
      "num_tokens": 198136384.0,
      "step": 95485
    },
    {
      "entropy": 5.693365621566772,
      "epoch": 8.662010159651668,
      "grad_norm": 1.3984375,
      "learning_rate": 6.992081663744194e-05,
      "loss": 4.8203,
      "mean_token_accuracy": 0.2327402725815773,
      "num_tokens": 198147368.0,
      "step": 95490
    },
    {
      "entropy": 5.724269437789917,
      "epoch": 8.662463715529753,
      "grad_norm": 1.5859375,
      "learning_rate": 6.990750791005019e-05,
      "loss": 4.8906,
      "mean_token_accuracy": 0.2321058079600334,
      "num_tokens": 198158273.0,
      "step": 95495
    },
    {
      "entropy": 5.698768568038941,
      "epoch": 8.662917271407837,
      "grad_norm": 1.46875,
      "learning_rate": 6.989420342402344e-05,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23580595403909682,
      "num_tokens": 198167941.0,
      "step": 95500
    },
    {
      "entropy": 5.709750413894653,
      "epoch": 8.663370827285922,
      "grad_norm": 1.703125,
      "learning_rate": 6.988090317963686e-05,
      "loss": 4.7901,
      "mean_token_accuracy": 0.24311989545822144,
      "num_tokens": 198177363.0,
      "step": 95505
    },
    {
      "entropy": 5.617508268356323,
      "epoch": 8.663824383164005,
      "grad_norm": 1.4296875,
      "learning_rate": 6.986760717716541e-05,
      "loss": 4.728,
      "mean_token_accuracy": 0.24515056759119033,
      "num_tokens": 198188924.0,
      "step": 95510
    },
    {
      "entropy": 5.656024551391601,
      "epoch": 8.66427793904209,
      "grad_norm": 1.5078125,
      "learning_rate": 6.985431541688417e-05,
      "loss": 4.7385,
      "mean_token_accuracy": 0.24705781638622284,
      "num_tokens": 198198931.0,
      "step": 95515
    },
    {
      "entropy": 5.694518947601319,
      "epoch": 8.664731494920174,
      "grad_norm": 1.46875,
      "learning_rate": 6.984102789906794e-05,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23536790609359742,
      "num_tokens": 198210506.0,
      "step": 95520
    },
    {
      "entropy": 5.662294721603393,
      "epoch": 8.665185050798259,
      "grad_norm": 1.515625,
      "learning_rate": 6.982774462399157e-05,
      "loss": 4.771,
      "mean_token_accuracy": 0.2411331057548523,
      "num_tokens": 198220195.0,
      "step": 95525
    },
    {
      "entropy": 5.757305574417114,
      "epoch": 8.665638606676342,
      "grad_norm": 1.390625,
      "learning_rate": 6.98144655919297e-05,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22875373661518097,
      "num_tokens": 198231203.0,
      "step": 95530
    },
    {
      "entropy": 5.696727180480957,
      "epoch": 8.666092162554428,
      "grad_norm": 1.5,
      "learning_rate": 6.980119080315697e-05,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2358471781015396,
      "num_tokens": 198241768.0,
      "step": 95535
    },
    {
      "entropy": 5.685167026519776,
      "epoch": 8.666545718432511,
      "grad_norm": 1.515625,
      "learning_rate": 6.97879202579479e-05,
      "loss": 4.7792,
      "mean_token_accuracy": 0.24346558153629302,
      "num_tokens": 198251566.0,
      "step": 95540
    },
    {
      "entropy": 5.664193439483642,
      "epoch": 8.666999274310594,
      "grad_norm": 1.6796875,
      "learning_rate": 6.977465395657692e-05,
      "loss": 4.7502,
      "mean_token_accuracy": 0.24499266296625138,
      "num_tokens": 198262059.0,
      "step": 95545
    },
    {
      "entropy": 5.701961374282837,
      "epoch": 8.66745283018868,
      "grad_norm": 1.5234375,
      "learning_rate": 6.976139189931842e-05,
      "loss": 4.7967,
      "mean_token_accuracy": 0.2373700648546219,
      "num_tokens": 198272708.0,
      "step": 95550
    },
    {
      "entropy": 5.764362049102783,
      "epoch": 8.667906386066763,
      "grad_norm": 1.5078125,
      "learning_rate": 6.974813408644663e-05,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2406025841832161,
      "num_tokens": 198282913.0,
      "step": 95555
    },
    {
      "entropy": 5.6928948879241945,
      "epoch": 8.668359941944848,
      "grad_norm": 1.59375,
      "learning_rate": 6.973488051823574e-05,
      "loss": 4.7699,
      "mean_token_accuracy": 0.2442864090204239,
      "num_tokens": 198292870.0,
      "step": 95560
    },
    {
      "entropy": 5.7182024955749515,
      "epoch": 8.668813497822931,
      "grad_norm": 1.7109375,
      "learning_rate": 6.972163119495983e-05,
      "loss": 4.8368,
      "mean_token_accuracy": 0.2320595845580101,
      "num_tokens": 198303700.0,
      "step": 95565
    },
    {
      "entropy": 5.71904411315918,
      "epoch": 8.669267053701017,
      "grad_norm": 1.40625,
      "learning_rate": 6.970838611689288e-05,
      "loss": 4.7924,
      "mean_token_accuracy": 0.24570005685091018,
      "num_tokens": 198313917.0,
      "step": 95570
    },
    {
      "entropy": 5.718447971343994,
      "epoch": 8.6697206095791,
      "grad_norm": 1.515625,
      "learning_rate": 6.969514528430884e-05,
      "loss": 4.9376,
      "mean_token_accuracy": 0.23122527599334716,
      "num_tokens": 198324347.0,
      "step": 95575
    },
    {
      "entropy": 5.69571475982666,
      "epoch": 8.670174165457185,
      "grad_norm": 1.6640625,
      "learning_rate": 6.968190869748145e-05,
      "loss": 4.8285,
      "mean_token_accuracy": 0.23774511069059373,
      "num_tokens": 198333878.0,
      "step": 95580
    },
    {
      "entropy": 5.67494158744812,
      "epoch": 8.670627721335268,
      "grad_norm": 1.5,
      "learning_rate": 6.966867635668456e-05,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24204938262701034,
      "num_tokens": 198344931.0,
      "step": 95585
    },
    {
      "entropy": 5.691828632354737,
      "epoch": 8.671081277213352,
      "grad_norm": 1.53125,
      "learning_rate": 6.96554482621918e-05,
      "loss": 4.8042,
      "mean_token_accuracy": 0.23921214789152145,
      "num_tokens": 198355604.0,
      "step": 95590
    },
    {
      "entropy": 5.65418963432312,
      "epoch": 8.671534833091437,
      "grad_norm": 1.53125,
      "learning_rate": 6.964222441427664e-05,
      "loss": 4.7865,
      "mean_token_accuracy": 0.23525188565254213,
      "num_tokens": 198365030.0,
      "step": 95595
    },
    {
      "entropy": 5.707654905319214,
      "epoch": 8.67198838896952,
      "grad_norm": 1.453125,
      "learning_rate": 6.962900481321266e-05,
      "loss": 4.7844,
      "mean_token_accuracy": 0.24284059554338455,
      "num_tokens": 198376205.0,
      "step": 95600
    },
    {
      "entropy": 5.672033357620239,
      "epoch": 8.672441944847606,
      "grad_norm": 1.5625,
      "learning_rate": 6.961578945927316e-05,
      "loss": 4.7705,
      "mean_token_accuracy": 0.24021416902542114,
      "num_tokens": 198386294.0,
      "step": 95605
    },
    {
      "entropy": 5.68266487121582,
      "epoch": 8.672895500725689,
      "grad_norm": 1.421875,
      "learning_rate": 6.960257835273151e-05,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2338083028793335,
      "num_tokens": 198397711.0,
      "step": 95610
    },
    {
      "entropy": 5.702973699569702,
      "epoch": 8.673349056603774,
      "grad_norm": 1.46875,
      "learning_rate": 6.958937149386081e-05,
      "loss": 4.8513,
      "mean_token_accuracy": 0.23053452223539353,
      "num_tokens": 198408788.0,
      "step": 95615
    },
    {
      "entropy": 5.712434244155884,
      "epoch": 8.673802612481857,
      "grad_norm": 1.5546875,
      "learning_rate": 6.957616888293429e-05,
      "loss": 4.8584,
      "mean_token_accuracy": 0.22972007542848588,
      "num_tokens": 198418823.0,
      "step": 95620
    },
    {
      "entropy": 5.718565273284912,
      "epoch": 8.674256168359943,
      "grad_norm": 1.59375,
      "learning_rate": 6.956297052022497e-05,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23295072168111802,
      "num_tokens": 198428697.0,
      "step": 95625
    },
    {
      "entropy": 5.691961860656738,
      "epoch": 8.674709724238026,
      "grad_norm": 1.5078125,
      "learning_rate": 6.954977640600577e-05,
      "loss": 4.7611,
      "mean_token_accuracy": 0.24164350479841232,
      "num_tokens": 198439889.0,
      "step": 95630
    },
    {
      "entropy": 5.724201107025147,
      "epoch": 8.67516328011611,
      "grad_norm": 1.4140625,
      "learning_rate": 6.953658654054952e-05,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23250602632761003,
      "num_tokens": 198450860.0,
      "step": 95635
    },
    {
      "entropy": 5.715107345581055,
      "epoch": 8.675616835994195,
      "grad_norm": 1.515625,
      "learning_rate": 6.952340092412905e-05,
      "loss": 4.7693,
      "mean_token_accuracy": 0.24008978456258773,
      "num_tokens": 198460853.0,
      "step": 95640
    },
    {
      "entropy": 5.733629655838013,
      "epoch": 8.676070391872278,
      "grad_norm": 1.421875,
      "learning_rate": 6.9510219557017e-05,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2291654035449028,
      "num_tokens": 198471140.0,
      "step": 95645
    },
    {
      "entropy": 5.675580739974976,
      "epoch": 8.676523947750363,
      "grad_norm": 1.5234375,
      "learning_rate": 6.949704243948591e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.22873929291963577,
      "num_tokens": 198481043.0,
      "step": 95650
    },
    {
      "entropy": 5.748955154418946,
      "epoch": 8.676977503628446,
      "grad_norm": 1.390625,
      "learning_rate": 6.948386957180842e-05,
      "loss": 4.8995,
      "mean_token_accuracy": 0.23176807016134263,
      "num_tokens": 198491739.0,
      "step": 95655
    },
    {
      "entropy": 5.697116708755493,
      "epoch": 8.677431059506532,
      "grad_norm": 1.4375,
      "learning_rate": 6.947070095425686e-05,
      "loss": 4.793,
      "mean_token_accuracy": 0.23604439795017243,
      "num_tokens": 198502115.0,
      "step": 95660
    },
    {
      "entropy": 5.709907102584839,
      "epoch": 8.677884615384615,
      "grad_norm": 1.59375,
      "learning_rate": 6.94575365871036e-05,
      "loss": 4.7519,
      "mean_token_accuracy": 0.24005540162324907,
      "num_tokens": 198511696.0,
      "step": 95665
    },
    {
      "entropy": 5.715519714355469,
      "epoch": 8.6783381712627,
      "grad_norm": 1.515625,
      "learning_rate": 6.944437647062087e-05,
      "loss": 4.8163,
      "mean_token_accuracy": 0.233382149040699,
      "num_tokens": 198522727.0,
      "step": 95670
    },
    {
      "entropy": 5.715220642089844,
      "epoch": 8.678791727140784,
      "grad_norm": 1.6015625,
      "learning_rate": 6.943122060508079e-05,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23687007576227187,
      "num_tokens": 198533125.0,
      "step": 95675
    },
    {
      "entropy": 5.710307264328003,
      "epoch": 8.679245283018869,
      "grad_norm": 1.6875,
      "learning_rate": 6.941806899075546e-05,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2296052411198616,
      "num_tokens": 198543359.0,
      "step": 95680
    },
    {
      "entropy": 5.722940158843994,
      "epoch": 8.679698838896952,
      "grad_norm": 1.515625,
      "learning_rate": 6.940492162791686e-05,
      "loss": 4.8573,
      "mean_token_accuracy": 0.23465347439050674,
      "num_tokens": 198554773.0,
      "step": 95685
    },
    {
      "entropy": 5.69421877861023,
      "epoch": 8.680152394775035,
      "grad_norm": 1.5546875,
      "learning_rate": 6.939177851683687e-05,
      "loss": 4.7937,
      "mean_token_accuracy": 0.23227905631065368,
      "num_tokens": 198565002.0,
      "step": 95690
    },
    {
      "entropy": 5.698229074478149,
      "epoch": 8.68060595065312,
      "grad_norm": 1.5,
      "learning_rate": 6.93786396577873e-05,
      "loss": 4.7643,
      "mean_token_accuracy": 0.237798710167408,
      "num_tokens": 198576042.0,
      "step": 95695
    },
    {
      "entropy": 5.764480924606323,
      "epoch": 8.681059506531204,
      "grad_norm": 1.4140625,
      "learning_rate": 6.936550505103983e-05,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22625718414783477,
      "num_tokens": 198587226.0,
      "step": 95700
    },
    {
      "entropy": 5.7553307056427006,
      "epoch": 8.681513062409289,
      "grad_norm": 1.65625,
      "learning_rate": 6.935237469686614e-05,
      "loss": 4.8631,
      "mean_token_accuracy": 0.2316207081079483,
      "num_tokens": 198598170.0,
      "step": 95705
    },
    {
      "entropy": 5.731992864608765,
      "epoch": 8.681966618287372,
      "grad_norm": 1.4453125,
      "learning_rate": 6.93392485955377e-05,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23802961856126786,
      "num_tokens": 198608663.0,
      "step": 95710
    },
    {
      "entropy": 5.7005023002624515,
      "epoch": 8.682420174165458,
      "grad_norm": 1.4765625,
      "learning_rate": 6.932612674732607e-05,
      "loss": 4.8194,
      "mean_token_accuracy": 0.23368869423866273,
      "num_tokens": 198618554.0,
      "step": 95715
    },
    {
      "entropy": 5.66581163406372,
      "epoch": 8.682873730043541,
      "grad_norm": 1.5859375,
      "learning_rate": 6.931300915250253e-05,
      "loss": 4.8286,
      "mean_token_accuracy": 0.2286982148885727,
      "num_tokens": 198629203.0,
      "step": 95720
    },
    {
      "entropy": 5.662327289581299,
      "epoch": 8.683327285921626,
      "grad_norm": 1.546875,
      "learning_rate": 6.929989581133837e-05,
      "loss": 4.7215,
      "mean_token_accuracy": 0.24052055925130844,
      "num_tokens": 198639358.0,
      "step": 95725
    },
    {
      "entropy": 5.674302959442139,
      "epoch": 8.68378084179971,
      "grad_norm": 1.5,
      "learning_rate": 6.928678672410482e-05,
      "loss": 4.7714,
      "mean_token_accuracy": 0.230841226875782,
      "num_tokens": 198650531.0,
      "step": 95730
    },
    {
      "entropy": 5.6490706443786625,
      "epoch": 8.684234397677795,
      "grad_norm": 1.4765625,
      "learning_rate": 6.92736818910729e-05,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2430252894759178,
      "num_tokens": 198660794.0,
      "step": 95735
    },
    {
      "entropy": 5.645511198043823,
      "epoch": 8.684687953555878,
      "grad_norm": 1.5546875,
      "learning_rate": 6.926058131251368e-05,
      "loss": 4.7134,
      "mean_token_accuracy": 0.24001210629940034,
      "num_tokens": 198671825.0,
      "step": 95740
    },
    {
      "entropy": 5.648058938980102,
      "epoch": 8.685141509433961,
      "grad_norm": 1.4921875,
      "learning_rate": 6.924748498869803e-05,
      "loss": 4.7592,
      "mean_token_accuracy": 0.2364823564887047,
      "num_tokens": 198681313.0,
      "step": 95745
    },
    {
      "entropy": 5.6433563232421875,
      "epoch": 8.685595065312047,
      "grad_norm": 1.5234375,
      "learning_rate": 6.923439291989686e-05,
      "loss": 4.7533,
      "mean_token_accuracy": 0.246211476624012,
      "num_tokens": 198691839.0,
      "step": 95750
    },
    {
      "entropy": 5.657346248626709,
      "epoch": 8.68604862119013,
      "grad_norm": 1.5859375,
      "learning_rate": 6.92213051063809e-05,
      "loss": 4.7315,
      "mean_token_accuracy": 0.23752365112304688,
      "num_tokens": 198702033.0,
      "step": 95755
    },
    {
      "entropy": 5.6849370956420895,
      "epoch": 8.686502177068215,
      "grad_norm": 1.4296875,
      "learning_rate": 6.920822154842075e-05,
      "loss": 4.7864,
      "mean_token_accuracy": 0.23375179320573808,
      "num_tokens": 198711713.0,
      "step": 95760
    },
    {
      "entropy": 5.722238302230835,
      "epoch": 8.686955732946299,
      "grad_norm": 1.421875,
      "learning_rate": 6.919514224628703e-05,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23461538851261138,
      "num_tokens": 198722863.0,
      "step": 95765
    },
    {
      "entropy": 5.714054822921753,
      "epoch": 8.687409288824384,
      "grad_norm": 1.6015625,
      "learning_rate": 6.918206720025024e-05,
      "loss": 4.8325,
      "mean_token_accuracy": 0.24078026562929153,
      "num_tokens": 198732752.0,
      "step": 95770
    },
    {
      "entropy": 5.738992309570312,
      "epoch": 8.687862844702467,
      "grad_norm": 1.5625,
      "learning_rate": 6.916899641058073e-05,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22659051418304443,
      "num_tokens": 198742776.0,
      "step": 95775
    },
    {
      "entropy": 5.712087106704712,
      "epoch": 8.688316400580552,
      "grad_norm": 1.4609375,
      "learning_rate": 6.915592987754878e-05,
      "loss": 4.8567,
      "mean_token_accuracy": 0.2364112675189972,
      "num_tokens": 198753749.0,
      "step": 95780
    },
    {
      "entropy": 5.740852880477905,
      "epoch": 8.688769956458636,
      "grad_norm": 1.484375,
      "learning_rate": 6.914286760142471e-05,
      "loss": 4.8248,
      "mean_token_accuracy": 0.2325400397181511,
      "num_tokens": 198763963.0,
      "step": 95785
    },
    {
      "entropy": 5.714280986785889,
      "epoch": 8.68922351233672,
      "grad_norm": 1.4296875,
      "learning_rate": 6.912980958247858e-05,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22908333986997603,
      "num_tokens": 198774680.0,
      "step": 95790
    },
    {
      "entropy": 5.6874415397644045,
      "epoch": 8.689677068214804,
      "grad_norm": 1.5703125,
      "learning_rate": 6.911675582098046e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.24296523332595826,
      "num_tokens": 198784660.0,
      "step": 95795
    },
    {
      "entropy": 5.69869475364685,
      "epoch": 8.690130624092888,
      "grad_norm": 1.6796875,
      "learning_rate": 6.91037063172003e-05,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24073857218027114,
      "num_tokens": 198794637.0,
      "step": 95800
    },
    {
      "entropy": 5.664351320266723,
      "epoch": 8.690584179970973,
      "grad_norm": 1.578125,
      "learning_rate": 6.909066107140796e-05,
      "loss": 4.6509,
      "mean_token_accuracy": 0.24761765897274018,
      "num_tokens": 198804588.0,
      "step": 95805
    },
    {
      "entropy": 5.650163269042968,
      "epoch": 8.691037735849056,
      "grad_norm": 1.5859375,
      "learning_rate": 6.907762008387321e-05,
      "loss": 4.7541,
      "mean_token_accuracy": 0.24311603903770446,
      "num_tokens": 198814489.0,
      "step": 95810
    },
    {
      "entropy": 5.6026959896087645,
      "epoch": 8.691491291727141,
      "grad_norm": 1.53125,
      "learning_rate": 6.90645833548657e-05,
      "loss": 4.6857,
      "mean_token_accuracy": 0.24959986358880998,
      "num_tokens": 198823643.0,
      "step": 95815
    },
    {
      "entropy": 5.608878326416016,
      "epoch": 8.691944847605225,
      "grad_norm": 1.5234375,
      "learning_rate": 6.905155088465515e-05,
      "loss": 4.7536,
      "mean_token_accuracy": 0.24197398126125336,
      "num_tokens": 198833832.0,
      "step": 95820
    },
    {
      "entropy": 5.734337520599365,
      "epoch": 8.69239840348331,
      "grad_norm": 1.640625,
      "learning_rate": 6.903852267351101e-05,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23243834376335143,
      "num_tokens": 198845050.0,
      "step": 95825
    },
    {
      "entropy": 5.663645172119141,
      "epoch": 8.692851959361393,
      "grad_norm": 1.640625,
      "learning_rate": 6.90254987217027e-05,
      "loss": 4.8133,
      "mean_token_accuracy": 0.2352204740047455,
      "num_tokens": 198855208.0,
      "step": 95830
    },
    {
      "entropy": 5.652681064605713,
      "epoch": 8.693305515239478,
      "grad_norm": 1.5,
      "learning_rate": 6.901247902949958e-05,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2394893929362297,
      "num_tokens": 198867015.0,
      "step": 95835
    },
    {
      "entropy": 5.621439838409424,
      "epoch": 8.693759071117562,
      "grad_norm": 1.4765625,
      "learning_rate": 6.899946359717089e-05,
      "loss": 4.7157,
      "mean_token_accuracy": 0.24023293256759642,
      "num_tokens": 198876713.0,
      "step": 95840
    },
    {
      "entropy": 5.645596837997436,
      "epoch": 8.694212626995647,
      "grad_norm": 1.421875,
      "learning_rate": 6.898645242498572e-05,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2514489471912384,
      "num_tokens": 198888163.0,
      "step": 95845
    },
    {
      "entropy": 5.654492664337158,
      "epoch": 8.69466618287373,
      "grad_norm": 1.4765625,
      "learning_rate": 6.897344551321324e-05,
      "loss": 4.8046,
      "mean_token_accuracy": 0.24320015013217927,
      "num_tokens": 198898973.0,
      "step": 95850
    },
    {
      "entropy": 5.686447858810425,
      "epoch": 8.695119738751814,
      "grad_norm": 1.59375,
      "learning_rate": 6.896044286212244e-05,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23579125255346298,
      "num_tokens": 198908489.0,
      "step": 95855
    },
    {
      "entropy": 5.783032083511353,
      "epoch": 8.695573294629899,
      "grad_norm": 1.40625,
      "learning_rate": 6.894744447198217e-05,
      "loss": 4.889,
      "mean_token_accuracy": 0.23282967805862426,
      "num_tokens": 198919345.0,
      "step": 95860
    },
    {
      "entropy": 5.695092391967774,
      "epoch": 8.696026850507982,
      "grad_norm": 1.625,
      "learning_rate": 6.893445034306125e-05,
      "loss": 4.781,
      "mean_token_accuracy": 0.24115582406520844,
      "num_tokens": 198930207.0,
      "step": 95865
    },
    {
      "entropy": 5.665045547485351,
      "epoch": 8.696480406386067,
      "grad_norm": 1.546875,
      "learning_rate": 6.892146047562844e-05,
      "loss": 4.7691,
      "mean_token_accuracy": 0.23650185465812684,
      "num_tokens": 198940127.0,
      "step": 95870
    },
    {
      "entropy": 5.7043262958526615,
      "epoch": 8.69693396226415,
      "grad_norm": 1.4609375,
      "learning_rate": 6.89084748699523e-05,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23335874378681182,
      "num_tokens": 198951316.0,
      "step": 95875
    },
    {
      "entropy": 5.720611619949341,
      "epoch": 8.697387518142236,
      "grad_norm": 1.5234375,
      "learning_rate": 6.889549352630142e-05,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23283210545778274,
      "num_tokens": 198961372.0,
      "step": 95880
    },
    {
      "entropy": 5.705486202239991,
      "epoch": 8.69784107402032,
      "grad_norm": 1.609375,
      "learning_rate": 6.888251644494429e-05,
      "loss": 4.8083,
      "mean_token_accuracy": 0.24186842143535614,
      "num_tokens": 198972213.0,
      "step": 95885
    },
    {
      "entropy": 5.638075304031372,
      "epoch": 8.698294629898404,
      "grad_norm": 1.5625,
      "learning_rate": 6.886954362614922e-05,
      "loss": 4.7282,
      "mean_token_accuracy": 0.23960114419460296,
      "num_tokens": 198982442.0,
      "step": 95890
    },
    {
      "entropy": 5.67402286529541,
      "epoch": 8.698748185776488,
      "grad_norm": 1.546875,
      "learning_rate": 6.885657507018453e-05,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23880548179149627,
      "num_tokens": 198991916.0,
      "step": 95895
    },
    {
      "entropy": 5.5943375587463375,
      "epoch": 8.699201741654571,
      "grad_norm": 1.3828125,
      "learning_rate": 6.884361077731839e-05,
      "loss": 4.7549,
      "mean_token_accuracy": 0.23399844616651536,
      "num_tokens": 199002673.0,
      "step": 95900
    },
    {
      "entropy": 5.6474439144134525,
      "epoch": 8.699655297532656,
      "grad_norm": 1.546875,
      "learning_rate": 6.883065074781892e-05,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23283493518829346,
      "num_tokens": 199012578.0,
      "step": 95905
    },
    {
      "entropy": 5.70696988105774,
      "epoch": 8.70010885341074,
      "grad_norm": 1.5078125,
      "learning_rate": 6.881769498195415e-05,
      "loss": 4.8012,
      "mean_token_accuracy": 0.23807062804698945,
      "num_tokens": 199022680.0,
      "step": 95910
    },
    {
      "entropy": 5.689529943466186,
      "epoch": 8.700562409288825,
      "grad_norm": 1.5625,
      "learning_rate": 6.880474347999191e-05,
      "loss": 4.8304,
      "mean_token_accuracy": 0.2363561913371086,
      "num_tokens": 199033065.0,
      "step": 95915
    },
    {
      "entropy": 5.675337266921997,
      "epoch": 8.701015965166908,
      "grad_norm": 1.4375,
      "learning_rate": 6.879179624220021e-05,
      "loss": 4.758,
      "mean_token_accuracy": 0.23835701495409012,
      "num_tokens": 199044318.0,
      "step": 95920
    },
    {
      "entropy": 5.649109363555908,
      "epoch": 8.701469521044993,
      "grad_norm": 1.5546875,
      "learning_rate": 6.877885326884666e-05,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23914112597703935,
      "num_tokens": 199054959.0,
      "step": 95925
    },
    {
      "entropy": 5.660922050476074,
      "epoch": 8.701923076923077,
      "grad_norm": 1.484375,
      "learning_rate": 6.876591456019901e-05,
      "loss": 4.7689,
      "mean_token_accuracy": 0.2328091874718666,
      "num_tokens": 199065387.0,
      "step": 95930
    },
    {
      "entropy": 5.699146842956543,
      "epoch": 8.702376632801162,
      "grad_norm": 1.6484375,
      "learning_rate": 6.87529801165248e-05,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23563579767942427,
      "num_tokens": 199075340.0,
      "step": 95935
    },
    {
      "entropy": 5.67115216255188,
      "epoch": 8.702830188679245,
      "grad_norm": 1.4609375,
      "learning_rate": 6.87400499380915e-05,
      "loss": 4.8289,
      "mean_token_accuracy": 0.2433947429060936,
      "num_tokens": 199086405.0,
      "step": 95940
    },
    {
      "entropy": 5.687764596939087,
      "epoch": 8.703283744557329,
      "grad_norm": 1.5234375,
      "learning_rate": 6.872712402516656e-05,
      "loss": 4.8201,
      "mean_token_accuracy": 0.22746991366147995,
      "num_tokens": 199098785.0,
      "step": 95945
    },
    {
      "entropy": 5.659663915634155,
      "epoch": 8.703737300435414,
      "grad_norm": 1.53125,
      "learning_rate": 6.871420237801718e-05,
      "loss": 4.7277,
      "mean_token_accuracy": 0.24250107407569885,
      "num_tokens": 199108526.0,
      "step": 95950
    },
    {
      "entropy": 5.665271949768067,
      "epoch": 8.704190856313497,
      "grad_norm": 1.453125,
      "learning_rate": 6.87012849969107e-05,
      "loss": 4.7551,
      "mean_token_accuracy": 0.2425840198993683,
      "num_tokens": 199118701.0,
      "step": 95955
    },
    {
      "entropy": 5.651819515228271,
      "epoch": 8.704644412191582,
      "grad_norm": 1.5859375,
      "learning_rate": 6.868837188211426e-05,
      "loss": 4.7363,
      "mean_token_accuracy": 0.2486264243721962,
      "num_tokens": 199128790.0,
      "step": 95960
    },
    {
      "entropy": 5.665106248855591,
      "epoch": 8.705097968069666,
      "grad_norm": 1.5625,
      "learning_rate": 6.867546303389485e-05,
      "loss": 4.7667,
      "mean_token_accuracy": 0.24090919494628907,
      "num_tokens": 199139628.0,
      "step": 95965
    },
    {
      "entropy": 5.690004539489746,
      "epoch": 8.70555152394775,
      "grad_norm": 1.4921875,
      "learning_rate": 6.866255845251944e-05,
      "loss": 4.8402,
      "mean_token_accuracy": 0.23223710358142852,
      "num_tokens": 199151299.0,
      "step": 95970
    },
    {
      "entropy": 5.692129421234131,
      "epoch": 8.706005079825834,
      "grad_norm": 1.46875,
      "learning_rate": 6.86496581382549e-05,
      "loss": 4.8393,
      "mean_token_accuracy": 0.22915642708539963,
      "num_tokens": 199161153.0,
      "step": 95975
    },
    {
      "entropy": 5.693355655670166,
      "epoch": 8.70645863570392,
      "grad_norm": 1.6328125,
      "learning_rate": 6.8636762091368e-05,
      "loss": 4.7674,
      "mean_token_accuracy": 0.23914378583431245,
      "num_tokens": 199171046.0,
      "step": 95980
    },
    {
      "entropy": 5.715702915191651,
      "epoch": 8.706912191582003,
      "grad_norm": 1.609375,
      "learning_rate": 6.862387031212542e-05,
      "loss": 4.8285,
      "mean_token_accuracy": 0.24435840547084808,
      "num_tokens": 199181324.0,
      "step": 95985
    },
    {
      "entropy": 5.687318754196167,
      "epoch": 8.707365747460088,
      "grad_norm": 1.5859375,
      "learning_rate": 6.861098280079383e-05,
      "loss": 4.834,
      "mean_token_accuracy": 0.2281220182776451,
      "num_tokens": 199192237.0,
      "step": 95990
    },
    {
      "entropy": 5.699472951889038,
      "epoch": 8.707819303338171,
      "grad_norm": 1.4453125,
      "learning_rate": 6.859809955763969e-05,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22849217057228088,
      "num_tokens": 199203725.0,
      "step": 95995
    },
    {
      "entropy": 5.631324768066406,
      "epoch": 8.708272859216255,
      "grad_norm": 1.5703125,
      "learning_rate": 6.858522058292948e-05,
      "loss": 4.7185,
      "mean_token_accuracy": 0.24287748336791992,
      "num_tokens": 199213366.0,
      "step": 96000
    },
    {
      "epoch": 8.708272859216255,
      "eval_entropy": 5.393428455542655,
      "eval_loss": 4.9093828201293945,
      "eval_mean_token_accuracy": 0.23876589754284508,
      "eval_num_tokens": 199213366.0,
      "eval_runtime": 20.7622,
      "eval_samples_per_second": 1703.094,
      "eval_steps_per_second": 212.887,
      "step": 96000
    },
    {
      "entropy": 5.66262731552124,
      "epoch": 8.70872641509434,
      "grad_norm": 1.375,
      "learning_rate": 6.857234587692952e-05,
      "loss": 4.8066,
      "mean_token_accuracy": 0.2283506438136101,
      "num_tokens": 199224220.0,
      "step": 96005
    },
    {
      "entropy": 5.632016801834107,
      "epoch": 8.709179970972423,
      "grad_norm": 1.546875,
      "learning_rate": 6.855947543990601e-05,
      "loss": 4.7523,
      "mean_token_accuracy": 0.24020805805921555,
      "num_tokens": 199234566.0,
      "step": 96010
    },
    {
      "entropy": 5.679271125793457,
      "epoch": 8.709633526850508,
      "grad_norm": 1.4296875,
      "learning_rate": 6.854660927212515e-05,
      "loss": 4.8372,
      "mean_token_accuracy": 0.23144153505563736,
      "num_tokens": 199245356.0,
      "step": 96015
    },
    {
      "entropy": 5.651326036453247,
      "epoch": 8.710087082728592,
      "grad_norm": 1.5625,
      "learning_rate": 6.853374737385303e-05,
      "loss": 4.7851,
      "mean_token_accuracy": 0.2358391985297203,
      "num_tokens": 199256289.0,
      "step": 96020
    },
    {
      "entropy": 5.680152225494385,
      "epoch": 8.710540638606677,
      "grad_norm": 1.5625,
      "learning_rate": 6.852088974535564e-05,
      "loss": 4.7823,
      "mean_token_accuracy": 0.24260251671075822,
      "num_tokens": 199266599.0,
      "step": 96025
    },
    {
      "entropy": 5.732037782669067,
      "epoch": 8.71099419448476,
      "grad_norm": 1.65625,
      "learning_rate": 6.85080363868989e-05,
      "loss": 4.7885,
      "mean_token_accuracy": 0.23821428567171096,
      "num_tokens": 199276327.0,
      "step": 96030
    },
    {
      "entropy": 5.692191171646118,
      "epoch": 8.711447750362845,
      "grad_norm": 1.6796875,
      "learning_rate": 6.849518729874856e-05,
      "loss": 4.9024,
      "mean_token_accuracy": 0.22486144304275513,
      "num_tokens": 199286358.0,
      "step": 96035
    },
    {
      "entropy": 5.700990676879883,
      "epoch": 8.711901306240929,
      "grad_norm": 1.6015625,
      "learning_rate": 6.848234248117036e-05,
      "loss": 4.8057,
      "mean_token_accuracy": 0.2407666727900505,
      "num_tokens": 199295660.0,
      "step": 96040
    },
    {
      "entropy": 5.663610601425171,
      "epoch": 8.712354862119014,
      "grad_norm": 1.5625,
      "learning_rate": 6.846950193442994e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2367233783006668,
      "num_tokens": 199305749.0,
      "step": 96045
    },
    {
      "entropy": 5.7170515060424805,
      "epoch": 8.712808417997097,
      "grad_norm": 1.4921875,
      "learning_rate": 6.845666565879285e-05,
      "loss": 4.8785,
      "mean_token_accuracy": 0.22683407813310624,
      "num_tokens": 199316453.0,
      "step": 96050
    },
    {
      "entropy": 5.67510576248169,
      "epoch": 8.71326197387518,
      "grad_norm": 1.5390625,
      "learning_rate": 6.844383365452454e-05,
      "loss": 4.8062,
      "mean_token_accuracy": 0.23996615707874297,
      "num_tokens": 199326358.0,
      "step": 96055
    },
    {
      "entropy": 5.680955600738526,
      "epoch": 8.713715529753266,
      "grad_norm": 1.484375,
      "learning_rate": 6.843100592189043e-05,
      "loss": 4.7958,
      "mean_token_accuracy": 0.23681056052446364,
      "num_tokens": 199337125.0,
      "step": 96060
    },
    {
      "entropy": 5.679425382614136,
      "epoch": 8.71416908563135,
      "grad_norm": 1.46875,
      "learning_rate": 6.841818246115572e-05,
      "loss": 4.8082,
      "mean_token_accuracy": 0.23601134419441222,
      "num_tokens": 199348357.0,
      "step": 96065
    },
    {
      "entropy": 5.653819179534912,
      "epoch": 8.714622641509434,
      "grad_norm": 1.5078125,
      "learning_rate": 6.840536327258566e-05,
      "loss": 4.8008,
      "mean_token_accuracy": 0.23044730871915817,
      "num_tokens": 199358770.0,
      "step": 96070
    },
    {
      "entropy": 5.651932430267334,
      "epoch": 8.715076197387518,
      "grad_norm": 1.6015625,
      "learning_rate": 6.839254835644534e-05,
      "loss": 4.7495,
      "mean_token_accuracy": 0.2425619825720787,
      "num_tokens": 199368474.0,
      "step": 96075
    },
    {
      "entropy": 5.687190389633178,
      "epoch": 8.715529753265603,
      "grad_norm": 1.46875,
      "learning_rate": 6.837973771299974e-05,
      "loss": 4.7713,
      "mean_token_accuracy": 0.23939911872148514,
      "num_tokens": 199378431.0,
      "step": 96080
    },
    {
      "entropy": 5.696781206130981,
      "epoch": 8.715983309143686,
      "grad_norm": 1.5625,
      "learning_rate": 6.83669313425138e-05,
      "loss": 4.8461,
      "mean_token_accuracy": 0.23755491822957991,
      "num_tokens": 199390226.0,
      "step": 96085
    },
    {
      "entropy": 5.678949928283691,
      "epoch": 8.716436865021771,
      "grad_norm": 1.5,
      "learning_rate": 6.835412924525242e-05,
      "loss": 4.8229,
      "mean_token_accuracy": 0.2388919249176979,
      "num_tokens": 199401253.0,
      "step": 96090
    },
    {
      "entropy": 5.671956014633179,
      "epoch": 8.716890420899855,
      "grad_norm": 1.5546875,
      "learning_rate": 6.834133142148032e-05,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23791813105344772,
      "num_tokens": 199411856.0,
      "step": 96095
    },
    {
      "entropy": 5.623066473007202,
      "epoch": 8.71734397677794,
      "grad_norm": 1.515625,
      "learning_rate": 6.832853787146212e-05,
      "loss": 4.6931,
      "mean_token_accuracy": 0.25044112652540207,
      "num_tokens": 199422357.0,
      "step": 96100
    },
    {
      "entropy": 5.760988044738769,
      "epoch": 8.717797532656023,
      "grad_norm": 1.6640625,
      "learning_rate": 6.831574859546243e-05,
      "loss": 4.8347,
      "mean_token_accuracy": 0.23482432067394257,
      "num_tokens": 199432387.0,
      "step": 96105
    },
    {
      "entropy": 5.6984639167785645,
      "epoch": 8.718251088534107,
      "grad_norm": 1.5390625,
      "learning_rate": 6.830296359374577e-05,
      "loss": 4.7924,
      "mean_token_accuracy": 0.2388578549027443,
      "num_tokens": 199443514.0,
      "step": 96110
    },
    {
      "entropy": 5.629387283325196,
      "epoch": 8.718704644412192,
      "grad_norm": 1.53125,
      "learning_rate": 6.829018286657643e-05,
      "loss": 4.7323,
      "mean_token_accuracy": 0.2415179580450058,
      "num_tokens": 199454616.0,
      "step": 96115
    },
    {
      "entropy": 5.6945901870727536,
      "epoch": 8.719158200290275,
      "grad_norm": 1.578125,
      "learning_rate": 6.827740641421879e-05,
      "loss": 4.7875,
      "mean_token_accuracy": 0.24310997873544693,
      "num_tokens": 199464820.0,
      "step": 96120
    },
    {
      "entropy": 5.716186618804931,
      "epoch": 8.71961175616836,
      "grad_norm": 1.640625,
      "learning_rate": 6.826463423693711e-05,
      "loss": 4.7996,
      "mean_token_accuracy": 0.23661217838525772,
      "num_tokens": 199474731.0,
      "step": 96125
    },
    {
      "entropy": 5.654679298400879,
      "epoch": 8.720065312046444,
      "grad_norm": 1.5859375,
      "learning_rate": 6.825186633499544e-05,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23764938861131668,
      "num_tokens": 199484802.0,
      "step": 96130
    },
    {
      "entropy": 5.6571465015411375,
      "epoch": 8.720518867924529,
      "grad_norm": 1.4375,
      "learning_rate": 6.823910270865787e-05,
      "loss": 4.7651,
      "mean_token_accuracy": 0.23874193876981736,
      "num_tokens": 199495064.0,
      "step": 96135
    },
    {
      "entropy": 5.657387113571167,
      "epoch": 8.720972423802612,
      "grad_norm": 1.453125,
      "learning_rate": 6.822634335818836e-05,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23537816256284713,
      "num_tokens": 199505002.0,
      "step": 96140
    },
    {
      "entropy": 5.644528532028199,
      "epoch": 8.721425979680697,
      "grad_norm": 1.515625,
      "learning_rate": 6.82135882838507e-05,
      "loss": 4.7571,
      "mean_token_accuracy": 0.2451302245259285,
      "num_tokens": 199515945.0,
      "step": 96145
    },
    {
      "entropy": 5.686708974838257,
      "epoch": 8.72187953555878,
      "grad_norm": 1.46875,
      "learning_rate": 6.820083748590878e-05,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23379101008176803,
      "num_tokens": 199526277.0,
      "step": 96150
    },
    {
      "entropy": 5.657414531707763,
      "epoch": 8.722333091436864,
      "grad_norm": 1.515625,
      "learning_rate": 6.818809096462623e-05,
      "loss": 4.743,
      "mean_token_accuracy": 0.24712992310523987,
      "num_tokens": 199537020.0,
      "step": 96155
    },
    {
      "entropy": 5.694050931930542,
      "epoch": 8.72278664731495,
      "grad_norm": 1.59375,
      "learning_rate": 6.817534872026668e-05,
      "loss": 4.8219,
      "mean_token_accuracy": 0.23909775614738465,
      "num_tokens": 199546796.0,
      "step": 96160
    },
    {
      "entropy": 5.66568808555603,
      "epoch": 8.723240203193033,
      "grad_norm": 1.4609375,
      "learning_rate": 6.81626107530936e-05,
      "loss": 4.7437,
      "mean_token_accuracy": 0.23706289976835251,
      "num_tokens": 199556921.0,
      "step": 96165
    },
    {
      "entropy": 5.690482234954834,
      "epoch": 8.723693759071118,
      "grad_norm": 1.6171875,
      "learning_rate": 6.814987706337044e-05,
      "loss": 4.8342,
      "mean_token_accuracy": 0.2373340681195259,
      "num_tokens": 199567982.0,
      "step": 96170
    },
    {
      "entropy": 5.669035816192627,
      "epoch": 8.724147314949201,
      "grad_norm": 1.4609375,
      "learning_rate": 6.813714765136055e-05,
      "loss": 4.8177,
      "mean_token_accuracy": 0.24155838787555695,
      "num_tokens": 199578528.0,
      "step": 96175
    },
    {
      "entropy": 5.6910919666290285,
      "epoch": 8.724600870827286,
      "grad_norm": 1.5703125,
      "learning_rate": 6.812442251732713e-05,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22941257059574127,
      "num_tokens": 199589767.0,
      "step": 96180
    },
    {
      "entropy": 5.69490065574646,
      "epoch": 8.72505442670537,
      "grad_norm": 1.59375,
      "learning_rate": 6.81117016615334e-05,
      "loss": 4.787,
      "mean_token_accuracy": 0.24083503782749177,
      "num_tokens": 199598977.0,
      "step": 96185
    },
    {
      "entropy": 5.650962257385254,
      "epoch": 8.725507982583455,
      "grad_norm": 1.75,
      "learning_rate": 6.809898508424239e-05,
      "loss": 4.7997,
      "mean_token_accuracy": 0.23647417575120927,
      "num_tokens": 199609241.0,
      "step": 96190
    },
    {
      "entropy": 5.697373390197754,
      "epoch": 8.725961538461538,
      "grad_norm": 1.578125,
      "learning_rate": 6.808627278571711e-05,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23481715321540833,
      "num_tokens": 199619075.0,
      "step": 96195
    },
    {
      "entropy": 5.658539438247681,
      "epoch": 8.726415094339622,
      "grad_norm": 1.4765625,
      "learning_rate": 6.807356476622041e-05,
      "loss": 4.7474,
      "mean_token_accuracy": 0.24118629544973375,
      "num_tokens": 199629998.0,
      "step": 96200
    },
    {
      "entropy": 5.586414957046509,
      "epoch": 8.726868650217707,
      "grad_norm": 1.4921875,
      "learning_rate": 6.806086102601514e-05,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2509779006242752,
      "num_tokens": 199640309.0,
      "step": 96205
    },
    {
      "entropy": 5.703933334350586,
      "epoch": 8.72732220609579,
      "grad_norm": 1.5,
      "learning_rate": 6.8048161565364e-05,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23104805648326873,
      "num_tokens": 199650551.0,
      "step": 96210
    },
    {
      "entropy": 5.6473469734191895,
      "epoch": 8.727775761973875,
      "grad_norm": 1.6015625,
      "learning_rate": 6.803546638452957e-05,
      "loss": 4.7902,
      "mean_token_accuracy": 0.23539258986711503,
      "num_tokens": 199659940.0,
      "step": 96215
    },
    {
      "entropy": 5.630548477172852,
      "epoch": 8.728229317851959,
      "grad_norm": 1.5625,
      "learning_rate": 6.802277548377449e-05,
      "loss": 4.7028,
      "mean_token_accuracy": 0.2444749429821968,
      "num_tokens": 199670649.0,
      "step": 96220
    },
    {
      "entropy": 5.7838012218475345,
      "epoch": 8.728682873730044,
      "grad_norm": 1.453125,
      "learning_rate": 6.801008886336116e-05,
      "loss": 4.9188,
      "mean_token_accuracy": 0.22070621252059935,
      "num_tokens": 199681318.0,
      "step": 96225
    },
    {
      "entropy": 5.784667062759399,
      "epoch": 8.729136429608127,
      "grad_norm": 1.5703125,
      "learning_rate": 6.799740652355192e-05,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2218153640627861,
      "num_tokens": 199691666.0,
      "step": 96230
    },
    {
      "entropy": 5.660522985458374,
      "epoch": 8.729589985486212,
      "grad_norm": 1.4921875,
      "learning_rate": 6.798472846460907e-05,
      "loss": 4.7468,
      "mean_token_accuracy": 0.24519306123256684,
      "num_tokens": 199702257.0,
      "step": 96235
    },
    {
      "entropy": 5.693199586868286,
      "epoch": 8.730043541364296,
      "grad_norm": 1.4453125,
      "learning_rate": 6.797205468679478e-05,
      "loss": 4.8165,
      "mean_token_accuracy": 0.24279074370861053,
      "num_tokens": 199712563.0,
      "step": 96240
    },
    {
      "entropy": 5.689426231384277,
      "epoch": 8.730497097242381,
      "grad_norm": 1.4765625,
      "learning_rate": 6.795938519037118e-05,
      "loss": 4.7852,
      "mean_token_accuracy": 0.24448681324720384,
      "num_tokens": 199723146.0,
      "step": 96245
    },
    {
      "entropy": 5.667449998855591,
      "epoch": 8.730950653120464,
      "grad_norm": 1.4375,
      "learning_rate": 6.794671997560018e-05,
      "loss": 4.798,
      "mean_token_accuracy": 0.23016846626996995,
      "num_tokens": 199732685.0,
      "step": 96250
    },
    {
      "entropy": 5.68251371383667,
      "epoch": 8.731404208998548,
      "grad_norm": 1.578125,
      "learning_rate": 6.793405904274386e-05,
      "loss": 4.8164,
      "mean_token_accuracy": 0.24152113050222396,
      "num_tokens": 199743644.0,
      "step": 96255
    },
    {
      "entropy": 5.666622161865234,
      "epoch": 8.731857764876633,
      "grad_norm": 1.5,
      "learning_rate": 6.792140239206393e-05,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23488992154598237,
      "num_tokens": 199754575.0,
      "step": 96260
    },
    {
      "entropy": 5.691819047927856,
      "epoch": 8.732311320754716,
      "grad_norm": 1.4453125,
      "learning_rate": 6.790875002382217e-05,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2392791047692299,
      "num_tokens": 199766189.0,
      "step": 96265
    },
    {
      "entropy": 5.637793254852295,
      "epoch": 8.732764876632801,
      "grad_norm": 1.5546875,
      "learning_rate": 6.789610193828026e-05,
      "loss": 4.7912,
      "mean_token_accuracy": 0.23194321244955063,
      "num_tokens": 199775589.0,
      "step": 96270
    },
    {
      "entropy": 5.699746704101562,
      "epoch": 8.733218432510885,
      "grad_norm": 1.4453125,
      "learning_rate": 6.78834581356997e-05,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23667455166578294,
      "num_tokens": 199786988.0,
      "step": 96275
    },
    {
      "entropy": 5.701726198196411,
      "epoch": 8.73367198838897,
      "grad_norm": 1.3359375,
      "learning_rate": 6.787081861634203e-05,
      "loss": 4.7685,
      "mean_token_accuracy": 0.23216334581375123,
      "num_tokens": 199797509.0,
      "step": 96280
    },
    {
      "entropy": 5.665735244750977,
      "epoch": 8.734125544267053,
      "grad_norm": 1.4609375,
      "learning_rate": 6.785818338046858e-05,
      "loss": 4.7757,
      "mean_token_accuracy": 0.23772511035203933,
      "num_tokens": 199808073.0,
      "step": 96285
    },
    {
      "entropy": 5.6242108821868895,
      "epoch": 8.734579100145138,
      "grad_norm": 1.59375,
      "learning_rate": 6.78455524283407e-05,
      "loss": 4.6999,
      "mean_token_accuracy": 0.24709651470184327,
      "num_tokens": 199818412.0,
      "step": 96290
    },
    {
      "entropy": 5.706109237670899,
      "epoch": 8.735032656023222,
      "grad_norm": 1.4453125,
      "learning_rate": 6.783292576021959e-05,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23500945568084716,
      "num_tokens": 199828415.0,
      "step": 96295
    },
    {
      "entropy": 5.715557622909546,
      "epoch": 8.735486211901307,
      "grad_norm": 1.3984375,
      "learning_rate": 6.78203033763664e-05,
      "loss": 4.8355,
      "mean_token_accuracy": 0.2329508990049362,
      "num_tokens": 199838781.0,
      "step": 96300
    },
    {
      "entropy": 5.628757429122925,
      "epoch": 8.73593976777939,
      "grad_norm": 1.4921875,
      "learning_rate": 6.780768527704209e-05,
      "loss": 4.6788,
      "mean_token_accuracy": 0.2434739053249359,
      "num_tokens": 199848300.0,
      "step": 96305
    },
    {
      "entropy": 5.661186695098877,
      "epoch": 8.736393323657474,
      "grad_norm": 1.484375,
      "learning_rate": 6.779507146250766e-05,
      "loss": 4.7623,
      "mean_token_accuracy": 0.23076913356781006,
      "num_tokens": 199858514.0,
      "step": 96310
    },
    {
      "entropy": 5.633873081207275,
      "epoch": 8.736846879535559,
      "grad_norm": 1.4765625,
      "learning_rate": 6.778246193302399e-05,
      "loss": 4.7735,
      "mean_token_accuracy": 0.2391260787844658,
      "num_tokens": 199869106.0,
      "step": 96315
    },
    {
      "entropy": 5.7211662292480465,
      "epoch": 8.737300435413642,
      "grad_norm": 1.46875,
      "learning_rate": 6.776985668885172e-05,
      "loss": 4.8315,
      "mean_token_accuracy": 0.23270571380853652,
      "num_tokens": 199878760.0,
      "step": 96320
    },
    {
      "entropy": 5.695219039916992,
      "epoch": 8.737753991291727,
      "grad_norm": 1.53125,
      "learning_rate": 6.775725573025168e-05,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23565389662981034,
      "num_tokens": 199890025.0,
      "step": 96325
    },
    {
      "entropy": 5.73369083404541,
      "epoch": 8.73820754716981,
      "grad_norm": 1.4765625,
      "learning_rate": 6.77446590574844e-05,
      "loss": 4.8457,
      "mean_token_accuracy": 0.23267795592546464,
      "num_tokens": 199900727.0,
      "step": 96330
    },
    {
      "entropy": 5.683148765563965,
      "epoch": 8.738661103047896,
      "grad_norm": 1.46875,
      "learning_rate": 6.773206667081039e-05,
      "loss": 4.8715,
      "mean_token_accuracy": 0.23493884205818177,
      "num_tokens": 199911175.0,
      "step": 96335
    },
    {
      "entropy": 5.61182279586792,
      "epoch": 8.73911465892598,
      "grad_norm": 1.59375,
      "learning_rate": 6.771947857049007e-05,
      "loss": 4.688,
      "mean_token_accuracy": 0.24665363430976867,
      "num_tokens": 199921577.0,
      "step": 96340
    },
    {
      "entropy": 5.7191667556762695,
      "epoch": 8.739568214804065,
      "grad_norm": 1.375,
      "learning_rate": 6.770689475678374e-05,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2342657834291458,
      "num_tokens": 199931872.0,
      "step": 96345
    },
    {
      "entropy": 5.669942617416382,
      "epoch": 8.740021770682148,
      "grad_norm": 1.5859375,
      "learning_rate": 6.769431522995163e-05,
      "loss": 4.7602,
      "mean_token_accuracy": 0.2486461356282234,
      "num_tokens": 199941937.0,
      "step": 96350
    },
    {
      "entropy": 5.694740867614746,
      "epoch": 8.740475326560233,
      "grad_norm": 1.4609375,
      "learning_rate": 6.768173999025393e-05,
      "loss": 4.8325,
      "mean_token_accuracy": 0.23603419512510299,
      "num_tokens": 199952778.0,
      "step": 96355
    },
    {
      "entropy": 5.699014949798584,
      "epoch": 8.740928882438316,
      "grad_norm": 1.6875,
      "learning_rate": 6.766916903795069e-05,
      "loss": 4.7957,
      "mean_token_accuracy": 0.23744803220033645,
      "num_tokens": 199963217.0,
      "step": 96360
    },
    {
      "entropy": 5.723055028915406,
      "epoch": 8.7413824383164,
      "grad_norm": 1.3984375,
      "learning_rate": 6.765660237330186e-05,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23378810286521912,
      "num_tokens": 199973341.0,
      "step": 96365
    },
    {
      "entropy": 5.701033210754394,
      "epoch": 8.741835994194485,
      "grad_norm": 1.6015625,
      "learning_rate": 6.764403999656732e-05,
      "loss": 4.8114,
      "mean_token_accuracy": 0.23558196276426316,
      "num_tokens": 199983541.0,
      "step": 96370
    },
    {
      "entropy": 5.719546318054199,
      "epoch": 8.742289550072568,
      "grad_norm": 1.59375,
      "learning_rate": 6.76314819080069e-05,
      "loss": 4.8263,
      "mean_token_accuracy": 0.2364787057042122,
      "num_tokens": 199994203.0,
      "step": 96375
    },
    {
      "entropy": 5.7709400177001955,
      "epoch": 8.742743105950654,
      "grad_norm": 1.453125,
      "learning_rate": 6.761892810788026e-05,
      "loss": 4.9117,
      "mean_token_accuracy": 0.22642490416765212,
      "num_tokens": 200004991.0,
      "step": 96380
    },
    {
      "entropy": 5.674531316757202,
      "epoch": 8.743196661828737,
      "grad_norm": 1.640625,
      "learning_rate": 6.7606378596447e-05,
      "loss": 4.8121,
      "mean_token_accuracy": 0.234504896402359,
      "num_tokens": 200015009.0,
      "step": 96385
    },
    {
      "entropy": 5.669355154037476,
      "epoch": 8.743650217706822,
      "grad_norm": 1.46875,
      "learning_rate": 6.75938333739667e-05,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23269417881965637,
      "num_tokens": 200025339.0,
      "step": 96390
    },
    {
      "entropy": 5.776067399978638,
      "epoch": 8.744103773584905,
      "grad_norm": 1.53125,
      "learning_rate": 6.75812924406988e-05,
      "loss": 4.9273,
      "mean_token_accuracy": 0.22622648626565933,
      "num_tokens": 200035814.0,
      "step": 96395
    },
    {
      "entropy": 5.670278739929199,
      "epoch": 8.74455732946299,
      "grad_norm": 1.546875,
      "learning_rate": 6.756875579690266e-05,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23840238749980927,
      "num_tokens": 200046323.0,
      "step": 96400
    },
    {
      "entropy": 5.646836423873902,
      "epoch": 8.745010885341074,
      "grad_norm": 1.53125,
      "learning_rate": 6.755622344283746e-05,
      "loss": 4.7065,
      "mean_token_accuracy": 0.23724051415920258,
      "num_tokens": 200056725.0,
      "step": 96405
    },
    {
      "entropy": 5.638754177093506,
      "epoch": 8.74546444121916,
      "grad_norm": 1.4609375,
      "learning_rate": 6.754369537876245e-05,
      "loss": 4.7312,
      "mean_token_accuracy": 0.24064146131277084,
      "num_tokens": 200067450.0,
      "step": 96410
    },
    {
      "entropy": 5.689153146743775,
      "epoch": 8.745917997097242,
      "grad_norm": 1.5625,
      "learning_rate": 6.753117160493666e-05,
      "loss": 4.799,
      "mean_token_accuracy": 0.23751013427972795,
      "num_tokens": 200077691.0,
      "step": 96415
    },
    {
      "entropy": 5.668054866790771,
      "epoch": 8.746371552975326,
      "grad_norm": 1.5625,
      "learning_rate": 6.751865212161909e-05,
      "loss": 4.7107,
      "mean_token_accuracy": 0.24917687028646468,
      "num_tokens": 200087794.0,
      "step": 96420
    },
    {
      "entropy": 5.6200902462005615,
      "epoch": 8.746825108853411,
      "grad_norm": 1.625,
      "learning_rate": 6.75061369290687e-05,
      "loss": 4.724,
      "mean_token_accuracy": 0.24920393377542496,
      "num_tokens": 200097899.0,
      "step": 96425
    },
    {
      "entropy": 5.689812278747558,
      "epoch": 8.747278664731494,
      "grad_norm": 1.5390625,
      "learning_rate": 6.749362602754427e-05,
      "loss": 4.836,
      "mean_token_accuracy": 0.2307540848851204,
      "num_tokens": 200109644.0,
      "step": 96430
    },
    {
      "entropy": 5.658649682998657,
      "epoch": 8.74773222060958,
      "grad_norm": 1.53125,
      "learning_rate": 6.748111941730454e-05,
      "loss": 4.7347,
      "mean_token_accuracy": 0.24153763204813003,
      "num_tokens": 200120518.0,
      "step": 96435
    },
    {
      "entropy": 5.660561513900757,
      "epoch": 8.748185776487663,
      "grad_norm": 1.3671875,
      "learning_rate": 6.746861709860815e-05,
      "loss": 4.7833,
      "mean_token_accuracy": 0.235795097053051,
      "num_tokens": 200130886.0,
      "step": 96440
    },
    {
      "entropy": 5.7350804805755615,
      "epoch": 8.748639332365748,
      "grad_norm": 1.640625,
      "learning_rate": 6.745611907171362e-05,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23677044212818146,
      "num_tokens": 200140466.0,
      "step": 96445
    },
    {
      "entropy": 5.69506139755249,
      "epoch": 8.749092888243831,
      "grad_norm": 1.390625,
      "learning_rate": 6.744362533687944e-05,
      "loss": 4.898,
      "mean_token_accuracy": 0.23188083618879318,
      "num_tokens": 200151424.0,
      "step": 96450
    },
    {
      "entropy": 5.702037620544433,
      "epoch": 8.749546444121917,
      "grad_norm": 1.71875,
      "learning_rate": 6.743113589436397e-05,
      "loss": 4.8349,
      "mean_token_accuracy": 0.22933679968118667,
      "num_tokens": 200161537.0,
      "step": 96455
    },
    {
      "entropy": 5.750099420547485,
      "epoch": 8.75,
      "grad_norm": 1.6015625,
      "learning_rate": 6.741865074442552e-05,
      "loss": 4.8706,
      "mean_token_accuracy": 0.23063631653785704,
      "num_tokens": 200171722.0,
      "step": 96460
    },
    {
      "entropy": 5.733180236816406,
      "epoch": 8.750453555878083,
      "grad_norm": 1.515625,
      "learning_rate": 6.74061698873223e-05,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2363736242055893,
      "num_tokens": 200181458.0,
      "step": 96465
    },
    {
      "entropy": 5.638527345657349,
      "epoch": 8.750907111756169,
      "grad_norm": 1.5390625,
      "learning_rate": 6.739369332331236e-05,
      "loss": 4.7617,
      "mean_token_accuracy": 0.2354294925928116,
      "num_tokens": 200192145.0,
      "step": 96470
    },
    {
      "entropy": 5.737450885772705,
      "epoch": 8.751360667634252,
      "grad_norm": 1.65625,
      "learning_rate": 6.738122105265375e-05,
      "loss": 4.8364,
      "mean_token_accuracy": 0.23425998240709306,
      "num_tokens": 200202107.0,
      "step": 96475
    },
    {
      "entropy": 5.692864942550659,
      "epoch": 8.751814223512337,
      "grad_norm": 1.265625,
      "learning_rate": 6.736875307560438e-05,
      "loss": 4.7719,
      "mean_token_accuracy": 0.24224441200494767,
      "num_tokens": 200214108.0,
      "step": 96480
    },
    {
      "entropy": 5.581885433197021,
      "epoch": 8.75226777939042,
      "grad_norm": 1.5,
      "learning_rate": 6.735628939242212e-05,
      "loss": 4.7094,
      "mean_token_accuracy": 0.24173057228326797,
      "num_tokens": 200224546.0,
      "step": 96485
    },
    {
      "entropy": 5.747790145874023,
      "epoch": 8.752721335268506,
      "grad_norm": 1.6484375,
      "learning_rate": 6.734383000336467e-05,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2335784003138542,
      "num_tokens": 200234104.0,
      "step": 96490
    },
    {
      "entropy": 5.617891120910644,
      "epoch": 8.753174891146589,
      "grad_norm": 1.65625,
      "learning_rate": 6.733137490868978e-05,
      "loss": 4.672,
      "mean_token_accuracy": 0.24834194481372834,
      "num_tokens": 200245021.0,
      "step": 96495
    },
    {
      "entropy": 5.696099948883057,
      "epoch": 8.753628447024674,
      "grad_norm": 1.5390625,
      "learning_rate": 6.731892410865498e-05,
      "loss": 4.7852,
      "mean_token_accuracy": 0.2333117827773094,
      "num_tokens": 200255178.0,
      "step": 96500
    },
    {
      "entropy": 5.639035034179687,
      "epoch": 8.754082002902758,
      "grad_norm": 1.4765625,
      "learning_rate": 6.730647760351772e-05,
      "loss": 4.7455,
      "mean_token_accuracy": 0.23701688647270203,
      "num_tokens": 200264848.0,
      "step": 96505
    },
    {
      "entropy": 5.712043952941895,
      "epoch": 8.75453555878084,
      "grad_norm": 1.6015625,
      "learning_rate": 6.729403539353547e-05,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2345828741788864,
      "num_tokens": 200275632.0,
      "step": 96510
    },
    {
      "entropy": 5.6222062587738035,
      "epoch": 8.754989114658926,
      "grad_norm": 1.4296875,
      "learning_rate": 6.728159747896548e-05,
      "loss": 4.7247,
      "mean_token_accuracy": 0.24603281617164613,
      "num_tokens": 200286528.0,
      "step": 96515
    },
    {
      "entropy": 5.650107431411743,
      "epoch": 8.75544267053701,
      "grad_norm": 1.4140625,
      "learning_rate": 6.7269163860065e-05,
      "loss": 4.7387,
      "mean_token_accuracy": 0.24272983223199845,
      "num_tokens": 200297413.0,
      "step": 96520
    },
    {
      "entropy": 5.700477218627929,
      "epoch": 8.755896226415095,
      "grad_norm": 1.515625,
      "learning_rate": 6.72567345370911e-05,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22735320627689362,
      "num_tokens": 200307967.0,
      "step": 96525
    },
    {
      "entropy": 5.711098718643188,
      "epoch": 8.756349782293178,
      "grad_norm": 1.4921875,
      "learning_rate": 6.72443095103009e-05,
      "loss": 4.7996,
      "mean_token_accuracy": 0.2310865953564644,
      "num_tokens": 200318649.0,
      "step": 96530
    },
    {
      "entropy": 5.7040893077850345,
      "epoch": 8.756803338171263,
      "grad_norm": 1.4921875,
      "learning_rate": 6.723188877995138e-05,
      "loss": 4.8278,
      "mean_token_accuracy": 0.2331450968980789,
      "num_tokens": 200328586.0,
      "step": 96535
    },
    {
      "entropy": 5.700654315948486,
      "epoch": 8.757256894049346,
      "grad_norm": 1.5390625,
      "learning_rate": 6.721947234629928e-05,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23878565281629563,
      "num_tokens": 200338944.0,
      "step": 96540
    },
    {
      "entropy": 5.706690883636474,
      "epoch": 8.757710449927432,
      "grad_norm": 1.6484375,
      "learning_rate": 6.720706020960145e-05,
      "loss": 4.8029,
      "mean_token_accuracy": 0.2335526764392853,
      "num_tokens": 200349130.0,
      "step": 96545
    },
    {
      "entropy": 5.677613115310669,
      "epoch": 8.758164005805515,
      "grad_norm": 1.4375,
      "learning_rate": 6.719465237011455e-05,
      "loss": 4.775,
      "mean_token_accuracy": 0.22763556838035584,
      "num_tokens": 200360245.0,
      "step": 96550
    },
    {
      "entropy": 5.712292098999024,
      "epoch": 8.7586175616836,
      "grad_norm": 1.40625,
      "learning_rate": 6.718224882809521e-05,
      "loss": 4.7855,
      "mean_token_accuracy": 0.23589649349451064,
      "num_tokens": 200370463.0,
      "step": 96555
    },
    {
      "entropy": 5.601830625534058,
      "epoch": 8.759071117561684,
      "grad_norm": 1.5703125,
      "learning_rate": 6.716984958379992e-05,
      "loss": 4.729,
      "mean_token_accuracy": 0.24622447788715363,
      "num_tokens": 200381022.0,
      "step": 96560
    },
    {
      "entropy": 5.733254384994507,
      "epoch": 8.759524673439767,
      "grad_norm": 1.5,
      "learning_rate": 6.715745463748512e-05,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23333143144845964,
      "num_tokens": 200391649.0,
      "step": 96565
    },
    {
      "entropy": 5.700889158248901,
      "epoch": 8.759978229317852,
      "grad_norm": 1.6953125,
      "learning_rate": 6.714506398940712e-05,
      "loss": 4.7796,
      "mean_token_accuracy": 0.2408888280391693,
      "num_tokens": 200401300.0,
      "step": 96570
    },
    {
      "entropy": 5.6011255264282225,
      "epoch": 8.760431785195935,
      "grad_norm": 1.375,
      "learning_rate": 6.713267763982214e-05,
      "loss": 4.693,
      "mean_token_accuracy": 0.2460313156247139,
      "num_tokens": 200412609.0,
      "step": 96575
    },
    {
      "entropy": 5.658655738830566,
      "epoch": 8.76088534107402,
      "grad_norm": 1.4375,
      "learning_rate": 6.712029558898639e-05,
      "loss": 4.8084,
      "mean_token_accuracy": 0.2324751004576683,
      "num_tokens": 200423861.0,
      "step": 96580
    },
    {
      "entropy": 5.641210508346558,
      "epoch": 8.761338896952104,
      "grad_norm": 1.5546875,
      "learning_rate": 6.710791783715585e-05,
      "loss": 4.7641,
      "mean_token_accuracy": 0.2458208382129669,
      "num_tokens": 200434582.0,
      "step": 96585
    },
    {
      "entropy": 5.689571952819824,
      "epoch": 8.76179245283019,
      "grad_norm": 1.5390625,
      "learning_rate": 6.709554438458657e-05,
      "loss": 4.7606,
      "mean_token_accuracy": 0.2397902563214302,
      "num_tokens": 200445440.0,
      "step": 96590
    },
    {
      "entropy": 5.776263809204101,
      "epoch": 8.762246008708273,
      "grad_norm": 1.4453125,
      "learning_rate": 6.708317523153441e-05,
      "loss": 4.9179,
      "mean_token_accuracy": 0.23209817558526993,
      "num_tokens": 200455745.0,
      "step": 96595
    },
    {
      "entropy": 5.713454055786133,
      "epoch": 8.762699564586358,
      "grad_norm": 1.46875,
      "learning_rate": 6.70708103782552e-05,
      "loss": 4.8415,
      "mean_token_accuracy": 0.2295996993780136,
      "num_tokens": 200466395.0,
      "step": 96600
    },
    {
      "entropy": 5.653805351257324,
      "epoch": 8.763153120464441,
      "grad_norm": 1.5703125,
      "learning_rate": 6.705844982500459e-05,
      "loss": 4.7863,
      "mean_token_accuracy": 0.24671195149421693,
      "num_tokens": 200477171.0,
      "step": 96605
    },
    {
      "entropy": 5.564399766921997,
      "epoch": 8.763606676342526,
      "grad_norm": 1.5234375,
      "learning_rate": 6.704609357203826e-05,
      "loss": 4.6141,
      "mean_token_accuracy": 0.2642362415790558,
      "num_tokens": 200487818.0,
      "step": 96610
    },
    {
      "entropy": 5.686679172515869,
      "epoch": 8.76406023222061,
      "grad_norm": 1.453125,
      "learning_rate": 6.703374161961168e-05,
      "loss": 4.8488,
      "mean_token_accuracy": 0.23372852057218552,
      "num_tokens": 200498276.0,
      "step": 96615
    },
    {
      "entropy": 5.712084102630615,
      "epoch": 8.764513788098693,
      "grad_norm": 1.7890625,
      "learning_rate": 6.702139396798029e-05,
      "loss": 4.8518,
      "mean_token_accuracy": 0.22817419916391374,
      "num_tokens": 200508134.0,
      "step": 96620
    },
    {
      "entropy": 5.638867473602295,
      "epoch": 8.764967343976778,
      "grad_norm": 1.4921875,
      "learning_rate": 6.70090506173995e-05,
      "loss": 4.7319,
      "mean_token_accuracy": 0.24644651412963867,
      "num_tokens": 200518573.0,
      "step": 96625
    },
    {
      "entropy": 5.706202840805053,
      "epoch": 8.765420899854862,
      "grad_norm": 1.5859375,
      "learning_rate": 6.699671156812457e-05,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2377387374639511,
      "num_tokens": 200528634.0,
      "step": 96630
    },
    {
      "entropy": 5.660369634628296,
      "epoch": 8.765874455732947,
      "grad_norm": 1.4453125,
      "learning_rate": 6.698437682041064e-05,
      "loss": 4.7638,
      "mean_token_accuracy": 0.2399592012166977,
      "num_tokens": 200539589.0,
      "step": 96635
    },
    {
      "entropy": 5.679997730255127,
      "epoch": 8.76632801161103,
      "grad_norm": 1.40625,
      "learning_rate": 6.697204637451279e-05,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23552077412605285,
      "num_tokens": 200550018.0,
      "step": 96640
    },
    {
      "entropy": 5.633405160903931,
      "epoch": 8.766781567489115,
      "grad_norm": 1.4921875,
      "learning_rate": 6.695972023068604e-05,
      "loss": 4.7164,
      "mean_token_accuracy": 0.24023498445749283,
      "num_tokens": 200560561.0,
      "step": 96645
    },
    {
      "entropy": 5.605841207504272,
      "epoch": 8.767235123367199,
      "grad_norm": 1.59375,
      "learning_rate": 6.694739838918526e-05,
      "loss": 4.6675,
      "mean_token_accuracy": 0.2507797345519066,
      "num_tokens": 200571287.0,
      "step": 96650
    },
    {
      "entropy": 5.680393838882447,
      "epoch": 8.767688679245284,
      "grad_norm": 1.4453125,
      "learning_rate": 6.693508085026532e-05,
      "loss": 4.8198,
      "mean_token_accuracy": 0.23128364682197572,
      "num_tokens": 200582109.0,
      "step": 96655
    },
    {
      "entropy": 5.592596912384034,
      "epoch": 8.768142235123367,
      "grad_norm": 1.4375,
      "learning_rate": 6.692276761418095e-05,
      "loss": 4.7491,
      "mean_token_accuracy": 0.24496504813432693,
      "num_tokens": 200592617.0,
      "step": 96660
    },
    {
      "entropy": 5.710501956939697,
      "epoch": 8.768595791001452,
      "grad_norm": 1.4296875,
      "learning_rate": 6.691045868118675e-05,
      "loss": 4.7598,
      "mean_token_accuracy": 0.2441597670316696,
      "num_tokens": 200602572.0,
      "step": 96665
    },
    {
      "entropy": 5.715892267227173,
      "epoch": 8.769049346879536,
      "grad_norm": 1.3828125,
      "learning_rate": 6.689815405153732e-05,
      "loss": 4.857,
      "mean_token_accuracy": 0.23108642399311066,
      "num_tokens": 200613786.0,
      "step": 96670
    },
    {
      "entropy": 5.7155379295349125,
      "epoch": 8.769502902757619,
      "grad_norm": 1.515625,
      "learning_rate": 6.688585372548708e-05,
      "loss": 4.8071,
      "mean_token_accuracy": 0.23838639557361602,
      "num_tokens": 200623575.0,
      "step": 96675
    },
    {
      "entropy": 5.745948743820191,
      "epoch": 8.769956458635704,
      "grad_norm": 1.5078125,
      "learning_rate": 6.687355770329043e-05,
      "loss": 4.8851,
      "mean_token_accuracy": 0.23194247037172316,
      "num_tokens": 200634060.0,
      "step": 96680
    },
    {
      "entropy": 5.6441551685333256,
      "epoch": 8.770410014513788,
      "grad_norm": 1.375,
      "learning_rate": 6.686126598520159e-05,
      "loss": 4.722,
      "mean_token_accuracy": 0.24019566923379898,
      "num_tokens": 200644617.0,
      "step": 96685
    },
    {
      "entropy": 5.707045316696167,
      "epoch": 8.770863570391873,
      "grad_norm": 1.5234375,
      "learning_rate": 6.684897857147485e-05,
      "loss": 4.7726,
      "mean_token_accuracy": 0.2333565816283226,
      "num_tokens": 200655302.0,
      "step": 96690
    },
    {
      "entropy": 5.7188982486724855,
      "epoch": 8.771317126269956,
      "grad_norm": 1.6328125,
      "learning_rate": 6.683669546236429e-05,
      "loss": 4.7945,
      "mean_token_accuracy": 0.24256370067596436,
      "num_tokens": 200665054.0,
      "step": 96695
    },
    {
      "entropy": 5.666860818862915,
      "epoch": 8.771770682148041,
      "grad_norm": 1.4375,
      "learning_rate": 6.682441665812389e-05,
      "loss": 4.8099,
      "mean_token_accuracy": 0.24099669307470323,
      "num_tokens": 200676168.0,
      "step": 96700
    },
    {
      "entropy": 5.668418025970459,
      "epoch": 8.772224238026125,
      "grad_norm": 1.453125,
      "learning_rate": 6.681214215900763e-05,
      "loss": 4.7732,
      "mean_token_accuracy": 0.24580787122249603,
      "num_tokens": 200686407.0,
      "step": 96705
    },
    {
      "entropy": 5.722747373580932,
      "epoch": 8.77267779390421,
      "grad_norm": 1.5859375,
      "learning_rate": 6.67998719652693e-05,
      "loss": 4.905,
      "mean_token_accuracy": 0.22375078946352006,
      "num_tokens": 200695926.0,
      "step": 96710
    },
    {
      "entropy": 5.724463844299317,
      "epoch": 8.773131349782293,
      "grad_norm": 1.6796875,
      "learning_rate": 6.678760607716268e-05,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23499290496110917,
      "num_tokens": 200705062.0,
      "step": 96715
    },
    {
      "entropy": 5.653147220611572,
      "epoch": 8.773584905660378,
      "grad_norm": 1.90625,
      "learning_rate": 6.677534449494139e-05,
      "loss": 4.7603,
      "mean_token_accuracy": 0.23409926593303682,
      "num_tokens": 200715835.0,
      "step": 96720
    },
    {
      "entropy": 5.701577281951904,
      "epoch": 8.774038461538462,
      "grad_norm": 1.5078125,
      "learning_rate": 6.676308721885907e-05,
      "loss": 4.8867,
      "mean_token_accuracy": 0.23007525652647018,
      "num_tokens": 200726800.0,
      "step": 96725
    },
    {
      "entropy": 5.639312601089477,
      "epoch": 8.774492017416545,
      "grad_norm": 1.3828125,
      "learning_rate": 6.675083424916917e-05,
      "loss": 4.6692,
      "mean_token_accuracy": 0.2441237077116966,
      "num_tokens": 200737058.0,
      "step": 96730
    },
    {
      "entropy": 5.656045627593994,
      "epoch": 8.77494557329463,
      "grad_norm": 1.59375,
      "learning_rate": 6.673858558612508e-05,
      "loss": 4.7516,
      "mean_token_accuracy": 0.24203009605407716,
      "num_tokens": 200747417.0,
      "step": 96735
    },
    {
      "entropy": 5.658394908905029,
      "epoch": 8.775399129172714,
      "grad_norm": 1.5703125,
      "learning_rate": 6.672634122998014e-05,
      "loss": 4.7519,
      "mean_token_accuracy": 0.24397179186344148,
      "num_tokens": 200758171.0,
      "step": 96740
    },
    {
      "entropy": 5.670050430297851,
      "epoch": 8.775852685050799,
      "grad_norm": 1.46875,
      "learning_rate": 6.671410118098752e-05,
      "loss": 4.7554,
      "mean_token_accuracy": 0.24257276803255082,
      "num_tokens": 200768512.0,
      "step": 96745
    },
    {
      "entropy": 5.660011434555054,
      "epoch": 8.776306240928882,
      "grad_norm": 1.359375,
      "learning_rate": 6.670186543940037e-05,
      "loss": 4.7784,
      "mean_token_accuracy": 0.23277491629123687,
      "num_tokens": 200779138.0,
      "step": 96750
    },
    {
      "entropy": 5.689007043838501,
      "epoch": 8.776759796806967,
      "grad_norm": 1.53125,
      "learning_rate": 6.668963400547168e-05,
      "loss": 4.8098,
      "mean_token_accuracy": 0.2349856272339821,
      "num_tokens": 200789643.0,
      "step": 96755
    },
    {
      "entropy": 5.722050619125366,
      "epoch": 8.77721335268505,
      "grad_norm": 1.5234375,
      "learning_rate": 6.667740687945447e-05,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2379545897245407,
      "num_tokens": 200800493.0,
      "step": 96760
    },
    {
      "entropy": 5.67942705154419,
      "epoch": 8.777666908563136,
      "grad_norm": 1.53125,
      "learning_rate": 6.66651840616016e-05,
      "loss": 4.8519,
      "mean_token_accuracy": 0.23473345637321472,
      "num_tokens": 200810328.0,
      "step": 96765
    },
    {
      "entropy": 5.6826425075531,
      "epoch": 8.77812046444122,
      "grad_norm": 1.4140625,
      "learning_rate": 6.665296555216581e-05,
      "loss": 4.7699,
      "mean_token_accuracy": 0.23420975655317305,
      "num_tokens": 200820508.0,
      "step": 96770
    },
    {
      "entropy": 5.70816330909729,
      "epoch": 8.778574020319303,
      "grad_norm": 1.546875,
      "learning_rate": 6.664075135139976e-05,
      "loss": 4.7413,
      "mean_token_accuracy": 0.24380262941122055,
      "num_tokens": 200831505.0,
      "step": 96775
    },
    {
      "entropy": 5.621404838562012,
      "epoch": 8.779027576197388,
      "grad_norm": 1.453125,
      "learning_rate": 6.662854145955609e-05,
      "loss": 4.746,
      "mean_token_accuracy": 0.23557863980531693,
      "num_tokens": 200841736.0,
      "step": 96780
    },
    {
      "entropy": 5.644989156723023,
      "epoch": 8.779481132075471,
      "grad_norm": 1.5390625,
      "learning_rate": 6.661633587688727e-05,
      "loss": 4.7883,
      "mean_token_accuracy": 0.23906248360872268,
      "num_tokens": 200852528.0,
      "step": 96785
    },
    {
      "entropy": 5.71359486579895,
      "epoch": 8.779934687953556,
      "grad_norm": 1.5078125,
      "learning_rate": 6.66041346036457e-05,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23254574239253997,
      "num_tokens": 200863964.0,
      "step": 96790
    },
    {
      "entropy": 5.716502046585083,
      "epoch": 8.78038824383164,
      "grad_norm": 1.3828125,
      "learning_rate": 6.659193764008375e-05,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23117168098688126,
      "num_tokens": 200874371.0,
      "step": 96795
    },
    {
      "entropy": 5.645013999938965,
      "epoch": 8.780841799709725,
      "grad_norm": 1.4765625,
      "learning_rate": 6.657974498645366e-05,
      "loss": 4.7525,
      "mean_token_accuracy": 0.24861744046211243,
      "num_tokens": 200885177.0,
      "step": 96800
    },
    {
      "entropy": 5.695674562454224,
      "epoch": 8.781295355587808,
      "grad_norm": 1.6015625,
      "learning_rate": 6.656755664300754e-05,
      "loss": 4.7714,
      "mean_token_accuracy": 0.2420763909816742,
      "num_tokens": 200895014.0,
      "step": 96805
    },
    {
      "entropy": 5.732991361618042,
      "epoch": 8.781748911465893,
      "grad_norm": 1.546875,
      "learning_rate": 6.655537260999748e-05,
      "loss": 4.8843,
      "mean_token_accuracy": 0.23181844055652617,
      "num_tokens": 200905579.0,
      "step": 96810
    },
    {
      "entropy": 5.614050436019897,
      "epoch": 8.782202467343977,
      "grad_norm": 1.53125,
      "learning_rate": 6.654319288767543e-05,
      "loss": 4.6737,
      "mean_token_accuracy": 0.252563413977623,
      "num_tokens": 200915873.0,
      "step": 96815
    },
    {
      "entropy": 5.754844617843628,
      "epoch": 8.78265602322206,
      "grad_norm": 1.515625,
      "learning_rate": 6.653101747629326e-05,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22599680870771408,
      "num_tokens": 200926674.0,
      "step": 96820
    },
    {
      "entropy": 5.684358406066894,
      "epoch": 8.783109579100145,
      "grad_norm": 1.5703125,
      "learning_rate": 6.651884637610272e-05,
      "loss": 4.8347,
      "mean_token_accuracy": 0.23591038435697556,
      "num_tokens": 200938375.0,
      "step": 96825
    },
    {
      "entropy": 5.694739866256714,
      "epoch": 8.783563134978229,
      "grad_norm": 1.453125,
      "learning_rate": 6.650667958735563e-05,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2322750747203827,
      "num_tokens": 200948666.0,
      "step": 96830
    },
    {
      "entropy": 5.638801860809326,
      "epoch": 8.784016690856314,
      "grad_norm": 1.5859375,
      "learning_rate": 6.649451711030352e-05,
      "loss": 4.7126,
      "mean_token_accuracy": 0.24012874066829681,
      "num_tokens": 200959124.0,
      "step": 96835
    },
    {
      "entropy": 5.6723987579345705,
      "epoch": 8.784470246734397,
      "grad_norm": 1.53125,
      "learning_rate": 6.648235894519795e-05,
      "loss": 4.7488,
      "mean_token_accuracy": 0.23577870726585387,
      "num_tokens": 200968350.0,
      "step": 96840
    },
    {
      "entropy": 5.660201692581177,
      "epoch": 8.784923802612482,
      "grad_norm": 1.4921875,
      "learning_rate": 6.647020509229032e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.24435027837753295,
      "num_tokens": 200978336.0,
      "step": 96845
    },
    {
      "entropy": 5.6667886734008786,
      "epoch": 8.785377358490566,
      "grad_norm": 1.40625,
      "learning_rate": 6.6458055551832e-05,
      "loss": 4.7296,
      "mean_token_accuracy": 0.2458815574645996,
      "num_tokens": 200988616.0,
      "step": 96850
    },
    {
      "entropy": 5.681587982177734,
      "epoch": 8.78583091436865,
      "grad_norm": 1.296875,
      "learning_rate": 6.644591032407417e-05,
      "loss": 4.8473,
      "mean_token_accuracy": 0.23775320053100585,
      "num_tokens": 201000434.0,
      "step": 96855
    },
    {
      "entropy": 5.753593492507934,
      "epoch": 8.786284470246734,
      "grad_norm": 1.59375,
      "learning_rate": 6.643376940926813e-05,
      "loss": 4.8502,
      "mean_token_accuracy": 0.2302565649151802,
      "num_tokens": 201011157.0,
      "step": 96860
    },
    {
      "entropy": 5.684244060516358,
      "epoch": 8.78673802612482,
      "grad_norm": 1.640625,
      "learning_rate": 6.642163280766486e-05,
      "loss": 4.7641,
      "mean_token_accuracy": 0.24008131474256517,
      "num_tokens": 201020444.0,
      "step": 96865
    },
    {
      "entropy": 5.580636930465698,
      "epoch": 8.787191582002903,
      "grad_norm": 1.6640625,
      "learning_rate": 6.64095005195154e-05,
      "loss": 4.7036,
      "mean_token_accuracy": 0.2394087165594101,
      "num_tokens": 201030819.0,
      "step": 96870
    },
    {
      "entropy": 5.698754930496216,
      "epoch": 8.787645137880986,
      "grad_norm": 1.53125,
      "learning_rate": 6.63973725450706e-05,
      "loss": 4.8117,
      "mean_token_accuracy": 0.23579001426696777,
      "num_tokens": 201041204.0,
      "step": 96875
    },
    {
      "entropy": 5.741617870330811,
      "epoch": 8.788098693759071,
      "grad_norm": 1.4765625,
      "learning_rate": 6.63852488845813e-05,
      "loss": 4.949,
      "mean_token_accuracy": 0.22584914416074753,
      "num_tokens": 201051706.0,
      "step": 96880
    },
    {
      "entropy": 5.658806037902832,
      "epoch": 8.788552249637155,
      "grad_norm": 1.53125,
      "learning_rate": 6.637312953829824e-05,
      "loss": 4.7573,
      "mean_token_accuracy": 0.24173886179924012,
      "num_tokens": 201062138.0,
      "step": 96885
    },
    {
      "entropy": 5.646897268295288,
      "epoch": 8.78900580551524,
      "grad_norm": 1.390625,
      "learning_rate": 6.636101450647198e-05,
      "loss": 4.733,
      "mean_token_accuracy": 0.23880858421325685,
      "num_tokens": 201072847.0,
      "step": 96890
    },
    {
      "entropy": 5.685939598083496,
      "epoch": 8.789459361393323,
      "grad_norm": 1.4921875,
      "learning_rate": 6.634890378935313e-05,
      "loss": 4.8,
      "mean_token_accuracy": 0.24006254822015763,
      "num_tokens": 201082072.0,
      "step": 96895
    },
    {
      "entropy": 5.699086761474609,
      "epoch": 8.789912917271408,
      "grad_norm": 1.6875,
      "learning_rate": 6.633679738719212e-05,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23700127601623536,
      "num_tokens": 201092880.0,
      "step": 96900
    },
    {
      "entropy": 5.717922019958496,
      "epoch": 8.790366473149492,
      "grad_norm": 1.671875,
      "learning_rate": 6.632469530023932e-05,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22737907022237777,
      "num_tokens": 201102693.0,
      "step": 96905
    },
    {
      "entropy": 5.7436034202575685,
      "epoch": 8.790820029027577,
      "grad_norm": 1.671875,
      "learning_rate": 6.6312597528745e-05,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2359866291284561,
      "num_tokens": 201112422.0,
      "step": 96910
    },
    {
      "entropy": 5.788537693023682,
      "epoch": 8.79127358490566,
      "grad_norm": 1.5,
      "learning_rate": 6.630050407295932e-05,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22599430680274962,
      "num_tokens": 201122491.0,
      "step": 96915
    },
    {
      "entropy": 5.640388774871826,
      "epoch": 8.791727140783745,
      "grad_norm": 1.546875,
      "learning_rate": 6.628841493313242e-05,
      "loss": 4.7267,
      "mean_token_accuracy": 0.24243979156017303,
      "num_tokens": 201132518.0,
      "step": 96920
    },
    {
      "entropy": 5.655674314498901,
      "epoch": 8.792180696661829,
      "grad_norm": 1.5234375,
      "learning_rate": 6.627633010951424e-05,
      "loss": 4.7677,
      "mean_token_accuracy": 0.23835130780935287,
      "num_tokens": 201143166.0,
      "step": 96925
    },
    {
      "entropy": 5.6943995475769045,
      "epoch": 8.792634252539912,
      "grad_norm": 1.546875,
      "learning_rate": 6.62642496023548e-05,
      "loss": 4.8048,
      "mean_token_accuracy": 0.23505332320928574,
      "num_tokens": 201153786.0,
      "step": 96930
    },
    {
      "entropy": 5.676356935501099,
      "epoch": 8.793087808417997,
      "grad_norm": 1.609375,
      "learning_rate": 6.625217341190384e-05,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23128860145807267,
      "num_tokens": 201164192.0,
      "step": 96935
    },
    {
      "entropy": 5.672725105285645,
      "epoch": 8.79354136429608,
      "grad_norm": 1.53125,
      "learning_rate": 6.624010153841116e-05,
      "loss": 4.7459,
      "mean_token_accuracy": 0.2453038603067398,
      "num_tokens": 201174395.0,
      "step": 96940
    },
    {
      "entropy": 5.7298369884490965,
      "epoch": 8.793994920174166,
      "grad_norm": 1.59375,
      "learning_rate": 6.622803398212637e-05,
      "loss": 4.9122,
      "mean_token_accuracy": 0.22682723253965378,
      "num_tokens": 201184233.0,
      "step": 96945
    },
    {
      "entropy": 5.654462623596191,
      "epoch": 8.79444847605225,
      "grad_norm": 1.484375,
      "learning_rate": 6.621597074329902e-05,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23732666671276093,
      "num_tokens": 201194534.0,
      "step": 96950
    },
    {
      "entropy": 5.708859872817993,
      "epoch": 8.794902031930334,
      "grad_norm": 1.5234375,
      "learning_rate": 6.620391182217861e-05,
      "loss": 4.8019,
      "mean_token_accuracy": 0.23580418527126312,
      "num_tokens": 201205397.0,
      "step": 96955
    },
    {
      "entropy": 5.662790536880493,
      "epoch": 8.795355587808418,
      "grad_norm": 1.6015625,
      "learning_rate": 6.619185721901448e-05,
      "loss": 4.6957,
      "mean_token_accuracy": 0.24460689574480057,
      "num_tokens": 201215484.0,
      "step": 96960
    },
    {
      "entropy": 5.628923797607422,
      "epoch": 8.795809143686503,
      "grad_norm": 1.6875,
      "learning_rate": 6.617980693405604e-05,
      "loss": 4.7196,
      "mean_token_accuracy": 0.2382511854171753,
      "num_tokens": 201224615.0,
      "step": 96965
    },
    {
      "entropy": 5.7255393981933596,
      "epoch": 8.796262699564586,
      "grad_norm": 1.546875,
      "learning_rate": 6.616776096755232e-05,
      "loss": 4.7944,
      "mean_token_accuracy": 0.24688465148210526,
      "num_tokens": 201235320.0,
      "step": 96970
    },
    {
      "entropy": 5.657316398620606,
      "epoch": 8.796716255442671,
      "grad_norm": 1.515625,
      "learning_rate": 6.615571931975256e-05,
      "loss": 4.8188,
      "mean_token_accuracy": 0.23429181426763535,
      "num_tokens": 201246202.0,
      "step": 96975
    },
    {
      "entropy": 5.669037866592407,
      "epoch": 8.797169811320755,
      "grad_norm": 1.6328125,
      "learning_rate": 6.614368199090575e-05,
      "loss": 4.8128,
      "mean_token_accuracy": 0.24134217351675033,
      "num_tokens": 201256076.0,
      "step": 96980
    },
    {
      "entropy": 5.687069129943848,
      "epoch": 8.797623367198838,
      "grad_norm": 1.6875,
      "learning_rate": 6.613164898126077e-05,
      "loss": 4.7874,
      "mean_token_accuracy": 0.2405286580324173,
      "num_tokens": 201265182.0,
      "step": 96985
    },
    {
      "entropy": 5.705921173095703,
      "epoch": 8.798076923076923,
      "grad_norm": 1.4921875,
      "learning_rate": 6.611962029106655e-05,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2374078243970871,
      "num_tokens": 201275698.0,
      "step": 96990
    },
    {
      "entropy": 5.693662548065186,
      "epoch": 8.798530478955007,
      "grad_norm": 1.5,
      "learning_rate": 6.610759592057181e-05,
      "loss": 4.8341,
      "mean_token_accuracy": 0.23899353444576263,
      "num_tokens": 201286214.0,
      "step": 96995
    },
    {
      "entropy": 5.65715651512146,
      "epoch": 8.798984034833092,
      "grad_norm": 1.609375,
      "learning_rate": 6.609557587002522e-05,
      "loss": 4.7919,
      "mean_token_accuracy": 0.2313505232334137,
      "num_tokens": 201297418.0,
      "step": 97000
    },
    {
      "entropy": 5.682501888275146,
      "epoch": 8.799437590711175,
      "grad_norm": 1.4140625,
      "learning_rate": 6.608356013967534e-05,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2406748652458191,
      "num_tokens": 201308045.0,
      "step": 97005
    },
    {
      "entropy": 5.679693365097046,
      "epoch": 8.79989114658926,
      "grad_norm": 1.6015625,
      "learning_rate": 6.60715487297707e-05,
      "loss": 4.8068,
      "mean_token_accuracy": 0.23881423473358154,
      "num_tokens": 201318706.0,
      "step": 97010
    },
    {
      "entropy": 5.638021850585938,
      "epoch": 8.800344702467344,
      "grad_norm": 1.4765625,
      "learning_rate": 6.605954164055967e-05,
      "loss": 4.7717,
      "mean_token_accuracy": 0.2409937560558319,
      "num_tokens": 201329904.0,
      "step": 97015
    },
    {
      "entropy": 5.639017248153687,
      "epoch": 8.800798258345429,
      "grad_norm": 1.4296875,
      "learning_rate": 6.604753887229054e-05,
      "loss": 4.7279,
      "mean_token_accuracy": 0.2440046936273575,
      "num_tokens": 201340320.0,
      "step": 97020
    },
    {
      "entropy": 5.559109449386597,
      "epoch": 8.801251814223512,
      "grad_norm": 1.5,
      "learning_rate": 6.603554042521154e-05,
      "loss": 4.6128,
      "mean_token_accuracy": 0.25584068298339846,
      "num_tokens": 201351593.0,
      "step": 97025
    },
    {
      "entropy": 5.6879112243652346,
      "epoch": 8.801705370101596,
      "grad_norm": 1.46875,
      "learning_rate": 6.602354629957086e-05,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23776120096445083,
      "num_tokens": 201363070.0,
      "step": 97030
    },
    {
      "entropy": 5.72509388923645,
      "epoch": 8.80215892597968,
      "grad_norm": 1.578125,
      "learning_rate": 6.601155649561649e-05,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2371584579348564,
      "num_tokens": 201373467.0,
      "step": 97035
    },
    {
      "entropy": 5.622357320785523,
      "epoch": 8.802612481857764,
      "grad_norm": 1.53125,
      "learning_rate": 6.599957101359638e-05,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23180890679359437,
      "num_tokens": 201383914.0,
      "step": 97040
    },
    {
      "entropy": 5.731077003479004,
      "epoch": 8.80306603773585,
      "grad_norm": 1.390625,
      "learning_rate": 6.598758985375841e-05,
      "loss": 4.812,
      "mean_token_accuracy": 0.2401813641190529,
      "num_tokens": 201394453.0,
      "step": 97045
    },
    {
      "entropy": 5.731385517120361,
      "epoch": 8.803519593613933,
      "grad_norm": 1.5,
      "learning_rate": 6.597561301635033e-05,
      "loss": 4.8768,
      "mean_token_accuracy": 0.23553863763809205,
      "num_tokens": 201405896.0,
      "step": 97050
    },
    {
      "entropy": 5.6880615234375,
      "epoch": 8.803973149492018,
      "grad_norm": 1.609375,
      "learning_rate": 6.59636405016198e-05,
      "loss": 4.7411,
      "mean_token_accuracy": 0.25054390728473663,
      "num_tokens": 201416440.0,
      "step": 97055
    },
    {
      "entropy": 5.708688116073608,
      "epoch": 8.804426705370101,
      "grad_norm": 1.484375,
      "learning_rate": 6.595167230981452e-05,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23624253869056702,
      "num_tokens": 201426260.0,
      "step": 97060
    },
    {
      "entropy": 5.705337905883789,
      "epoch": 8.804880261248186,
      "grad_norm": 1.46875,
      "learning_rate": 6.593970844118189e-05,
      "loss": 4.7871,
      "mean_token_accuracy": 0.22985775768756866,
      "num_tokens": 201435571.0,
      "step": 97065
    },
    {
      "entropy": 5.65234956741333,
      "epoch": 8.80533381712627,
      "grad_norm": 1.4921875,
      "learning_rate": 6.592774889596939e-05,
      "loss": 4.7995,
      "mean_token_accuracy": 0.23281140327453614,
      "num_tokens": 201445951.0,
      "step": 97070
    },
    {
      "entropy": 5.720153379440307,
      "epoch": 8.805787373004353,
      "grad_norm": 1.53125,
      "learning_rate": 6.591579367442431e-05,
      "loss": 4.8319,
      "mean_token_accuracy": 0.2321472406387329,
      "num_tokens": 201455497.0,
      "step": 97075
    },
    {
      "entropy": 5.690278816223144,
      "epoch": 8.806240928882438,
      "grad_norm": 1.5390625,
      "learning_rate": 6.59038427767939e-05,
      "loss": 4.8306,
      "mean_token_accuracy": 0.23360862880945205,
      "num_tokens": 201465794.0,
      "step": 97080
    },
    {
      "entropy": 5.718770742416382,
      "epoch": 8.806694484760522,
      "grad_norm": 1.4140625,
      "learning_rate": 6.589189620332529e-05,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2308204412460327,
      "num_tokens": 201475726.0,
      "step": 97085
    },
    {
      "entropy": 5.739223051071167,
      "epoch": 8.807148040638607,
      "grad_norm": 1.65625,
      "learning_rate": 6.587995395426555e-05,
      "loss": 4.8534,
      "mean_token_accuracy": 0.23543656915426253,
      "num_tokens": 201485148.0,
      "step": 97090
    },
    {
      "entropy": 5.674309349060058,
      "epoch": 8.80760159651669,
      "grad_norm": 1.4921875,
      "learning_rate": 6.586801602986166e-05,
      "loss": 4.7768,
      "mean_token_accuracy": 0.23629170209169387,
      "num_tokens": 201496032.0,
      "step": 97095
    },
    {
      "entropy": 5.656707715988159,
      "epoch": 8.808055152394775,
      "grad_norm": 1.640625,
      "learning_rate": 6.585608243036051e-05,
      "loss": 4.7822,
      "mean_token_accuracy": 0.2359201893210411,
      "num_tokens": 201506331.0,
      "step": 97100
    },
    {
      "entropy": 5.658959865570068,
      "epoch": 8.808508708272859,
      "grad_norm": 1.515625,
      "learning_rate": 6.584415315600886e-05,
      "loss": 4.7856,
      "mean_token_accuracy": 0.24321311414241792,
      "num_tokens": 201516945.0,
      "step": 97105
    },
    {
      "entropy": 5.690653371810913,
      "epoch": 8.808962264150944,
      "grad_norm": 1.6015625,
      "learning_rate": 6.583222820705344e-05,
      "loss": 4.8214,
      "mean_token_accuracy": 0.23610632866621017,
      "num_tokens": 201527656.0,
      "step": 97110
    },
    {
      "entropy": 5.665946578979492,
      "epoch": 8.809415820029027,
      "grad_norm": 1.671875,
      "learning_rate": 6.58203075837408e-05,
      "loss": 4.7765,
      "mean_token_accuracy": 0.2422817602753639,
      "num_tokens": 201537982.0,
      "step": 97115
    },
    {
      "entropy": 5.707540035247803,
      "epoch": 8.809869375907112,
      "grad_norm": 1.46875,
      "learning_rate": 6.580839128631755e-05,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2396424412727356,
      "num_tokens": 201548356.0,
      "step": 97120
    },
    {
      "entropy": 5.719807052612305,
      "epoch": 8.810322931785196,
      "grad_norm": 1.78125,
      "learning_rate": 6.579647931503001e-05,
      "loss": 4.8354,
      "mean_token_accuracy": 0.2358831897377968,
      "num_tokens": 201559033.0,
      "step": 97125
    },
    {
      "entropy": 5.708458709716797,
      "epoch": 8.81077648766328,
      "grad_norm": 1.3984375,
      "learning_rate": 6.578457167012466e-05,
      "loss": 4.7587,
      "mean_token_accuracy": 0.23986868113279342,
      "num_tokens": 201569324.0,
      "step": 97130
    },
    {
      "entropy": 5.6987205982208256,
      "epoch": 8.811230043541364,
      "grad_norm": 1.5625,
      "learning_rate": 6.577266835184765e-05,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23711021691560746,
      "num_tokens": 201580058.0,
      "step": 97135
    },
    {
      "entropy": 5.648358869552612,
      "epoch": 8.811683599419448,
      "grad_norm": 1.3671875,
      "learning_rate": 6.576076936044519e-05,
      "loss": 4.7374,
      "mean_token_accuracy": 0.24831056147813796,
      "num_tokens": 201591845.0,
      "step": 97140
    },
    {
      "entropy": 5.729258108139038,
      "epoch": 8.812137155297533,
      "grad_norm": 1.4921875,
      "learning_rate": 6.574887469616334e-05,
      "loss": 4.8502,
      "mean_token_accuracy": 0.23874690383672714,
      "num_tokens": 201602564.0,
      "step": 97145
    },
    {
      "entropy": 5.630048608779907,
      "epoch": 8.812590711175616,
      "grad_norm": 1.59375,
      "learning_rate": 6.57369843592481e-05,
      "loss": 4.7886,
      "mean_token_accuracy": 0.23501349836587906,
      "num_tokens": 201613080.0,
      "step": 97150
    },
    {
      "entropy": 5.7224196434021,
      "epoch": 8.813044267053701,
      "grad_norm": 1.59375,
      "learning_rate": 6.572509834994529e-05,
      "loss": 4.8319,
      "mean_token_accuracy": 0.23577648997306824,
      "num_tokens": 201622791.0,
      "step": 97155
    },
    {
      "entropy": 5.684481763839722,
      "epoch": 8.813497822931785,
      "grad_norm": 1.609375,
      "learning_rate": 6.571321666850082e-05,
      "loss": 4.7785,
      "mean_token_accuracy": 0.23976372331380844,
      "num_tokens": 201634293.0,
      "step": 97160
    },
    {
      "entropy": 5.765559530258178,
      "epoch": 8.81395137880987,
      "grad_norm": 1.5703125,
      "learning_rate": 6.570133931516035e-05,
      "loss": 4.879,
      "mean_token_accuracy": 0.2274692952632904,
      "num_tokens": 201644281.0,
      "step": 97165
    },
    {
      "entropy": 5.653521490097046,
      "epoch": 8.814404934687953,
      "grad_norm": 1.5234375,
      "learning_rate": 6.568946629016952e-05,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23669057935476304,
      "num_tokens": 201655144.0,
      "step": 97170
    },
    {
      "entropy": 5.665448236465454,
      "epoch": 8.814858490566039,
      "grad_norm": 1.59375,
      "learning_rate": 6.567759759377388e-05,
      "loss": 4.7882,
      "mean_token_accuracy": 0.2406125158071518,
      "num_tokens": 201664745.0,
      "step": 97175
    },
    {
      "entropy": 5.69126615524292,
      "epoch": 8.815312046444122,
      "grad_norm": 1.421875,
      "learning_rate": 6.566573322621885e-05,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23664076030254363,
      "num_tokens": 201675648.0,
      "step": 97180
    },
    {
      "entropy": 5.678478765487671,
      "epoch": 8.815765602322205,
      "grad_norm": 1.53125,
      "learning_rate": 6.565387318774982e-05,
      "loss": 4.7491,
      "mean_token_accuracy": 0.2357524186372757,
      "num_tokens": 201685134.0,
      "step": 97185
    },
    {
      "entropy": 5.676448678970337,
      "epoch": 8.81621915820029,
      "grad_norm": 1.5234375,
      "learning_rate": 6.5642017478612e-05,
      "loss": 4.7535,
      "mean_token_accuracy": 0.23330202847719192,
      "num_tokens": 201695839.0,
      "step": 97190
    },
    {
      "entropy": 5.745116710662842,
      "epoch": 8.816672714078374,
      "grad_norm": 1.5234375,
      "learning_rate": 6.563016609905064e-05,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23658130317926407,
      "num_tokens": 201706567.0,
      "step": 97195
    },
    {
      "entropy": 5.689292049407959,
      "epoch": 8.817126269956459,
      "grad_norm": 1.453125,
      "learning_rate": 6.56183190493108e-05,
      "loss": 4.8017,
      "mean_token_accuracy": 0.23148776292800904,
      "num_tokens": 201717618.0,
      "step": 97200
    },
    {
      "entropy": 5.711903142929077,
      "epoch": 8.817579825834542,
      "grad_norm": 1.3828125,
      "learning_rate": 6.560647632963748e-05,
      "loss": 4.799,
      "mean_token_accuracy": 0.24072165191173553,
      "num_tokens": 201728862.0,
      "step": 97205
    },
    {
      "entropy": 5.644222640991211,
      "epoch": 8.818033381712628,
      "grad_norm": 1.4921875,
      "learning_rate": 6.559463794027558e-05,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2378619909286499,
      "num_tokens": 201739190.0,
      "step": 97210
    },
    {
      "entropy": 5.629842376708984,
      "epoch": 8.818486937590711,
      "grad_norm": 1.609375,
      "learning_rate": 6.558280388146993e-05,
      "loss": 4.7402,
      "mean_token_accuracy": 0.2420770436525345,
      "num_tokens": 201749892.0,
      "step": 97215
    },
    {
      "entropy": 5.588505601882934,
      "epoch": 8.818940493468796,
      "grad_norm": 1.5390625,
      "learning_rate": 6.557097415346529e-05,
      "loss": 4.7365,
      "mean_token_accuracy": 0.23522330522537233,
      "num_tokens": 201759763.0,
      "step": 97220
    },
    {
      "entropy": 5.623520946502685,
      "epoch": 8.81939404934688,
      "grad_norm": 1.5078125,
      "learning_rate": 6.55591487565062e-05,
      "loss": 4.7279,
      "mean_token_accuracy": 0.24070168286561966,
      "num_tokens": 201771302.0,
      "step": 97225
    },
    {
      "entropy": 5.678734397888183,
      "epoch": 8.819847605224965,
      "grad_norm": 1.5625,
      "learning_rate": 6.554732769083732e-05,
      "loss": 4.8147,
      "mean_token_accuracy": 0.24287722110748292,
      "num_tokens": 201782335.0,
      "step": 97230
    },
    {
      "entropy": 5.696629047393799,
      "epoch": 8.820301161103048,
      "grad_norm": 1.65625,
      "learning_rate": 6.553551095670311e-05,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23926049768924712,
      "num_tokens": 201792715.0,
      "step": 97235
    },
    {
      "entropy": 5.6252950668334964,
      "epoch": 8.820754716981131,
      "grad_norm": 1.453125,
      "learning_rate": 6.552369855434789e-05,
      "loss": 4.8054,
      "mean_token_accuracy": 0.2355431377887726,
      "num_tokens": 201803456.0,
      "step": 97240
    },
    {
      "entropy": 5.688415002822876,
      "epoch": 8.821208272859216,
      "grad_norm": 1.6796875,
      "learning_rate": 6.551189048401597e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.2373250886797905,
      "num_tokens": 201813497.0,
      "step": 97245
    },
    {
      "entropy": 5.717879724502564,
      "epoch": 8.8216618287373,
      "grad_norm": 1.5390625,
      "learning_rate": 6.550008674595155e-05,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23979779928922654,
      "num_tokens": 201824690.0,
      "step": 97250
    },
    {
      "entropy": 5.6372314453125,
      "epoch": 8.822115384615385,
      "grad_norm": 1.421875,
      "learning_rate": 6.548828734039873e-05,
      "loss": 4.7481,
      "mean_token_accuracy": 0.24021052420139313,
      "num_tokens": 201835365.0,
      "step": 97255
    },
    {
      "entropy": 5.690332269668579,
      "epoch": 8.822568940493468,
      "grad_norm": 1.4296875,
      "learning_rate": 6.547649226760146e-05,
      "loss": 4.7484,
      "mean_token_accuracy": 0.24257175028324127,
      "num_tokens": 201846220.0,
      "step": 97260
    },
    {
      "entropy": 5.706616830825806,
      "epoch": 8.823022496371554,
      "grad_norm": 1.515625,
      "learning_rate": 6.546470152780377e-05,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2389391764998436,
      "num_tokens": 201857461.0,
      "step": 97265
    },
    {
      "entropy": 5.6455528259277346,
      "epoch": 8.823476052249637,
      "grad_norm": 1.515625,
      "learning_rate": 6.545291512124942e-05,
      "loss": 4.7172,
      "mean_token_accuracy": 0.2463550314307213,
      "num_tokens": 201867891.0,
      "step": 97270
    },
    {
      "entropy": 5.697177982330322,
      "epoch": 8.823929608127722,
      "grad_norm": 1.4921875,
      "learning_rate": 6.544113304818221e-05,
      "loss": 4.6703,
      "mean_token_accuracy": 0.24889078587293625,
      "num_tokens": 201877805.0,
      "step": 97275
    },
    {
      "entropy": 5.628636598587036,
      "epoch": 8.824383164005805,
      "grad_norm": 1.5859375,
      "learning_rate": 6.542935530884577e-05,
      "loss": 4.7245,
      "mean_token_accuracy": 0.24289537966251373,
      "num_tokens": 201888416.0,
      "step": 97280
    },
    {
      "entropy": 5.674315929412842,
      "epoch": 8.82483671988389,
      "grad_norm": 1.6640625,
      "learning_rate": 6.541758190348365e-05,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23686033487319946,
      "num_tokens": 201898468.0,
      "step": 97285
    },
    {
      "entropy": 5.707389879226684,
      "epoch": 8.825290275761974,
      "grad_norm": 1.46875,
      "learning_rate": 6.540581283233937e-05,
      "loss": 4.8236,
      "mean_token_accuracy": 0.23485027253627777,
      "num_tokens": 201909164.0,
      "step": 97290
    },
    {
      "entropy": 5.631258535385132,
      "epoch": 8.825743831640057,
      "grad_norm": 1.546875,
      "learning_rate": 6.539404809565625e-05,
      "loss": 4.7457,
      "mean_token_accuracy": 0.24283572137355805,
      "num_tokens": 201919224.0,
      "step": 97295
    },
    {
      "entropy": 5.685454988479615,
      "epoch": 8.826197387518143,
      "grad_norm": 1.5,
      "learning_rate": 6.538228769367766e-05,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23920728415250778,
      "num_tokens": 201929389.0,
      "step": 97300
    },
    {
      "entropy": 5.681279516220092,
      "epoch": 8.826650943396226,
      "grad_norm": 1.5859375,
      "learning_rate": 6.537053162664679e-05,
      "loss": 4.8464,
      "mean_token_accuracy": 0.2367883339524269,
      "num_tokens": 201939308.0,
      "step": 97305
    },
    {
      "entropy": 5.770686483383178,
      "epoch": 8.827104499274311,
      "grad_norm": 1.5,
      "learning_rate": 6.535877989480671e-05,
      "loss": 4.8552,
      "mean_token_accuracy": 0.2368529662489891,
      "num_tokens": 201949644.0,
      "step": 97310
    },
    {
      "entropy": 5.665755748748779,
      "epoch": 8.827558055152394,
      "grad_norm": 1.515625,
      "learning_rate": 6.534703249840052e-05,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23654358237981796,
      "num_tokens": 201960746.0,
      "step": 97315
    },
    {
      "entropy": 5.690921020507813,
      "epoch": 8.82801161103048,
      "grad_norm": 1.515625,
      "learning_rate": 6.53352894376711e-05,
      "loss": 4.7899,
      "mean_token_accuracy": 0.23784760683774947,
      "num_tokens": 201970850.0,
      "step": 97320
    },
    {
      "entropy": 5.604172801971435,
      "epoch": 8.828465166908563,
      "grad_norm": 1.421875,
      "learning_rate": 6.532355071286134e-05,
      "loss": 4.7508,
      "mean_token_accuracy": 0.23504789173603058,
      "num_tokens": 201981706.0,
      "step": 97325
    },
    {
      "entropy": 5.728294038772583,
      "epoch": 8.828918722786648,
      "grad_norm": 1.453125,
      "learning_rate": 6.531181632421394e-05,
      "loss": 4.8056,
      "mean_token_accuracy": 0.23706458061933516,
      "num_tokens": 201991700.0,
      "step": 97330
    },
    {
      "entropy": 5.68754825592041,
      "epoch": 8.829372278664732,
      "grad_norm": 1.640625,
      "learning_rate": 6.530008627197166e-05,
      "loss": 4.7814,
      "mean_token_accuracy": 0.23788429498672486,
      "num_tokens": 202001215.0,
      "step": 97335
    },
    {
      "entropy": 5.646896934509277,
      "epoch": 8.829825834542815,
      "grad_norm": 1.5390625,
      "learning_rate": 6.528836055637702e-05,
      "loss": 4.7762,
      "mean_token_accuracy": 0.2411225140094757,
      "num_tokens": 202010615.0,
      "step": 97340
    },
    {
      "entropy": 5.654959058761596,
      "epoch": 8.8302793904209,
      "grad_norm": 1.6484375,
      "learning_rate": 6.527663917767253e-05,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23725188076496123,
      "num_tokens": 202022136.0,
      "step": 97345
    },
    {
      "entropy": 5.72183780670166,
      "epoch": 8.830732946298983,
      "grad_norm": 1.796875,
      "learning_rate": 6.526492213610058e-05,
      "loss": 4.7884,
      "mean_token_accuracy": 0.23236973583698273,
      "num_tokens": 202031686.0,
      "step": 97350
    },
    {
      "entropy": 5.675318193435669,
      "epoch": 8.831186502177069,
      "grad_norm": 1.546875,
      "learning_rate": 6.525320943190351e-05,
      "loss": 4.8258,
      "mean_token_accuracy": 0.23112691938877106,
      "num_tokens": 202041765.0,
      "step": 97355
    },
    {
      "entropy": 5.709316635131836,
      "epoch": 8.831640058055152,
      "grad_norm": 1.5546875,
      "learning_rate": 6.524150106532346e-05,
      "loss": 4.838,
      "mean_token_accuracy": 0.24007219523191453,
      "num_tokens": 202051730.0,
      "step": 97360
    },
    {
      "entropy": 5.723689842224121,
      "epoch": 8.832093613933237,
      "grad_norm": 1.4765625,
      "learning_rate": 6.522979703660265e-05,
      "loss": 4.8841,
      "mean_token_accuracy": 0.23579794615507127,
      "num_tokens": 202062559.0,
      "step": 97365
    },
    {
      "entropy": 5.77278733253479,
      "epoch": 8.83254716981132,
      "grad_norm": 1.6015625,
      "learning_rate": 6.52180973459831e-05,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22836913019418717,
      "num_tokens": 202073044.0,
      "step": 97370
    },
    {
      "entropy": 5.653805446624756,
      "epoch": 8.833000725689406,
      "grad_norm": 1.5859375,
      "learning_rate": 6.520640199370677e-05,
      "loss": 4.7695,
      "mean_token_accuracy": 0.24506785869598388,
      "num_tokens": 202082311.0,
      "step": 97375
    },
    {
      "entropy": 5.655465412139892,
      "epoch": 8.833454281567489,
      "grad_norm": 1.375,
      "learning_rate": 6.51947109800155e-05,
      "loss": 4.749,
      "mean_token_accuracy": 0.2404848501086235,
      "num_tokens": 202092113.0,
      "step": 97380
    },
    {
      "entropy": 5.74980936050415,
      "epoch": 8.833907837445572,
      "grad_norm": 1.4609375,
      "learning_rate": 6.518302430515106e-05,
      "loss": 4.8588,
      "mean_token_accuracy": 0.23316131085157393,
      "num_tokens": 202101790.0,
      "step": 97385
    },
    {
      "entropy": 5.692536020278931,
      "epoch": 8.834361393323658,
      "grad_norm": 1.4453125,
      "learning_rate": 6.517134196935518e-05,
      "loss": 4.7766,
      "mean_token_accuracy": 0.24107228964567184,
      "num_tokens": 202111608.0,
      "step": 97390
    },
    {
      "entropy": 5.679561614990234,
      "epoch": 8.834814949201741,
      "grad_norm": 1.515625,
      "learning_rate": 6.515966397286938e-05,
      "loss": 4.7448,
      "mean_token_accuracy": 0.2394416004419327,
      "num_tokens": 202121491.0,
      "step": 97395
    },
    {
      "entropy": 5.642597103118897,
      "epoch": 8.835268505079826,
      "grad_norm": 1.6171875,
      "learning_rate": 6.514799031593523e-05,
      "loss": 4.7476,
      "mean_token_accuracy": 0.24048522114753723,
      "num_tokens": 202131763.0,
      "step": 97400
    },
    {
      "entropy": 5.6675543785095215,
      "epoch": 8.83572206095791,
      "grad_norm": 1.46875,
      "learning_rate": 6.513632099879408e-05,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2445211261510849,
      "num_tokens": 202141363.0,
      "step": 97405
    },
    {
      "entropy": 5.631484460830689,
      "epoch": 8.836175616835995,
      "grad_norm": 1.4921875,
      "learning_rate": 6.512465602168734e-05,
      "loss": 4.7373,
      "mean_token_accuracy": 0.24479094743728638,
      "num_tokens": 202150588.0,
      "step": 97410
    },
    {
      "entropy": 5.7261463642120365,
      "epoch": 8.836629172714078,
      "grad_norm": 1.578125,
      "learning_rate": 6.511299538485617e-05,
      "loss": 4.861,
      "mean_token_accuracy": 0.2308421403169632,
      "num_tokens": 202161009.0,
      "step": 97415
    },
    {
      "entropy": 5.645843648910523,
      "epoch": 8.837082728592163,
      "grad_norm": 1.6328125,
      "learning_rate": 6.510133908854174e-05,
      "loss": 4.7607,
      "mean_token_accuracy": 0.23326776921749115,
      "num_tokens": 202170083.0,
      "step": 97420
    },
    {
      "entropy": 5.696034145355225,
      "epoch": 8.837536284470247,
      "grad_norm": 1.546875,
      "learning_rate": 6.508968713298507e-05,
      "loss": 4.745,
      "mean_token_accuracy": 0.2550994291901588,
      "num_tokens": 202179984.0,
      "step": 97425
    },
    {
      "entropy": 5.669994354248047,
      "epoch": 8.837989840348332,
      "grad_norm": 1.3984375,
      "learning_rate": 6.507803951842719e-05,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2334341287612915,
      "num_tokens": 202190019.0,
      "step": 97430
    },
    {
      "entropy": 5.672575044631958,
      "epoch": 8.838443396226415,
      "grad_norm": 1.46875,
      "learning_rate": 6.506639624510897e-05,
      "loss": 4.7711,
      "mean_token_accuracy": 0.23973947763442993,
      "num_tokens": 202200360.0,
      "step": 97435
    },
    {
      "entropy": 5.688237857818604,
      "epoch": 8.838896952104498,
      "grad_norm": 1.5546875,
      "learning_rate": 6.505475731327115e-05,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2406380742788315,
      "num_tokens": 202212295.0,
      "step": 97440
    },
    {
      "entropy": 5.716217756271362,
      "epoch": 8.839350507982584,
      "grad_norm": 1.640625,
      "learning_rate": 6.504312272315448e-05,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23436512500047685,
      "num_tokens": 202222119.0,
      "step": 97445
    },
    {
      "entropy": 5.637020921707153,
      "epoch": 8.839804063860667,
      "grad_norm": 1.5234375,
      "learning_rate": 6.503149247499951e-05,
      "loss": 4.7365,
      "mean_token_accuracy": 0.24357970356941222,
      "num_tokens": 202231609.0,
      "step": 97450
    },
    {
      "entropy": 5.699280452728272,
      "epoch": 8.840257619738752,
      "grad_norm": 1.46875,
      "learning_rate": 6.501986656904675e-05,
      "loss": 4.8061,
      "mean_token_accuracy": 0.2371782049536705,
      "num_tokens": 202241700.0,
      "step": 97455
    },
    {
      "entropy": 5.708913135528564,
      "epoch": 8.840711175616836,
      "grad_norm": 1.6015625,
      "learning_rate": 6.500824500553669e-05,
      "loss": 4.8171,
      "mean_token_accuracy": 0.23675849735736848,
      "num_tokens": 202252112.0,
      "step": 97460
    },
    {
      "entropy": 5.617816400527954,
      "epoch": 8.84116473149492,
      "grad_norm": 1.5546875,
      "learning_rate": 6.499662778470965e-05,
      "loss": 4.6738,
      "mean_token_accuracy": 0.2448143482208252,
      "num_tokens": 202262655.0,
      "step": 97465
    },
    {
      "entropy": 5.70119800567627,
      "epoch": 8.841618287373004,
      "grad_norm": 1.4375,
      "learning_rate": 6.498501490680585e-05,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23118699043989183,
      "num_tokens": 202273475.0,
      "step": 97470
    },
    {
      "entropy": 5.670724773406983,
      "epoch": 8.84207184325109,
      "grad_norm": 1.390625,
      "learning_rate": 6.497340637206548e-05,
      "loss": 4.7687,
      "mean_token_accuracy": 0.23774069994688035,
      "num_tokens": 202283464.0,
      "step": 97475
    },
    {
      "entropy": 5.656654739379883,
      "epoch": 8.842525399129173,
      "grad_norm": 1.46875,
      "learning_rate": 6.496180218072856e-05,
      "loss": 4.7202,
      "mean_token_accuracy": 0.24402388483285903,
      "num_tokens": 202294281.0,
      "step": 97480
    },
    {
      "entropy": 5.673524332046509,
      "epoch": 8.842978955007258,
      "grad_norm": 1.5390625,
      "learning_rate": 6.49502023330351e-05,
      "loss": 4.7703,
      "mean_token_accuracy": 0.22868161052465438,
      "num_tokens": 202304458.0,
      "step": 97485
    },
    {
      "entropy": 5.707862997055054,
      "epoch": 8.843432510885341,
      "grad_norm": 1.5546875,
      "learning_rate": 6.493860682922497e-05,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23489751070737838,
      "num_tokens": 202315234.0,
      "step": 97490
    },
    {
      "entropy": 5.686643981933594,
      "epoch": 8.843886066763424,
      "grad_norm": 1.5390625,
      "learning_rate": 6.492701566953798e-05,
      "loss": 4.8192,
      "mean_token_accuracy": 0.2371927335858345,
      "num_tokens": 202325079.0,
      "step": 97495
    },
    {
      "entropy": 5.712758636474609,
      "epoch": 8.84433962264151,
      "grad_norm": 1.6328125,
      "learning_rate": 6.491542885421387e-05,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22979606837034225,
      "num_tokens": 202334511.0,
      "step": 97500
    },
    {
      "entropy": 5.707519292831421,
      "epoch": 8.844793178519593,
      "grad_norm": 1.546875,
      "learning_rate": 6.490384638349221e-05,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2448781058192253,
      "num_tokens": 202344160.0,
      "step": 97505
    },
    {
      "entropy": 5.662712478637696,
      "epoch": 8.845246734397678,
      "grad_norm": 1.53125,
      "learning_rate": 6.489226825761256e-05,
      "loss": 4.7641,
      "mean_token_accuracy": 0.23811529129743575,
      "num_tokens": 202354190.0,
      "step": 97510
    },
    {
      "entropy": 5.639169216156006,
      "epoch": 8.845700290275762,
      "grad_norm": 1.5546875,
      "learning_rate": 6.488069447681434e-05,
      "loss": 4.7428,
      "mean_token_accuracy": 0.24442884474992752,
      "num_tokens": 202365004.0,
      "step": 97515
    },
    {
      "entropy": 5.712614488601685,
      "epoch": 8.846153846153847,
      "grad_norm": 1.4453125,
      "learning_rate": 6.486912504133691e-05,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22781440913677214,
      "num_tokens": 202375720.0,
      "step": 97520
    },
    {
      "entropy": 5.653118038177491,
      "epoch": 8.84660740203193,
      "grad_norm": 1.5546875,
      "learning_rate": 6.485755995141947e-05,
      "loss": 4.7933,
      "mean_token_accuracy": 0.24329691082239152,
      "num_tokens": 202385331.0,
      "step": 97525
    },
    {
      "entropy": 5.655610990524292,
      "epoch": 8.847060957910015,
      "grad_norm": 1.3515625,
      "learning_rate": 6.484599920730128e-05,
      "loss": 4.7771,
      "mean_token_accuracy": 0.2394275799393654,
      "num_tokens": 202396981.0,
      "step": 97530
    },
    {
      "entropy": 5.741682052612305,
      "epoch": 8.847514513788099,
      "grad_norm": 1.4140625,
      "learning_rate": 6.483444280922137e-05,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22869793325662613,
      "num_tokens": 202408569.0,
      "step": 97535
    },
    {
      "entropy": 5.731461811065674,
      "epoch": 8.847968069666184,
      "grad_norm": 1.453125,
      "learning_rate": 6.482289075741876e-05,
      "loss": 4.8205,
      "mean_token_accuracy": 0.24251444190740584,
      "num_tokens": 202419373.0,
      "step": 97540
    },
    {
      "entropy": 5.6541100025177,
      "epoch": 8.848421625544267,
      "grad_norm": 1.453125,
      "learning_rate": 6.48113430521323e-05,
      "loss": 4.7236,
      "mean_token_accuracy": 0.24292613714933395,
      "num_tokens": 202430194.0,
      "step": 97545
    },
    {
      "entropy": 5.658222246170044,
      "epoch": 8.84887518142235,
      "grad_norm": 1.5703125,
      "learning_rate": 6.479979969360088e-05,
      "loss": 4.7437,
      "mean_token_accuracy": 0.2413261666893959,
      "num_tokens": 202440429.0,
      "step": 97550
    },
    {
      "entropy": 5.738159561157227,
      "epoch": 8.849328737300436,
      "grad_norm": 1.546875,
      "learning_rate": 6.478826068206311e-05,
      "loss": 4.8297,
      "mean_token_accuracy": 0.24196949005126953,
      "num_tokens": 202450973.0,
      "step": 97555
    },
    {
      "entropy": 5.688739347457886,
      "epoch": 8.849782293178519,
      "grad_norm": 1.5078125,
      "learning_rate": 6.477672601775766e-05,
      "loss": 4.7903,
      "mean_token_accuracy": 0.23221752047538757,
      "num_tokens": 202461410.0,
      "step": 97560
    },
    {
      "entropy": 5.692725372314453,
      "epoch": 8.850235849056604,
      "grad_norm": 1.4375,
      "learning_rate": 6.476519570092315e-05,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23267467319965363,
      "num_tokens": 202472661.0,
      "step": 97565
    },
    {
      "entropy": 5.717580127716064,
      "epoch": 8.850689404934688,
      "grad_norm": 1.5234375,
      "learning_rate": 6.475366973179796e-05,
      "loss": 4.8733,
      "mean_token_accuracy": 0.23037417531013488,
      "num_tokens": 202483784.0,
      "step": 97570
    },
    {
      "entropy": 5.683550500869751,
      "epoch": 8.851142960812773,
      "grad_norm": 1.5234375,
      "learning_rate": 6.474214811062043e-05,
      "loss": 4.8075,
      "mean_token_accuracy": 0.24153592735528945,
      "num_tokens": 202494106.0,
      "step": 97575
    },
    {
      "entropy": 5.672946929931641,
      "epoch": 8.851596516690856,
      "grad_norm": 1.453125,
      "learning_rate": 6.473063083762886e-05,
      "loss": 4.7623,
      "mean_token_accuracy": 0.24293074905872344,
      "num_tokens": 202504797.0,
      "step": 97580
    },
    {
      "entropy": 5.663831758499145,
      "epoch": 8.852050072568941,
      "grad_norm": 1.34375,
      "learning_rate": 6.471911791306145e-05,
      "loss": 4.7397,
      "mean_token_accuracy": 0.24733283668756484,
      "num_tokens": 202515102.0,
      "step": 97585
    },
    {
      "entropy": 5.666518020629883,
      "epoch": 8.852503628447025,
      "grad_norm": 1.4375,
      "learning_rate": 6.470760933715626e-05,
      "loss": 4.7794,
      "mean_token_accuracy": 0.23863689452409745,
      "num_tokens": 202525450.0,
      "step": 97590
    },
    {
      "entropy": 5.754364109039306,
      "epoch": 8.852957184325108,
      "grad_norm": 1.609375,
      "learning_rate": 6.469610511015127e-05,
      "loss": 4.912,
      "mean_token_accuracy": 0.22355742305517196,
      "num_tokens": 202536257.0,
      "step": 97595
    },
    {
      "entropy": 5.719266939163208,
      "epoch": 8.853410740203193,
      "grad_norm": 1.515625,
      "learning_rate": 6.468460523228445e-05,
      "loss": 4.786,
      "mean_token_accuracy": 0.23918561041355133,
      "num_tokens": 202545189.0,
      "step": 97600
    },
    {
      "entropy": 5.609397935867309,
      "epoch": 8.853864296081277,
      "grad_norm": 1.53125,
      "learning_rate": 6.467310970379356e-05,
      "loss": 4.7257,
      "mean_token_accuracy": 0.24496104270219804,
      "num_tokens": 202554929.0,
      "step": 97605
    },
    {
      "entropy": 5.686703634262085,
      "epoch": 8.854317851959362,
      "grad_norm": 1.5234375,
      "learning_rate": 6.46616185249164e-05,
      "loss": 4.7851,
      "mean_token_accuracy": 0.24857719540596007,
      "num_tokens": 202564875.0,
      "step": 97610
    },
    {
      "entropy": 5.647556781768799,
      "epoch": 8.854771407837445,
      "grad_norm": 1.515625,
      "learning_rate": 6.465013169589056e-05,
      "loss": 4.7529,
      "mean_token_accuracy": 0.24463108330965042,
      "num_tokens": 202575135.0,
      "step": 97615
    },
    {
      "entropy": 5.7195127487182615,
      "epoch": 8.85522496371553,
      "grad_norm": 1.671875,
      "learning_rate": 6.463864921695358e-05,
      "loss": 4.8206,
      "mean_token_accuracy": 0.24039374887943268,
      "num_tokens": 202584834.0,
      "step": 97620
    },
    {
      "entropy": 5.629875040054321,
      "epoch": 8.855678519593614,
      "grad_norm": 1.6328125,
      "learning_rate": 6.462717108834295e-05,
      "loss": 4.7415,
      "mean_token_accuracy": 0.23694879561662674,
      "num_tokens": 202594417.0,
      "step": 97625
    },
    {
      "entropy": 5.676660633087158,
      "epoch": 8.856132075471699,
      "grad_norm": 1.546875,
      "learning_rate": 6.461569731029599e-05,
      "loss": 4.7534,
      "mean_token_accuracy": 0.23420110642910003,
      "num_tokens": 202604203.0,
      "step": 97630
    },
    {
      "entropy": 5.645672464370728,
      "epoch": 8.856585631349782,
      "grad_norm": 1.4453125,
      "learning_rate": 6.460422788305007e-05,
      "loss": 4.774,
      "mean_token_accuracy": 0.24065130203962326,
      "num_tokens": 202614733.0,
      "step": 97635
    },
    {
      "entropy": 5.734382200241089,
      "epoch": 8.857039187227866,
      "grad_norm": 1.609375,
      "learning_rate": 6.459276280684232e-05,
      "loss": 4.861,
      "mean_token_accuracy": 0.23008860796689987,
      "num_tokens": 202625113.0,
      "step": 97640
    },
    {
      "entropy": 5.6696840763092045,
      "epoch": 8.85749274310595,
      "grad_norm": 1.515625,
      "learning_rate": 6.458130208190986e-05,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24403538852930068,
      "num_tokens": 202635303.0,
      "step": 97645
    },
    {
      "entropy": 5.671565294265747,
      "epoch": 8.857946298984034,
      "grad_norm": 1.5546875,
      "learning_rate": 6.456984570848969e-05,
      "loss": 4.7409,
      "mean_token_accuracy": 0.23396212011575698,
      "num_tokens": 202645155.0,
      "step": 97650
    },
    {
      "entropy": 5.633208227157593,
      "epoch": 8.85839985486212,
      "grad_norm": 1.6484375,
      "learning_rate": 6.455839368681874e-05,
      "loss": 4.7127,
      "mean_token_accuracy": 0.2463879853487015,
      "num_tokens": 202656535.0,
      "step": 97655
    },
    {
      "entropy": 5.647749662399292,
      "epoch": 8.858853410740203,
      "grad_norm": 1.5078125,
      "learning_rate": 6.45469460171338e-05,
      "loss": 4.7145,
      "mean_token_accuracy": 0.2425408437848091,
      "num_tokens": 202666412.0,
      "step": 97660
    },
    {
      "entropy": 5.655143594741821,
      "epoch": 8.859306966618288,
      "grad_norm": 1.75,
      "learning_rate": 6.453550269967166e-05,
      "loss": 4.7626,
      "mean_token_accuracy": 0.2405415505170822,
      "num_tokens": 202676204.0,
      "step": 97665
    },
    {
      "entropy": 5.730618715286255,
      "epoch": 8.859760522496371,
      "grad_norm": 1.4921875,
      "learning_rate": 6.4524063734669e-05,
      "loss": 4.8177,
      "mean_token_accuracy": 0.24335146993398665,
      "num_tokens": 202687273.0,
      "step": 97670
    },
    {
      "entropy": 5.632700777053833,
      "epoch": 8.860214078374456,
      "grad_norm": 1.609375,
      "learning_rate": 6.45126291223623e-05,
      "loss": 4.7882,
      "mean_token_accuracy": 0.23390913456678392,
      "num_tokens": 202697033.0,
      "step": 97675
    },
    {
      "entropy": 5.681014347076416,
      "epoch": 8.86066763425254,
      "grad_norm": 1.46875,
      "learning_rate": 6.450119886298809e-05,
      "loss": 4.7387,
      "mean_token_accuracy": 0.24646396040916443,
      "num_tokens": 202707645.0,
      "step": 97680
    },
    {
      "entropy": 5.6451311111450195,
      "epoch": 8.861121190130625,
      "grad_norm": 1.609375,
      "learning_rate": 6.448977295678273e-05,
      "loss": 4.7434,
      "mean_token_accuracy": 0.23517927378416062,
      "num_tokens": 202717915.0,
      "step": 97685
    },
    {
      "entropy": 5.650540065765381,
      "epoch": 8.861574746008708,
      "grad_norm": 1.4609375,
      "learning_rate": 6.447835140398249e-05,
      "loss": 4.7319,
      "mean_token_accuracy": 0.24258334040641785,
      "num_tokens": 202728223.0,
      "step": 97690
    },
    {
      "entropy": 5.636365747451782,
      "epoch": 8.862028301886792,
      "grad_norm": 1.6875,
      "learning_rate": 6.446693420482358e-05,
      "loss": 4.689,
      "mean_token_accuracy": 0.24320616722106933,
      "num_tokens": 202737764.0,
      "step": 97695
    },
    {
      "entropy": 5.646965742111206,
      "epoch": 8.862481857764877,
      "grad_norm": 1.6484375,
      "learning_rate": 6.445552135954213e-05,
      "loss": 4.7702,
      "mean_token_accuracy": 0.2341623455286026,
      "num_tokens": 202748635.0,
      "step": 97700
    },
    {
      "entropy": 5.71028356552124,
      "epoch": 8.86293541364296,
      "grad_norm": 1.59375,
      "learning_rate": 6.444411286837417e-05,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22762349992990494,
      "num_tokens": 202758419.0,
      "step": 97705
    },
    {
      "entropy": 5.6804896831512455,
      "epoch": 8.863388969521045,
      "grad_norm": 1.5546875,
      "learning_rate": 6.44327087315556e-05,
      "loss": 4.7215,
      "mean_token_accuracy": 0.23762225061655046,
      "num_tokens": 202768258.0,
      "step": 97710
    },
    {
      "entropy": 5.673408126831054,
      "epoch": 8.863842525399129,
      "grad_norm": 1.4609375,
      "learning_rate": 6.442130894932229e-05,
      "loss": 4.8273,
      "mean_token_accuracy": 0.23150912225246428,
      "num_tokens": 202778974.0,
      "step": 97715
    },
    {
      "entropy": 5.627245950698852,
      "epoch": 8.864296081277214,
      "grad_norm": 1.4609375,
      "learning_rate": 6.440991352190995e-05,
      "loss": 4.6798,
      "mean_token_accuracy": 0.2410773143172264,
      "num_tokens": 202788876.0,
      "step": 97720
    },
    {
      "entropy": 5.736932754516602,
      "epoch": 8.864749637155297,
      "grad_norm": 1.546875,
      "learning_rate": 6.439852244955427e-05,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2259435012936592,
      "num_tokens": 202799706.0,
      "step": 97725
    },
    {
      "entropy": 5.661042261123657,
      "epoch": 8.865203193033382,
      "grad_norm": 1.6796875,
      "learning_rate": 6.438713573249078e-05,
      "loss": 4.7856,
      "mean_token_accuracy": 0.23656613677740096,
      "num_tokens": 202810078.0,
      "step": 97730
    },
    {
      "entropy": 5.644256353378296,
      "epoch": 8.865656748911466,
      "grad_norm": 1.6875,
      "learning_rate": 6.437575337095505e-05,
      "loss": 4.7765,
      "mean_token_accuracy": 0.23807260394096375,
      "num_tokens": 202820315.0,
      "step": 97735
    },
    {
      "entropy": 5.599646329879761,
      "epoch": 8.86611030478955,
      "grad_norm": 1.4453125,
      "learning_rate": 6.436437536518239e-05,
      "loss": 4.6781,
      "mean_token_accuracy": 0.24733121693134308,
      "num_tokens": 202830050.0,
      "step": 97740
    },
    {
      "entropy": 5.738531541824341,
      "epoch": 8.866563860667634,
      "grad_norm": 1.703125,
      "learning_rate": 6.435300171540813e-05,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23247140794992446,
      "num_tokens": 202840557.0,
      "step": 97745
    },
    {
      "entropy": 5.703372192382813,
      "epoch": 8.867017416545718,
      "grad_norm": 1.5234375,
      "learning_rate": 6.434163242186748e-05,
      "loss": 4.7694,
      "mean_token_accuracy": 0.24030234664678574,
      "num_tokens": 202851027.0,
      "step": 97750
    },
    {
      "entropy": 5.710016965866089,
      "epoch": 8.867470972423803,
      "grad_norm": 1.5078125,
      "learning_rate": 6.433026748479555e-05,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2309641048312187,
      "num_tokens": 202861579.0,
      "step": 97755
    },
    {
      "entropy": 5.688867378234863,
      "epoch": 8.867924528301886,
      "grad_norm": 1.609375,
      "learning_rate": 6.431890690442736e-05,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2429543599486351,
      "num_tokens": 202871541.0,
      "step": 97760
    },
    {
      "entropy": 5.672736120223999,
      "epoch": 8.868378084179971,
      "grad_norm": 1.5546875,
      "learning_rate": 6.430755068099784e-05,
      "loss": 4.7721,
      "mean_token_accuracy": 0.22902362793684006,
      "num_tokens": 202881985.0,
      "step": 97765
    },
    {
      "entropy": 5.698618459701538,
      "epoch": 8.868831640058055,
      "grad_norm": 1.4375,
      "learning_rate": 6.429619881474189e-05,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23201297372579574,
      "num_tokens": 202892150.0,
      "step": 97770
    },
    {
      "entropy": 5.741258144378662,
      "epoch": 8.86928519593614,
      "grad_norm": 1.5078125,
      "learning_rate": 6.428485130589424e-05,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22782656848430632,
      "num_tokens": 202904367.0,
      "step": 97775
    },
    {
      "entropy": 5.809531497955322,
      "epoch": 8.869738751814223,
      "grad_norm": 1.5,
      "learning_rate": 6.427350815468959e-05,
      "loss": 4.9348,
      "mean_token_accuracy": 0.23004272282123567,
      "num_tokens": 202914695.0,
      "step": 97780
    },
    {
      "entropy": 5.666261482238769,
      "epoch": 8.870192307692308,
      "grad_norm": 1.46875,
      "learning_rate": 6.426216936136245e-05,
      "loss": 4.773,
      "mean_token_accuracy": 0.24560380429029466,
      "num_tokens": 202924096.0,
      "step": 97785
    },
    {
      "entropy": 5.64280514717102,
      "epoch": 8.870645863570392,
      "grad_norm": 1.5625,
      "learning_rate": 6.425083492614737e-05,
      "loss": 4.7765,
      "mean_token_accuracy": 0.2383841961622238,
      "num_tokens": 202935048.0,
      "step": 97790
    },
    {
      "entropy": 5.74262490272522,
      "epoch": 8.871099419448477,
      "grad_norm": 1.4453125,
      "learning_rate": 6.423950484927875e-05,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23554763495922088,
      "num_tokens": 202945111.0,
      "step": 97795
    },
    {
      "entropy": 5.704035377502441,
      "epoch": 8.87155297532656,
      "grad_norm": 1.6015625,
      "learning_rate": 6.42281791309908e-05,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23580415844917296,
      "num_tokens": 202955697.0,
      "step": 97800
    },
    {
      "entropy": 5.70127010345459,
      "epoch": 8.872006531204644,
      "grad_norm": 1.6015625,
      "learning_rate": 6.421685777151789e-05,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23382191807031633,
      "num_tokens": 202966130.0,
      "step": 97805
    },
    {
      "entropy": 5.688915395736695,
      "epoch": 8.872460087082729,
      "grad_norm": 1.609375,
      "learning_rate": 6.420554077109404e-05,
      "loss": 4.69,
      "mean_token_accuracy": 0.2394665852189064,
      "num_tokens": 202975662.0,
      "step": 97810
    },
    {
      "entropy": 5.678547716140747,
      "epoch": 8.872913642960812,
      "grad_norm": 1.4453125,
      "learning_rate": 6.41942281299534e-05,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23472709953784943,
      "num_tokens": 202986250.0,
      "step": 97815
    },
    {
      "entropy": 5.659782266616821,
      "epoch": 8.873367198838897,
      "grad_norm": 1.5078125,
      "learning_rate": 6.418291984832978e-05,
      "loss": 4.7858,
      "mean_token_accuracy": 0.23855352252721787,
      "num_tokens": 202996710.0,
      "step": 97820
    },
    {
      "entropy": 5.7007286071777346,
      "epoch": 8.87382075471698,
      "grad_norm": 1.6640625,
      "learning_rate": 6.417161592645705e-05,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23777731508016586,
      "num_tokens": 203006654.0,
      "step": 97825
    },
    {
      "entropy": 5.643151330947876,
      "epoch": 8.874274310595066,
      "grad_norm": 1.7265625,
      "learning_rate": 6.416031636456909e-05,
      "loss": 4.6965,
      "mean_token_accuracy": 0.24351665675640105,
      "num_tokens": 203016879.0,
      "step": 97830
    },
    {
      "entropy": 5.70547137260437,
      "epoch": 8.87472786647315,
      "grad_norm": 1.5,
      "learning_rate": 6.41490211628995e-05,
      "loss": 4.8475,
      "mean_token_accuracy": 0.23623892813920974,
      "num_tokens": 203027851.0,
      "step": 97835
    },
    {
      "entropy": 5.650147867202759,
      "epoch": 8.875181422351234,
      "grad_norm": 1.4765625,
      "learning_rate": 6.41377303216819e-05,
      "loss": 4.7067,
      "mean_token_accuracy": 0.24769876003265381,
      "num_tokens": 203038174.0,
      "step": 97840
    },
    {
      "entropy": 5.637833881378174,
      "epoch": 8.875634978229318,
      "grad_norm": 1.546875,
      "learning_rate": 6.412644384114978e-05,
      "loss": 4.7236,
      "mean_token_accuracy": 0.24031900316476823,
      "num_tokens": 203048614.0,
      "step": 97845
    },
    {
      "entropy": 5.738307762145996,
      "epoch": 8.876088534107403,
      "grad_norm": 1.6328125,
      "learning_rate": 6.411516172153652e-05,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2330142468214035,
      "num_tokens": 203058878.0,
      "step": 97850
    },
    {
      "entropy": 5.658227062225341,
      "epoch": 8.876542089985486,
      "grad_norm": 1.640625,
      "learning_rate": 6.410388396307546e-05,
      "loss": 4.7878,
      "mean_token_accuracy": 0.23641326874494553,
      "num_tokens": 203069415.0,
      "step": 97855
    },
    {
      "entropy": 5.76826663017273,
      "epoch": 8.87699564586357,
      "grad_norm": 1.4921875,
      "learning_rate": 6.409261056599979e-05,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22450321912765503,
      "num_tokens": 203079226.0,
      "step": 97860
    },
    {
      "entropy": 5.706547021865845,
      "epoch": 8.877449201741655,
      "grad_norm": 1.4453125,
      "learning_rate": 6.40813415305427e-05,
      "loss": 4.8184,
      "mean_token_accuracy": 0.23440183997154235,
      "num_tokens": 203089666.0,
      "step": 97865
    },
    {
      "entropy": 5.6803642272949215,
      "epoch": 8.877902757619738,
      "grad_norm": 1.515625,
      "learning_rate": 6.407007685693722e-05,
      "loss": 4.7628,
      "mean_token_accuracy": 0.23809519410133362,
      "num_tokens": 203100412.0,
      "step": 97870
    },
    {
      "entropy": 5.732563257217407,
      "epoch": 8.878356313497823,
      "grad_norm": 1.640625,
      "learning_rate": 6.40588165454163e-05,
      "loss": 4.8292,
      "mean_token_accuracy": 0.23690660297870636,
      "num_tokens": 203110808.0,
      "step": 97875
    },
    {
      "entropy": 5.749230527877808,
      "epoch": 8.878809869375907,
      "grad_norm": 1.6640625,
      "learning_rate": 6.404756059621283e-05,
      "loss": 4.9495,
      "mean_token_accuracy": 0.23352828919887542,
      "num_tokens": 203121425.0,
      "step": 97880
    },
    {
      "entropy": 5.604851961135864,
      "epoch": 8.879263425253992,
      "grad_norm": 1.59375,
      "learning_rate": 6.403630900955955e-05,
      "loss": 4.6968,
      "mean_token_accuracy": 0.24628683775663376,
      "num_tokens": 203131413.0,
      "step": 97885
    },
    {
      "entropy": 5.666931009292602,
      "epoch": 8.879716981132075,
      "grad_norm": 1.59375,
      "learning_rate": 6.402506178568915e-05,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23650390803813934,
      "num_tokens": 203141270.0,
      "step": 97890
    },
    {
      "entropy": 5.671171188354492,
      "epoch": 8.88017053701016,
      "grad_norm": 1.671875,
      "learning_rate": 6.40138189248342e-05,
      "loss": 4.7635,
      "mean_token_accuracy": 0.23844099789857864,
      "num_tokens": 203152062.0,
      "step": 97895
    },
    {
      "entropy": 5.753640031814575,
      "epoch": 8.880624092888244,
      "grad_norm": 1.578125,
      "learning_rate": 6.400258042722729e-05,
      "loss": 4.8499,
      "mean_token_accuracy": 0.23531690537929534,
      "num_tokens": 203161868.0,
      "step": 97900
    },
    {
      "entropy": 5.721957349777222,
      "epoch": 8.881077648766327,
      "grad_norm": 1.5859375,
      "learning_rate": 6.399134629310078e-05,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23869109898805618,
      "num_tokens": 203171633.0,
      "step": 97905
    },
    {
      "entropy": 5.6936705112457275,
      "epoch": 8.881531204644412,
      "grad_norm": 1.5859375,
      "learning_rate": 6.398011652268699e-05,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23544104993343354,
      "num_tokens": 203183283.0,
      "step": 97910
    },
    {
      "entropy": 5.720013523101807,
      "epoch": 8.881984760522496,
      "grad_norm": 1.53125,
      "learning_rate": 6.396889111621818e-05,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2319570004940033,
      "num_tokens": 203193670.0,
      "step": 97915
    },
    {
      "entropy": 5.698061990737915,
      "epoch": 8.882438316400581,
      "grad_norm": 1.4375,
      "learning_rate": 6.395767007392649e-05,
      "loss": 4.825,
      "mean_token_accuracy": 0.236500883102417,
      "num_tokens": 203203203.0,
      "step": 97920
    },
    {
      "entropy": 5.675648498535156,
      "epoch": 8.882891872278664,
      "grad_norm": 1.5703125,
      "learning_rate": 6.394645339604392e-05,
      "loss": 4.7495,
      "mean_token_accuracy": 0.23962987065315247,
      "num_tokens": 203213323.0,
      "step": 97925
    },
    {
      "entropy": 5.6104989528656,
      "epoch": 8.88334542815675,
      "grad_norm": 1.53125,
      "learning_rate": 6.393524108280246e-05,
      "loss": 4.6996,
      "mean_token_accuracy": 0.24205637872219085,
      "num_tokens": 203223200.0,
      "step": 97930
    },
    {
      "entropy": 5.669977235794067,
      "epoch": 8.883798984034833,
      "grad_norm": 1.4453125,
      "learning_rate": 6.392403313443405e-05,
      "loss": 4.7849,
      "mean_token_accuracy": 0.24157272279262543,
      "num_tokens": 203234207.0,
      "step": 97935
    },
    {
      "entropy": 5.697441244125367,
      "epoch": 8.884252539912918,
      "grad_norm": 1.5390625,
      "learning_rate": 6.391282955117043e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2345607325434685,
      "num_tokens": 203244584.0,
      "step": 97940
    },
    {
      "entropy": 5.719444799423218,
      "epoch": 8.884706095791001,
      "grad_norm": 1.6015625,
      "learning_rate": 6.390163033324327e-05,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2293451890349388,
      "num_tokens": 203255720.0,
      "step": 97945
    },
    {
      "entropy": 5.735246467590332,
      "epoch": 8.885159651669085,
      "grad_norm": 1.5703125,
      "learning_rate": 6.389043548088418e-05,
      "loss": 4.8049,
      "mean_token_accuracy": 0.24053987711668015,
      "num_tokens": 203266301.0,
      "step": 97950
    },
    {
      "entropy": 5.764325141906738,
      "epoch": 8.88561320754717,
      "grad_norm": 1.4453125,
      "learning_rate": 6.38792449943247e-05,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22775816768407822,
      "num_tokens": 203277548.0,
      "step": 97955
    },
    {
      "entropy": 5.716698217391968,
      "epoch": 8.886066763425253,
      "grad_norm": 1.5,
      "learning_rate": 6.38680588737962e-05,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23761253356933593,
      "num_tokens": 203288921.0,
      "step": 97960
    },
    {
      "entropy": 5.768531703948975,
      "epoch": 8.886520319303338,
      "grad_norm": 1.46875,
      "learning_rate": 6.385687711953006e-05,
      "loss": 4.885,
      "mean_token_accuracy": 0.2312289297580719,
      "num_tokens": 203299763.0,
      "step": 97965
    },
    {
      "entropy": 5.629726886749268,
      "epoch": 8.886973875181422,
      "grad_norm": 1.5703125,
      "learning_rate": 6.38456997317575e-05,
      "loss": 4.7106,
      "mean_token_accuracy": 0.2459884464740753,
      "num_tokens": 203310030.0,
      "step": 97970
    },
    {
      "entropy": 5.728412866592407,
      "epoch": 8.887427431059507,
      "grad_norm": 1.5703125,
      "learning_rate": 6.38345267107097e-05,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23427428007125856,
      "num_tokens": 203320428.0,
      "step": 97975
    },
    {
      "entropy": 5.720850038528442,
      "epoch": 8.88788098693759,
      "grad_norm": 1.6640625,
      "learning_rate": 6.382335805661771e-05,
      "loss": 4.7575,
      "mean_token_accuracy": 0.23908771723508834,
      "num_tokens": 203329718.0,
      "step": 97980
    },
    {
      "entropy": 5.734273529052734,
      "epoch": 8.888334542815675,
      "grad_norm": 1.421875,
      "learning_rate": 6.381219376971247e-05,
      "loss": 4.9134,
      "mean_token_accuracy": 0.23655234575271605,
      "num_tokens": 203340495.0,
      "step": 97985
    },
    {
      "entropy": 5.695552778244019,
      "epoch": 8.888788098693759,
      "grad_norm": 1.5234375,
      "learning_rate": 6.380103385022489e-05,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24921115338802338,
      "num_tokens": 203351241.0,
      "step": 97990
    },
    {
      "entropy": 5.6641254901885985,
      "epoch": 8.889241654571844,
      "grad_norm": 1.5234375,
      "learning_rate": 6.37898782983857e-05,
      "loss": 4.7398,
      "mean_token_accuracy": 0.23787085711956024,
      "num_tokens": 203361598.0,
      "step": 97995
    },
    {
      "entropy": 5.7450777053833,
      "epoch": 8.889695210449927,
      "grad_norm": 1.484375,
      "learning_rate": 6.377872711442569e-05,
      "loss": 4.8514,
      "mean_token_accuracy": 0.23686182498931885,
      "num_tokens": 203373172.0,
      "step": 98000
    },
    {
      "entropy": 5.740471029281617,
      "epoch": 8.89014876632801,
      "grad_norm": 1.71875,
      "learning_rate": 6.376758029857543e-05,
      "loss": 4.8482,
      "mean_token_accuracy": 0.23642073273658754,
      "num_tokens": 203382953.0,
      "step": 98005
    },
    {
      "entropy": 5.663624000549317,
      "epoch": 8.890602322206096,
      "grad_norm": 1.515625,
      "learning_rate": 6.375643785106545e-05,
      "loss": 4.7467,
      "mean_token_accuracy": 0.24230839014053346,
      "num_tokens": 203393826.0,
      "step": 98010
    },
    {
      "entropy": 5.618016338348388,
      "epoch": 8.89105587808418,
      "grad_norm": 1.5546875,
      "learning_rate": 6.374529977212614e-05,
      "loss": 4.6793,
      "mean_token_accuracy": 0.2500133901834488,
      "num_tokens": 203403774.0,
      "step": 98015
    },
    {
      "entropy": 5.689151430130005,
      "epoch": 8.891509433962264,
      "grad_norm": 1.4375,
      "learning_rate": 6.373416606198789e-05,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2399270385503769,
      "num_tokens": 203414296.0,
      "step": 98020
    },
    {
      "entropy": 5.718053817749023,
      "epoch": 8.891962989840348,
      "grad_norm": 1.5859375,
      "learning_rate": 6.37230367208809e-05,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2400914952158928,
      "num_tokens": 203424886.0,
      "step": 98025
    },
    {
      "entropy": 5.726640224456787,
      "epoch": 8.892416545718433,
      "grad_norm": 1.6015625,
      "learning_rate": 6.371191174903535e-05,
      "loss": 4.8701,
      "mean_token_accuracy": 0.23983149528503417,
      "num_tokens": 203435136.0,
      "step": 98030
    },
    {
      "entropy": 5.62253155708313,
      "epoch": 8.892870101596516,
      "grad_norm": 1.484375,
      "learning_rate": 6.370079114668132e-05,
      "loss": 4.7485,
      "mean_token_accuracy": 0.23747865110635757,
      "num_tokens": 203446241.0,
      "step": 98035
    },
    {
      "entropy": 5.599478340148925,
      "epoch": 8.893323657474602,
      "grad_norm": 1.546875,
      "learning_rate": 6.368967491404878e-05,
      "loss": 4.6915,
      "mean_token_accuracy": 0.2465820476412773,
      "num_tokens": 203455991.0,
      "step": 98040
    },
    {
      "entropy": 5.712496614456176,
      "epoch": 8.893777213352685,
      "grad_norm": 1.609375,
      "learning_rate": 6.367856305136761e-05,
      "loss": 4.8403,
      "mean_token_accuracy": 0.23465651422739028,
      "num_tokens": 203466659.0,
      "step": 98045
    },
    {
      "entropy": 5.684095144271851,
      "epoch": 8.89423076923077,
      "grad_norm": 1.609375,
      "learning_rate": 6.366745555886759e-05,
      "loss": 4.7623,
      "mean_token_accuracy": 0.23530927449464797,
      "num_tokens": 203477139.0,
      "step": 98050
    },
    {
      "entropy": 5.769555759429932,
      "epoch": 8.894684325108853,
      "grad_norm": 1.4140625,
      "learning_rate": 6.365635243677848e-05,
      "loss": 4.9355,
      "mean_token_accuracy": 0.22728936821222306,
      "num_tokens": 203487091.0,
      "step": 98055
    },
    {
      "entropy": 5.680793237686157,
      "epoch": 8.895137880986937,
      "grad_norm": 1.53125,
      "learning_rate": 6.364525368532984e-05,
      "loss": 4.8189,
      "mean_token_accuracy": 0.24179269671440123,
      "num_tokens": 203498153.0,
      "step": 98060
    },
    {
      "entropy": 5.775969219207764,
      "epoch": 8.895591436865022,
      "grad_norm": 1.6171875,
      "learning_rate": 6.363415930475116e-05,
      "loss": 4.8758,
      "mean_token_accuracy": 0.2282691240310669,
      "num_tokens": 203508591.0,
      "step": 98065
    },
    {
      "entropy": 5.784748458862305,
      "epoch": 8.896044992743105,
      "grad_norm": 1.71875,
      "learning_rate": 6.362306929527196e-05,
      "loss": 4.9123,
      "mean_token_accuracy": 0.22266198843717575,
      "num_tokens": 203517988.0,
      "step": 98070
    },
    {
      "entropy": 5.73214807510376,
      "epoch": 8.89649854862119,
      "grad_norm": 1.4765625,
      "learning_rate": 6.361198365712157e-05,
      "loss": 4.8615,
      "mean_token_accuracy": 0.23469115197658538,
      "num_tokens": 203528589.0,
      "step": 98075
    },
    {
      "entropy": 5.717279720306396,
      "epoch": 8.896952104499274,
      "grad_norm": 1.4765625,
      "learning_rate": 6.360090239052924e-05,
      "loss": 4.8408,
      "mean_token_accuracy": 0.2410322070121765,
      "num_tokens": 203539933.0,
      "step": 98080
    },
    {
      "entropy": 5.723220443725586,
      "epoch": 8.897405660377359,
      "grad_norm": 1.515625,
      "learning_rate": 6.358982549572411e-05,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22948981821537018,
      "num_tokens": 203550801.0,
      "step": 98085
    },
    {
      "entropy": 5.749911880493164,
      "epoch": 8.897859216255442,
      "grad_norm": 1.390625,
      "learning_rate": 6.357875297293524e-05,
      "loss": 4.8488,
      "mean_token_accuracy": 0.23508855998516082,
      "num_tokens": 203561842.0,
      "step": 98090
    },
    {
      "entropy": 5.666502523422241,
      "epoch": 8.898312772133528,
      "grad_norm": 1.5546875,
      "learning_rate": 6.356768482239168e-05,
      "loss": 4.747,
      "mean_token_accuracy": 0.23452934324741365,
      "num_tokens": 203571731.0,
      "step": 98095
    },
    {
      "entropy": 5.703822040557862,
      "epoch": 8.898766328011611,
      "grad_norm": 1.5390625,
      "learning_rate": 6.355662104432221e-05,
      "loss": 4.7205,
      "mean_token_accuracy": 0.24842463582754135,
      "num_tokens": 203582012.0,
      "step": 98100
    },
    {
      "entropy": 5.7344873428344725,
      "epoch": 8.899219883889696,
      "grad_norm": 1.515625,
      "learning_rate": 6.354556163895572e-05,
      "loss": 4.8612,
      "mean_token_accuracy": 0.2322070077061653,
      "num_tokens": 203592882.0,
      "step": 98105
    },
    {
      "entropy": 5.720294427871704,
      "epoch": 8.89967343976778,
      "grad_norm": 1.546875,
      "learning_rate": 6.353450660652095e-05,
      "loss": 4.8979,
      "mean_token_accuracy": 0.23226162046194077,
      "num_tokens": 203604385.0,
      "step": 98110
    },
    {
      "entropy": 5.708036708831787,
      "epoch": 8.900126995645863,
      "grad_norm": 1.515625,
      "learning_rate": 6.352345594724642e-05,
      "loss": 4.8559,
      "mean_token_accuracy": 0.23258899748325348,
      "num_tokens": 203615670.0,
      "step": 98115
    },
    {
      "entropy": 5.757740926742554,
      "epoch": 8.900580551523948,
      "grad_norm": 1.4921875,
      "learning_rate": 6.351240966136073e-05,
      "loss": 4.884,
      "mean_token_accuracy": 0.23193887174129485,
      "num_tokens": 203625910.0,
      "step": 98120
    },
    {
      "entropy": 5.703135538101196,
      "epoch": 8.901034107402031,
      "grad_norm": 1.6484375,
      "learning_rate": 6.35013677490923e-05,
      "loss": 4.8047,
      "mean_token_accuracy": 0.23704902082681656,
      "num_tokens": 203636791.0,
      "step": 98125
    },
    {
      "entropy": 5.6986290454864506,
      "epoch": 8.901487663280117,
      "grad_norm": 1.5234375,
      "learning_rate": 6.349033021066951e-05,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23816209733486177,
      "num_tokens": 203647551.0,
      "step": 98130
    },
    {
      "entropy": 5.715506744384766,
      "epoch": 8.9019412191582,
      "grad_norm": 1.6484375,
      "learning_rate": 6.347929704632052e-05,
      "loss": 4.7921,
      "mean_token_accuracy": 0.23719525337219238,
      "num_tokens": 203658217.0,
      "step": 98135
    },
    {
      "entropy": 5.72891936302185,
      "epoch": 8.902394775036285,
      "grad_norm": 1.3984375,
      "learning_rate": 6.346826825627362e-05,
      "loss": 4.8278,
      "mean_token_accuracy": 0.22955704778432845,
      "num_tokens": 203669132.0,
      "step": 98140
    },
    {
      "entropy": 5.693762159347534,
      "epoch": 8.902848330914368,
      "grad_norm": 1.5859375,
      "learning_rate": 6.345724384075682e-05,
      "loss": 4.8039,
      "mean_token_accuracy": 0.23464658856391907,
      "num_tokens": 203679895.0,
      "step": 98145
    },
    {
      "entropy": 5.689338111877442,
      "epoch": 8.903301886792454,
      "grad_norm": 1.5625,
      "learning_rate": 6.344622379999815e-05,
      "loss": 4.8061,
      "mean_token_accuracy": 0.235416416823864,
      "num_tokens": 203690468.0,
      "step": 98150
    },
    {
      "entropy": 5.749297952651977,
      "epoch": 8.903755442670537,
      "grad_norm": 1.6015625,
      "learning_rate": 6.343520813422548e-05,
      "loss": 4.834,
      "mean_token_accuracy": 0.23381888568401338,
      "num_tokens": 203700206.0,
      "step": 98155
    },
    {
      "entropy": 5.701363134384155,
      "epoch": 8.904208998548622,
      "grad_norm": 1.75,
      "learning_rate": 6.342419684366658e-05,
      "loss": 4.7813,
      "mean_token_accuracy": 0.24026089906692505,
      "num_tokens": 203710815.0,
      "step": 98160
    },
    {
      "entropy": 5.658713865280151,
      "epoch": 8.904662554426706,
      "grad_norm": 1.4296875,
      "learning_rate": 6.34131899285492e-05,
      "loss": 4.7785,
      "mean_token_accuracy": 0.23838384300470353,
      "num_tokens": 203721329.0,
      "step": 98165
    },
    {
      "entropy": 5.6754469871521,
      "epoch": 8.905116110304789,
      "grad_norm": 1.390625,
      "learning_rate": 6.340218738910101e-05,
      "loss": 4.8165,
      "mean_token_accuracy": 0.2389571949839592,
      "num_tokens": 203731537.0,
      "step": 98170
    },
    {
      "entropy": 5.710326099395752,
      "epoch": 8.905569666182874,
      "grad_norm": 1.5546875,
      "learning_rate": 6.339118922554951e-05,
      "loss": 4.7875,
      "mean_token_accuracy": 0.23552779257297515,
      "num_tokens": 203740952.0,
      "step": 98175
    },
    {
      "entropy": 5.63844313621521,
      "epoch": 8.906023222060957,
      "grad_norm": 1.546875,
      "learning_rate": 6.338019543812211e-05,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2429698660969734,
      "num_tokens": 203750950.0,
      "step": 98180
    },
    {
      "entropy": 5.720400476455689,
      "epoch": 8.906476777939043,
      "grad_norm": 1.4609375,
      "learning_rate": 6.336920602704622e-05,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22726186066865922,
      "num_tokens": 203762209.0,
      "step": 98185
    },
    {
      "entropy": 5.548933219909668,
      "epoch": 8.906930333817126,
      "grad_norm": 1.4140625,
      "learning_rate": 6.335822099254909e-05,
      "loss": 4.6193,
      "mean_token_accuracy": 0.24797439873218535,
      "num_tokens": 203772753.0,
      "step": 98190
    },
    {
      "entropy": 5.6955650806427,
      "epoch": 8.907383889695211,
      "grad_norm": 1.5625,
      "learning_rate": 6.334724033485788e-05,
      "loss": 4.8114,
      "mean_token_accuracy": 0.24044343084096909,
      "num_tokens": 203783109.0,
      "step": 98195
    },
    {
      "entropy": 5.685979557037354,
      "epoch": 8.907837445573294,
      "grad_norm": 1.5,
      "learning_rate": 6.333626405419963e-05,
      "loss": 4.733,
      "mean_token_accuracy": 0.24025966823101044,
      "num_tokens": 203793276.0,
      "step": 98200
    },
    {
      "entropy": 5.752237462997437,
      "epoch": 8.908291001451378,
      "grad_norm": 1.5859375,
      "learning_rate": 6.332529215080139e-05,
      "loss": 4.8981,
      "mean_token_accuracy": 0.23107083588838578,
      "num_tokens": 203804489.0,
      "step": 98205
    },
    {
      "entropy": 5.792819929122925,
      "epoch": 8.908744557329463,
      "grad_norm": 1.4375,
      "learning_rate": 6.331432462489009e-05,
      "loss": 4.9191,
      "mean_token_accuracy": 0.22669755518436432,
      "num_tokens": 203816396.0,
      "step": 98210
    },
    {
      "entropy": 5.671675443649292,
      "epoch": 8.909198113207546,
      "grad_norm": 1.390625,
      "learning_rate": 6.330336147669248e-05,
      "loss": 4.7836,
      "mean_token_accuracy": 0.2342408135533333,
      "num_tokens": 203828106.0,
      "step": 98215
    },
    {
      "entropy": 5.754123878479004,
      "epoch": 8.909651669085632,
      "grad_norm": 1.6171875,
      "learning_rate": 6.32924027064353e-05,
      "loss": 4.8628,
      "mean_token_accuracy": 0.23058492988348006,
      "num_tokens": 203837993.0,
      "step": 98220
    },
    {
      "entropy": 5.7172167778015135,
      "epoch": 8.910105224963715,
      "grad_norm": 1.5234375,
      "learning_rate": 6.328144831434518e-05,
      "loss": 4.7865,
      "mean_token_accuracy": 0.23595905900001526,
      "num_tokens": 203847624.0,
      "step": 98225
    },
    {
      "entropy": 5.648945283889771,
      "epoch": 8.9105587808418,
      "grad_norm": 1.546875,
      "learning_rate": 6.327049830064868e-05,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23817362636327744,
      "num_tokens": 203858472.0,
      "step": 98230
    },
    {
      "entropy": 5.749978876113891,
      "epoch": 8.911012336719883,
      "grad_norm": 1.4921875,
      "learning_rate": 6.325955266557216e-05,
      "loss": 4.8824,
      "mean_token_accuracy": 0.23300011307001114,
      "num_tokens": 203870570.0,
      "step": 98235
    },
    {
      "entropy": 5.677467346191406,
      "epoch": 8.911465892597969,
      "grad_norm": 1.5390625,
      "learning_rate": 6.324861140934212e-05,
      "loss": 4.6851,
      "mean_token_accuracy": 0.24098197221755982,
      "num_tokens": 203881093.0,
      "step": 98240
    },
    {
      "entropy": 5.696637678146362,
      "epoch": 8.911919448476052,
      "grad_norm": 1.5546875,
      "learning_rate": 6.323767453218478e-05,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23413354754447938,
      "num_tokens": 203891326.0,
      "step": 98245
    },
    {
      "entropy": 5.738678979873657,
      "epoch": 8.912373004354137,
      "grad_norm": 1.3984375,
      "learning_rate": 6.322674203432626e-05,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22513746470212936,
      "num_tokens": 203901943.0,
      "step": 98250
    },
    {
      "entropy": 5.667449140548706,
      "epoch": 8.91282656023222,
      "grad_norm": 1.3828125,
      "learning_rate": 6.321581391599268e-05,
      "loss": 4.7674,
      "mean_token_accuracy": 0.24087972640991212,
      "num_tokens": 203912452.0,
      "step": 98255
    },
    {
      "entropy": 5.713769960403442,
      "epoch": 8.913280116110304,
      "grad_norm": 1.5625,
      "learning_rate": 6.320489017741e-05,
      "loss": 4.8839,
      "mean_token_accuracy": 0.2305188849568367,
      "num_tokens": 203923954.0,
      "step": 98260
    },
    {
      "entropy": 5.702348327636718,
      "epoch": 8.913733671988389,
      "grad_norm": 1.4375,
      "learning_rate": 6.319397081880417e-05,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23997247070074082,
      "num_tokens": 203934103.0,
      "step": 98265
    },
    {
      "entropy": 5.71597056388855,
      "epoch": 8.914187227866472,
      "grad_norm": 1.609375,
      "learning_rate": 6.318305584040104e-05,
      "loss": 4.7836,
      "mean_token_accuracy": 0.2377261996269226,
      "num_tokens": 203944861.0,
      "step": 98270
    },
    {
      "entropy": 5.686637687683105,
      "epoch": 8.914640783744558,
      "grad_norm": 1.640625,
      "learning_rate": 6.317214524242628e-05,
      "loss": 4.7577,
      "mean_token_accuracy": 0.24050681591033934,
      "num_tokens": 203954634.0,
      "step": 98275
    },
    {
      "entropy": 5.679332876205445,
      "epoch": 8.915094339622641,
      "grad_norm": 1.3046875,
      "learning_rate": 6.316123902510552e-05,
      "loss": 4.7808,
      "mean_token_accuracy": 0.23794894367456437,
      "num_tokens": 203965806.0,
      "step": 98280
    },
    {
      "entropy": 5.687490367889405,
      "epoch": 8.915547895500726,
      "grad_norm": 1.59375,
      "learning_rate": 6.315033718866434e-05,
      "loss": 4.7638,
      "mean_token_accuracy": 0.2476590946316719,
      "num_tokens": 203975917.0,
      "step": 98285
    },
    {
      "entropy": 5.726072835922241,
      "epoch": 8.91600145137881,
      "grad_norm": 1.4765625,
      "learning_rate": 6.313943973332817e-05,
      "loss": 4.8412,
      "mean_token_accuracy": 0.22965929210186004,
      "num_tokens": 203985749.0,
      "step": 98290
    },
    {
      "entropy": 5.650608253479004,
      "epoch": 8.916455007256895,
      "grad_norm": 1.6015625,
      "learning_rate": 6.312854665932234e-05,
      "loss": 4.7556,
      "mean_token_accuracy": 0.24310659617185593,
      "num_tokens": 203997578.0,
      "step": 98295
    },
    {
      "entropy": 5.692889595031739,
      "epoch": 8.916908563134978,
      "grad_norm": 1.4296875,
      "learning_rate": 6.311765796687219e-05,
      "loss": 4.779,
      "mean_token_accuracy": 0.23701490759849547,
      "num_tokens": 204007824.0,
      "step": 98300
    },
    {
      "entropy": 5.811197233200073,
      "epoch": 8.917362119013063,
      "grad_norm": 1.4609375,
      "learning_rate": 6.310677365620286e-05,
      "loss": 4.9634,
      "mean_token_accuracy": 0.22848581075668334,
      "num_tokens": 204018272.0,
      "step": 98305
    },
    {
      "entropy": 5.703503942489624,
      "epoch": 8.917815674891147,
      "grad_norm": 1.6171875,
      "learning_rate": 6.309589372753946e-05,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23362531661987304,
      "num_tokens": 204029656.0,
      "step": 98310
    },
    {
      "entropy": 5.638335418701172,
      "epoch": 8.91826923076923,
      "grad_norm": 1.421875,
      "learning_rate": 6.308501818110697e-05,
      "loss": 4.7327,
      "mean_token_accuracy": 0.2405376136302948,
      "num_tokens": 204039768.0,
      "step": 98315
    },
    {
      "entropy": 5.768176507949829,
      "epoch": 8.918722786647315,
      "grad_norm": 1.5390625,
      "learning_rate": 6.307414701713032e-05,
      "loss": 4.78,
      "mean_token_accuracy": 0.2448711723089218,
      "num_tokens": 204050056.0,
      "step": 98320
    },
    {
      "entropy": 5.724326181411743,
      "epoch": 8.919176342525398,
      "grad_norm": 1.4921875,
      "learning_rate": 6.306328023583428e-05,
      "loss": 4.8645,
      "mean_token_accuracy": 0.23072660118341445,
      "num_tokens": 204060867.0,
      "step": 98325
    },
    {
      "entropy": 5.657086324691773,
      "epoch": 8.919629898403484,
      "grad_norm": 1.609375,
      "learning_rate": 6.30524178374436e-05,
      "loss": 4.7755,
      "mean_token_accuracy": 0.24297548979520797,
      "num_tokens": 204071237.0,
      "step": 98330
    },
    {
      "entropy": 5.627369785308838,
      "epoch": 8.920083454281567,
      "grad_norm": 1.59375,
      "learning_rate": 6.304155982218298e-05,
      "loss": 4.7038,
      "mean_token_accuracy": 0.24394398480653762,
      "num_tokens": 204081925.0,
      "step": 98335
    },
    {
      "entropy": 5.655172967910767,
      "epoch": 8.920537010159652,
      "grad_norm": 1.640625,
      "learning_rate": 6.303070619027688e-05,
      "loss": 4.6791,
      "mean_token_accuracy": 0.24667007774114608,
      "num_tokens": 204091735.0,
      "step": 98340
    },
    {
      "entropy": 5.693485355377197,
      "epoch": 8.920990566037736,
      "grad_norm": 1.6640625,
      "learning_rate": 6.301985694194982e-05,
      "loss": 4.7744,
      "mean_token_accuracy": 0.23613267689943312,
      "num_tokens": 204101309.0,
      "step": 98345
    },
    {
      "entropy": 5.6664121627807615,
      "epoch": 8.92144412191582,
      "grad_norm": 1.5859375,
      "learning_rate": 6.300901207742613e-05,
      "loss": 4.7794,
      "mean_token_accuracy": 0.24063094109296798,
      "num_tokens": 204111725.0,
      "step": 98350
    },
    {
      "entropy": 5.684040975570679,
      "epoch": 8.921897677793904,
      "grad_norm": 1.4921875,
      "learning_rate": 6.29981715969301e-05,
      "loss": 4.7832,
      "mean_token_accuracy": 0.24187856018543244,
      "num_tokens": 204121845.0,
      "step": 98355
    },
    {
      "entropy": 5.728105545043945,
      "epoch": 8.92235123367199,
      "grad_norm": 1.40625,
      "learning_rate": 6.298733550068589e-05,
      "loss": 4.8815,
      "mean_token_accuracy": 0.23439652025699614,
      "num_tokens": 204132267.0,
      "step": 98360
    },
    {
      "entropy": 5.658990573883057,
      "epoch": 8.922804789550073,
      "grad_norm": 1.453125,
      "learning_rate": 6.29765037889176e-05,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23854935318231582,
      "num_tokens": 204142929.0,
      "step": 98365
    },
    {
      "entropy": 5.690318584442139,
      "epoch": 8.923258345428156,
      "grad_norm": 1.6171875,
      "learning_rate": 6.296567646184924e-05,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23609996140003203,
      "num_tokens": 204153640.0,
      "step": 98370
    },
    {
      "entropy": 5.714227533340454,
      "epoch": 8.923711901306241,
      "grad_norm": 1.4453125,
      "learning_rate": 6.295485351970475e-05,
      "loss": 4.8124,
      "mean_token_accuracy": 0.2332911640405655,
      "num_tokens": 204164488.0,
      "step": 98375
    },
    {
      "entropy": 5.6715521812438965,
      "epoch": 8.924165457184325,
      "grad_norm": 1.4609375,
      "learning_rate": 6.294403496270793e-05,
      "loss": 4.7186,
      "mean_token_accuracy": 0.2465948775410652,
      "num_tokens": 204174814.0,
      "step": 98380
    },
    {
      "entropy": 5.706851387023926,
      "epoch": 8.92461901306241,
      "grad_norm": 1.546875,
      "learning_rate": 6.29332207910825e-05,
      "loss": 4.8086,
      "mean_token_accuracy": 0.2323472797870636,
      "num_tokens": 204184772.0,
      "step": 98385
    },
    {
      "entropy": 5.719767904281616,
      "epoch": 8.925072568940493,
      "grad_norm": 1.59375,
      "learning_rate": 6.292241100505207e-05,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2268697738647461,
      "num_tokens": 204194843.0,
      "step": 98390
    },
    {
      "entropy": 5.654260778427124,
      "epoch": 8.925526124818578,
      "grad_norm": 1.640625,
      "learning_rate": 6.291160560484026e-05,
      "loss": 4.7609,
      "mean_token_accuracy": 0.24163622111082078,
      "num_tokens": 204204817.0,
      "step": 98395
    },
    {
      "entropy": 5.674897384643555,
      "epoch": 8.925979680696662,
      "grad_norm": 1.4140625,
      "learning_rate": 6.290080459067045e-05,
      "loss": 4.7576,
      "mean_token_accuracy": 0.23944975286722184,
      "num_tokens": 204214735.0,
      "step": 98400
    },
    {
      "entropy": 5.6911579132080075,
      "epoch": 8.926433236574747,
      "grad_norm": 1.5078125,
      "learning_rate": 6.289000796276608e-05,
      "loss": 4.7763,
      "mean_token_accuracy": 0.23380666524171828,
      "num_tokens": 204224808.0,
      "step": 98405
    },
    {
      "entropy": 5.6455176830291744,
      "epoch": 8.92688679245283,
      "grad_norm": 1.625,
      "learning_rate": 6.28792157213504e-05,
      "loss": 4.8325,
      "mean_token_accuracy": 0.23674753308296204,
      "num_tokens": 204234767.0,
      "step": 98410
    },
    {
      "entropy": 5.739429330825805,
      "epoch": 8.927340348330915,
      "grad_norm": 1.4140625,
      "learning_rate": 6.286842786664661e-05,
      "loss": 4.8763,
      "mean_token_accuracy": 0.23224921077489852,
      "num_tokens": 204245482.0,
      "step": 98415
    },
    {
      "entropy": 5.686124181747436,
      "epoch": 8.927793904208999,
      "grad_norm": 1.546875,
      "learning_rate": 6.285764439887776e-05,
      "loss": 4.7753,
      "mean_token_accuracy": 0.24003576934337617,
      "num_tokens": 204256365.0,
      "step": 98420
    },
    {
      "entropy": 5.7316083908081055,
      "epoch": 8.928247460087082,
      "grad_norm": 1.53125,
      "learning_rate": 6.284686531826688e-05,
      "loss": 4.8307,
      "mean_token_accuracy": 0.2353966936469078,
      "num_tokens": 204266912.0,
      "step": 98425
    },
    {
      "entropy": 5.6877930641174315,
      "epoch": 8.928701015965167,
      "grad_norm": 1.6796875,
      "learning_rate": 6.283609062503694e-05,
      "loss": 4.8052,
      "mean_token_accuracy": 0.2405453234910965,
      "num_tokens": 204277335.0,
      "step": 98430
    },
    {
      "entropy": 5.64551796913147,
      "epoch": 8.92915457184325,
      "grad_norm": 1.5703125,
      "learning_rate": 6.282532031941064e-05,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2464692085981369,
      "num_tokens": 204287537.0,
      "step": 98435
    },
    {
      "entropy": 5.673730278015137,
      "epoch": 8.929608127721336,
      "grad_norm": 1.546875,
      "learning_rate": 6.281455440161081e-05,
      "loss": 4.7926,
      "mean_token_accuracy": 0.2362107142806053,
      "num_tokens": 204297568.0,
      "step": 98440
    },
    {
      "entropy": 5.713086080551148,
      "epoch": 8.930061683599419,
      "grad_norm": 1.5859375,
      "learning_rate": 6.280379287186009e-05,
      "loss": 4.7613,
      "mean_token_accuracy": 0.24698010385036467,
      "num_tokens": 204306992.0,
      "step": 98445
    },
    {
      "entropy": 5.733016777038574,
      "epoch": 8.930515239477504,
      "grad_norm": 1.4609375,
      "learning_rate": 6.279303573038101e-05,
      "loss": 4.8024,
      "mean_token_accuracy": 0.22916921824216843,
      "num_tokens": 204317385.0,
      "step": 98450
    },
    {
      "entropy": 5.69451150894165,
      "epoch": 8.930968795355588,
      "grad_norm": 1.546875,
      "learning_rate": 6.278228297739599e-05,
      "loss": 4.808,
      "mean_token_accuracy": 0.23236481547355653,
      "num_tokens": 204327205.0,
      "step": 98455
    },
    {
      "entropy": 5.672597789764405,
      "epoch": 8.931422351233673,
      "grad_norm": 1.484375,
      "learning_rate": 6.277153461312749e-05,
      "loss": 4.7426,
      "mean_token_accuracy": 0.24251834005117417,
      "num_tokens": 204337139.0,
      "step": 98460
    },
    {
      "entropy": 5.690902757644653,
      "epoch": 8.931875907111756,
      "grad_norm": 1.515625,
      "learning_rate": 6.27607906377977e-05,
      "loss": 4.7673,
      "mean_token_accuracy": 0.2411469802260399,
      "num_tokens": 204347673.0,
      "step": 98465
    },
    {
      "entropy": 5.721477127075195,
      "epoch": 8.93232946298984,
      "grad_norm": 1.5,
      "learning_rate": 6.275005105162883e-05,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22409103512763978,
      "num_tokens": 204359122.0,
      "step": 98470
    },
    {
      "entropy": 5.683357763290405,
      "epoch": 8.932783018867925,
      "grad_norm": 1.515625,
      "learning_rate": 6.273931585484301e-05,
      "loss": 4.7894,
      "mean_token_accuracy": 0.23698287308216096,
      "num_tokens": 204369420.0,
      "step": 98475
    },
    {
      "entropy": 5.73177318572998,
      "epoch": 8.933236574746008,
      "grad_norm": 1.4609375,
      "learning_rate": 6.272858504766223e-05,
      "loss": 4.847,
      "mean_token_accuracy": 0.2381119340658188,
      "num_tokens": 204379373.0,
      "step": 98480
    },
    {
      "entropy": 5.630163288116455,
      "epoch": 8.933690130624093,
      "grad_norm": 1.4921875,
      "learning_rate": 6.27178586303084e-05,
      "loss": 4.7311,
      "mean_token_accuracy": 0.24227388799190522,
      "num_tokens": 204388749.0,
      "step": 98485
    },
    {
      "entropy": 5.677283430099488,
      "epoch": 8.934143686502177,
      "grad_norm": 1.625,
      "learning_rate": 6.270713660300336e-05,
      "loss": 4.7878,
      "mean_token_accuracy": 0.2451644092798233,
      "num_tokens": 204399180.0,
      "step": 98490
    },
    {
      "entropy": 5.703753709793091,
      "epoch": 8.934597242380262,
      "grad_norm": 1.578125,
      "learning_rate": 6.269641896596884e-05,
      "loss": 4.8342,
      "mean_token_accuracy": 0.23342112302780152,
      "num_tokens": 204409458.0,
      "step": 98495
    },
    {
      "entropy": 5.722093534469605,
      "epoch": 8.935050798258345,
      "grad_norm": 1.4296875,
      "learning_rate": 6.268570571942642e-05,
      "loss": 4.8451,
      "mean_token_accuracy": 0.23447066992521287,
      "num_tokens": 204419563.0,
      "step": 98500
    },
    {
      "entropy": 5.675559234619141,
      "epoch": 8.93550435413643,
      "grad_norm": 1.53125,
      "learning_rate": 6.267499686359778e-05,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23194992989301683,
      "num_tokens": 204429329.0,
      "step": 98505
    },
    {
      "entropy": 5.7511780738830565,
      "epoch": 8.935957910014514,
      "grad_norm": 1.5,
      "learning_rate": 6.266429239870428e-05,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2335835501551628,
      "num_tokens": 204440269.0,
      "step": 98510
    },
    {
      "entropy": 5.649885654449463,
      "epoch": 8.936411465892597,
      "grad_norm": 1.421875,
      "learning_rate": 6.265359232496734e-05,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23312572836875917,
      "num_tokens": 204450644.0,
      "step": 98515
    },
    {
      "entropy": 5.764443349838257,
      "epoch": 8.936865021770682,
      "grad_norm": 1.453125,
      "learning_rate": 6.264289664260821e-05,
      "loss": 4.8737,
      "mean_token_accuracy": 0.2296822562813759,
      "num_tokens": 204460044.0,
      "step": 98520
    },
    {
      "entropy": 5.753687238693237,
      "epoch": 8.937318577648766,
      "grad_norm": 1.640625,
      "learning_rate": 6.263220535184811e-05,
      "loss": 4.816,
      "mean_token_accuracy": 0.23574181646108627,
      "num_tokens": 204469993.0,
      "step": 98525
    },
    {
      "entropy": 5.70249080657959,
      "epoch": 8.93777213352685,
      "grad_norm": 1.515625,
      "learning_rate": 6.262151845290808e-05,
      "loss": 4.7602,
      "mean_token_accuracy": 0.2392909497022629,
      "num_tokens": 204480357.0,
      "step": 98530
    },
    {
      "entropy": 5.665193796157837,
      "epoch": 8.938225689404934,
      "grad_norm": 1.5859375,
      "learning_rate": 6.261083594600916e-05,
      "loss": 4.799,
      "mean_token_accuracy": 0.23538624197244645,
      "num_tokens": 204491333.0,
      "step": 98535
    },
    {
      "entropy": 5.728129482269287,
      "epoch": 8.93867924528302,
      "grad_norm": 1.46875,
      "learning_rate": 6.260015783137234e-05,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23102068305015563,
      "num_tokens": 204502315.0,
      "step": 98540
    },
    {
      "entropy": 5.716320514678955,
      "epoch": 8.939132801161103,
      "grad_norm": 1.671875,
      "learning_rate": 6.258948410921836e-05,
      "loss": 4.8511,
      "mean_token_accuracy": 0.2424340009689331,
      "num_tokens": 204512841.0,
      "step": 98545
    },
    {
      "entropy": 5.682498788833618,
      "epoch": 8.939586357039188,
      "grad_norm": 1.6328125,
      "learning_rate": 6.257881477976797e-05,
      "loss": 4.7554,
      "mean_token_accuracy": 0.24522095918655396,
      "num_tokens": 204522212.0,
      "step": 98550
    },
    {
      "entropy": 5.680400419235229,
      "epoch": 8.940039912917271,
      "grad_norm": 1.3984375,
      "learning_rate": 6.256814984324182e-05,
      "loss": 4.8302,
      "mean_token_accuracy": 0.2267964407801628,
      "num_tokens": 204534157.0,
      "step": 98555
    },
    {
      "entropy": 5.6657508373260494,
      "epoch": 8.940493468795356,
      "grad_norm": 1.625,
      "learning_rate": 6.255748929986047e-05,
      "loss": 4.7179,
      "mean_token_accuracy": 0.24065175354480745,
      "num_tokens": 204543724.0,
      "step": 98560
    },
    {
      "entropy": 5.689940452575684,
      "epoch": 8.94094702467344,
      "grad_norm": 1.3984375,
      "learning_rate": 6.254683314984437e-05,
      "loss": 4.7931,
      "mean_token_accuracy": 0.23967828452587128,
      "num_tokens": 204554856.0,
      "step": 98565
    },
    {
      "entropy": 5.6936592102050785,
      "epoch": 8.941400580551523,
      "grad_norm": 1.4140625,
      "learning_rate": 6.253618139341388e-05,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2384667456150055,
      "num_tokens": 204565985.0,
      "step": 98570
    },
    {
      "entropy": 5.691290998458863,
      "epoch": 8.941854136429608,
      "grad_norm": 1.4375,
      "learning_rate": 6.252553403078933e-05,
      "loss": 4.8185,
      "mean_token_accuracy": 0.24495327472686768,
      "num_tokens": 204575650.0,
      "step": 98575
    },
    {
      "entropy": 5.667672443389892,
      "epoch": 8.942307692307692,
      "grad_norm": 1.4921875,
      "learning_rate": 6.251489106219088e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.24197472035884857,
      "num_tokens": 204586165.0,
      "step": 98580
    },
    {
      "entropy": 5.68800048828125,
      "epoch": 8.942761248185777,
      "grad_norm": 1.53125,
      "learning_rate": 6.250425248783864e-05,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23662813007831573,
      "num_tokens": 204596880.0,
      "step": 98585
    },
    {
      "entropy": 5.694296884536743,
      "epoch": 8.94321480406386,
      "grad_norm": 1.46875,
      "learning_rate": 6.249361830795261e-05,
      "loss": 4.8979,
      "mean_token_accuracy": 0.23291819244623185,
      "num_tokens": 204608584.0,
      "step": 98590
    },
    {
      "entropy": 5.724698257446289,
      "epoch": 8.943668359941945,
      "grad_norm": 1.5234375,
      "learning_rate": 6.24829885227527e-05,
      "loss": 4.8523,
      "mean_token_accuracy": 0.22771232426166535,
      "num_tokens": 204618003.0,
      "step": 98595
    },
    {
      "entropy": 5.680739641189575,
      "epoch": 8.944121915820029,
      "grad_norm": 1.453125,
      "learning_rate": 6.247236313245874e-05,
      "loss": 4.7867,
      "mean_token_accuracy": 0.22941488921642303,
      "num_tokens": 204628297.0,
      "step": 98600
    },
    {
      "entropy": 5.713666009902954,
      "epoch": 8.944575471698114,
      "grad_norm": 1.515625,
      "learning_rate": 6.246174213729042e-05,
      "loss": 4.8263,
      "mean_token_accuracy": 0.23952470272779464,
      "num_tokens": 204638924.0,
      "step": 98605
    },
    {
      "entropy": 5.692615032196045,
      "epoch": 8.945029027576197,
      "grad_norm": 1.6015625,
      "learning_rate": 6.245112553746748e-05,
      "loss": 4.8063,
      "mean_token_accuracy": 0.2433663010597229,
      "num_tokens": 204648876.0,
      "step": 98610
    },
    {
      "entropy": 5.685476398468017,
      "epoch": 8.945482583454282,
      "grad_norm": 1.484375,
      "learning_rate": 6.244051333320943e-05,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23788537979125976,
      "num_tokens": 204658844.0,
      "step": 98615
    },
    {
      "entropy": 5.74513521194458,
      "epoch": 8.945936139332366,
      "grad_norm": 1.4921875,
      "learning_rate": 6.242990552473574e-05,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23196338266134262,
      "num_tokens": 204668378.0,
      "step": 98620
    },
    {
      "entropy": 5.718401718139648,
      "epoch": 8.94638969521045,
      "grad_norm": 1.6484375,
      "learning_rate": 6.241930211226576e-05,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23870845139026642,
      "num_tokens": 204678181.0,
      "step": 98625
    },
    {
      "entropy": 5.7401327133178714,
      "epoch": 8.946843251088534,
      "grad_norm": 1.484375,
      "learning_rate": 6.240870309601879e-05,
      "loss": 4.8615,
      "mean_token_accuracy": 0.2287153646349907,
      "num_tokens": 204688448.0,
      "step": 98630
    },
    {
      "entropy": 5.710846900939941,
      "epoch": 8.947296806966618,
      "grad_norm": 1.7109375,
      "learning_rate": 6.2398108476214e-05,
      "loss": 4.8574,
      "mean_token_accuracy": 0.2387128248810768,
      "num_tokens": 204697754.0,
      "step": 98635
    },
    {
      "entropy": 5.720006513595581,
      "epoch": 8.947750362844703,
      "grad_norm": 1.5234375,
      "learning_rate": 6.23875182530705e-05,
      "loss": 4.853,
      "mean_token_accuracy": 0.23890553414821625,
      "num_tokens": 204707758.0,
      "step": 98640
    },
    {
      "entropy": 5.6461536407470705,
      "epoch": 8.948203918722786,
      "grad_norm": 1.546875,
      "learning_rate": 6.237693242680732e-05,
      "loss": 4.7134,
      "mean_token_accuracy": 0.24522225111722945,
      "num_tokens": 204717617.0,
      "step": 98645
    },
    {
      "entropy": 5.68659987449646,
      "epoch": 8.948657474600871,
      "grad_norm": 1.6015625,
      "learning_rate": 6.236635099764336e-05,
      "loss": 4.752,
      "mean_token_accuracy": 0.23861591517925262,
      "num_tokens": 204727157.0,
      "step": 98650
    },
    {
      "entropy": 5.6935422897338865,
      "epoch": 8.949111030478955,
      "grad_norm": 1.5078125,
      "learning_rate": 6.235577396579744e-05,
      "loss": 4.7425,
      "mean_token_accuracy": 0.23980752825737,
      "num_tokens": 204736577.0,
      "step": 98655
    },
    {
      "entropy": 5.667499256134033,
      "epoch": 8.94956458635704,
      "grad_norm": 1.6171875,
      "learning_rate": 6.234520133148833e-05,
      "loss": 4.8016,
      "mean_token_accuracy": 0.236936292052269,
      "num_tokens": 204746985.0,
      "step": 98660
    },
    {
      "entropy": 5.706916809082031,
      "epoch": 8.950018142235123,
      "grad_norm": 1.453125,
      "learning_rate": 6.233463309493462e-05,
      "loss": 4.8041,
      "mean_token_accuracy": 0.2340630277991295,
      "num_tokens": 204757641.0,
      "step": 98665
    },
    {
      "entropy": 5.679394006729126,
      "epoch": 8.950471698113208,
      "grad_norm": 1.46875,
      "learning_rate": 6.232406925635488e-05,
      "loss": 4.8025,
      "mean_token_accuracy": 0.23825526386499404,
      "num_tokens": 204767832.0,
      "step": 98670
    },
    {
      "entropy": 5.74031138420105,
      "epoch": 8.950925253991292,
      "grad_norm": 1.4921875,
      "learning_rate": 6.231350981596762e-05,
      "loss": 4.8233,
      "mean_token_accuracy": 0.23535453528165817,
      "num_tokens": 204776839.0,
      "step": 98675
    },
    {
      "entropy": 5.694750165939331,
      "epoch": 8.951378809869375,
      "grad_norm": 1.53125,
      "learning_rate": 6.230295477399115e-05,
      "loss": 4.7337,
      "mean_token_accuracy": 0.24150235950946808,
      "num_tokens": 204786644.0,
      "step": 98680
    },
    {
      "entropy": 5.760227489471435,
      "epoch": 8.95183236574746,
      "grad_norm": 1.5859375,
      "learning_rate": 6.22924041306438e-05,
      "loss": 4.8741,
      "mean_token_accuracy": 0.23299241662025452,
      "num_tokens": 204797223.0,
      "step": 98685
    },
    {
      "entropy": 5.652632522583008,
      "epoch": 8.952285921625544,
      "grad_norm": 1.5390625,
      "learning_rate": 6.22818578861437e-05,
      "loss": 4.7712,
      "mean_token_accuracy": 0.2435129776597023,
      "num_tokens": 204807254.0,
      "step": 98690
    },
    {
      "entropy": 5.72221827507019,
      "epoch": 8.952739477503629,
      "grad_norm": 1.59375,
      "learning_rate": 6.227131604070901e-05,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23310426026582717,
      "num_tokens": 204817248.0,
      "step": 98695
    },
    {
      "entropy": 5.733519792556763,
      "epoch": 8.953193033381712,
      "grad_norm": 1.453125,
      "learning_rate": 6.226077859455769e-05,
      "loss": 4.8006,
      "mean_token_accuracy": 0.24292484372854234,
      "num_tokens": 204827608.0,
      "step": 98700
    },
    {
      "entropy": 5.708644390106201,
      "epoch": 8.953646589259797,
      "grad_norm": 1.5234375,
      "learning_rate": 6.22502455479077e-05,
      "loss": 4.8564,
      "mean_token_accuracy": 0.23433154672384263,
      "num_tokens": 204837923.0,
      "step": 98705
    },
    {
      "entropy": 5.653563928604126,
      "epoch": 8.95410014513788,
      "grad_norm": 1.4921875,
      "learning_rate": 6.223971690097685e-05,
      "loss": 4.7722,
      "mean_token_accuracy": 0.24552104473114014,
      "num_tokens": 204849249.0,
      "step": 98710
    },
    {
      "entropy": 5.624060583114624,
      "epoch": 8.954553701015966,
      "grad_norm": 1.5625,
      "learning_rate": 6.222919265398287e-05,
      "loss": 4.7273,
      "mean_token_accuracy": 0.24432912468910217,
      "num_tokens": 204859084.0,
      "step": 98715
    },
    {
      "entropy": 5.747919273376465,
      "epoch": 8.95500725689405,
      "grad_norm": 1.84375,
      "learning_rate": 6.22186728071434e-05,
      "loss": 4.8188,
      "mean_token_accuracy": 0.2291925758123398,
      "num_tokens": 204870097.0,
      "step": 98720
    },
    {
      "entropy": 5.675084114074707,
      "epoch": 8.955460812772134,
      "grad_norm": 1.46875,
      "learning_rate": 6.220815736067601e-05,
      "loss": 4.7453,
      "mean_token_accuracy": 0.23850262761116028,
      "num_tokens": 204879701.0,
      "step": 98725
    },
    {
      "entropy": 5.684398365020752,
      "epoch": 8.955914368650218,
      "grad_norm": 1.328125,
      "learning_rate": 6.219764631479815e-05,
      "loss": 4.8227,
      "mean_token_accuracy": 0.23179997205734254,
      "num_tokens": 204890586.0,
      "step": 98730
    },
    {
      "entropy": 5.737507343292236,
      "epoch": 8.956367924528301,
      "grad_norm": 1.390625,
      "learning_rate": 6.218713966972717e-05,
      "loss": 4.8593,
      "mean_token_accuracy": 0.22633358091115952,
      "num_tokens": 204900426.0,
      "step": 98735
    },
    {
      "entropy": 5.5515388488769535,
      "epoch": 8.956821480406386,
      "grad_norm": 1.65625,
      "learning_rate": 6.21766374256804e-05,
      "loss": 4.6583,
      "mean_token_accuracy": 0.24794104546308518,
      "num_tokens": 204911578.0,
      "step": 98740
    },
    {
      "entropy": 5.69596004486084,
      "epoch": 8.95727503628447,
      "grad_norm": 1.625,
      "learning_rate": 6.2166139582875e-05,
      "loss": 4.838,
      "mean_token_accuracy": 0.23784897476434708,
      "num_tokens": 204921566.0,
      "step": 98745
    },
    {
      "entropy": 5.602304840087891,
      "epoch": 8.957728592162555,
      "grad_norm": 1.4609375,
      "learning_rate": 6.215564614152806e-05,
      "loss": 4.712,
      "mean_token_accuracy": 0.2496707707643509,
      "num_tokens": 204931452.0,
      "step": 98750
    },
    {
      "entropy": 5.643250465393066,
      "epoch": 8.958182148040638,
      "grad_norm": 1.4921875,
      "learning_rate": 6.214515710185661e-05,
      "loss": 4.7002,
      "mean_token_accuracy": 0.24306079000234604,
      "num_tokens": 204941947.0,
      "step": 98755
    },
    {
      "entropy": 5.656443119049072,
      "epoch": 8.958635703918723,
      "grad_norm": 1.6015625,
      "learning_rate": 6.213467246407753e-05,
      "loss": 4.7347,
      "mean_token_accuracy": 0.23562268316745758,
      "num_tokens": 204952211.0,
      "step": 98760
    },
    {
      "entropy": 5.712298822402954,
      "epoch": 8.959089259796807,
      "grad_norm": 1.5390625,
      "learning_rate": 6.212419222840772e-05,
      "loss": 4.8912,
      "mean_token_accuracy": 0.23576502203941346,
      "num_tokens": 204963266.0,
      "step": 98765
    },
    {
      "entropy": 5.732945680618286,
      "epoch": 8.959542815674892,
      "grad_norm": 1.59375,
      "learning_rate": 6.211371639506376e-05,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23224656730890275,
      "num_tokens": 204973886.0,
      "step": 98770
    },
    {
      "entropy": 5.696603107452392,
      "epoch": 8.959996371552975,
      "grad_norm": 1.625,
      "learning_rate": 6.210324496426248e-05,
      "loss": 4.7791,
      "mean_token_accuracy": 0.23985721468925475,
      "num_tokens": 204983523.0,
      "step": 98775
    },
    {
      "entropy": 5.685753440856933,
      "epoch": 8.960449927431059,
      "grad_norm": 1.4296875,
      "learning_rate": 6.209277793622033e-05,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23183367103338243,
      "num_tokens": 204994073.0,
      "step": 98780
    },
    {
      "entropy": 5.706664562225342,
      "epoch": 8.960903483309144,
      "grad_norm": 1.6796875,
      "learning_rate": 6.208231531115378e-05,
      "loss": 4.75,
      "mean_token_accuracy": 0.24150972664356232,
      "num_tokens": 205003399.0,
      "step": 98785
    },
    {
      "entropy": 5.711711835861206,
      "epoch": 8.961357039187227,
      "grad_norm": 1.5546875,
      "learning_rate": 6.207185708927922e-05,
      "loss": 4.8282,
      "mean_token_accuracy": 0.234585802257061,
      "num_tokens": 205014152.0,
      "step": 98790
    },
    {
      "entropy": 5.7489725112915036,
      "epoch": 8.961810595065312,
      "grad_norm": 1.640625,
      "learning_rate": 6.20614032708129e-05,
      "loss": 4.861,
      "mean_token_accuracy": 0.24130417853593827,
      "num_tokens": 205025326.0,
      "step": 98795
    },
    {
      "entropy": 5.669000148773193,
      "epoch": 8.962264150943396,
      "grad_norm": 1.484375,
      "learning_rate": 6.205095385597102e-05,
      "loss": 4.7855,
      "mean_token_accuracy": 0.2361120879650116,
      "num_tokens": 205036564.0,
      "step": 98800
    },
    {
      "entropy": 5.693169450759887,
      "epoch": 8.962717706821481,
      "grad_norm": 1.53125,
      "learning_rate": 6.204050884496968e-05,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23665838092565536,
      "num_tokens": 205046720.0,
      "step": 98805
    },
    {
      "entropy": 5.648668050765991,
      "epoch": 8.963171262699564,
      "grad_norm": 1.5859375,
      "learning_rate": 6.20300682380249e-05,
      "loss": 4.8085,
      "mean_token_accuracy": 0.23505089730024337,
      "num_tokens": 205058270.0,
      "step": 98810
    },
    {
      "entropy": 5.715304231643676,
      "epoch": 8.96362481857765,
      "grad_norm": 1.5,
      "learning_rate": 6.201963203535259e-05,
      "loss": 4.8048,
      "mean_token_accuracy": 0.23296733647584916,
      "num_tokens": 205068687.0,
      "step": 98815
    },
    {
      "entropy": 5.737999963760376,
      "epoch": 8.964078374455733,
      "grad_norm": 1.484375,
      "learning_rate": 6.200920023716853e-05,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2317593902349472,
      "num_tokens": 205078357.0,
      "step": 98820
    },
    {
      "entropy": 5.750287389755249,
      "epoch": 8.964531930333816,
      "grad_norm": 1.4453125,
      "learning_rate": 6.199877284368851e-05,
      "loss": 4.931,
      "mean_token_accuracy": 0.22654947638511658,
      "num_tokens": 205089379.0,
      "step": 98825
    },
    {
      "entropy": 5.676682758331299,
      "epoch": 8.964985486211901,
      "grad_norm": 1.5859375,
      "learning_rate": 6.198834985512815e-05,
      "loss": 4.7797,
      "mean_token_accuracy": 0.23992180228233337,
      "num_tokens": 205099584.0,
      "step": 98830
    },
    {
      "entropy": 5.671371221542358,
      "epoch": 8.965439042089985,
      "grad_norm": 1.546875,
      "learning_rate": 6.197793127170296e-05,
      "loss": 4.8147,
      "mean_token_accuracy": 0.23742879182100296,
      "num_tokens": 205110636.0,
      "step": 98835
    },
    {
      "entropy": 5.722956991195678,
      "epoch": 8.96589259796807,
      "grad_norm": 1.5390625,
      "learning_rate": 6.196751709362848e-05,
      "loss": 4.9023,
      "mean_token_accuracy": 0.2342892289161682,
      "num_tokens": 205121812.0,
      "step": 98840
    },
    {
      "entropy": 5.671295261383056,
      "epoch": 8.966346153846153,
      "grad_norm": 1.5078125,
      "learning_rate": 6.195710732112001e-05,
      "loss": 4.8198,
      "mean_token_accuracy": 0.23945742547512056,
      "num_tokens": 205132014.0,
      "step": 98845
    },
    {
      "entropy": 5.708183145523071,
      "epoch": 8.966799709724238,
      "grad_norm": 1.6015625,
      "learning_rate": 6.194670195439287e-05,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23706755191087722,
      "num_tokens": 205141807.0,
      "step": 98850
    },
    {
      "entropy": 5.777433109283447,
      "epoch": 8.967253265602322,
      "grad_norm": 1.609375,
      "learning_rate": 6.19363009936622e-05,
      "loss": 4.9121,
      "mean_token_accuracy": 0.23037055134773254,
      "num_tokens": 205153217.0,
      "step": 98855
    },
    {
      "entropy": 5.740661239624023,
      "epoch": 8.967706821480407,
      "grad_norm": 1.4765625,
      "learning_rate": 6.192590443914314e-05,
      "loss": 4.8645,
      "mean_token_accuracy": 0.23638711720705033,
      "num_tokens": 205163403.0,
      "step": 98860
    },
    {
      "entropy": 5.699515438079834,
      "epoch": 8.96816037735849,
      "grad_norm": 1.5703125,
      "learning_rate": 6.191551229105068e-05,
      "loss": 4.8278,
      "mean_token_accuracy": 0.23340329080820083,
      "num_tokens": 205173549.0,
      "step": 98865
    },
    {
      "entropy": 5.654272842407226,
      "epoch": 8.968613933236576,
      "grad_norm": 1.4921875,
      "learning_rate": 6.190512454959969e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.23299044668674468,
      "num_tokens": 205183984.0,
      "step": 98870
    },
    {
      "entropy": 5.689266300201416,
      "epoch": 8.969067489114659,
      "grad_norm": 1.609375,
      "learning_rate": 6.189474121500506e-05,
      "loss": 4.782,
      "mean_token_accuracy": 0.23786310255527496,
      "num_tokens": 205194009.0,
      "step": 98875
    },
    {
      "entropy": 5.700791835784912,
      "epoch": 8.969521044992742,
      "grad_norm": 1.390625,
      "learning_rate": 6.188436228748149e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2328411266207695,
      "num_tokens": 205204867.0,
      "step": 98880
    },
    {
      "entropy": 5.703082942962647,
      "epoch": 8.969974600870827,
      "grad_norm": 1.625,
      "learning_rate": 6.187398776724359e-05,
      "loss": 4.772,
      "mean_token_accuracy": 0.24387530088424683,
      "num_tokens": 205214124.0,
      "step": 98885
    },
    {
      "entropy": 5.700757551193237,
      "epoch": 8.97042815674891,
      "grad_norm": 1.375,
      "learning_rate": 6.186361765450595e-05,
      "loss": 4.8393,
      "mean_token_accuracy": 0.23605289608240126,
      "num_tokens": 205225356.0,
      "step": 98890
    },
    {
      "entropy": 5.701473999023437,
      "epoch": 8.970881712626996,
      "grad_norm": 1.6171875,
      "learning_rate": 6.185325194948302e-05,
      "loss": 4.7947,
      "mean_token_accuracy": 0.23690915405750274,
      "num_tokens": 205235631.0,
      "step": 98895
    },
    {
      "entropy": 5.6818609714508055,
      "epoch": 8.97133526850508,
      "grad_norm": 1.7734375,
      "learning_rate": 6.184289065238916e-05,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23908189833164215,
      "num_tokens": 205245648.0,
      "step": 98900
    },
    {
      "entropy": 5.67523717880249,
      "epoch": 8.971788824383164,
      "grad_norm": 1.609375,
      "learning_rate": 6.183253376343859e-05,
      "loss": 4.7357,
      "mean_token_accuracy": 0.23937047272920609,
      "num_tokens": 205255519.0,
      "step": 98905
    },
    {
      "entropy": 5.661187267303466,
      "epoch": 8.972242380261248,
      "grad_norm": 1.484375,
      "learning_rate": 6.182218128284557e-05,
      "loss": 4.7672,
      "mean_token_accuracy": 0.24218235313892364,
      "num_tokens": 205266546.0,
      "step": 98910
    },
    {
      "entropy": 5.702623462677002,
      "epoch": 8.972695936139333,
      "grad_norm": 1.53125,
      "learning_rate": 6.181183321082417e-05,
      "loss": 4.837,
      "mean_token_accuracy": 0.23634012788534164,
      "num_tokens": 205276582.0,
      "step": 98915
    },
    {
      "entropy": 5.665572261810302,
      "epoch": 8.973149492017416,
      "grad_norm": 1.6484375,
      "learning_rate": 6.180148954758842e-05,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2306506484746933,
      "num_tokens": 205287591.0,
      "step": 98920
    },
    {
      "entropy": 5.716396760940552,
      "epoch": 8.973603047895502,
      "grad_norm": 1.578125,
      "learning_rate": 6.179115029335218e-05,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23969324678182602,
      "num_tokens": 205297109.0,
      "step": 98925
    },
    {
      "entropy": 5.658834314346313,
      "epoch": 8.974056603773585,
      "grad_norm": 1.453125,
      "learning_rate": 6.178081544832926e-05,
      "loss": 4.702,
      "mean_token_accuracy": 0.24413192868232728,
      "num_tokens": 205308011.0,
      "step": 98930
    },
    {
      "entropy": 5.685218858718872,
      "epoch": 8.974510159651668,
      "grad_norm": 1.5703125,
      "learning_rate": 6.177048501273343e-05,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2399345561861992,
      "num_tokens": 205318127.0,
      "step": 98935
    },
    {
      "entropy": 5.64957823753357,
      "epoch": 8.974963715529753,
      "grad_norm": 1.5078125,
      "learning_rate": 6.176015898677827e-05,
      "loss": 4.7793,
      "mean_token_accuracy": 0.23777881264686584,
      "num_tokens": 205329697.0,
      "step": 98940
    },
    {
      "entropy": 5.7233898639678955,
      "epoch": 8.975417271407837,
      "grad_norm": 1.4453125,
      "learning_rate": 6.174983737067738e-05,
      "loss": 4.8142,
      "mean_token_accuracy": 0.23697328418493271,
      "num_tokens": 205340425.0,
      "step": 98945
    },
    {
      "entropy": 5.6295814037323,
      "epoch": 8.975870827285922,
      "grad_norm": 1.6171875,
      "learning_rate": 6.173952016464423e-05,
      "loss": 4.6733,
      "mean_token_accuracy": 0.25352311283349993,
      "num_tokens": 205350097.0,
      "step": 98950
    },
    {
      "entropy": 5.7016847133636475,
      "epoch": 8.976324383164005,
      "grad_norm": 1.453125,
      "learning_rate": 6.172920736889211e-05,
      "loss": 4.7974,
      "mean_token_accuracy": 0.2352410227060318,
      "num_tokens": 205360556.0,
      "step": 98955
    },
    {
      "entropy": 5.676326656341553,
      "epoch": 8.97677793904209,
      "grad_norm": 1.5078125,
      "learning_rate": 6.171889898363437e-05,
      "loss": 4.7862,
      "mean_token_accuracy": 0.23583896458148956,
      "num_tokens": 205370398.0,
      "step": 98960
    },
    {
      "entropy": 5.704813623428345,
      "epoch": 8.977231494920174,
      "grad_norm": 1.609375,
      "learning_rate": 6.170859500908412e-05,
      "loss": 4.8502,
      "mean_token_accuracy": 0.23385756611824035,
      "num_tokens": 205381589.0,
      "step": 98965
    },
    {
      "entropy": 5.651411342620849,
      "epoch": 8.977685050798259,
      "grad_norm": 1.484375,
      "learning_rate": 6.169829544545448e-05,
      "loss": 4.7375,
      "mean_token_accuracy": 0.24603208601474763,
      "num_tokens": 205391790.0,
      "step": 98970
    },
    {
      "entropy": 5.705541372299194,
      "epoch": 8.978138606676342,
      "grad_norm": 1.5390625,
      "learning_rate": 6.168800029295842e-05,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23222616463899612,
      "num_tokens": 205402137.0,
      "step": 98975
    },
    {
      "entropy": 5.631842565536499,
      "epoch": 8.978592162554428,
      "grad_norm": 1.5703125,
      "learning_rate": 6.16777095518089e-05,
      "loss": 4.7098,
      "mean_token_accuracy": 0.24494522213935851,
      "num_tokens": 205412128.0,
      "step": 98980
    },
    {
      "entropy": 5.630285167694092,
      "epoch": 8.979045718432511,
      "grad_norm": 1.671875,
      "learning_rate": 6.166742322221872e-05,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2340015545487404,
      "num_tokens": 205421537.0,
      "step": 98985
    },
    {
      "entropy": 5.714199876785278,
      "epoch": 8.979499274310594,
      "grad_norm": 1.453125,
      "learning_rate": 6.165714130440059e-05,
      "loss": 4.8116,
      "mean_token_accuracy": 0.23061299473047256,
      "num_tokens": 205433335.0,
      "step": 98990
    },
    {
      "entropy": 5.651383590698242,
      "epoch": 8.97995283018868,
      "grad_norm": 1.6015625,
      "learning_rate": 6.164686379856714e-05,
      "loss": 4.7513,
      "mean_token_accuracy": 0.24282229840755462,
      "num_tokens": 205442829.0,
      "step": 98995
    },
    {
      "entropy": 5.727269124984741,
      "epoch": 8.980406386066763,
      "grad_norm": 1.578125,
      "learning_rate": 6.16365907049309e-05,
      "loss": 4.9344,
      "mean_token_accuracy": 0.22747056931257248,
      "num_tokens": 205452667.0,
      "step": 99000
    },
    {
      "epoch": 8.980406386066763,
      "eval_entropy": 5.3918111889610465,
      "eval_loss": 4.90824031829834,
      "eval_mean_token_accuracy": 0.23883985562663002,
      "eval_num_tokens": 205452667.0,
      "eval_runtime": 20.7261,
      "eval_samples_per_second": 1706.059,
      "eval_steps_per_second": 213.257,
      "step": 99000
    },
    {
      "entropy": 5.650223302841186,
      "epoch": 8.980859941944848,
      "grad_norm": 1.5,
      "learning_rate": 6.16263220237043e-05,
      "loss": 4.7633,
      "mean_token_accuracy": 0.24300053119659423,
      "num_tokens": 205463417.0,
      "step": 99005
    },
    {
      "entropy": 5.739115524291992,
      "epoch": 8.981313497822931,
      "grad_norm": 1.6328125,
      "learning_rate": 6.161605775509979e-05,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23468674570322037,
      "num_tokens": 205473822.0,
      "step": 99010
    },
    {
      "entropy": 5.693098402023315,
      "epoch": 8.981767053701017,
      "grad_norm": 1.546875,
      "learning_rate": 6.160579789932958e-05,
      "loss": 4.8402,
      "mean_token_accuracy": 0.23034343123435974,
      "num_tokens": 205484667.0,
      "step": 99015
    },
    {
      "entropy": 5.632887601852417,
      "epoch": 8.9822206095791,
      "grad_norm": 1.4375,
      "learning_rate": 6.159554245660583e-05,
      "loss": 4.7566,
      "mean_token_accuracy": 0.24207305461168288,
      "num_tokens": 205495352.0,
      "step": 99020
    },
    {
      "entropy": 5.670588636398316,
      "epoch": 8.982674165457185,
      "grad_norm": 1.734375,
      "learning_rate": 6.158529142714068e-05,
      "loss": 4.7193,
      "mean_token_accuracy": 0.25243948549032214,
      "num_tokens": 205505716.0,
      "step": 99025
    },
    {
      "entropy": 5.719659805297852,
      "epoch": 8.983127721335268,
      "grad_norm": 1.5703125,
      "learning_rate": 6.157504481114606e-05,
      "loss": 4.8777,
      "mean_token_accuracy": 0.23350590467453003,
      "num_tokens": 205515496.0,
      "step": 99030
    },
    {
      "entropy": 5.657127952575683,
      "epoch": 8.983581277213352,
      "grad_norm": 1.46875,
      "learning_rate": 6.156480260883392e-05,
      "loss": 4.7798,
      "mean_token_accuracy": 0.23818355649709702,
      "num_tokens": 205525973.0,
      "step": 99035
    },
    {
      "entropy": 5.701015949249268,
      "epoch": 8.984034833091437,
      "grad_norm": 1.5234375,
      "learning_rate": 6.1554564820416e-05,
      "loss": 4.8069,
      "mean_token_accuracy": 0.24027952253818513,
      "num_tokens": 205535577.0,
      "step": 99040
    },
    {
      "entropy": 5.679419136047363,
      "epoch": 8.98448838896952,
      "grad_norm": 1.71875,
      "learning_rate": 6.15443314461041e-05,
      "loss": 4.7974,
      "mean_token_accuracy": 0.22774591892957688,
      "num_tokens": 205545775.0,
      "step": 99045
    },
    {
      "entropy": 5.674267768859863,
      "epoch": 8.984941944847606,
      "grad_norm": 1.6328125,
      "learning_rate": 6.153410248610983e-05,
      "loss": 4.7381,
      "mean_token_accuracy": 0.2431922361254692,
      "num_tokens": 205555403.0,
      "step": 99050
    },
    {
      "entropy": 5.629307222366333,
      "epoch": 8.985395500725689,
      "grad_norm": 1.46875,
      "learning_rate": 6.152387794064472e-05,
      "loss": 4.7383,
      "mean_token_accuracy": 0.2400689899921417,
      "num_tokens": 205564931.0,
      "step": 99055
    },
    {
      "entropy": 5.659494829177857,
      "epoch": 8.985849056603774,
      "grad_norm": 1.59375,
      "learning_rate": 6.151365780992023e-05,
      "loss": 4.8103,
      "mean_token_accuracy": 0.23199834674596786,
      "num_tokens": 205575443.0,
      "step": 99060
    },
    {
      "entropy": 5.749416017532349,
      "epoch": 8.986302612481857,
      "grad_norm": 1.734375,
      "learning_rate": 6.150344209414766e-05,
      "loss": 4.8813,
      "mean_token_accuracy": 0.23122409582138062,
      "num_tokens": 205584711.0,
      "step": 99065
    },
    {
      "entropy": 5.685802268981933,
      "epoch": 8.986756168359943,
      "grad_norm": 1.453125,
      "learning_rate": 6.149323079353834e-05,
      "loss": 4.7832,
      "mean_token_accuracy": 0.24147934317588807,
      "num_tokens": 205595563.0,
      "step": 99070
    },
    {
      "entropy": 5.687224912643432,
      "epoch": 8.987209724238026,
      "grad_norm": 1.671875,
      "learning_rate": 6.148302390830338e-05,
      "loss": 4.8062,
      "mean_token_accuracy": 0.2351991981267929,
      "num_tokens": 205605743.0,
      "step": 99075
    },
    {
      "entropy": 5.717566108703613,
      "epoch": 8.98766328011611,
      "grad_norm": 1.4765625,
      "learning_rate": 6.147282143865391e-05,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2348471537232399,
      "num_tokens": 205617662.0,
      "step": 99080
    },
    {
      "entropy": 5.728404998779297,
      "epoch": 8.988116835994195,
      "grad_norm": 1.46875,
      "learning_rate": 6.146262338480094e-05,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2410639241337776,
      "num_tokens": 205629086.0,
      "step": 99085
    },
    {
      "entropy": 5.630046367645264,
      "epoch": 8.988570391872278,
      "grad_norm": 1.625,
      "learning_rate": 6.145242974695531e-05,
      "loss": 4.6868,
      "mean_token_accuracy": 0.24642462879419327,
      "num_tokens": 205639316.0,
      "step": 99090
    },
    {
      "entropy": 5.682599306106567,
      "epoch": 8.989023947750363,
      "grad_norm": 1.40625,
      "learning_rate": 6.14422405253279e-05,
      "loss": 4.8573,
      "mean_token_accuracy": 0.22760229110717772,
      "num_tokens": 205650472.0,
      "step": 99095
    },
    {
      "entropy": 5.721733283996582,
      "epoch": 8.989477503628446,
      "grad_norm": 1.59375,
      "learning_rate": 6.143205572012929e-05,
      "loss": 4.8467,
      "mean_token_accuracy": 0.24022649675607682,
      "num_tokens": 205661558.0,
      "step": 99100
    },
    {
      "entropy": 5.655652236938477,
      "epoch": 8.989931059506532,
      "grad_norm": 1.609375,
      "learning_rate": 6.142187533157024e-05,
      "loss": 4.7243,
      "mean_token_accuracy": 0.24860402792692185,
      "num_tokens": 205670695.0,
      "step": 99105
    },
    {
      "entropy": 5.726513624191284,
      "epoch": 8.990384615384615,
      "grad_norm": 1.546875,
      "learning_rate": 6.141169935986124e-05,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2420755982398987,
      "num_tokens": 205681061.0,
      "step": 99110
    },
    {
      "entropy": 5.704431962966919,
      "epoch": 8.9908381712627,
      "grad_norm": 1.5625,
      "learning_rate": 6.140152780521272e-05,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2397311583161354,
      "num_tokens": 205690845.0,
      "step": 99115
    },
    {
      "entropy": 5.717966794967651,
      "epoch": 8.991291727140784,
      "grad_norm": 1.4375,
      "learning_rate": 6.139136066783504e-05,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22572096288204194,
      "num_tokens": 205700742.0,
      "step": 99120
    },
    {
      "entropy": 5.63898286819458,
      "epoch": 8.991745283018869,
      "grad_norm": 1.4375,
      "learning_rate": 6.138119794793844e-05,
      "loss": 4.7322,
      "mean_token_accuracy": 0.239504611492157,
      "num_tokens": 205711424.0,
      "step": 99125
    },
    {
      "entropy": 5.740510368347168,
      "epoch": 8.992198838896952,
      "grad_norm": 1.6484375,
      "learning_rate": 6.13710396457331e-05,
      "loss": 4.8577,
      "mean_token_accuracy": 0.24077919572591783,
      "num_tokens": 205721027.0,
      "step": 99130
    },
    {
      "entropy": 5.705829715728759,
      "epoch": 8.992652394775035,
      "grad_norm": 1.5625,
      "learning_rate": 6.13608857614291e-05,
      "loss": 4.8083,
      "mean_token_accuracy": 0.23720188587903976,
      "num_tokens": 205730884.0,
      "step": 99135
    },
    {
      "entropy": 5.612936973571777,
      "epoch": 8.99310595065312,
      "grad_norm": 1.484375,
      "learning_rate": 6.135073629523642e-05,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2551376298069954,
      "num_tokens": 205741334.0,
      "step": 99140
    },
    {
      "entropy": 5.654494571685791,
      "epoch": 8.993559506531204,
      "grad_norm": 1.6796875,
      "learning_rate": 6.134059124736497e-05,
      "loss": 4.7534,
      "mean_token_accuracy": 0.23874111473560333,
      "num_tokens": 205751024.0,
      "step": 99145
    },
    {
      "entropy": 5.700309228897095,
      "epoch": 8.994013062409289,
      "grad_norm": 1.625,
      "learning_rate": 6.133045061802453e-05,
      "loss": 4.818,
      "mean_token_accuracy": 0.23340051621198654,
      "num_tokens": 205761157.0,
      "step": 99150
    },
    {
      "entropy": 5.57951135635376,
      "epoch": 8.994466618287372,
      "grad_norm": 1.5390625,
      "learning_rate": 6.132031440742485e-05,
      "loss": 4.6937,
      "mean_token_accuracy": 0.24841332137584687,
      "num_tokens": 205771869.0,
      "step": 99155
    },
    {
      "entropy": 5.700020599365234,
      "epoch": 8.994920174165458,
      "grad_norm": 1.453125,
      "learning_rate": 6.131018261577551e-05,
      "loss": 4.7722,
      "mean_token_accuracy": 0.24189165830612183,
      "num_tokens": 205782066.0,
      "step": 99160
    },
    {
      "entropy": 5.697604703903198,
      "epoch": 8.995373730043541,
      "grad_norm": 1.609375,
      "learning_rate": 6.130005524328603e-05,
      "loss": 4.8356,
      "mean_token_accuracy": 0.23529251515865326,
      "num_tokens": 205791792.0,
      "step": 99165
    },
    {
      "entropy": 5.7116889476776125,
      "epoch": 8.995827285921626,
      "grad_norm": 1.65625,
      "learning_rate": 6.128993229016583e-05,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22781676054000854,
      "num_tokens": 205801597.0,
      "step": 99170
    },
    {
      "entropy": 5.650073385238647,
      "epoch": 8.99628084179971,
      "grad_norm": 1.53125,
      "learning_rate": 6.127981375662433e-05,
      "loss": 4.702,
      "mean_token_accuracy": 0.24998276978731154,
      "num_tokens": 205811755.0,
      "step": 99175
    },
    {
      "entropy": 5.671375894546509,
      "epoch": 8.996734397677795,
      "grad_norm": 1.3984375,
      "learning_rate": 6.126969964287078e-05,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24785353243350983,
      "num_tokens": 205821849.0,
      "step": 99180
    },
    {
      "entropy": 5.7383500099182125,
      "epoch": 8.997187953555878,
      "grad_norm": 1.640625,
      "learning_rate": 6.125958994911426e-05,
      "loss": 4.8573,
      "mean_token_accuracy": 0.23228156864643096,
      "num_tokens": 205832056.0,
      "step": 99185
    },
    {
      "entropy": 5.67652473449707,
      "epoch": 8.997641509433961,
      "grad_norm": 1.4453125,
      "learning_rate": 6.124948467556392e-05,
      "loss": 4.7295,
      "mean_token_accuracy": 0.2479681998491287,
      "num_tokens": 205842710.0,
      "step": 99190
    },
    {
      "entropy": 5.729760551452637,
      "epoch": 8.998095065312047,
      "grad_norm": 1.40625,
      "learning_rate": 6.123938382242869e-05,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2334763988852501,
      "num_tokens": 205853228.0,
      "step": 99195
    },
    {
      "entropy": 5.674992895126342,
      "epoch": 8.99854862119013,
      "grad_norm": 1.546875,
      "learning_rate": 6.122928738991747e-05,
      "loss": 4.7874,
      "mean_token_accuracy": 0.2409123569726944,
      "num_tokens": 205863726.0,
      "step": 99200
    },
    {
      "entropy": 5.6535602569580075,
      "epoch": 8.999002177068215,
      "grad_norm": 1.53125,
      "learning_rate": 6.121919537823902e-05,
      "loss": 4.6888,
      "mean_token_accuracy": 0.2415306895971298,
      "num_tokens": 205873368.0,
      "step": 99205
    },
    {
      "entropy": 5.672261762619018,
      "epoch": 8.999455732946299,
      "grad_norm": 1.4921875,
      "learning_rate": 6.120910778760213e-05,
      "loss": 4.7785,
      "mean_token_accuracy": 0.23829688727855683,
      "num_tokens": 205882546.0,
      "step": 99210
    },
    {
      "entropy": 5.65552396774292,
      "epoch": 8.999909288824384,
      "grad_norm": 1.453125,
      "learning_rate": 6.119902461821539e-05,
      "loss": 4.753,
      "mean_token_accuracy": 0.23880380243062974,
      "num_tokens": 205893090.0,
      "step": 99215
    },
    {
      "entropy": 5.7065027236938475,
      "epoch": 9.000362844702467,
      "grad_norm": 1.515625,
      "learning_rate": 6.11889458702873e-05,
      "loss": 4.8255,
      "mean_token_accuracy": 0.238831090927124,
      "num_tokens": 205902847.0,
      "step": 99220
    },
    {
      "entropy": 5.624173784255982,
      "epoch": 9.000816400580552,
      "grad_norm": 1.4921875,
      "learning_rate": 6.11788715440263e-05,
      "loss": 4.7513,
      "mean_token_accuracy": 0.23749978840351105,
      "num_tokens": 205913443.0,
      "step": 99225
    },
    {
      "entropy": 5.690079879760742,
      "epoch": 9.001269956458636,
      "grad_norm": 1.4921875,
      "learning_rate": 6.11688016396407e-05,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23157079070806502,
      "num_tokens": 205923264.0,
      "step": 99230
    },
    {
      "entropy": 5.656507873535157,
      "epoch": 9.00172351233672,
      "grad_norm": 1.671875,
      "learning_rate": 6.115873615733877e-05,
      "loss": 4.6994,
      "mean_token_accuracy": 0.2462252840399742,
      "num_tokens": 205932700.0,
      "step": 99235
    },
    {
      "entropy": 5.655551862716675,
      "epoch": 9.002177068214804,
      "grad_norm": 1.546875,
      "learning_rate": 6.114867509732866e-05,
      "loss": 4.7181,
      "mean_token_accuracy": 0.24109613448381423,
      "num_tokens": 205942949.0,
      "step": 99240
    },
    {
      "entropy": 5.706796789169312,
      "epoch": 9.002630624092888,
      "grad_norm": 1.4453125,
      "learning_rate": 6.113861845981846e-05,
      "loss": 4.758,
      "mean_token_accuracy": 0.23630611300468446,
      "num_tokens": 205952518.0,
      "step": 99245
    },
    {
      "entropy": 5.635228252410888,
      "epoch": 9.003084179970973,
      "grad_norm": 1.5234375,
      "learning_rate": 6.112856624501614e-05,
      "loss": 4.7306,
      "mean_token_accuracy": 0.24215932190418243,
      "num_tokens": 205962814.0,
      "step": 99250
    },
    {
      "entropy": 5.665156650543213,
      "epoch": 9.003537735849056,
      "grad_norm": 1.5859375,
      "learning_rate": 6.111851845312957e-05,
      "loss": 4.779,
      "mean_token_accuracy": 0.23953672349452973,
      "num_tokens": 205973014.0,
      "step": 99255
    },
    {
      "entropy": 5.663276481628418,
      "epoch": 9.003991291727141,
      "grad_norm": 1.4375,
      "learning_rate": 6.110847508436653e-05,
      "loss": 4.7769,
      "mean_token_accuracy": 0.2382173404097557,
      "num_tokens": 205982812.0,
      "step": 99260
    },
    {
      "entropy": 5.708706903457641,
      "epoch": 9.004444847605225,
      "grad_norm": 1.484375,
      "learning_rate": 6.109843613893476e-05,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2317374035716057,
      "num_tokens": 205993015.0,
      "step": 99265
    },
    {
      "entropy": 5.743025922775269,
      "epoch": 9.00489840348331,
      "grad_norm": 1.640625,
      "learning_rate": 6.10884016170418e-05,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2264157071709633,
      "num_tokens": 206003365.0,
      "step": 99270
    },
    {
      "entropy": 5.673341083526611,
      "epoch": 9.005351959361393,
      "grad_norm": 1.4765625,
      "learning_rate": 6.107837151889521e-05,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2406823679804802,
      "num_tokens": 206014214.0,
      "step": 99275
    },
    {
      "entropy": 5.64697642326355,
      "epoch": 9.005805515239478,
      "grad_norm": 1.5625,
      "learning_rate": 6.106834584470242e-05,
      "loss": 4.7175,
      "mean_token_accuracy": 0.24797309041023255,
      "num_tokens": 206024810.0,
      "step": 99280
    },
    {
      "entropy": 5.6506914615631105,
      "epoch": 9.006259071117562,
      "grad_norm": 1.484375,
      "learning_rate": 6.105832459467075e-05,
      "loss": 4.753,
      "mean_token_accuracy": 0.23316276371479033,
      "num_tokens": 206034625.0,
      "step": 99285
    },
    {
      "entropy": 5.662105464935303,
      "epoch": 9.006712626995645,
      "grad_norm": 1.5,
      "learning_rate": 6.104830776900747e-05,
      "loss": 4.7568,
      "mean_token_accuracy": 0.23484475016593934,
      "num_tokens": 206044698.0,
      "step": 99290
    },
    {
      "entropy": 5.6611274719238285,
      "epoch": 9.00716618287373,
      "grad_norm": 1.4375,
      "learning_rate": 6.103829536791969e-05,
      "loss": 4.8045,
      "mean_token_accuracy": 0.23748826682567598,
      "num_tokens": 206055930.0,
      "step": 99295
    },
    {
      "entropy": 5.717660188674927,
      "epoch": 9.007619738751814,
      "grad_norm": 1.5546875,
      "learning_rate": 6.102828739161449e-05,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22645417600870132,
      "num_tokens": 206066272.0,
      "step": 99300
    },
    {
      "entropy": 5.7064002513885494,
      "epoch": 9.008073294629899,
      "grad_norm": 1.421875,
      "learning_rate": 6.101828384029881e-05,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23827534765005112,
      "num_tokens": 206076514.0,
      "step": 99305
    },
    {
      "entropy": 5.68279275894165,
      "epoch": 9.008526850507982,
      "grad_norm": 1.4765625,
      "learning_rate": 6.100828471417956e-05,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2345763698220253,
      "num_tokens": 206086942.0,
      "step": 99310
    },
    {
      "entropy": 5.5869001865386965,
      "epoch": 9.008980406386067,
      "grad_norm": 1.34375,
      "learning_rate": 6.0998290013463536e-05,
      "loss": 4.7348,
      "mean_token_accuracy": 0.24408368915319442,
      "num_tokens": 206097785.0,
      "step": 99315
    },
    {
      "entropy": 5.787119388580322,
      "epoch": 9.00943396226415,
      "grad_norm": 1.6875,
      "learning_rate": 6.098829973835738e-05,
      "loss": 4.9277,
      "mean_token_accuracy": 0.2224715992808342,
      "num_tokens": 206108431.0,
      "step": 99320
    },
    {
      "entropy": 5.680783033370972,
      "epoch": 9.009887518142236,
      "grad_norm": 1.515625,
      "learning_rate": 6.0978313889067744e-05,
      "loss": 4.8016,
      "mean_token_accuracy": 0.24031831622123717,
      "num_tokens": 206118089.0,
      "step": 99325
    },
    {
      "entropy": 5.71709098815918,
      "epoch": 9.01034107402032,
      "grad_norm": 1.3671875,
      "learning_rate": 6.096833246580111e-05,
      "loss": 4.7969,
      "mean_token_accuracy": 0.24076632410287857,
      "num_tokens": 206128503.0,
      "step": 99330
    },
    {
      "entropy": 5.684155130386353,
      "epoch": 9.010794629898404,
      "grad_norm": 1.453125,
      "learning_rate": 6.0958355468763895e-05,
      "loss": 4.7818,
      "mean_token_accuracy": 0.23053330183029175,
      "num_tokens": 206138507.0,
      "step": 99335
    },
    {
      "entropy": 5.7912578105926515,
      "epoch": 9.011248185776488,
      "grad_norm": 1.4609375,
      "learning_rate": 6.094838289816241e-05,
      "loss": 4.8731,
      "mean_token_accuracy": 0.2339527890086174,
      "num_tokens": 206150561.0,
      "step": 99340
    },
    {
      "entropy": 5.728863525390625,
      "epoch": 9.011701741654571,
      "grad_norm": 1.6328125,
      "learning_rate": 6.0938414754202945e-05,
      "loss": 4.8431,
      "mean_token_accuracy": 0.23256101608276367,
      "num_tokens": 206160537.0,
      "step": 99345
    },
    {
      "entropy": 5.719253063201904,
      "epoch": 9.012155297532656,
      "grad_norm": 1.4296875,
      "learning_rate": 6.092845103709163e-05,
      "loss": 4.8021,
      "mean_token_accuracy": 0.22999551594257356,
      "num_tokens": 206170326.0,
      "step": 99350
    },
    {
      "entropy": 5.722294473648072,
      "epoch": 9.01260885341074,
      "grad_norm": 1.484375,
      "learning_rate": 6.0918491747034486e-05,
      "loss": 4.8343,
      "mean_token_accuracy": 0.23843088150024414,
      "num_tokens": 206180400.0,
      "step": 99355
    },
    {
      "entropy": 5.625561285018921,
      "epoch": 9.013062409288825,
      "grad_norm": 1.3984375,
      "learning_rate": 6.090853688423746e-05,
      "loss": 4.7622,
      "mean_token_accuracy": 0.23923192024230958,
      "num_tokens": 206191893.0,
      "step": 99360
    },
    {
      "entropy": 5.692044448852539,
      "epoch": 9.013515965166908,
      "grad_norm": 1.53125,
      "learning_rate": 6.089858644890649e-05,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23453183770179747,
      "num_tokens": 206202565.0,
      "step": 99365
    },
    {
      "entropy": 5.731924819946289,
      "epoch": 9.013969521044993,
      "grad_norm": 1.625,
      "learning_rate": 6.08886404412473e-05,
      "loss": 4.855,
      "mean_token_accuracy": 0.22344399690628053,
      "num_tokens": 206212283.0,
      "step": 99370
    },
    {
      "entropy": 5.697425031661988,
      "epoch": 9.014423076923077,
      "grad_norm": 1.46875,
      "learning_rate": 6.087869886146555e-05,
      "loss": 4.7775,
      "mean_token_accuracy": 0.2359024003148079,
      "num_tokens": 206222877.0,
      "step": 99375
    },
    {
      "entropy": 5.623607921600342,
      "epoch": 9.014876632801162,
      "grad_norm": 1.421875,
      "learning_rate": 6.086876170976689e-05,
      "loss": 4.6921,
      "mean_token_accuracy": 0.2461788311600685,
      "num_tokens": 206233471.0,
      "step": 99380
    },
    {
      "entropy": 5.682684135437012,
      "epoch": 9.015330188679245,
      "grad_norm": 1.46875,
      "learning_rate": 6.085882898635683e-05,
      "loss": 4.7747,
      "mean_token_accuracy": 0.2388574168086052,
      "num_tokens": 206244260.0,
      "step": 99385
    },
    {
      "entropy": 5.679981279373169,
      "epoch": 9.01578374455733,
      "grad_norm": 1.5234375,
      "learning_rate": 6.084890069144074e-05,
      "loss": 4.7682,
      "mean_token_accuracy": 0.24255430996417998,
      "num_tokens": 206254326.0,
      "step": 99390
    },
    {
      "entropy": 5.69969367980957,
      "epoch": 9.016237300435414,
      "grad_norm": 1.546875,
      "learning_rate": 6.083897682522395e-05,
      "loss": 4.8126,
      "mean_token_accuracy": 0.24227603226900102,
      "num_tokens": 206264513.0,
      "step": 99395
    },
    {
      "entropy": 5.773894786834717,
      "epoch": 9.016690856313497,
      "grad_norm": 1.5625,
      "learning_rate": 6.08290573879117e-05,
      "loss": 4.8922,
      "mean_token_accuracy": 0.2340390533208847,
      "num_tokens": 206274025.0,
      "step": 99400
    },
    {
      "entropy": 5.69315094947815,
      "epoch": 9.017144412191582,
      "grad_norm": 1.53125,
      "learning_rate": 6.081914237970913e-05,
      "loss": 4.811,
      "mean_token_accuracy": 0.23451215028762817,
      "num_tokens": 206284726.0,
      "step": 99405
    },
    {
      "entropy": 5.691719913482666,
      "epoch": 9.017597968069666,
      "grad_norm": 1.4375,
      "learning_rate": 6.0809231800821234e-05,
      "loss": 4.7785,
      "mean_token_accuracy": 0.23314594626426696,
      "num_tokens": 206294244.0,
      "step": 99410
    },
    {
      "entropy": 5.6947766780853275,
      "epoch": 9.01805152394775,
      "grad_norm": 1.4765625,
      "learning_rate": 6.079932565145306e-05,
      "loss": 4.7887,
      "mean_token_accuracy": 0.2396070957183838,
      "num_tokens": 206304926.0,
      "step": 99415
    },
    {
      "entropy": 5.71451506614685,
      "epoch": 9.018505079825834,
      "grad_norm": 1.5390625,
      "learning_rate": 6.07894239318094e-05,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2271556094288826,
      "num_tokens": 206315366.0,
      "step": 99420
    },
    {
      "entropy": 5.712350749969483,
      "epoch": 9.01895863570392,
      "grad_norm": 1.609375,
      "learning_rate": 6.0779526642095034e-05,
      "loss": 4.8164,
      "mean_token_accuracy": 0.23991331905126573,
      "num_tokens": 206325732.0,
      "step": 99425
    },
    {
      "entropy": 5.707175064086914,
      "epoch": 9.019412191582003,
      "grad_norm": 1.515625,
      "learning_rate": 6.0769633782514656e-05,
      "loss": 4.8084,
      "mean_token_accuracy": 0.23912521898746492,
      "num_tokens": 206336312.0,
      "step": 99430
    },
    {
      "entropy": 5.582872724533081,
      "epoch": 9.019865747460088,
      "grad_norm": 1.4765625,
      "learning_rate": 6.075974535327285e-05,
      "loss": 4.6833,
      "mean_token_accuracy": 0.257763983309269,
      "num_tokens": 206347981.0,
      "step": 99435
    },
    {
      "entropy": 5.615431308746338,
      "epoch": 9.020319303338171,
      "grad_norm": 1.40625,
      "learning_rate": 6.074986135457407e-05,
      "loss": 4.7271,
      "mean_token_accuracy": 0.24360353201627732,
      "num_tokens": 206358486.0,
      "step": 99440
    },
    {
      "entropy": 5.73903260231018,
      "epoch": 9.020772859216255,
      "grad_norm": 1.46875,
      "learning_rate": 6.073998178662277e-05,
      "loss": 4.8711,
      "mean_token_accuracy": 0.2304803878068924,
      "num_tokens": 206368662.0,
      "step": 99445
    },
    {
      "entropy": 5.728040456771851,
      "epoch": 9.02122641509434,
      "grad_norm": 1.4453125,
      "learning_rate": 6.073010664962324e-05,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22933581173419954,
      "num_tokens": 206378857.0,
      "step": 99450
    },
    {
      "entropy": 5.722315311431885,
      "epoch": 9.021679970972423,
      "grad_norm": 1.4453125,
      "learning_rate": 6.072023594377973e-05,
      "loss": 4.851,
      "mean_token_accuracy": 0.23387980312108994,
      "num_tokens": 206389400.0,
      "step": 99455
    },
    {
      "entropy": 5.698324251174927,
      "epoch": 9.022133526850508,
      "grad_norm": 1.625,
      "learning_rate": 6.071036966929635e-05,
      "loss": 4.7532,
      "mean_token_accuracy": 0.2440732464194298,
      "num_tokens": 206399904.0,
      "step": 99460
    },
    {
      "entropy": 5.693975734710693,
      "epoch": 9.022587082728592,
      "grad_norm": 1.46875,
      "learning_rate": 6.0700507826377105e-05,
      "loss": 4.8326,
      "mean_token_accuracy": 0.23497824519872665,
      "num_tokens": 206410027.0,
      "step": 99465
    },
    {
      "entropy": 5.66781177520752,
      "epoch": 9.023040638606677,
      "grad_norm": 1.59375,
      "learning_rate": 6.0690650415225994e-05,
      "loss": 4.7859,
      "mean_token_accuracy": 0.2438651829957962,
      "num_tokens": 206419303.0,
      "step": 99470
    },
    {
      "entropy": 5.709674119949341,
      "epoch": 9.02349419448476,
      "grad_norm": 1.7265625,
      "learning_rate": 6.0680797436046844e-05,
      "loss": 4.7863,
      "mean_token_accuracy": 0.23869001269340515,
      "num_tokens": 206430670.0,
      "step": 99475
    },
    {
      "entropy": 5.697412347793579,
      "epoch": 9.023947750362845,
      "grad_norm": 1.640625,
      "learning_rate": 6.0670948889043376e-05,
      "loss": 4.7515,
      "mean_token_accuracy": 0.24204047173261642,
      "num_tokens": 206440076.0,
      "step": 99480
    },
    {
      "entropy": 5.689492082595825,
      "epoch": 9.024401306240929,
      "grad_norm": 1.6328125,
      "learning_rate": 6.0661104774419346e-05,
      "loss": 4.8184,
      "mean_token_accuracy": 0.2380094885826111,
      "num_tokens": 206450016.0,
      "step": 99485
    },
    {
      "entropy": 5.6958108901977536,
      "epoch": 9.024854862119014,
      "grad_norm": 1.6171875,
      "learning_rate": 6.0651265092378264e-05,
      "loss": 4.7754,
      "mean_token_accuracy": 0.23468477725982667,
      "num_tokens": 206459629.0,
      "step": 99490
    },
    {
      "entropy": 5.627122449874878,
      "epoch": 9.025308417997097,
      "grad_norm": 1.421875,
      "learning_rate": 6.0641429843123664e-05,
      "loss": 4.7719,
      "mean_token_accuracy": 0.23516761660575866,
      "num_tokens": 206470727.0,
      "step": 99495
    },
    {
      "entropy": 5.682817602157593,
      "epoch": 9.02576197387518,
      "grad_norm": 1.578125,
      "learning_rate": 6.063159902685892e-05,
      "loss": 4.7638,
      "mean_token_accuracy": 0.2345254972577095,
      "num_tokens": 206480446.0,
      "step": 99500
    },
    {
      "entropy": 5.6009886264801025,
      "epoch": 9.026215529753266,
      "grad_norm": 1.421875,
      "learning_rate": 6.062177264378732e-05,
      "loss": 4.6804,
      "mean_token_accuracy": 0.24297241866588593,
      "num_tokens": 206491752.0,
      "step": 99505
    },
    {
      "entropy": 5.656447982788086,
      "epoch": 9.02666908563135,
      "grad_norm": 1.5625,
      "learning_rate": 6.0611950694112084e-05,
      "loss": 4.7739,
      "mean_token_accuracy": 0.24018792510032655,
      "num_tokens": 206501689.0,
      "step": 99510
    },
    {
      "entropy": 5.670787000656128,
      "epoch": 9.027122641509434,
      "grad_norm": 1.5546875,
      "learning_rate": 6.060213317803634e-05,
      "loss": 4.8081,
      "mean_token_accuracy": 0.2397380381822586,
      "num_tokens": 206512092.0,
      "step": 99515
    },
    {
      "entropy": 5.683790779113769,
      "epoch": 9.027576197387518,
      "grad_norm": 1.4765625,
      "learning_rate": 6.059232009576315e-05,
      "loss": 4.7813,
      "mean_token_accuracy": 0.23429138213396072,
      "num_tokens": 206523136.0,
      "step": 99520
    },
    {
      "entropy": 5.705468225479126,
      "epoch": 9.028029753265603,
      "grad_norm": 1.5859375,
      "learning_rate": 6.0582511447495415e-05,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23916378915309905,
      "num_tokens": 206532951.0,
      "step": 99525
    },
    {
      "entropy": 5.696725702285766,
      "epoch": 9.028483309143686,
      "grad_norm": 1.7265625,
      "learning_rate": 6.057270723343598e-05,
      "loss": 4.832,
      "mean_token_accuracy": 0.23498983532190323,
      "num_tokens": 206542426.0,
      "step": 99530
    },
    {
      "entropy": 5.641983509063721,
      "epoch": 9.028936865021771,
      "grad_norm": 1.4765625,
      "learning_rate": 6.056290745378759e-05,
      "loss": 4.7553,
      "mean_token_accuracy": 0.2414061576128006,
      "num_tokens": 206554288.0,
      "step": 99535
    },
    {
      "entropy": 5.654215908050537,
      "epoch": 9.029390420899855,
      "grad_norm": 1.5390625,
      "learning_rate": 6.0553112108752906e-05,
      "loss": 4.7273,
      "mean_token_accuracy": 0.2398099943995476,
      "num_tokens": 206565137.0,
      "step": 99540
    },
    {
      "entropy": 5.583559322357178,
      "epoch": 9.02984397677794,
      "grad_norm": 1.515625,
      "learning_rate": 6.054332119853452e-05,
      "loss": 4.6452,
      "mean_token_accuracy": 0.2572304308414459,
      "num_tokens": 206576195.0,
      "step": 99545
    },
    {
      "entropy": 5.672413110733032,
      "epoch": 9.030297532656023,
      "grad_norm": 1.53125,
      "learning_rate": 6.05335347233349e-05,
      "loss": 4.7536,
      "mean_token_accuracy": 0.23370904922485353,
      "num_tokens": 206585714.0,
      "step": 99550
    },
    {
      "entropy": 5.708003139495849,
      "epoch": 9.030751088534107,
      "grad_norm": 1.4765625,
      "learning_rate": 6.052375268335644e-05,
      "loss": 4.7645,
      "mean_token_accuracy": 0.23738510459661483,
      "num_tokens": 206596171.0,
      "step": 99555
    },
    {
      "entropy": 5.718052721023559,
      "epoch": 9.031204644412192,
      "grad_norm": 1.6328125,
      "learning_rate": 6.0513975078801435e-05,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2380854606628418,
      "num_tokens": 206605807.0,
      "step": 99560
    },
    {
      "entropy": 5.6391912460327145,
      "epoch": 9.031658200290275,
      "grad_norm": 1.4921875,
      "learning_rate": 6.05042019098721e-05,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24534859657287597,
      "num_tokens": 206616502.0,
      "step": 99565
    },
    {
      "entropy": 5.681663751602173,
      "epoch": 9.03211175616836,
      "grad_norm": 1.3984375,
      "learning_rate": 6.049443317677051e-05,
      "loss": 4.718,
      "mean_token_accuracy": 0.23971494734287263,
      "num_tokens": 206626778.0,
      "step": 99570
    },
    {
      "entropy": 5.733637475967408,
      "epoch": 9.032565312046444,
      "grad_norm": 1.4375,
      "learning_rate": 6.048466887969867e-05,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22899944484233856,
      "num_tokens": 206638977.0,
      "step": 99575
    },
    {
      "entropy": 5.698170900344849,
      "epoch": 9.033018867924529,
      "grad_norm": 1.515625,
      "learning_rate": 6.047490901885856e-05,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23611014038324357,
      "num_tokens": 206649773.0,
      "step": 99580
    },
    {
      "entropy": 5.674515342712402,
      "epoch": 9.033472423802612,
      "grad_norm": 1.4765625,
      "learning_rate": 6.046515359445202e-05,
      "loss": 4.7566,
      "mean_token_accuracy": 0.24901281148195267,
      "num_tokens": 206660963.0,
      "step": 99585
    },
    {
      "entropy": 5.690186023712158,
      "epoch": 9.033925979680697,
      "grad_norm": 1.4140625,
      "learning_rate": 6.0455402606680765e-05,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2400102585554123,
      "num_tokens": 206672038.0,
      "step": 99590
    },
    {
      "entropy": 5.754874229431152,
      "epoch": 9.03437953555878,
      "grad_norm": 1.546875,
      "learning_rate": 6.0445656055746494e-05,
      "loss": 4.9003,
      "mean_token_accuracy": 0.23687430769205092,
      "num_tokens": 206682243.0,
      "step": 99595
    },
    {
      "entropy": 5.629057216644287,
      "epoch": 9.034833091436864,
      "grad_norm": 1.5859375,
      "learning_rate": 6.0435913941850667e-05,
      "loss": 4.7091,
      "mean_token_accuracy": 0.2433842360973358,
      "num_tokens": 206691257.0,
      "step": 99600
    },
    {
      "entropy": 5.727337408065796,
      "epoch": 9.03528664731495,
      "grad_norm": 1.5859375,
      "learning_rate": 6.042617626519482e-05,
      "loss": 4.7961,
      "mean_token_accuracy": 0.24113346487283707,
      "num_tokens": 206701203.0,
      "step": 99605
    },
    {
      "entropy": 5.685672283172607,
      "epoch": 9.035740203193033,
      "grad_norm": 1.5546875,
      "learning_rate": 6.0416443025980375e-05,
      "loss": 4.7375,
      "mean_token_accuracy": 0.23940136283636093,
      "num_tokens": 206711355.0,
      "step": 99610
    },
    {
      "entropy": 5.692949008941651,
      "epoch": 9.036193759071118,
      "grad_norm": 1.6640625,
      "learning_rate": 6.040671422440855e-05,
      "loss": 4.7776,
      "mean_token_accuracy": 0.24091841876506806,
      "num_tokens": 206720682.0,
      "step": 99615
    },
    {
      "entropy": 5.691609144210815,
      "epoch": 9.036647314949201,
      "grad_norm": 1.5390625,
      "learning_rate": 6.039698986068056e-05,
      "loss": 4.8062,
      "mean_token_accuracy": 0.23322667479515075,
      "num_tokens": 206731681.0,
      "step": 99620
    },
    {
      "entropy": 5.5783548831939695,
      "epoch": 9.037100870827286,
      "grad_norm": 1.546875,
      "learning_rate": 6.0387269934997514e-05,
      "loss": 4.6794,
      "mean_token_accuracy": 0.242159341275692,
      "num_tokens": 206741438.0,
      "step": 99625
    },
    {
      "entropy": 5.6560173511505125,
      "epoch": 9.03755442670537,
      "grad_norm": 1.546875,
      "learning_rate": 6.037755444756041e-05,
      "loss": 4.7619,
      "mean_token_accuracy": 0.23933223783969879,
      "num_tokens": 206751980.0,
      "step": 99630
    },
    {
      "entropy": 5.650340604782104,
      "epoch": 9.038007982583455,
      "grad_norm": 1.4609375,
      "learning_rate": 6.036784339857015e-05,
      "loss": 4.7941,
      "mean_token_accuracy": 0.23841725438833236,
      "num_tokens": 206762130.0,
      "step": 99635
    },
    {
      "entropy": 5.682325077056885,
      "epoch": 9.038461538461538,
      "grad_norm": 1.6171875,
      "learning_rate": 6.035813678822759e-05,
      "loss": 4.8365,
      "mean_token_accuracy": 0.23638948053121567,
      "num_tokens": 206772491.0,
      "step": 99640
    },
    {
      "entropy": 5.663561725616455,
      "epoch": 9.038915094339623,
      "grad_norm": 1.34375,
      "learning_rate": 6.034843461673348e-05,
      "loss": 4.8145,
      "mean_token_accuracy": 0.24305939674377441,
      "num_tokens": 206782973.0,
      "step": 99645
    },
    {
      "entropy": 5.712639617919922,
      "epoch": 9.039368650217707,
      "grad_norm": 1.4765625,
      "learning_rate": 6.033873688428845e-05,
      "loss": 4.8569,
      "mean_token_accuracy": 0.22908298820257186,
      "num_tokens": 206794203.0,
      "step": 99650
    },
    {
      "entropy": 5.811124467849732,
      "epoch": 9.03982220609579,
      "grad_norm": 1.5390625,
      "learning_rate": 6.0329043591093007e-05,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21902026236057281,
      "num_tokens": 206805272.0,
      "step": 99655
    },
    {
      "entropy": 5.649543428421021,
      "epoch": 9.040275761973875,
      "grad_norm": 1.4609375,
      "learning_rate": 6.031935473734768e-05,
      "loss": 4.7285,
      "mean_token_accuracy": 0.24756884276866914,
      "num_tokens": 206815777.0,
      "step": 99660
    },
    {
      "entropy": 5.709858036041259,
      "epoch": 9.040729317851959,
      "grad_norm": 1.3671875,
      "learning_rate": 6.03096703232528e-05,
      "loss": 4.8099,
      "mean_token_accuracy": 0.23154754787683487,
      "num_tokens": 206827890.0,
      "step": 99665
    },
    {
      "entropy": 5.675235414505005,
      "epoch": 9.041182873730044,
      "grad_norm": 1.515625,
      "learning_rate": 6.0299990349008624e-05,
      "loss": 4.7218,
      "mean_token_accuracy": 0.2421045795083046,
      "num_tokens": 206838586.0,
      "step": 99670
    },
    {
      "entropy": 5.604146480560303,
      "epoch": 9.041636429608127,
      "grad_norm": 1.5390625,
      "learning_rate": 6.029031481481536e-05,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2537127435207367,
      "num_tokens": 206849062.0,
      "step": 99675
    },
    {
      "entropy": 5.663531732559204,
      "epoch": 9.042089985486212,
      "grad_norm": 1.5546875,
      "learning_rate": 6.028064372087311e-05,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23989193737506867,
      "num_tokens": 206859576.0,
      "step": 99680
    },
    {
      "entropy": 5.717100191116333,
      "epoch": 9.042543541364296,
      "grad_norm": 1.421875,
      "learning_rate": 6.027097706738187e-05,
      "loss": 4.8588,
      "mean_token_accuracy": 0.23232247531414033,
      "num_tokens": 206870291.0,
      "step": 99685
    },
    {
      "entropy": 5.681991958618164,
      "epoch": 9.042997097242381,
      "grad_norm": 1.5859375,
      "learning_rate": 6.0261314854541546e-05,
      "loss": 4.755,
      "mean_token_accuracy": 0.2436453104019165,
      "num_tokens": 206880466.0,
      "step": 99690
    },
    {
      "entropy": 5.6631941318511965,
      "epoch": 9.043450653120464,
      "grad_norm": 1.59375,
      "learning_rate": 6.025165708255196e-05,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2423094093799591,
      "num_tokens": 206890419.0,
      "step": 99695
    },
    {
      "entropy": 5.702508878707886,
      "epoch": 9.043904208998548,
      "grad_norm": 1.4375,
      "learning_rate": 6.024200375161283e-05,
      "loss": 4.8091,
      "mean_token_accuracy": 0.23402215242385865,
      "num_tokens": 206900135.0,
      "step": 99700
    },
    {
      "entropy": 5.72024359703064,
      "epoch": 9.044357764876633,
      "grad_norm": 1.5703125,
      "learning_rate": 6.0232354861923783e-05,
      "loss": 4.9308,
      "mean_token_accuracy": 0.22427567094564438,
      "num_tokens": 206911290.0,
      "step": 99705
    },
    {
      "entropy": 5.70617265701294,
      "epoch": 9.044811320754716,
      "grad_norm": 1.53125,
      "learning_rate": 6.022271041368433e-05,
      "loss": 4.797,
      "mean_token_accuracy": 0.23185423165559768,
      "num_tokens": 206922027.0,
      "step": 99710
    },
    {
      "entropy": 5.719803953170777,
      "epoch": 9.045264876632801,
      "grad_norm": 1.5859375,
      "learning_rate": 6.0213070407093995e-05,
      "loss": 4.8364,
      "mean_token_accuracy": 0.2307243049144745,
      "num_tokens": 206933090.0,
      "step": 99715
    },
    {
      "entropy": 5.796301746368409,
      "epoch": 9.045718432510885,
      "grad_norm": 1.625,
      "learning_rate": 6.02034348423521e-05,
      "loss": 4.9364,
      "mean_token_accuracy": 0.22253721654415132,
      "num_tokens": 206943266.0,
      "step": 99720
    },
    {
      "entropy": 5.740170478820801,
      "epoch": 9.04617198838897,
      "grad_norm": 1.7109375,
      "learning_rate": 6.019380371965789e-05,
      "loss": 4.8595,
      "mean_token_accuracy": 0.23571766912937164,
      "num_tokens": 206954090.0,
      "step": 99725
    },
    {
      "entropy": 5.698361015319824,
      "epoch": 9.046625544267053,
      "grad_norm": 1.4453125,
      "learning_rate": 6.0184177039210594e-05,
      "loss": 4.8417,
      "mean_token_accuracy": 0.22935863584280014,
      "num_tokens": 206966449.0,
      "step": 99730
    },
    {
      "entropy": 5.634753084182739,
      "epoch": 9.047079100145138,
      "grad_norm": 1.46875,
      "learning_rate": 6.017455480120923e-05,
      "loss": 4.7602,
      "mean_token_accuracy": 0.23069472908973693,
      "num_tokens": 206976920.0,
      "step": 99735
    },
    {
      "entropy": 5.639634466171264,
      "epoch": 9.047532656023222,
      "grad_norm": 1.5078125,
      "learning_rate": 6.016493700585283e-05,
      "loss": 4.7152,
      "mean_token_accuracy": 0.24668673127889634,
      "num_tokens": 206987927.0,
      "step": 99740
    },
    {
      "entropy": 5.6597813129425045,
      "epoch": 9.047986211901307,
      "grad_norm": 1.609375,
      "learning_rate": 6.015532365334024e-05,
      "loss": 4.7414,
      "mean_token_accuracy": 0.23338496834039688,
      "num_tokens": 206997563.0,
      "step": 99745
    },
    {
      "entropy": 5.712761259078979,
      "epoch": 9.04843976777939,
      "grad_norm": 1.7265625,
      "learning_rate": 6.014571474387033e-05,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2256507471203804,
      "num_tokens": 207008056.0,
      "step": 99750
    },
    {
      "entropy": 5.633073377609253,
      "epoch": 9.048893323657474,
      "grad_norm": 1.5390625,
      "learning_rate": 6.0136110277641804e-05,
      "loss": 4.6989,
      "mean_token_accuracy": 0.243741674721241,
      "num_tokens": 207018027.0,
      "step": 99755
    },
    {
      "entropy": 5.634958124160766,
      "epoch": 9.049346879535559,
      "grad_norm": 1.4140625,
      "learning_rate": 6.012651025485327e-05,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23915574103593826,
      "num_tokens": 207028346.0,
      "step": 99760
    },
    {
      "entropy": 5.565905332565308,
      "epoch": 9.049800435413642,
      "grad_norm": 1.4140625,
      "learning_rate": 6.011691467570326e-05,
      "loss": 4.6409,
      "mean_token_accuracy": 0.24727030396461486,
      "num_tokens": 207038682.0,
      "step": 99765
    },
    {
      "entropy": 5.64340033531189,
      "epoch": 9.050253991291727,
      "grad_norm": 1.4921875,
      "learning_rate": 6.01073235403902e-05,
      "loss": 4.7596,
      "mean_token_accuracy": 0.2415357545018196,
      "num_tokens": 207048625.0,
      "step": 99770
    },
    {
      "entropy": 5.64664511680603,
      "epoch": 9.05070754716981,
      "grad_norm": 1.4375,
      "learning_rate": 6.0097736849112485e-05,
      "loss": 4.6884,
      "mean_token_accuracy": 0.2411164551973343,
      "num_tokens": 207058138.0,
      "step": 99775
    },
    {
      "entropy": 5.734752511978149,
      "epoch": 9.051161103047896,
      "grad_norm": 1.53125,
      "learning_rate": 6.008815460206827e-05,
      "loss": 4.8871,
      "mean_token_accuracy": 0.23466873914003372,
      "num_tokens": 207068427.0,
      "step": 99780
    },
    {
      "entropy": 5.6491998672485355,
      "epoch": 9.05161465892598,
      "grad_norm": 1.4453125,
      "learning_rate": 6.007857679945583e-05,
      "loss": 4.7487,
      "mean_token_accuracy": 0.2380714699625969,
      "num_tokens": 207079409.0,
      "step": 99785
    },
    {
      "entropy": 5.732165765762329,
      "epoch": 9.052068214804065,
      "grad_norm": 1.484375,
      "learning_rate": 6.006900344147318e-05,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2235308527946472,
      "num_tokens": 207090056.0,
      "step": 99790
    },
    {
      "entropy": 5.632514953613281,
      "epoch": 9.052521770682148,
      "grad_norm": 1.5546875,
      "learning_rate": 6.005943452831831e-05,
      "loss": 4.6758,
      "mean_token_accuracy": 0.24233659356832504,
      "num_tokens": 207100166.0,
      "step": 99795
    },
    {
      "entropy": 5.655752086639405,
      "epoch": 9.052975326560233,
      "grad_norm": 1.453125,
      "learning_rate": 6.004987006018911e-05,
      "loss": 4.7327,
      "mean_token_accuracy": 0.24419941157102584,
      "num_tokens": 207110513.0,
      "step": 99800
    },
    {
      "entropy": 5.640229225158691,
      "epoch": 9.053428882438316,
      "grad_norm": 1.59375,
      "learning_rate": 6.004031003728338e-05,
      "loss": 4.7598,
      "mean_token_accuracy": 0.25066733956336973,
      "num_tokens": 207121109.0,
      "step": 99805
    },
    {
      "entropy": 5.684616899490356,
      "epoch": 9.0538824383164,
      "grad_norm": 1.59375,
      "learning_rate": 6.003075445979877e-05,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2355464518070221,
      "num_tokens": 207130596.0,
      "step": 99810
    },
    {
      "entropy": 5.689698791503906,
      "epoch": 9.054335994194485,
      "grad_norm": 1.5625,
      "learning_rate": 6.0021203327932995e-05,
      "loss": 4.7944,
      "mean_token_accuracy": 0.235946361720562,
      "num_tokens": 207141287.0,
      "step": 99815
    },
    {
      "entropy": 5.7556480884552,
      "epoch": 9.054789550072568,
      "grad_norm": 1.5390625,
      "learning_rate": 6.001165664188351e-05,
      "loss": 4.8882,
      "mean_token_accuracy": 0.23969465643167495,
      "num_tokens": 207152498.0,
      "step": 99820
    },
    {
      "entropy": 5.7272491455078125,
      "epoch": 9.055243105950654,
      "grad_norm": 1.5703125,
      "learning_rate": 6.0002114401847756e-05,
      "loss": 4.8294,
      "mean_token_accuracy": 0.23507805168628693,
      "num_tokens": 207162493.0,
      "step": 99825
    },
    {
      "entropy": 5.689385461807251,
      "epoch": 9.055696661828737,
      "grad_norm": 1.5625,
      "learning_rate": 5.9992576608023045e-05,
      "loss": 4.7739,
      "mean_token_accuracy": 0.2398223578929901,
      "num_tokens": 207172582.0,
      "step": 99830
    },
    {
      "entropy": 5.692436456680298,
      "epoch": 9.056150217706822,
      "grad_norm": 1.65625,
      "learning_rate": 5.9983043260606656e-05,
      "loss": 4.757,
      "mean_token_accuracy": 0.24292390942573547,
      "num_tokens": 207183926.0,
      "step": 99835
    },
    {
      "entropy": 5.639887762069702,
      "epoch": 9.056603773584905,
      "grad_norm": 1.5390625,
      "learning_rate": 5.99735143597957e-05,
      "loss": 4.7097,
      "mean_token_accuracy": 0.25027861893177034,
      "num_tokens": 207193238.0,
      "step": 99840
    },
    {
      "entropy": 5.669734811782837,
      "epoch": 9.05705732946299,
      "grad_norm": 1.515625,
      "learning_rate": 5.9963989905787266e-05,
      "loss": 4.7654,
      "mean_token_accuracy": 0.24664880335330963,
      "num_tokens": 207204362.0,
      "step": 99845
    },
    {
      "entropy": 5.610396814346314,
      "epoch": 9.057510885341074,
      "grad_norm": 1.5,
      "learning_rate": 5.9954469898778324e-05,
      "loss": 4.7365,
      "mean_token_accuracy": 0.2453558087348938,
      "num_tokens": 207215263.0,
      "step": 99850
    },
    {
      "entropy": 5.6923134326934814,
      "epoch": 9.057964441219157,
      "grad_norm": 1.4765625,
      "learning_rate": 5.994495433896576e-05,
      "loss": 4.7964,
      "mean_token_accuracy": 0.23601904958486558,
      "num_tokens": 207225495.0,
      "step": 99855
    },
    {
      "entropy": 5.6547243118286135,
      "epoch": 9.058417997097242,
      "grad_norm": 1.6328125,
      "learning_rate": 5.993544322654633e-05,
      "loss": 4.8126,
      "mean_token_accuracy": 0.24315714091062546,
      "num_tokens": 207235967.0,
      "step": 99860
    },
    {
      "entropy": 5.717603349685669,
      "epoch": 9.058871552975326,
      "grad_norm": 1.5234375,
      "learning_rate": 5.992593656171676e-05,
      "loss": 4.7874,
      "mean_token_accuracy": 0.24019608199596404,
      "num_tokens": 207245661.0,
      "step": 99865
    },
    {
      "entropy": 5.716050434112549,
      "epoch": 9.059325108853411,
      "grad_norm": 1.6328125,
      "learning_rate": 5.99164343446736e-05,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23402196764945984,
      "num_tokens": 207255939.0,
      "step": 99870
    },
    {
      "entropy": 5.704310941696167,
      "epoch": 9.059778664731494,
      "grad_norm": 1.484375,
      "learning_rate": 5.9906936575613415e-05,
      "loss": 4.8327,
      "mean_token_accuracy": 0.2357384279370308,
      "num_tokens": 207265900.0,
      "step": 99875
    },
    {
      "entropy": 5.679752969741822,
      "epoch": 9.06023222060958,
      "grad_norm": 1.46875,
      "learning_rate": 5.989744325473254e-05,
      "loss": 4.8288,
      "mean_token_accuracy": 0.22768653482198714,
      "num_tokens": 207276003.0,
      "step": 99880
    },
    {
      "entropy": 5.667050552368164,
      "epoch": 9.060685776487663,
      "grad_norm": 1.59375,
      "learning_rate": 5.98879543822274e-05,
      "loss": 4.7638,
      "mean_token_accuracy": 0.23665805011987687,
      "num_tokens": 207286195.0,
      "step": 99885
    },
    {
      "entropy": 5.675888299942017,
      "epoch": 9.061139332365748,
      "grad_norm": 1.4140625,
      "learning_rate": 5.987846995829417e-05,
      "loss": 4.7909,
      "mean_token_accuracy": 0.2369275912642479,
      "num_tokens": 207296268.0,
      "step": 99890
    },
    {
      "entropy": 5.686682939529419,
      "epoch": 9.061592888243831,
      "grad_norm": 1.453125,
      "learning_rate": 5.9868989983129e-05,
      "loss": 4.7459,
      "mean_token_accuracy": 0.23483624160289765,
      "num_tokens": 207305889.0,
      "step": 99895
    },
    {
      "entropy": 5.6876567840576175,
      "epoch": 9.062046444121917,
      "grad_norm": 1.6484375,
      "learning_rate": 5.985951445692795e-05,
      "loss": 4.8194,
      "mean_token_accuracy": 0.23774021863937378,
      "num_tokens": 207315464.0,
      "step": 99900
    },
    {
      "entropy": 5.7028380870819095,
      "epoch": 9.0625,
      "grad_norm": 1.4375,
      "learning_rate": 5.985004337988697e-05,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23659011125564575,
      "num_tokens": 207325676.0,
      "step": 99905
    },
    {
      "entropy": 5.674267196655274,
      "epoch": 9.062953555878083,
      "grad_norm": 1.546875,
      "learning_rate": 5.9840576752201895e-05,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23529048562049865,
      "num_tokens": 207336452.0,
      "step": 99910
    },
    {
      "entropy": 5.714535665512085,
      "epoch": 9.063407111756169,
      "grad_norm": 1.609375,
      "learning_rate": 5.983111457406852e-05,
      "loss": 4.8817,
      "mean_token_accuracy": 0.23163031786680222,
      "num_tokens": 207347426.0,
      "step": 99915
    },
    {
      "entropy": 5.723910856246948,
      "epoch": 9.063860667634252,
      "grad_norm": 1.5234375,
      "learning_rate": 5.982165684568254e-05,
      "loss": 4.877,
      "mean_token_accuracy": 0.23122018426656724,
      "num_tokens": 207357631.0,
      "step": 99920
    },
    {
      "entropy": 5.684440279006958,
      "epoch": 9.064314223512337,
      "grad_norm": 1.5546875,
      "learning_rate": 5.981220356723953e-05,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2329295501112938,
      "num_tokens": 207368367.0,
      "step": 99925
    },
    {
      "entropy": 5.6826732635498045,
      "epoch": 9.06476777939042,
      "grad_norm": 1.625,
      "learning_rate": 5.9802754738935e-05,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23834897875785827,
      "num_tokens": 207377866.0,
      "step": 99930
    },
    {
      "entropy": 5.6592474460601805,
      "epoch": 9.065221335268506,
      "grad_norm": 1.3671875,
      "learning_rate": 5.979331036096432e-05,
      "loss": 4.7443,
      "mean_token_accuracy": 0.239008629322052,
      "num_tokens": 207388517.0,
      "step": 99935
    },
    {
      "entropy": 5.658218193054199,
      "epoch": 9.065674891146589,
      "grad_norm": 1.546875,
      "learning_rate": 5.978387043352283e-05,
      "loss": 4.771,
      "mean_token_accuracy": 0.24149406850337982,
      "num_tokens": 207398645.0,
      "step": 99940
    },
    {
      "entropy": 5.727023458480835,
      "epoch": 9.066128447024674,
      "grad_norm": 1.4765625,
      "learning_rate": 5.977443495680576e-05,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23061234205961229,
      "num_tokens": 207409355.0,
      "step": 99945
    },
    {
      "entropy": 5.692784643173217,
      "epoch": 9.066582002902758,
      "grad_norm": 1.546875,
      "learning_rate": 5.9765003931008196e-05,
      "loss": 4.7839,
      "mean_token_accuracy": 0.24231608361005783,
      "num_tokens": 207419653.0,
      "step": 99950
    },
    {
      "entropy": 5.6140439987182615,
      "epoch": 9.067035558780843,
      "grad_norm": 1.53125,
      "learning_rate": 5.975557735632524e-05,
      "loss": 4.7066,
      "mean_token_accuracy": 0.24461737424135208,
      "num_tokens": 207429767.0,
      "step": 99955
    },
    {
      "entropy": 5.68747992515564,
      "epoch": 9.067489114658926,
      "grad_norm": 1.578125,
      "learning_rate": 5.9746155232951744e-05,
      "loss": 4.7854,
      "mean_token_accuracy": 0.2426292821764946,
      "num_tokens": 207440110.0,
      "step": 99960
    },
    {
      "entropy": 5.72259521484375,
      "epoch": 9.06794267053701,
      "grad_norm": 1.3515625,
      "learning_rate": 5.973673756108264e-05,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2375742271542549,
      "num_tokens": 207451481.0,
      "step": 99965
    },
    {
      "entropy": 5.7323261260986325,
      "epoch": 9.068396226415095,
      "grad_norm": 1.5625,
      "learning_rate": 5.972732434091266e-05,
      "loss": 4.8385,
      "mean_token_accuracy": 0.23655507266521453,
      "num_tokens": 207462413.0,
      "step": 99970
    },
    {
      "entropy": 5.640608882904052,
      "epoch": 9.068849782293178,
      "grad_norm": 1.78125,
      "learning_rate": 5.971791557263642e-05,
      "loss": 4.6898,
      "mean_token_accuracy": 0.250791509449482,
      "num_tokens": 207471106.0,
      "step": 99975
    },
    {
      "entropy": 5.677517127990723,
      "epoch": 9.069303338171263,
      "grad_norm": 1.625,
      "learning_rate": 5.9708511256448606e-05,
      "loss": 4.7713,
      "mean_token_accuracy": 0.23739639073610305,
      "num_tokens": 207481288.0,
      "step": 99980
    },
    {
      "entropy": 5.750288438796997,
      "epoch": 9.069756894049346,
      "grad_norm": 1.46875,
      "learning_rate": 5.969911139254361e-05,
      "loss": 4.8994,
      "mean_token_accuracy": 0.23379762917757035,
      "num_tokens": 207492134.0,
      "step": 99985
    },
    {
      "entropy": 5.70576810836792,
      "epoch": 9.070210449927432,
      "grad_norm": 1.640625,
      "learning_rate": 5.968971598111588e-05,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22906061112880707,
      "num_tokens": 207503419.0,
      "step": 99990
    },
    {
      "entropy": 5.607445526123047,
      "epoch": 9.070664005805515,
      "grad_norm": 1.5390625,
      "learning_rate": 5.968032502235967e-05,
      "loss": 4.7171,
      "mean_token_accuracy": 0.24659982621669768,
      "num_tokens": 207513318.0,
      "step": 99995
    },
    {
      "entropy": 5.650836944580078,
      "epoch": 9.0711175616836,
      "grad_norm": 1.4921875,
      "learning_rate": 5.967093851646922e-05,
      "loss": 4.7367,
      "mean_token_accuracy": 0.2442232772707939,
      "num_tokens": 207523884.0,
      "step": 100000
    },
    {
      "entropy": 5.624355363845825,
      "epoch": 9.071571117561684,
      "grad_norm": 1.59375,
      "learning_rate": 5.966155646363864e-05,
      "loss": 4.7325,
      "mean_token_accuracy": 0.24159343987703324,
      "num_tokens": 207533005.0,
      "step": 100005
    },
    {
      "entropy": 5.674727964401245,
      "epoch": 9.072024673439767,
      "grad_norm": 1.4921875,
      "learning_rate": 5.965217886406192e-05,
      "loss": 4.806,
      "mean_token_accuracy": 0.23236578106880187,
      "num_tokens": 207544140.0,
      "step": 100010
    },
    {
      "entropy": 5.714066505432129,
      "epoch": 9.072478229317852,
      "grad_norm": 1.5546875,
      "learning_rate": 5.964280571793304e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.23918845057487487,
      "num_tokens": 207554020.0,
      "step": 100015
    },
    {
      "entropy": 5.779437398910522,
      "epoch": 9.072931785195935,
      "grad_norm": 1.5703125,
      "learning_rate": 5.963343702544582e-05,
      "loss": 4.8433,
      "mean_token_accuracy": 0.22898097038269044,
      "num_tokens": 207564180.0,
      "step": 100020
    },
    {
      "entropy": 5.665875673294067,
      "epoch": 9.07338534107402,
      "grad_norm": 1.546875,
      "learning_rate": 5.962407278679399e-05,
      "loss": 4.7028,
      "mean_token_accuracy": 0.24262305945158005,
      "num_tokens": 207574941.0,
      "step": 100025
    },
    {
      "entropy": 5.695754528045654,
      "epoch": 9.073838896952104,
      "grad_norm": 1.3203125,
      "learning_rate": 5.961471300217124e-05,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2328286200761795,
      "num_tokens": 207585964.0,
      "step": 100030
    },
    {
      "entropy": 5.68075795173645,
      "epoch": 9.07429245283019,
      "grad_norm": 1.4453125,
      "learning_rate": 5.96053576717711e-05,
      "loss": 4.7497,
      "mean_token_accuracy": 0.24051846712827682,
      "num_tokens": 207596153.0,
      "step": 100035
    },
    {
      "entropy": 5.684451913833618,
      "epoch": 9.074746008708273,
      "grad_norm": 1.5859375,
      "learning_rate": 5.9596006795787054e-05,
      "loss": 4.7877,
      "mean_token_accuracy": 0.23953595608472825,
      "num_tokens": 207605859.0,
      "step": 100040
    },
    {
      "entropy": 5.689827919006348,
      "epoch": 9.075199564586358,
      "grad_norm": 1.609375,
      "learning_rate": 5.958666037441246e-05,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23629736304283142,
      "num_tokens": 207616291.0,
      "step": 100045
    },
    {
      "entropy": 5.6781120777130125,
      "epoch": 9.075653120464441,
      "grad_norm": 1.4609375,
      "learning_rate": 5.9577318407840644e-05,
      "loss": 4.8583,
      "mean_token_accuracy": 0.2278430923819542,
      "num_tokens": 207627519.0,
      "step": 100050
    },
    {
      "entropy": 5.671184349060058,
      "epoch": 9.076106676342526,
      "grad_norm": 1.421875,
      "learning_rate": 5.956798089626477e-05,
      "loss": 4.7895,
      "mean_token_accuracy": 0.2331282064318657,
      "num_tokens": 207638928.0,
      "step": 100055
    },
    {
      "entropy": 5.705965566635132,
      "epoch": 9.07656023222061,
      "grad_norm": 1.4375,
      "learning_rate": 5.955864783987797e-05,
      "loss": 4.876,
      "mean_token_accuracy": 0.2222694933414459,
      "num_tokens": 207649927.0,
      "step": 100060
    },
    {
      "entropy": 5.6425539493560795,
      "epoch": 9.077013788098693,
      "grad_norm": 1.7109375,
      "learning_rate": 5.9549319238873226e-05,
      "loss": 4.7116,
      "mean_token_accuracy": 0.248233300447464,
      "num_tokens": 207660929.0,
      "step": 100065
    },
    {
      "entropy": 5.6331500053405765,
      "epoch": 9.077467343976778,
      "grad_norm": 1.484375,
      "learning_rate": 5.953999509344344e-05,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24510783702135086,
      "num_tokens": 207672126.0,
      "step": 100070
    },
    {
      "entropy": 5.7016149997711185,
      "epoch": 9.077920899854862,
      "grad_norm": 1.3359375,
      "learning_rate": 5.953067540378149e-05,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23239964097738267,
      "num_tokens": 207681915.0,
      "step": 100075
    },
    {
      "entropy": 5.663021516799927,
      "epoch": 9.078374455732947,
      "grad_norm": 1.515625,
      "learning_rate": 5.9521360170080034e-05,
      "loss": 4.7207,
      "mean_token_accuracy": 0.2401480868458748,
      "num_tokens": 207692645.0,
      "step": 100080
    },
    {
      "entropy": 5.71295166015625,
      "epoch": 9.07882801161103,
      "grad_norm": 1.515625,
      "learning_rate": 5.951204939253181e-05,
      "loss": 4.8325,
      "mean_token_accuracy": 0.23051868081092836,
      "num_tokens": 207703420.0,
      "step": 100085
    },
    {
      "entropy": 5.692414426803589,
      "epoch": 9.079281567489115,
      "grad_norm": 1.6015625,
      "learning_rate": 5.9502743071329295e-05,
      "loss": 4.7648,
      "mean_token_accuracy": 0.24386604875326157,
      "num_tokens": 207712814.0,
      "step": 100090
    },
    {
      "entropy": 5.665658664703369,
      "epoch": 9.079735123367199,
      "grad_norm": 1.375,
      "learning_rate": 5.9493441206665e-05,
      "loss": 4.7981,
      "mean_token_accuracy": 0.23033979088068007,
      "num_tokens": 207723770.0,
      "step": 100095
    },
    {
      "entropy": 5.665177631378174,
      "epoch": 9.080188679245284,
      "grad_norm": 1.5703125,
      "learning_rate": 5.948414379873123e-05,
      "loss": 4.7656,
      "mean_token_accuracy": 0.23638556897640228,
      "num_tokens": 207734179.0,
      "step": 100100
    },
    {
      "entropy": 5.714535903930664,
      "epoch": 9.080642235123367,
      "grad_norm": 1.421875,
      "learning_rate": 5.9474850847720276e-05,
      "loss": 4.8145,
      "mean_token_accuracy": 0.24276482611894606,
      "num_tokens": 207746203.0,
      "step": 100105
    },
    {
      "entropy": 5.694674301147461,
      "epoch": 9.081095791001452,
      "grad_norm": 1.4921875,
      "learning_rate": 5.9465562353824315e-05,
      "loss": 4.773,
      "mean_token_accuracy": 0.23896609097719193,
      "num_tokens": 207756726.0,
      "step": 100110
    },
    {
      "entropy": 5.6757392406463625,
      "epoch": 9.081549346879536,
      "grad_norm": 1.6640625,
      "learning_rate": 5.9456278317235484e-05,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23717350959777833,
      "num_tokens": 207766545.0,
      "step": 100115
    },
    {
      "entropy": 5.706217861175537,
      "epoch": 9.082002902757619,
      "grad_norm": 1.4140625,
      "learning_rate": 5.944699873814575e-05,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23504034131765367,
      "num_tokens": 207777195.0,
      "step": 100120
    },
    {
      "entropy": 5.685645198822021,
      "epoch": 9.082456458635704,
      "grad_norm": 1.4921875,
      "learning_rate": 5.9437723616747e-05,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23316654562950134,
      "num_tokens": 207787146.0,
      "step": 100125
    },
    {
      "entropy": 5.6561441898345945,
      "epoch": 9.082910014513788,
      "grad_norm": 1.5078125,
      "learning_rate": 5.942845295323106e-05,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23877521008253097,
      "num_tokens": 207797577.0,
      "step": 100130
    },
    {
      "entropy": 5.670863389968872,
      "epoch": 9.083363570391873,
      "grad_norm": 1.4296875,
      "learning_rate": 5.9419186747789644e-05,
      "loss": 4.8,
      "mean_token_accuracy": 0.23915615677833557,
      "num_tokens": 207808574.0,
      "step": 100135
    },
    {
      "entropy": 5.624763679504395,
      "epoch": 9.083817126269956,
      "grad_norm": 1.546875,
      "learning_rate": 5.94099250006144e-05,
      "loss": 4.7551,
      "mean_token_accuracy": 0.22590074837207794,
      "num_tokens": 207819138.0,
      "step": 100140
    },
    {
      "entropy": 5.649492359161377,
      "epoch": 9.084270682148041,
      "grad_norm": 1.421875,
      "learning_rate": 5.940066771189678e-05,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23175816237926483,
      "num_tokens": 207830348.0,
      "step": 100145
    },
    {
      "entropy": 5.775713205337524,
      "epoch": 9.084724238026125,
      "grad_norm": 1.546875,
      "learning_rate": 5.939141488182836e-05,
      "loss": 4.9022,
      "mean_token_accuracy": 0.23516566157341004,
      "num_tokens": 207840826.0,
      "step": 100150
    },
    {
      "entropy": 5.625623989105224,
      "epoch": 9.08517779390421,
      "grad_norm": 1.4921875,
      "learning_rate": 5.9382166510600394e-05,
      "loss": 4.688,
      "mean_token_accuracy": 0.246150603890419,
      "num_tokens": 207851266.0,
      "step": 100155
    },
    {
      "entropy": 5.667692565917969,
      "epoch": 9.085631349782293,
      "grad_norm": 1.5703125,
      "learning_rate": 5.9372922598404195e-05,
      "loss": 4.733,
      "mean_token_accuracy": 0.2463781625032425,
      "num_tokens": 207861189.0,
      "step": 100160
    },
    {
      "entropy": 5.682514810562134,
      "epoch": 9.086084905660377,
      "grad_norm": 1.484375,
      "learning_rate": 5.936368314543089e-05,
      "loss": 4.7612,
      "mean_token_accuracy": 0.23821281343698503,
      "num_tokens": 207870626.0,
      "step": 100165
    },
    {
      "entropy": 5.683491373062134,
      "epoch": 9.086538461538462,
      "grad_norm": 1.3984375,
      "learning_rate": 5.935444815187157e-05,
      "loss": 4.7408,
      "mean_token_accuracy": 0.24625276923179626,
      "num_tokens": 207881019.0,
      "step": 100170
    },
    {
      "entropy": 5.659989690780639,
      "epoch": 9.086992017416545,
      "grad_norm": 1.625,
      "learning_rate": 5.934521761791724e-05,
      "loss": 4.7708,
      "mean_token_accuracy": 0.23684318065643312,
      "num_tokens": 207891923.0,
      "step": 100175
    },
    {
      "entropy": 5.721272420883179,
      "epoch": 9.08744557329463,
      "grad_norm": 1.4453125,
      "learning_rate": 5.9335991543758714e-05,
      "loss": 4.842,
      "mean_token_accuracy": 0.22948607355356215,
      "num_tokens": 207902307.0,
      "step": 100180
    },
    {
      "entropy": 5.6278791427612305,
      "epoch": 9.087899129172714,
      "grad_norm": 1.4765625,
      "learning_rate": 5.932676992958686e-05,
      "loss": 4.685,
      "mean_token_accuracy": 0.24777254909276963,
      "num_tokens": 207912131.0,
      "step": 100185
    },
    {
      "entropy": 5.691724157333374,
      "epoch": 9.088352685050799,
      "grad_norm": 1.59375,
      "learning_rate": 5.9317552775592364e-05,
      "loss": 4.7941,
      "mean_token_accuracy": 0.2359551101922989,
      "num_tokens": 207922448.0,
      "step": 100190
    },
    {
      "entropy": 5.680682802200318,
      "epoch": 9.088806240928882,
      "grad_norm": 1.375,
      "learning_rate": 5.9308340081965866e-05,
      "loss": 4.7858,
      "mean_token_accuracy": 0.243010513484478,
      "num_tokens": 207933810.0,
      "step": 100195
    },
    {
      "entropy": 5.701582765579223,
      "epoch": 9.089259796806967,
      "grad_norm": 1.4921875,
      "learning_rate": 5.9299131848897844e-05,
      "loss": 4.8142,
      "mean_token_accuracy": 0.23724171370267869,
      "num_tokens": 207944013.0,
      "step": 100200
    },
    {
      "entropy": 5.6808388233184814,
      "epoch": 9.08971335268505,
      "grad_norm": 1.5078125,
      "learning_rate": 5.9289928076578744e-05,
      "loss": 4.7646,
      "mean_token_accuracy": 0.24577628821134567,
      "num_tokens": 207954989.0,
      "step": 100205
    },
    {
      "entropy": 5.711607980728149,
      "epoch": 9.090166908563136,
      "grad_norm": 1.359375,
      "learning_rate": 5.9280728765198885e-05,
      "loss": 4.8473,
      "mean_token_accuracy": 0.23347595781087876,
      "num_tokens": 207966587.0,
      "step": 100210
    },
    {
      "entropy": 5.670630645751953,
      "epoch": 9.09062046444122,
      "grad_norm": 1.4375,
      "learning_rate": 5.927153391494852e-05,
      "loss": 4.7932,
      "mean_token_accuracy": 0.2375008448958397,
      "num_tokens": 207976215.0,
      "step": 100215
    },
    {
      "entropy": 5.726435613632202,
      "epoch": 9.091074020319303,
      "grad_norm": 1.375,
      "learning_rate": 5.926234352601781e-05,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2237009435892105,
      "num_tokens": 207987237.0,
      "step": 100220
    },
    {
      "entropy": 5.7026674270629885,
      "epoch": 9.091527576197388,
      "grad_norm": 1.609375,
      "learning_rate": 5.925315759859684e-05,
      "loss": 4.8042,
      "mean_token_accuracy": 0.2326200008392334,
      "num_tokens": 207998139.0,
      "step": 100225
    },
    {
      "entropy": 5.702468585968018,
      "epoch": 9.091981132075471,
      "grad_norm": 1.4765625,
      "learning_rate": 5.9243976132875534e-05,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23432100117206572,
      "num_tokens": 208008875.0,
      "step": 100230
    },
    {
      "entropy": 5.73428201675415,
      "epoch": 9.092434687953556,
      "grad_norm": 1.4453125,
      "learning_rate": 5.923479912904381e-05,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23601196110248565,
      "num_tokens": 208019967.0,
      "step": 100235
    },
    {
      "entropy": 5.685262727737427,
      "epoch": 9.09288824383164,
      "grad_norm": 1.515625,
      "learning_rate": 5.922562658729139e-05,
      "loss": 4.7316,
      "mean_token_accuracy": 0.24619584381580353,
      "num_tokens": 208029608.0,
      "step": 100240
    },
    {
      "entropy": 5.72658896446228,
      "epoch": 9.093341799709725,
      "grad_norm": 1.609375,
      "learning_rate": 5.9216458507808e-05,
      "loss": 4.7592,
      "mean_token_accuracy": 0.2378694847226143,
      "num_tokens": 208040394.0,
      "step": 100245
    },
    {
      "entropy": 5.6368061065673825,
      "epoch": 9.093795355587808,
      "grad_norm": 1.6328125,
      "learning_rate": 5.920729489078323e-05,
      "loss": 4.728,
      "mean_token_accuracy": 0.25065895169973373,
      "num_tokens": 208050274.0,
      "step": 100250
    },
    {
      "entropy": 5.676438331604004,
      "epoch": 9.094248911465893,
      "grad_norm": 1.5859375,
      "learning_rate": 5.91981357364066e-05,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23173537403345107,
      "num_tokens": 208061177.0,
      "step": 100255
    },
    {
      "entropy": 5.687043046951294,
      "epoch": 9.094702467343977,
      "grad_norm": 1.6015625,
      "learning_rate": 5.918898104486754e-05,
      "loss": 4.7845,
      "mean_token_accuracy": 0.2406245529651642,
      "num_tokens": 208071268.0,
      "step": 100260
    },
    {
      "entropy": 5.672791385650635,
      "epoch": 9.09515602322206,
      "grad_norm": 1.4375,
      "learning_rate": 5.917983081635532e-05,
      "loss": 4.8008,
      "mean_token_accuracy": 0.23255836665630342,
      "num_tokens": 208082628.0,
      "step": 100265
    },
    {
      "entropy": 5.674013233184814,
      "epoch": 9.095609579100145,
      "grad_norm": 1.46875,
      "learning_rate": 5.9170685051059205e-05,
      "loss": 4.808,
      "mean_token_accuracy": 0.23586974591016768,
      "num_tokens": 208093053.0,
      "step": 100270
    },
    {
      "entropy": 5.724401235580444,
      "epoch": 9.096063134978229,
      "grad_norm": 1.4453125,
      "learning_rate": 5.9161543749168336e-05,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22985825985670089,
      "num_tokens": 208102901.0,
      "step": 100275
    },
    {
      "entropy": 5.69744029045105,
      "epoch": 9.096516690856314,
      "grad_norm": 1.6875,
      "learning_rate": 5.915240691087173e-05,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23239680528640747,
      "num_tokens": 208112568.0,
      "step": 100280
    },
    {
      "entropy": 5.7041925430297855,
      "epoch": 9.096970246734397,
      "grad_norm": 1.5625,
      "learning_rate": 5.914327453635832e-05,
      "loss": 4.7879,
      "mean_token_accuracy": 0.23408142626285552,
      "num_tokens": 208123942.0,
      "step": 100285
    },
    {
      "entropy": 5.6866754531860355,
      "epoch": 9.097423802612482,
      "grad_norm": 1.609375,
      "learning_rate": 5.913414662581705e-05,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2407766878604889,
      "num_tokens": 208135066.0,
      "step": 100290
    },
    {
      "entropy": 5.686605024337768,
      "epoch": 9.097877358490566,
      "grad_norm": 1.5078125,
      "learning_rate": 5.912502317943663e-05,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2325182557106018,
      "num_tokens": 208146311.0,
      "step": 100295
    },
    {
      "entropy": 5.707991600036621,
      "epoch": 9.09833091436865,
      "grad_norm": 1.46875,
      "learning_rate": 5.911590419740574e-05,
      "loss": 4.8352,
      "mean_token_accuracy": 0.23429211229085922,
      "num_tokens": 208157230.0,
      "step": 100300
    },
    {
      "entropy": 5.7535875797271725,
      "epoch": 9.098784470246734,
      "grad_norm": 1.453125,
      "learning_rate": 5.910678967991294e-05,
      "loss": 4.8696,
      "mean_token_accuracy": 0.237437142431736,
      "num_tokens": 208167791.0,
      "step": 100305
    },
    {
      "entropy": 5.721281623840332,
      "epoch": 9.09923802612482,
      "grad_norm": 1.53125,
      "learning_rate": 5.909767962714678e-05,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23664887845516205,
      "num_tokens": 208178314.0,
      "step": 100310
    },
    {
      "entropy": 5.6444038391113285,
      "epoch": 9.099691582002903,
      "grad_norm": 1.7421875,
      "learning_rate": 5.9088574039295584e-05,
      "loss": 4.6855,
      "mean_token_accuracy": 0.24631609320640563,
      "num_tokens": 208187897.0,
      "step": 100315
    },
    {
      "entropy": 5.642755699157715,
      "epoch": 9.100145137880986,
      "grad_norm": 1.453125,
      "learning_rate": 5.907947291654773e-05,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24228805899620057,
      "num_tokens": 208198565.0,
      "step": 100320
    },
    {
      "entropy": 5.698670339584351,
      "epoch": 9.100598693759071,
      "grad_norm": 1.5859375,
      "learning_rate": 5.9070376259091384e-05,
      "loss": 4.7922,
      "mean_token_accuracy": 0.24018273055553435,
      "num_tokens": 208207642.0,
      "step": 100325
    },
    {
      "entropy": 5.688212966918945,
      "epoch": 9.101052249637155,
      "grad_norm": 1.609375,
      "learning_rate": 5.9061284067114696e-05,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23653533905744553,
      "num_tokens": 208217714.0,
      "step": 100330
    },
    {
      "entropy": 5.725703382492066,
      "epoch": 9.10150580551524,
      "grad_norm": 1.6015625,
      "learning_rate": 5.9052196340805664e-05,
      "loss": 4.7724,
      "mean_token_accuracy": 0.24129972159862517,
      "num_tokens": 208228893.0,
      "step": 100335
    },
    {
      "entropy": 5.636558151245117,
      "epoch": 9.101959361393323,
      "grad_norm": 1.46875,
      "learning_rate": 5.9043113080352276e-05,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2553533881902695,
      "num_tokens": 208239555.0,
      "step": 100340
    },
    {
      "entropy": 5.717797708511353,
      "epoch": 9.102412917271408,
      "grad_norm": 1.5703125,
      "learning_rate": 5.903403428594232e-05,
      "loss": 4.808,
      "mean_token_accuracy": 0.23862784802913667,
      "num_tokens": 208249126.0,
      "step": 100345
    },
    {
      "entropy": 5.691369676589966,
      "epoch": 9.102866473149492,
      "grad_norm": 1.578125,
      "learning_rate": 5.9024959957763524e-05,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23767454028129578,
      "num_tokens": 208259787.0,
      "step": 100350
    },
    {
      "entropy": 5.667048597335816,
      "epoch": 9.103320029027577,
      "grad_norm": 1.546875,
      "learning_rate": 5.901589009600367e-05,
      "loss": 4.7847,
      "mean_token_accuracy": 0.24172621965408325,
      "num_tokens": 208269999.0,
      "step": 100355
    },
    {
      "entropy": 5.710177230834961,
      "epoch": 9.10377358490566,
      "grad_norm": 1.4375,
      "learning_rate": 5.900682470085021e-05,
      "loss": 4.8626,
      "mean_token_accuracy": 0.22932997196912766,
      "num_tokens": 208281325.0,
      "step": 100360
    },
    {
      "entropy": 5.685444021224976,
      "epoch": 9.104227140783745,
      "grad_norm": 1.6640625,
      "learning_rate": 5.8997763772490644e-05,
      "loss": 4.7106,
      "mean_token_accuracy": 0.24266132861375808,
      "num_tokens": 208291577.0,
      "step": 100365
    },
    {
      "entropy": 5.692441082000732,
      "epoch": 9.104680696661829,
      "grad_norm": 1.5859375,
      "learning_rate": 5.898870731111239e-05,
      "loss": 4.7379,
      "mean_token_accuracy": 0.2406261995434761,
      "num_tokens": 208302268.0,
      "step": 100370
    },
    {
      "entropy": 5.751064252853394,
      "epoch": 9.105134252539912,
      "grad_norm": 1.46875,
      "learning_rate": 5.897965531690274e-05,
      "loss": 4.898,
      "mean_token_accuracy": 0.23187757283449173,
      "num_tokens": 208312473.0,
      "step": 100375
    },
    {
      "entropy": 5.7284118175506595,
      "epoch": 9.105587808417997,
      "grad_norm": 1.46875,
      "learning_rate": 5.8970607790048826e-05,
      "loss": 4.8712,
      "mean_token_accuracy": 0.2267760828137398,
      "num_tokens": 208322531.0,
      "step": 100380
    },
    {
      "entropy": 5.650135517120361,
      "epoch": 9.10604136429608,
      "grad_norm": 1.375,
      "learning_rate": 5.896156473073781e-05,
      "loss": 4.7541,
      "mean_token_accuracy": 0.23851576298475266,
      "num_tokens": 208334122.0,
      "step": 100385
    },
    {
      "entropy": 5.647714233398437,
      "epoch": 9.106494920174166,
      "grad_norm": 1.5390625,
      "learning_rate": 5.895252613915669e-05,
      "loss": 4.7149,
      "mean_token_accuracy": 0.2437903806567192,
      "num_tokens": 208343565.0,
      "step": 100390
    },
    {
      "entropy": 5.66266279220581,
      "epoch": 9.10694847605225,
      "grad_norm": 1.484375,
      "learning_rate": 5.894349201549236e-05,
      "loss": 4.7506,
      "mean_token_accuracy": 0.24659538567066192,
      "num_tokens": 208354972.0,
      "step": 100395
    },
    {
      "entropy": 5.639678239822388,
      "epoch": 9.107402031930334,
      "grad_norm": 1.4765625,
      "learning_rate": 5.8934462359931695e-05,
      "loss": 4.7583,
      "mean_token_accuracy": 0.23528607189655304,
      "num_tokens": 208364522.0,
      "step": 100400
    },
    {
      "entropy": 5.760233974456787,
      "epoch": 9.107855587808418,
      "grad_norm": 1.5625,
      "learning_rate": 5.8925437172661416e-05,
      "loss": 4.8913,
      "mean_token_accuracy": 0.23503908514976501,
      "num_tokens": 208375862.0,
      "step": 100405
    },
    {
      "entropy": 5.625657415390014,
      "epoch": 9.108309143686503,
      "grad_norm": 1.640625,
      "learning_rate": 5.891641645386815e-05,
      "loss": 4.752,
      "mean_token_accuracy": 0.245969820022583,
      "num_tokens": 208386054.0,
      "step": 100410
    },
    {
      "entropy": 5.673351192474366,
      "epoch": 9.108762699564586,
      "grad_norm": 1.4765625,
      "learning_rate": 5.890740020373843e-05,
      "loss": 4.7738,
      "mean_token_accuracy": 0.2378006622195244,
      "num_tokens": 208395803.0,
      "step": 100415
    },
    {
      "entropy": 5.695439100265503,
      "epoch": 9.10921625544267,
      "grad_norm": 1.6171875,
      "learning_rate": 5.8898388422458766e-05,
      "loss": 4.8835,
      "mean_token_accuracy": 0.23160387426614762,
      "num_tokens": 208406576.0,
      "step": 100420
    },
    {
      "entropy": 5.632154703140259,
      "epoch": 9.109669811320755,
      "grad_norm": 1.5390625,
      "learning_rate": 5.8889381110215494e-05,
      "loss": 4.7078,
      "mean_token_accuracy": 0.24309991300106049,
      "num_tokens": 208417633.0,
      "step": 100425
    },
    {
      "entropy": 5.711025714874268,
      "epoch": 9.110123367198838,
      "grad_norm": 2.109375,
      "learning_rate": 5.888037826719489e-05,
      "loss": 4.7832,
      "mean_token_accuracy": 0.23859310746192933,
      "num_tokens": 208427389.0,
      "step": 100430
    },
    {
      "entropy": 5.733399772644043,
      "epoch": 9.110576923076923,
      "grad_norm": 1.59375,
      "learning_rate": 5.887137989358316e-05,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2352724090218544,
      "num_tokens": 208437264.0,
      "step": 100435
    },
    {
      "entropy": 5.672470378875732,
      "epoch": 9.111030478955007,
      "grad_norm": 1.515625,
      "learning_rate": 5.8862385989566346e-05,
      "loss": 4.7073,
      "mean_token_accuracy": 0.2473442003130913,
      "num_tokens": 208447269.0,
      "step": 100440
    },
    {
      "entropy": 5.687772464752197,
      "epoch": 9.111484034833092,
      "grad_norm": 1.4765625,
      "learning_rate": 5.8853396555330444e-05,
      "loss": 4.6868,
      "mean_token_accuracy": 0.24534082412719727,
      "num_tokens": 208457566.0,
      "step": 100445
    },
    {
      "entropy": 5.667999362945556,
      "epoch": 9.111937590711175,
      "grad_norm": 1.671875,
      "learning_rate": 5.8844411591061426e-05,
      "loss": 4.8163,
      "mean_token_accuracy": 0.2439817816019058,
      "num_tokens": 208467858.0,
      "step": 100450
    },
    {
      "entropy": 5.675578832626343,
      "epoch": 9.11239114658926,
      "grad_norm": 1.5703125,
      "learning_rate": 5.883543109694504e-05,
      "loss": 4.823,
      "mean_token_accuracy": 0.2420311078429222,
      "num_tokens": 208478488.0,
      "step": 100455
    },
    {
      "entropy": 5.687390947341919,
      "epoch": 9.112844702467344,
      "grad_norm": 1.5546875,
      "learning_rate": 5.8826455073167045e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2445591002702713,
      "num_tokens": 208489351.0,
      "step": 100460
    },
    {
      "entropy": 5.720354318618774,
      "epoch": 9.113298258345429,
      "grad_norm": 1.4375,
      "learning_rate": 5.881748351991302e-05,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2419312432408333,
      "num_tokens": 208499748.0,
      "step": 100465
    },
    {
      "entropy": 5.661149168014527,
      "epoch": 9.113751814223512,
      "grad_norm": 1.5859375,
      "learning_rate": 5.8808516437368554e-05,
      "loss": 4.7323,
      "mean_token_accuracy": 0.24337441474199295,
      "num_tokens": 208509192.0,
      "step": 100470
    },
    {
      "entropy": 5.6915857791900635,
      "epoch": 9.114205370101596,
      "grad_norm": 1.5546875,
      "learning_rate": 5.879955382571903e-05,
      "loss": 4.8037,
      "mean_token_accuracy": 0.23801302313804626,
      "num_tokens": 208520032.0,
      "step": 100475
    },
    {
      "entropy": 5.655043745040894,
      "epoch": 9.11465892597968,
      "grad_norm": 1.7265625,
      "learning_rate": 5.879059568514981e-05,
      "loss": 4.7365,
      "mean_token_accuracy": 0.2480214238166809,
      "num_tokens": 208530315.0,
      "step": 100480
    },
    {
      "entropy": 5.6282659530639645,
      "epoch": 9.115112481857764,
      "grad_norm": 1.578125,
      "learning_rate": 5.878164201584621e-05,
      "loss": 4.687,
      "mean_token_accuracy": 0.2498370051383972,
      "num_tokens": 208540192.0,
      "step": 100485
    },
    {
      "entropy": 5.685244941711426,
      "epoch": 9.11556603773585,
      "grad_norm": 1.53125,
      "learning_rate": 5.877269281799334e-05,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23203280419111252,
      "num_tokens": 208551204.0,
      "step": 100490
    },
    {
      "entropy": 5.70385103225708,
      "epoch": 9.116019593613933,
      "grad_norm": 1.5390625,
      "learning_rate": 5.876374809177629e-05,
      "loss": 4.7317,
      "mean_token_accuracy": 0.244780133664608,
      "num_tokens": 208560799.0,
      "step": 100495
    },
    {
      "entropy": 5.607622385025024,
      "epoch": 9.116473149492018,
      "grad_norm": 1.5078125,
      "learning_rate": 5.875480783738004e-05,
      "loss": 4.7109,
      "mean_token_accuracy": 0.2424223616719246,
      "num_tokens": 208570870.0,
      "step": 100500
    },
    {
      "entropy": 5.648723077774048,
      "epoch": 9.116926705370101,
      "grad_norm": 1.5546875,
      "learning_rate": 5.8745872054989467e-05,
      "loss": 4.7172,
      "mean_token_accuracy": 0.24905333518981934,
      "num_tokens": 208580052.0,
      "step": 100505
    },
    {
      "entropy": 5.6347133159637455,
      "epoch": 9.117380261248186,
      "grad_norm": 1.59375,
      "learning_rate": 5.873694074478935e-05,
      "loss": 4.721,
      "mean_token_accuracy": 0.2451910614967346,
      "num_tokens": 208589391.0,
      "step": 100510
    },
    {
      "entropy": 5.650158786773682,
      "epoch": 9.11783381712627,
      "grad_norm": 1.421875,
      "learning_rate": 5.8728013906964384e-05,
      "loss": 4.796,
      "mean_token_accuracy": 0.2408634439110756,
      "num_tokens": 208598998.0,
      "step": 100515
    },
    {
      "entropy": 5.669038105010986,
      "epoch": 9.118287373004355,
      "grad_norm": 1.5703125,
      "learning_rate": 5.871909154169924e-05,
      "loss": 4.7688,
      "mean_token_accuracy": 0.240908282995224,
      "num_tokens": 208608673.0,
      "step": 100520
    },
    {
      "entropy": 5.6927882671356205,
      "epoch": 9.118740928882438,
      "grad_norm": 1.6328125,
      "learning_rate": 5.8710173649178396e-05,
      "loss": 4.7763,
      "mean_token_accuracy": 0.24018320441246033,
      "num_tokens": 208618685.0,
      "step": 100525
    },
    {
      "entropy": 5.678165340423584,
      "epoch": 9.119194484760522,
      "grad_norm": 1.7109375,
      "learning_rate": 5.870126022958628e-05,
      "loss": 4.7459,
      "mean_token_accuracy": 0.2534042552113533,
      "num_tokens": 208628260.0,
      "step": 100530
    },
    {
      "entropy": 5.687825727462768,
      "epoch": 9.119648040638607,
      "grad_norm": 1.359375,
      "learning_rate": 5.8692351283107224e-05,
      "loss": 4.735,
      "mean_token_accuracy": 0.24149706363677978,
      "num_tokens": 208638709.0,
      "step": 100535
    },
    {
      "entropy": 5.6267674446105955,
      "epoch": 9.12010159651669,
      "grad_norm": 1.484375,
      "learning_rate": 5.868344680992547e-05,
      "loss": 4.7367,
      "mean_token_accuracy": 0.2376599207520485,
      "num_tokens": 208649413.0,
      "step": 100540
    },
    {
      "entropy": 5.643406772613526,
      "epoch": 9.120555152394775,
      "grad_norm": 1.46875,
      "learning_rate": 5.8674546810225145e-05,
      "loss": 4.7255,
      "mean_token_accuracy": 0.24763089418411255,
      "num_tokens": 208659628.0,
      "step": 100545
    },
    {
      "entropy": 5.689602518081665,
      "epoch": 9.121008708272859,
      "grad_norm": 1.53125,
      "learning_rate": 5.8665651284190297e-05,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2349516674876213,
      "num_tokens": 208669678.0,
      "step": 100550
    },
    {
      "entropy": 5.665335607528687,
      "epoch": 9.121462264150944,
      "grad_norm": 1.4921875,
      "learning_rate": 5.8656760232004934e-05,
      "loss": 4.7766,
      "mean_token_accuracy": 0.24294260442256926,
      "num_tokens": 208679490.0,
      "step": 100555
    },
    {
      "entropy": 5.64510326385498,
      "epoch": 9.121915820029027,
      "grad_norm": 1.5859375,
      "learning_rate": 5.8647873653852905e-05,
      "loss": 4.7638,
      "mean_token_accuracy": 0.24101511985063553,
      "num_tokens": 208689958.0,
      "step": 100560
    },
    {
      "entropy": 5.642427444458008,
      "epoch": 9.122369375907112,
      "grad_norm": 1.59375,
      "learning_rate": 5.863899154991799e-05,
      "loss": 4.7315,
      "mean_token_accuracy": 0.24628621339797974,
      "num_tokens": 208700124.0,
      "step": 100565
    },
    {
      "entropy": 5.646360731124878,
      "epoch": 9.122822931785196,
      "grad_norm": 1.421875,
      "learning_rate": 5.863011392038384e-05,
      "loss": 4.7774,
      "mean_token_accuracy": 0.23598040640354156,
      "num_tokens": 208710661.0,
      "step": 100570
    },
    {
      "entropy": 5.639346551895142,
      "epoch": 9.12327648766328,
      "grad_norm": 1.5625,
      "learning_rate": 5.862124076543409e-05,
      "loss": 4.7543,
      "mean_token_accuracy": 0.24012050181627273,
      "num_tokens": 208720547.0,
      "step": 100575
    },
    {
      "entropy": 5.705797290802002,
      "epoch": 9.123730043541364,
      "grad_norm": 1.4140625,
      "learning_rate": 5.8612372085252224e-05,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23980868458747864,
      "num_tokens": 208731203.0,
      "step": 100580
    },
    {
      "entropy": 5.687053394317627,
      "epoch": 9.124183599419448,
      "grad_norm": 1.53125,
      "learning_rate": 5.8603507880021595e-05,
      "loss": 4.7901,
      "mean_token_accuracy": 0.24235571175813675,
      "num_tokens": 208740600.0,
      "step": 100585
    },
    {
      "entropy": 5.6751861572265625,
      "epoch": 9.124637155297533,
      "grad_norm": 1.5703125,
      "learning_rate": 5.8594648149925626e-05,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22604013234376907,
      "num_tokens": 208749721.0,
      "step": 100590
    },
    {
      "entropy": 5.666849374771118,
      "epoch": 9.125090711175616,
      "grad_norm": 1.4296875,
      "learning_rate": 5.858579289514746e-05,
      "loss": 4.7849,
      "mean_token_accuracy": 0.23242483288049698,
      "num_tokens": 208759984.0,
      "step": 100595
    },
    {
      "entropy": 5.712153196334839,
      "epoch": 9.125544267053701,
      "grad_norm": 1.5859375,
      "learning_rate": 5.857694211587025e-05,
      "loss": 4.818,
      "mean_token_accuracy": 0.2352354794740677,
      "num_tokens": 208769960.0,
      "step": 100600
    },
    {
      "entropy": 5.73536787033081,
      "epoch": 9.125997822931785,
      "grad_norm": 1.515625,
      "learning_rate": 5.8568095812277036e-05,
      "loss": 4.8987,
      "mean_token_accuracy": 0.23339225351810455,
      "num_tokens": 208780068.0,
      "step": 100605
    },
    {
      "entropy": 5.721828031539917,
      "epoch": 9.12645137880987,
      "grad_norm": 1.546875,
      "learning_rate": 5.855925398455074e-05,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2390957295894623,
      "num_tokens": 208790595.0,
      "step": 100610
    },
    {
      "entropy": 5.6502313137054445,
      "epoch": 9.126904934687953,
      "grad_norm": 1.5859375,
      "learning_rate": 5.855041663287424e-05,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2445285588502884,
      "num_tokens": 208800705.0,
      "step": 100615
    },
    {
      "entropy": 5.657447052001953,
      "epoch": 9.127358490566039,
      "grad_norm": 1.421875,
      "learning_rate": 5.854158375743025e-05,
      "loss": 4.8115,
      "mean_token_accuracy": 0.2363432303071022,
      "num_tokens": 208811964.0,
      "step": 100620
    },
    {
      "entropy": 5.60571699142456,
      "epoch": 9.127812046444122,
      "grad_norm": 1.515625,
      "learning_rate": 5.853275535840149e-05,
      "loss": 4.7514,
      "mean_token_accuracy": 0.247608882188797,
      "num_tokens": 208822606.0,
      "step": 100625
    },
    {
      "entropy": 5.6531274795532225,
      "epoch": 9.128265602322205,
      "grad_norm": 1.515625,
      "learning_rate": 5.852393143597051e-05,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2528596669435501,
      "num_tokens": 208832229.0,
      "step": 100630
    },
    {
      "entropy": 5.7923379898071286,
      "epoch": 9.12871915820029,
      "grad_norm": 1.4453125,
      "learning_rate": 5.851511199031982e-05,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22487399429082872,
      "num_tokens": 208842606.0,
      "step": 100635
    },
    {
      "entropy": 5.672113370895386,
      "epoch": 9.129172714078374,
      "grad_norm": 1.4609375,
      "learning_rate": 5.850629702163177e-05,
      "loss": 4.7383,
      "mean_token_accuracy": 0.23686694353818893,
      "num_tokens": 208853093.0,
      "step": 100640
    },
    {
      "entropy": 5.747565507888794,
      "epoch": 9.129626269956459,
      "grad_norm": 1.53125,
      "learning_rate": 5.849748653008863e-05,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23817726820707322,
      "num_tokens": 208862590.0,
      "step": 100645
    },
    {
      "entropy": 5.748446893692017,
      "epoch": 9.130079825834542,
      "grad_norm": 1.4765625,
      "learning_rate": 5.848868051587266e-05,
      "loss": 4.8789,
      "mean_token_accuracy": 0.23511074036359786,
      "num_tokens": 208872115.0,
      "step": 100650
    },
    {
      "entropy": 5.666124868392944,
      "epoch": 9.130533381712628,
      "grad_norm": 1.703125,
      "learning_rate": 5.847987897916595e-05,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23594016432762147,
      "num_tokens": 208882730.0,
      "step": 100655
    },
    {
      "entropy": 5.664472341537476,
      "epoch": 9.130986937590711,
      "grad_norm": 1.4921875,
      "learning_rate": 5.847108192015054e-05,
      "loss": 4.7645,
      "mean_token_accuracy": 0.2404848262667656,
      "num_tokens": 208892500.0,
      "step": 100660
    },
    {
      "entropy": 5.661363506317139,
      "epoch": 9.131440493468796,
      "grad_norm": 1.4921875,
      "learning_rate": 5.846228933900831e-05,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23633595556020737,
      "num_tokens": 208902628.0,
      "step": 100665
    },
    {
      "entropy": 5.633824396133423,
      "epoch": 9.13189404934688,
      "grad_norm": 1.5625,
      "learning_rate": 5.845350123592112e-05,
      "loss": 4.7103,
      "mean_token_accuracy": 0.256799179315567,
      "num_tokens": 208912374.0,
      "step": 100670
    },
    {
      "entropy": 5.722471809387207,
      "epoch": 9.132347605224965,
      "grad_norm": 1.578125,
      "learning_rate": 5.844471761107071e-05,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2344381630420685,
      "num_tokens": 208922207.0,
      "step": 100675
    },
    {
      "entropy": 5.7745472431182865,
      "epoch": 9.132801161103048,
      "grad_norm": 1.6484375,
      "learning_rate": 5.843593846463873e-05,
      "loss": 4.9306,
      "mean_token_accuracy": 0.23311521410942077,
      "num_tokens": 208931871.0,
      "step": 100680
    },
    {
      "entropy": 5.6635246753692625,
      "epoch": 9.133254716981131,
      "grad_norm": 1.6953125,
      "learning_rate": 5.842716379680669e-05,
      "loss": 4.729,
      "mean_token_accuracy": 0.2568617552518845,
      "num_tokens": 208942446.0,
      "step": 100685
    },
    {
      "entropy": 5.679634618759155,
      "epoch": 9.133708272859216,
      "grad_norm": 1.59375,
      "learning_rate": 5.841839360775614e-05,
      "loss": 4.7216,
      "mean_token_accuracy": 0.24603582322597503,
      "num_tokens": 208951415.0,
      "step": 100690
    },
    {
      "entropy": 5.62731785774231,
      "epoch": 9.1341618287373,
      "grad_norm": 1.609375,
      "learning_rate": 5.84096278976684e-05,
      "loss": 4.709,
      "mean_token_accuracy": 0.24286814481019975,
      "num_tokens": 208960832.0,
      "step": 100695
    },
    {
      "entropy": 5.736389446258545,
      "epoch": 9.134615384615385,
      "grad_norm": 1.53125,
      "learning_rate": 5.8400866666724744e-05,
      "loss": 4.8291,
      "mean_token_accuracy": 0.2400259405374527,
      "num_tokens": 208972554.0,
      "step": 100700
    },
    {
      "entropy": 5.738260221481323,
      "epoch": 9.135068940493468,
      "grad_norm": 1.515625,
      "learning_rate": 5.839210991510635e-05,
      "loss": 4.8801,
      "mean_token_accuracy": 0.2354903981089592,
      "num_tokens": 208982544.0,
      "step": 100705
    },
    {
      "entropy": 5.687513160705566,
      "epoch": 9.135522496371554,
      "grad_norm": 1.6015625,
      "learning_rate": 5.8383357642994335e-05,
      "loss": 4.7364,
      "mean_token_accuracy": 0.24640734493732452,
      "num_tokens": 208992589.0,
      "step": 100710
    },
    {
      "entropy": 5.639460039138794,
      "epoch": 9.135976052249637,
      "grad_norm": 1.4765625,
      "learning_rate": 5.837460985056968e-05,
      "loss": 4.7561,
      "mean_token_accuracy": 0.2409678190946579,
      "num_tokens": 209003389.0,
      "step": 100715
    },
    {
      "entropy": 5.714088201522827,
      "epoch": 9.136429608127722,
      "grad_norm": 1.4765625,
      "learning_rate": 5.836586653801327e-05,
      "loss": 4.8656,
      "mean_token_accuracy": 0.2386513754725456,
      "num_tokens": 209013655.0,
      "step": 100720
    },
    {
      "entropy": 5.725073146820068,
      "epoch": 9.136883164005805,
      "grad_norm": 1.3515625,
      "learning_rate": 5.835712770550597e-05,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2316157877445221,
      "num_tokens": 209024397.0,
      "step": 100725
    },
    {
      "entropy": 5.67950668334961,
      "epoch": 9.137336719883889,
      "grad_norm": 1.484375,
      "learning_rate": 5.834839335322847e-05,
      "loss": 4.7556,
      "mean_token_accuracy": 0.24789571911096572,
      "num_tokens": 209034123.0,
      "step": 100730
    },
    {
      "entropy": 5.601857471466064,
      "epoch": 9.137790275761974,
      "grad_norm": 1.4453125,
      "learning_rate": 5.83396634813614e-05,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2456949383020401,
      "num_tokens": 209044632.0,
      "step": 100735
    },
    {
      "entropy": 5.684118032455444,
      "epoch": 9.138243831640057,
      "grad_norm": 1.640625,
      "learning_rate": 5.8330938090085327e-05,
      "loss": 4.8212,
      "mean_token_accuracy": 0.23082095831632615,
      "num_tokens": 209054216.0,
      "step": 100740
    },
    {
      "entropy": 5.663836145401001,
      "epoch": 9.138697387518143,
      "grad_norm": 1.703125,
      "learning_rate": 5.8322217179580636e-05,
      "loss": 4.783,
      "mean_token_accuracy": 0.2423141598701477,
      "num_tokens": 209064037.0,
      "step": 100745
    },
    {
      "entropy": 5.742472791671753,
      "epoch": 9.139150943396226,
      "grad_norm": 1.6171875,
      "learning_rate": 5.831350075002773e-05,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2196595326066017,
      "num_tokens": 209075943.0,
      "step": 100750
    },
    {
      "entropy": 5.632445478439331,
      "epoch": 9.139604499274311,
      "grad_norm": 1.625,
      "learning_rate": 5.830478880160683e-05,
      "loss": 4.7661,
      "mean_token_accuracy": 0.23800324648618698,
      "num_tokens": 209086201.0,
      "step": 100755
    },
    {
      "entropy": 5.725170278549195,
      "epoch": 9.140058055152394,
      "grad_norm": 1.4921875,
      "learning_rate": 5.829608133449813e-05,
      "loss": 4.8069,
      "mean_token_accuracy": 0.23363422155380248,
      "num_tokens": 209096568.0,
      "step": 100760
    },
    {
      "entropy": 5.662164211273193,
      "epoch": 9.14051161103048,
      "grad_norm": 1.765625,
      "learning_rate": 5.828737834888169e-05,
      "loss": 4.751,
      "mean_token_accuracy": 0.23829906880855561,
      "num_tokens": 209106484.0,
      "step": 100765
    },
    {
      "entropy": 5.719658422470093,
      "epoch": 9.140965166908563,
      "grad_norm": 1.6484375,
      "learning_rate": 5.8278679844937494e-05,
      "loss": 4.8453,
      "mean_token_accuracy": 0.23315465152263642,
      "num_tokens": 209118593.0,
      "step": 100770
    },
    {
      "entropy": 5.625981283187866,
      "epoch": 9.141418722786648,
      "grad_norm": 1.6953125,
      "learning_rate": 5.826998582284543e-05,
      "loss": 4.7173,
      "mean_token_accuracy": 0.24447306841611863,
      "num_tokens": 209128827.0,
      "step": 100775
    },
    {
      "entropy": 5.692573404312133,
      "epoch": 9.141872278664732,
      "grad_norm": 1.6015625,
      "learning_rate": 5.826129628278527e-05,
      "loss": 4.754,
      "mean_token_accuracy": 0.24325598180294036,
      "num_tokens": 209139057.0,
      "step": 100780
    },
    {
      "entropy": 5.762549638748169,
      "epoch": 9.142325834542815,
      "grad_norm": 1.5078125,
      "learning_rate": 5.825261122493675e-05,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2312890037894249,
      "num_tokens": 209149162.0,
      "step": 100785
    },
    {
      "entropy": 5.767986106872558,
      "epoch": 9.1427793904209,
      "grad_norm": 1.5703125,
      "learning_rate": 5.824393064947943e-05,
      "loss": 4.891,
      "mean_token_accuracy": 0.2376765325665474,
      "num_tokens": 209159557.0,
      "step": 100790
    },
    {
      "entropy": 5.667447853088379,
      "epoch": 9.143232946298983,
      "grad_norm": 1.59375,
      "learning_rate": 5.82352545565929e-05,
      "loss": 4.7772,
      "mean_token_accuracy": 0.2446452111005783,
      "num_tokens": 209168542.0,
      "step": 100795
    },
    {
      "entropy": 5.6121961116790775,
      "epoch": 9.143686502177069,
      "grad_norm": 1.4296875,
      "learning_rate": 5.822658294645654e-05,
      "loss": 4.6569,
      "mean_token_accuracy": 0.25043363124132156,
      "num_tokens": 209179188.0,
      "step": 100800
    },
    {
      "entropy": 5.690994548797607,
      "epoch": 9.144140058055152,
      "grad_norm": 1.484375,
      "learning_rate": 5.8217915819249716e-05,
      "loss": 4.8018,
      "mean_token_accuracy": 0.23506174832582474,
      "num_tokens": 209190203.0,
      "step": 100805
    },
    {
      "entropy": 5.72690978050232,
      "epoch": 9.144593613933237,
      "grad_norm": 1.40625,
      "learning_rate": 5.820925317515159e-05,
      "loss": 4.845,
      "mean_token_accuracy": 0.23228400498628615,
      "num_tokens": 209200807.0,
      "step": 100810
    },
    {
      "entropy": 5.714183521270752,
      "epoch": 9.14504716981132,
      "grad_norm": 1.6875,
      "learning_rate": 5.8200595014341345e-05,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2262909933924675,
      "num_tokens": 209211139.0,
      "step": 100815
    },
    {
      "entropy": 5.676371145248413,
      "epoch": 9.145500725689406,
      "grad_norm": 1.609375,
      "learning_rate": 5.819194133699806e-05,
      "loss": 4.785,
      "mean_token_accuracy": 0.24173903912305833,
      "num_tokens": 209221512.0,
      "step": 100820
    },
    {
      "entropy": 5.722708654403687,
      "epoch": 9.145954281567489,
      "grad_norm": 1.5,
      "learning_rate": 5.818329214330068e-05,
      "loss": 4.8258,
      "mean_token_accuracy": 0.23304129391908646,
      "num_tokens": 209231454.0,
      "step": 100825
    },
    {
      "entropy": 5.620232486724854,
      "epoch": 9.146407837445572,
      "grad_norm": 1.5625,
      "learning_rate": 5.8174647433428066e-05,
      "loss": 4.7104,
      "mean_token_accuracy": 0.24325830340385438,
      "num_tokens": 209240814.0,
      "step": 100830
    },
    {
      "entropy": 5.770507383346557,
      "epoch": 9.146861393323658,
      "grad_norm": 1.5546875,
      "learning_rate": 5.816600720755899e-05,
      "loss": 4.8851,
      "mean_token_accuracy": 0.23613614439964295,
      "num_tokens": 209250673.0,
      "step": 100835
    },
    {
      "entropy": 5.638852500915528,
      "epoch": 9.147314949201741,
      "grad_norm": 1.546875,
      "learning_rate": 5.815737146587215e-05,
      "loss": 4.7331,
      "mean_token_accuracy": 0.24341936260461808,
      "num_tokens": 209260855.0,
      "step": 100840
    },
    {
      "entropy": 5.626227903366089,
      "epoch": 9.147768505079826,
      "grad_norm": 1.5078125,
      "learning_rate": 5.814874020854613e-05,
      "loss": 4.6702,
      "mean_token_accuracy": 0.24526024907827376,
      "num_tokens": 209271335.0,
      "step": 100845
    },
    {
      "entropy": 5.6444295883178714,
      "epoch": 9.14822206095791,
      "grad_norm": 1.4609375,
      "learning_rate": 5.8140113435759384e-05,
      "loss": 4.7241,
      "mean_token_accuracy": 0.2423413410782814,
      "num_tokens": 209280793.0,
      "step": 100850
    },
    {
      "entropy": 5.704995107650757,
      "epoch": 9.148675616835995,
      "grad_norm": 1.53125,
      "learning_rate": 5.8131491147690385e-05,
      "loss": 4.7833,
      "mean_token_accuracy": 0.24261248260736465,
      "num_tokens": 209290003.0,
      "step": 100855
    },
    {
      "entropy": 5.746371698379517,
      "epoch": 9.149129172714078,
      "grad_norm": 1.609375,
      "learning_rate": 5.8122873344517436e-05,
      "loss": 4.9208,
      "mean_token_accuracy": 0.23145721554756166,
      "num_tokens": 209300542.0,
      "step": 100860
    },
    {
      "entropy": 5.627858018875122,
      "epoch": 9.149582728592163,
      "grad_norm": 1.4609375,
      "learning_rate": 5.8114260026418704e-05,
      "loss": 4.7509,
      "mean_token_accuracy": 0.24546722322702408,
      "num_tokens": 209311212.0,
      "step": 100865
    },
    {
      "entropy": 5.672444105148315,
      "epoch": 9.150036284470247,
      "grad_norm": 1.640625,
      "learning_rate": 5.810565119357236e-05,
      "loss": 4.7394,
      "mean_token_accuracy": 0.24032308906316757,
      "num_tokens": 209321597.0,
      "step": 100870
    },
    {
      "entropy": 5.7065101146698,
      "epoch": 9.150489840348332,
      "grad_norm": 1.5078125,
      "learning_rate": 5.8097046846156426e-05,
      "loss": 4.8196,
      "mean_token_accuracy": 0.24180746227502822,
      "num_tokens": 209331854.0,
      "step": 100875
    },
    {
      "entropy": 5.690631771087647,
      "epoch": 9.150943396226415,
      "grad_norm": 1.546875,
      "learning_rate": 5.8088446984348816e-05,
      "loss": 4.849,
      "mean_token_accuracy": 0.23577394783496858,
      "num_tokens": 209343776.0,
      "step": 100880
    },
    {
      "entropy": 5.734452867507935,
      "epoch": 9.151396952104498,
      "grad_norm": 1.5703125,
      "learning_rate": 5.807985160832738e-05,
      "loss": 4.8459,
      "mean_token_accuracy": 0.23387833535671235,
      "num_tokens": 209353487.0,
      "step": 100885
    },
    {
      "entropy": 5.756203556060791,
      "epoch": 9.151850507982584,
      "grad_norm": 1.4140625,
      "learning_rate": 5.807126071826992e-05,
      "loss": 4.8704,
      "mean_token_accuracy": 0.23589495867490767,
      "num_tokens": 209364689.0,
      "step": 100890
    },
    {
      "entropy": 5.704490756988525,
      "epoch": 9.152304063860667,
      "grad_norm": 1.5,
      "learning_rate": 5.8062674314354064e-05,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23984191119670867,
      "num_tokens": 209374316.0,
      "step": 100895
    },
    {
      "entropy": 5.659917974472046,
      "epoch": 9.152757619738752,
      "grad_norm": 1.546875,
      "learning_rate": 5.8054092396757377e-05,
      "loss": 4.7413,
      "mean_token_accuracy": 0.2458326503634453,
      "num_tokens": 209384561.0,
      "step": 100900
    },
    {
      "entropy": 5.682030057907104,
      "epoch": 9.153211175616836,
      "grad_norm": 1.453125,
      "learning_rate": 5.804551496565737e-05,
      "loss": 4.7521,
      "mean_token_accuracy": 0.23744242638349533,
      "num_tokens": 209395971.0,
      "step": 100905
    },
    {
      "entropy": 5.751417446136474,
      "epoch": 9.15366473149492,
      "grad_norm": 1.40625,
      "learning_rate": 5.8036942021231384e-05,
      "loss": 4.889,
      "mean_token_accuracy": 0.23264411091804504,
      "num_tokens": 209406265.0,
      "step": 100910
    },
    {
      "entropy": 5.639829778671265,
      "epoch": 9.154118287373004,
      "grad_norm": 1.5703125,
      "learning_rate": 5.802837356365672e-05,
      "loss": 4.7501,
      "mean_token_accuracy": 0.23884500414133072,
      "num_tokens": 209416571.0,
      "step": 100915
    },
    {
      "entropy": 5.683706617355346,
      "epoch": 9.15457184325109,
      "grad_norm": 1.5703125,
      "learning_rate": 5.8019809593110584e-05,
      "loss": 4.7628,
      "mean_token_accuracy": 0.24442636221647263,
      "num_tokens": 209427991.0,
      "step": 100920
    },
    {
      "entropy": 5.7485206604003904,
      "epoch": 9.155025399129173,
      "grad_norm": 1.4921875,
      "learning_rate": 5.801125010977007e-05,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22974785417318344,
      "num_tokens": 209438382.0,
      "step": 100925
    },
    {
      "entropy": 5.773911571502685,
      "epoch": 9.155478955007258,
      "grad_norm": 1.46875,
      "learning_rate": 5.8002695113812205e-05,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23166837692260742,
      "num_tokens": 209449208.0,
      "step": 100930
    },
    {
      "entropy": 5.740772819519043,
      "epoch": 9.155932510885341,
      "grad_norm": 1.53125,
      "learning_rate": 5.7994144605413914e-05,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2363950252532959,
      "num_tokens": 209459315.0,
      "step": 100935
    },
    {
      "entropy": 5.725565338134766,
      "epoch": 9.156386066763424,
      "grad_norm": 1.421875,
      "learning_rate": 5.7985598584752025e-05,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2327394053339958,
      "num_tokens": 209468971.0,
      "step": 100940
    },
    {
      "entropy": 5.659408903121948,
      "epoch": 9.15683962264151,
      "grad_norm": 1.5,
      "learning_rate": 5.797705705200323e-05,
      "loss": 4.733,
      "mean_token_accuracy": 0.23992914855480194,
      "num_tokens": 209478453.0,
      "step": 100945
    },
    {
      "entropy": 5.668317699432373,
      "epoch": 9.157293178519593,
      "grad_norm": 1.5703125,
      "learning_rate": 5.7968520007344176e-05,
      "loss": 4.7492,
      "mean_token_accuracy": 0.24402296394109727,
      "num_tokens": 209488199.0,
      "step": 100950
    },
    {
      "entropy": 5.688090181350708,
      "epoch": 9.157746734397678,
      "grad_norm": 1.5546875,
      "learning_rate": 5.795998745095147e-05,
      "loss": 4.8252,
      "mean_token_accuracy": 0.23492904007434845,
      "num_tokens": 209500308.0,
      "step": 100955
    },
    {
      "entropy": 5.689174222946167,
      "epoch": 9.158200290275762,
      "grad_norm": 1.515625,
      "learning_rate": 5.795145938300154e-05,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23843558877706528,
      "num_tokens": 209510864.0,
      "step": 100960
    },
    {
      "entropy": 5.652224397659301,
      "epoch": 9.158653846153847,
      "grad_norm": 1.5625,
      "learning_rate": 5.794293580367073e-05,
      "loss": 4.7803,
      "mean_token_accuracy": 0.23966164141893387,
      "num_tokens": 209521420.0,
      "step": 100965
    },
    {
      "entropy": 5.649929189682007,
      "epoch": 9.15910740203193,
      "grad_norm": 1.46875,
      "learning_rate": 5.7934416713135317e-05,
      "loss": 4.8076,
      "mean_token_accuracy": 0.2364630177617073,
      "num_tokens": 209531768.0,
      "step": 100970
    },
    {
      "entropy": 5.677593660354614,
      "epoch": 9.159560957910015,
      "grad_norm": 1.6640625,
      "learning_rate": 5.7925902111571484e-05,
      "loss": 4.7821,
      "mean_token_accuracy": 0.24677774906158448,
      "num_tokens": 209541280.0,
      "step": 100975
    },
    {
      "entropy": 5.726515579223633,
      "epoch": 9.160014513788099,
      "grad_norm": 1.46875,
      "learning_rate": 5.791739199915532e-05,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2346196472644806,
      "num_tokens": 209552302.0,
      "step": 100980
    },
    {
      "entropy": 5.706981611251831,
      "epoch": 9.160468069666184,
      "grad_norm": 1.4375,
      "learning_rate": 5.790888637606277e-05,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2335013121366501,
      "num_tokens": 209563198.0,
      "step": 100985
    },
    {
      "entropy": 5.698384714126587,
      "epoch": 9.160921625544267,
      "grad_norm": 1.4296875,
      "learning_rate": 5.790038524246982e-05,
      "loss": 4.8329,
      "mean_token_accuracy": 0.23392653465270996,
      "num_tokens": 209573625.0,
      "step": 100990
    },
    {
      "entropy": 5.592269515991211,
      "epoch": 9.16137518142235,
      "grad_norm": 1.5625,
      "learning_rate": 5.789188859855221e-05,
      "loss": 4.6676,
      "mean_token_accuracy": 0.24677026122808457,
      "num_tokens": 209583471.0,
      "step": 100995
    },
    {
      "entropy": 5.68595142364502,
      "epoch": 9.161828737300436,
      "grad_norm": 1.6328125,
      "learning_rate": 5.788339644448567e-05,
      "loss": 4.8069,
      "mean_token_accuracy": 0.23636351227760316,
      "num_tokens": 209594452.0,
      "step": 101000
    },
    {
      "entropy": 5.620609092712402,
      "epoch": 9.162282293178519,
      "grad_norm": 1.5625,
      "learning_rate": 5.787490878044582e-05,
      "loss": 4.7566,
      "mean_token_accuracy": 0.24785009324550628,
      "num_tokens": 209604775.0,
      "step": 101005
    },
    {
      "entropy": 5.638047885894776,
      "epoch": 9.162735849056604,
      "grad_norm": 1.53125,
      "learning_rate": 5.786642560660818e-05,
      "loss": 4.719,
      "mean_token_accuracy": 0.23940953910350798,
      "num_tokens": 209614653.0,
      "step": 101010
    },
    {
      "entropy": 5.603724670410156,
      "epoch": 9.163189404934688,
      "grad_norm": 1.6953125,
      "learning_rate": 5.7857946923148215e-05,
      "loss": 4.7135,
      "mean_token_accuracy": 0.2413689225912094,
      "num_tokens": 209624787.0,
      "step": 101015
    },
    {
      "entropy": 5.628799915313721,
      "epoch": 9.163642960812773,
      "grad_norm": 1.484375,
      "learning_rate": 5.784947273024119e-05,
      "loss": 4.6828,
      "mean_token_accuracy": 0.25184241086244585,
      "num_tokens": 209636615.0,
      "step": 101020
    },
    {
      "entropy": 5.7179811000823975,
      "epoch": 9.164096516690856,
      "grad_norm": 1.546875,
      "learning_rate": 5.784100302806244e-05,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23068761378526687,
      "num_tokens": 209646489.0,
      "step": 101025
    },
    {
      "entropy": 5.619156122207642,
      "epoch": 9.164550072568941,
      "grad_norm": 1.46875,
      "learning_rate": 5.783253781678709e-05,
      "loss": 4.728,
      "mean_token_accuracy": 0.24238058030605317,
      "num_tokens": 209656537.0,
      "step": 101030
    },
    {
      "entropy": 5.646154832839966,
      "epoch": 9.165003628447025,
      "grad_norm": 1.71875,
      "learning_rate": 5.782407709659019e-05,
      "loss": 4.7386,
      "mean_token_accuracy": 0.2439085900783539,
      "num_tokens": 209667228.0,
      "step": 101035
    },
    {
      "entropy": 5.69858078956604,
      "epoch": 9.165457184325108,
      "grad_norm": 1.6015625,
      "learning_rate": 5.781562086764674e-05,
      "loss": 4.7791,
      "mean_token_accuracy": 0.24609634131193162,
      "num_tokens": 209677189.0,
      "step": 101040
    },
    {
      "entropy": 5.72174186706543,
      "epoch": 9.165910740203193,
      "grad_norm": 1.390625,
      "learning_rate": 5.780716913013156e-05,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2332765981554985,
      "num_tokens": 209687374.0,
      "step": 101045
    },
    {
      "entropy": 5.666788578033447,
      "epoch": 9.166364296081277,
      "grad_norm": 1.4765625,
      "learning_rate": 5.7798721884219506e-05,
      "loss": 4.8031,
      "mean_token_accuracy": 0.2381643310189247,
      "num_tokens": 209698442.0,
      "step": 101050
    },
    {
      "entropy": 5.676121425628662,
      "epoch": 9.166817851959362,
      "grad_norm": 1.359375,
      "learning_rate": 5.7790279130085176e-05,
      "loss": 4.7903,
      "mean_token_accuracy": 0.23687006682157516,
      "num_tokens": 209710799.0,
      "step": 101055
    },
    {
      "entropy": 5.6613516330719,
      "epoch": 9.167271407837445,
      "grad_norm": 1.484375,
      "learning_rate": 5.778184086790328e-05,
      "loss": 4.7602,
      "mean_token_accuracy": 0.23647212982177734,
      "num_tokens": 209720565.0,
      "step": 101060
    },
    {
      "entropy": 5.7563223361969,
      "epoch": 9.16772496371553,
      "grad_norm": 1.4296875,
      "learning_rate": 5.777340709784825e-05,
      "loss": 4.8877,
      "mean_token_accuracy": 0.22949588745832444,
      "num_tokens": 209731269.0,
      "step": 101065
    },
    {
      "entropy": 5.622594356536865,
      "epoch": 9.168178519593614,
      "grad_norm": 1.4140625,
      "learning_rate": 5.776497782009453e-05,
      "loss": 4.7494,
      "mean_token_accuracy": 0.2410190299153328,
      "num_tokens": 209742609.0,
      "step": 101070
    },
    {
      "entropy": 5.666272068023682,
      "epoch": 9.168632075471699,
      "grad_norm": 1.5625,
      "learning_rate": 5.7756553034816395e-05,
      "loss": 4.7882,
      "mean_token_accuracy": 0.23363217413425447,
      "num_tokens": 209752661.0,
      "step": 101075
    },
    {
      "entropy": 5.6807348251342775,
      "epoch": 9.169085631349782,
      "grad_norm": 1.6484375,
      "learning_rate": 5.774813274218815e-05,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23345238864421844,
      "num_tokens": 209761525.0,
      "step": 101080
    },
    {
      "entropy": 5.709030389785767,
      "epoch": 9.169539187227867,
      "grad_norm": 1.4609375,
      "learning_rate": 5.7739716942383855e-05,
      "loss": 4.8712,
      "mean_token_accuracy": 0.23077893555164336,
      "num_tokens": 209772232.0,
      "step": 101085
    },
    {
      "entropy": 5.710816287994385,
      "epoch": 9.16999274310595,
      "grad_norm": 1.5390625,
      "learning_rate": 5.773130563557754e-05,
      "loss": 4.7938,
      "mean_token_accuracy": 0.23493459224700927,
      "num_tokens": 209782265.0,
      "step": 101090
    },
    {
      "entropy": 5.668586683273316,
      "epoch": 9.170446298984034,
      "grad_norm": 1.5703125,
      "learning_rate": 5.772289882194323e-05,
      "loss": 4.8075,
      "mean_token_accuracy": 0.2386668309569359,
      "num_tokens": 209792181.0,
      "step": 101095
    },
    {
      "entropy": 5.7974072933197025,
      "epoch": 9.17089985486212,
      "grad_norm": 1.7578125,
      "learning_rate": 5.771449650165474e-05,
      "loss": 4.9253,
      "mean_token_accuracy": 0.23308760523796082,
      "num_tokens": 209801963.0,
      "step": 101100
    },
    {
      "entropy": 5.69154953956604,
      "epoch": 9.171353410740203,
      "grad_norm": 1.625,
      "learning_rate": 5.770609867488587e-05,
      "loss": 4.8416,
      "mean_token_accuracy": 0.23474406749010085,
      "num_tokens": 209811932.0,
      "step": 101105
    },
    {
      "entropy": 5.670922183990479,
      "epoch": 9.171806966618288,
      "grad_norm": 1.4453125,
      "learning_rate": 5.76977053418102e-05,
      "loss": 4.7609,
      "mean_token_accuracy": 0.2391236588358879,
      "num_tokens": 209823150.0,
      "step": 101110
    },
    {
      "entropy": 5.742044401168823,
      "epoch": 9.172260522496371,
      "grad_norm": 1.6328125,
      "learning_rate": 5.76893165026014e-05,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22733384370803833,
      "num_tokens": 209832662.0,
      "step": 101115
    },
    {
      "entropy": 5.694376802444458,
      "epoch": 9.172714078374456,
      "grad_norm": 1.6484375,
      "learning_rate": 5.768093215743288e-05,
      "loss": 4.7763,
      "mean_token_accuracy": 0.23759540766477585,
      "num_tokens": 209842269.0,
      "step": 101120
    },
    {
      "entropy": 5.658236980438232,
      "epoch": 9.17316763425254,
      "grad_norm": 1.59375,
      "learning_rate": 5.767255230647803e-05,
      "loss": 4.689,
      "mean_token_accuracy": 0.24771223813295365,
      "num_tokens": 209851869.0,
      "step": 101125
    },
    {
      "entropy": 5.637798547744751,
      "epoch": 9.173621190130625,
      "grad_norm": 1.546875,
      "learning_rate": 5.7664176949910225e-05,
      "loss": 4.7054,
      "mean_token_accuracy": 0.23811432719230652,
      "num_tokens": 209862017.0,
      "step": 101130
    },
    {
      "entropy": 5.650383329391479,
      "epoch": 9.174074746008708,
      "grad_norm": 1.46875,
      "learning_rate": 5.765580608790262e-05,
      "loss": 4.7515,
      "mean_token_accuracy": 0.23540754467248917,
      "num_tokens": 209871637.0,
      "step": 101135
    },
    {
      "entropy": 5.7556558609008786,
      "epoch": 9.174528301886792,
      "grad_norm": 1.46875,
      "learning_rate": 5.764743972062835e-05,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22845862358808516,
      "num_tokens": 209883629.0,
      "step": 101140
    },
    {
      "entropy": 5.62816596031189,
      "epoch": 9.174981857764877,
      "grad_norm": 1.6875,
      "learning_rate": 5.7639077848260405e-05,
      "loss": 4.742,
      "mean_token_accuracy": 0.23468622118234633,
      "num_tokens": 209893767.0,
      "step": 101145
    },
    {
      "entropy": 5.631982660293579,
      "epoch": 9.17543541364296,
      "grad_norm": 1.6953125,
      "learning_rate": 5.763072047097171e-05,
      "loss": 4.7063,
      "mean_token_accuracy": 0.25030582696199416,
      "num_tokens": 209904102.0,
      "step": 101150
    },
    {
      "entropy": 5.736003971099853,
      "epoch": 9.175888969521045,
      "grad_norm": 1.7265625,
      "learning_rate": 5.762236758893507e-05,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22675665020942687,
      "num_tokens": 209913941.0,
      "step": 101155
    },
    {
      "entropy": 5.656425476074219,
      "epoch": 9.176342525399129,
      "grad_norm": 1.578125,
      "learning_rate": 5.7614019202323306e-05,
      "loss": 4.7097,
      "mean_token_accuracy": 0.24470312744379044,
      "num_tokens": 209924514.0,
      "step": 101160
    },
    {
      "entropy": 5.684019613265991,
      "epoch": 9.176796081277214,
      "grad_norm": 1.4609375,
      "learning_rate": 5.760567531130904e-05,
      "loss": 4.6962,
      "mean_token_accuracy": 0.2465697169303894,
      "num_tokens": 209935219.0,
      "step": 101165
    },
    {
      "entropy": 5.6977986812591555,
      "epoch": 9.177249637155297,
      "grad_norm": 1.5390625,
      "learning_rate": 5.759733591606478e-05,
      "loss": 4.7689,
      "mean_token_accuracy": 0.2439318984746933,
      "num_tokens": 209945059.0,
      "step": 101170
    },
    {
      "entropy": 5.7281452178955075,
      "epoch": 9.177703193033382,
      "grad_norm": 1.6640625,
      "learning_rate": 5.758900101676301e-05,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2307477742433548,
      "num_tokens": 209954905.0,
      "step": 101175
    },
    {
      "entropy": 5.683432102203369,
      "epoch": 9.178156748911466,
      "grad_norm": 1.53125,
      "learning_rate": 5.758067061357612e-05,
      "loss": 4.8478,
      "mean_token_accuracy": 0.23078984916210174,
      "num_tokens": 209965251.0,
      "step": 101180
    },
    {
      "entropy": 5.730329608917236,
      "epoch": 9.17861030478955,
      "grad_norm": 1.4375,
      "learning_rate": 5.7572344706676354e-05,
      "loss": 4.8278,
      "mean_token_accuracy": 0.22826884239912032,
      "num_tokens": 209976225.0,
      "step": 101185
    },
    {
      "entropy": 5.68026294708252,
      "epoch": 9.179063860667634,
      "grad_norm": 1.46875,
      "learning_rate": 5.756402329623589e-05,
      "loss": 4.79,
      "mean_token_accuracy": 0.23308126926422118,
      "num_tokens": 209986379.0,
      "step": 101190
    },
    {
      "entropy": 5.711122226715088,
      "epoch": 9.179517416545718,
      "grad_norm": 1.53125,
      "learning_rate": 5.755570638242687e-05,
      "loss": 4.8506,
      "mean_token_accuracy": 0.23224893659353257,
      "num_tokens": 209996994.0,
      "step": 101195
    },
    {
      "entropy": 5.666979646682739,
      "epoch": 9.179970972423803,
      "grad_norm": 1.5546875,
      "learning_rate": 5.7547393965421216e-05,
      "loss": 4.7767,
      "mean_token_accuracy": 0.23988394141197206,
      "num_tokens": 210007589.0,
      "step": 101200
    },
    {
      "entropy": 5.691450929641723,
      "epoch": 9.180424528301886,
      "grad_norm": 1.5078125,
      "learning_rate": 5.75390860453909e-05,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2378301963210106,
      "num_tokens": 210018735.0,
      "step": 101205
    },
    {
      "entropy": 5.655982208251953,
      "epoch": 9.180878084179971,
      "grad_norm": 1.5,
      "learning_rate": 5.75307826225077e-05,
      "loss": 4.7691,
      "mean_token_accuracy": 0.23987587839365004,
      "num_tokens": 210029887.0,
      "step": 101210
    },
    {
      "entropy": 5.614995002746582,
      "epoch": 9.181331640058055,
      "grad_norm": 1.6484375,
      "learning_rate": 5.752248369694332e-05,
      "loss": 4.7379,
      "mean_token_accuracy": 0.24645347595214845,
      "num_tokens": 210039625.0,
      "step": 101215
    },
    {
      "entropy": 5.686473083496094,
      "epoch": 9.18178519593614,
      "grad_norm": 1.5703125,
      "learning_rate": 5.7514189268869395e-05,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22902671843767167,
      "num_tokens": 210049202.0,
      "step": 101220
    },
    {
      "entropy": 5.652646827697754,
      "epoch": 9.182238751814223,
      "grad_norm": 1.5078125,
      "learning_rate": 5.7505899338457444e-05,
      "loss": 4.7515,
      "mean_token_accuracy": 0.23584742844104767,
      "num_tokens": 210058680.0,
      "step": 101225
    },
    {
      "entropy": 5.63798246383667,
      "epoch": 9.182692307692308,
      "grad_norm": 1.4765625,
      "learning_rate": 5.749761390587897e-05,
      "loss": 4.7064,
      "mean_token_accuracy": 0.24310127943754195,
      "num_tokens": 210068306.0,
      "step": 101230
    },
    {
      "entropy": 5.769039058685303,
      "epoch": 9.183145863570392,
      "grad_norm": 1.890625,
      "learning_rate": 5.7489332971305237e-05,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2262225016951561,
      "num_tokens": 210079652.0,
      "step": 101235
    },
    {
      "entropy": 5.655624294281006,
      "epoch": 9.183599419448477,
      "grad_norm": 1.609375,
      "learning_rate": 5.7481056534907514e-05,
      "loss": 4.695,
      "mean_token_accuracy": 0.24523125290870668,
      "num_tokens": 210090213.0,
      "step": 101240
    },
    {
      "entropy": 5.620161342620849,
      "epoch": 9.18405297532656,
      "grad_norm": 1.59375,
      "learning_rate": 5.747278459685698e-05,
      "loss": 4.6858,
      "mean_token_accuracy": 0.2508385986089706,
      "num_tokens": 210099947.0,
      "step": 101245
    },
    {
      "entropy": 5.664350843429565,
      "epoch": 9.184506531204644,
      "grad_norm": 1.6328125,
      "learning_rate": 5.746451715732468e-05,
      "loss": 4.7766,
      "mean_token_accuracy": 0.243143792450428,
      "num_tokens": 210110787.0,
      "step": 101250
    },
    {
      "entropy": 5.660854434967041,
      "epoch": 9.184960087082729,
      "grad_norm": 1.609375,
      "learning_rate": 5.7456254216481606e-05,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2473118022084236,
      "num_tokens": 210120668.0,
      "step": 101255
    },
    {
      "entropy": 5.671524715423584,
      "epoch": 9.185413642960812,
      "grad_norm": 1.59375,
      "learning_rate": 5.7447995774498666e-05,
      "loss": 4.786,
      "mean_token_accuracy": 0.23350586295127868,
      "num_tokens": 210131740.0,
      "step": 101260
    },
    {
      "entropy": 5.646691608428955,
      "epoch": 9.185867198838897,
      "grad_norm": 1.46875,
      "learning_rate": 5.743974183154657e-05,
      "loss": 4.7066,
      "mean_token_accuracy": 0.24629868417978287,
      "num_tokens": 210141572.0,
      "step": 101265
    },
    {
      "entropy": 5.7268139839172365,
      "epoch": 9.18632075471698,
      "grad_norm": 1.5234375,
      "learning_rate": 5.743149238779609e-05,
      "loss": 4.8149,
      "mean_token_accuracy": 0.24317592978477479,
      "num_tokens": 210151855.0,
      "step": 101270
    },
    {
      "entropy": 5.693928050994873,
      "epoch": 9.186774310595066,
      "grad_norm": 1.71875,
      "learning_rate": 5.7423247443417755e-05,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2360013961791992,
      "num_tokens": 210161793.0,
      "step": 101275
    },
    {
      "entropy": 5.717860746383667,
      "epoch": 9.18722786647315,
      "grad_norm": 1.578125,
      "learning_rate": 5.741500699858213e-05,
      "loss": 4.8904,
      "mean_token_accuracy": 0.23440296351909637,
      "num_tokens": 210172856.0,
      "step": 101280
    },
    {
      "entropy": 5.667159605026245,
      "epoch": 9.187681422351234,
      "grad_norm": 1.578125,
      "learning_rate": 5.7406771053459564e-05,
      "loss": 4.7959,
      "mean_token_accuracy": 0.24162923842668532,
      "num_tokens": 210183268.0,
      "step": 101285
    },
    {
      "entropy": 5.67816686630249,
      "epoch": 9.188134978229318,
      "grad_norm": 1.5703125,
      "learning_rate": 5.739853960822047e-05,
      "loss": 4.7621,
      "mean_token_accuracy": 0.23607310354709626,
      "num_tokens": 210193923.0,
      "step": 101290
    },
    {
      "entropy": 5.669769620895385,
      "epoch": 9.188588534107401,
      "grad_norm": 1.46875,
      "learning_rate": 5.7390312663035014e-05,
      "loss": 4.7686,
      "mean_token_accuracy": 0.2319034978747368,
      "num_tokens": 210205101.0,
      "step": 101295
    },
    {
      "entropy": 5.69979567527771,
      "epoch": 9.189042089985486,
      "grad_norm": 1.46875,
      "learning_rate": 5.738209021807336e-05,
      "loss": 4.7728,
      "mean_token_accuracy": 0.2421009957790375,
      "num_tokens": 210215328.0,
      "step": 101300
    },
    {
      "entropy": 5.705284643173218,
      "epoch": 9.18949564586357,
      "grad_norm": 1.5,
      "learning_rate": 5.7373872273505515e-05,
      "loss": 4.787,
      "mean_token_accuracy": 0.23192712515592576,
      "num_tokens": 210225484.0,
      "step": 101305
    },
    {
      "entropy": 5.714508581161499,
      "epoch": 9.189949201741655,
      "grad_norm": 1.6484375,
      "learning_rate": 5.736565882950147e-05,
      "loss": 4.7788,
      "mean_token_accuracy": 0.23792957961559297,
      "num_tokens": 210235073.0,
      "step": 101310
    },
    {
      "entropy": 5.6487743854522705,
      "epoch": 9.190402757619738,
      "grad_norm": 1.4453125,
      "learning_rate": 5.735744988623106e-05,
      "loss": 4.7548,
      "mean_token_accuracy": 0.24869346767663955,
      "num_tokens": 210245723.0,
      "step": 101315
    },
    {
      "entropy": 5.636103296279908,
      "epoch": 9.190856313497823,
      "grad_norm": 1.59375,
      "learning_rate": 5.734924544386402e-05,
      "loss": 4.7073,
      "mean_token_accuracy": 0.24907068610191346,
      "num_tokens": 210255707.0,
      "step": 101320
    },
    {
      "entropy": 5.641573333740235,
      "epoch": 9.191309869375907,
      "grad_norm": 1.453125,
      "learning_rate": 5.734104550257007e-05,
      "loss": 4.7379,
      "mean_token_accuracy": 0.24030677378177642,
      "num_tokens": 210266090.0,
      "step": 101325
    },
    {
      "entropy": 5.6957940578460695,
      "epoch": 9.191763425253992,
      "grad_norm": 1.453125,
      "learning_rate": 5.7332850062518754e-05,
      "loss": 4.7522,
      "mean_token_accuracy": 0.24585239142179488,
      "num_tokens": 210276695.0,
      "step": 101330
    },
    {
      "entropy": 5.686233997344971,
      "epoch": 9.192216981132075,
      "grad_norm": 1.4609375,
      "learning_rate": 5.7324659123879576e-05,
      "loss": 4.7967,
      "mean_token_accuracy": 0.2431512027978897,
      "num_tokens": 210287427.0,
      "step": 101335
    },
    {
      "entropy": 5.70320315361023,
      "epoch": 9.19267053701016,
      "grad_norm": 1.5625,
      "learning_rate": 5.7316472686821926e-05,
      "loss": 4.8103,
      "mean_token_accuracy": 0.23388008922338485,
      "num_tokens": 210298158.0,
      "step": 101340
    },
    {
      "entropy": 5.653908538818359,
      "epoch": 9.193124092888244,
      "grad_norm": 1.53125,
      "learning_rate": 5.730829075151508e-05,
      "loss": 4.7578,
      "mean_token_accuracy": 0.23812895268201828,
      "num_tokens": 210307869.0,
      "step": 101345
    },
    {
      "entropy": 5.740224170684814,
      "epoch": 9.193577648766327,
      "grad_norm": 1.4921875,
      "learning_rate": 5.7300113318128275e-05,
      "loss": 4.8949,
      "mean_token_accuracy": 0.2343326434493065,
      "num_tokens": 210318381.0,
      "step": 101350
    },
    {
      "entropy": 5.7335169315338135,
      "epoch": 9.194031204644412,
      "grad_norm": 1.3671875,
      "learning_rate": 5.7291940386830544e-05,
      "loss": 4.8334,
      "mean_token_accuracy": 0.237962706387043,
      "num_tokens": 210328995.0,
      "step": 101355
    },
    {
      "entropy": 5.632558012008667,
      "epoch": 9.194484760522496,
      "grad_norm": 1.5078125,
      "learning_rate": 5.728377195779102e-05,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2469208374619484,
      "num_tokens": 210340349.0,
      "step": 101360
    },
    {
      "entropy": 5.726474857330322,
      "epoch": 9.194938316400581,
      "grad_norm": 1.6328125,
      "learning_rate": 5.727560803117856e-05,
      "loss": 4.8746,
      "mean_token_accuracy": 0.23938533067703247,
      "num_tokens": 210351042.0,
      "step": 101365
    },
    {
      "entropy": 5.712001848220825,
      "epoch": 9.195391872278664,
      "grad_norm": 1.40625,
      "learning_rate": 5.726744860716202e-05,
      "loss": 4.8062,
      "mean_token_accuracy": 0.23805576413869858,
      "num_tokens": 210362973.0,
      "step": 101370
    },
    {
      "entropy": 5.673559379577637,
      "epoch": 9.19584542815675,
      "grad_norm": 1.5,
      "learning_rate": 5.725929368591011e-05,
      "loss": 4.8125,
      "mean_token_accuracy": 0.2427704930305481,
      "num_tokens": 210375028.0,
      "step": 101375
    },
    {
      "entropy": 5.638546705245972,
      "epoch": 9.196298984034833,
      "grad_norm": 1.5,
      "learning_rate": 5.725114326759148e-05,
      "loss": 4.7176,
      "mean_token_accuracy": 0.23843868523836137,
      "num_tokens": 210385575.0,
      "step": 101380
    },
    {
      "entropy": 5.696994304656982,
      "epoch": 9.196752539912918,
      "grad_norm": 1.328125,
      "learning_rate": 5.7242997352374724e-05,
      "loss": 4.7554,
      "mean_token_accuracy": 0.23989488631486894,
      "num_tokens": 210396434.0,
      "step": 101385
    },
    {
      "entropy": 5.748320817947388,
      "epoch": 9.197206095791001,
      "grad_norm": 1.6171875,
      "learning_rate": 5.7234855940428243e-05,
      "loss": 4.7956,
      "mean_token_accuracy": 0.24614300727844238,
      "num_tokens": 210406254.0,
      "step": 101390
    },
    {
      "entropy": 5.738400268554687,
      "epoch": 9.197659651669087,
      "grad_norm": 1.4296875,
      "learning_rate": 5.722671903192046e-05,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2275507405400276,
      "num_tokens": 210416861.0,
      "step": 101395
    },
    {
      "entropy": 5.734044694900513,
      "epoch": 9.19811320754717,
      "grad_norm": 1.640625,
      "learning_rate": 5.721858662701964e-05,
      "loss": 4.8625,
      "mean_token_accuracy": 0.23719168454408646,
      "num_tokens": 210427314.0,
      "step": 101400
    },
    {
      "entropy": 5.668119144439697,
      "epoch": 9.198566763425253,
      "grad_norm": 1.546875,
      "learning_rate": 5.721045872589392e-05,
      "loss": 4.739,
      "mean_token_accuracy": 0.24330947995185853,
      "num_tokens": 210438155.0,
      "step": 101405
    },
    {
      "entropy": 5.716606330871582,
      "epoch": 9.199020319303338,
      "grad_norm": 1.59375,
      "learning_rate": 5.720233532871142e-05,
      "loss": 4.8074,
      "mean_token_accuracy": 0.24204583317041398,
      "num_tokens": 210448984.0,
      "step": 101410
    },
    {
      "entropy": 5.709401273727417,
      "epoch": 9.199473875181422,
      "grad_norm": 1.640625,
      "learning_rate": 5.719421643564013e-05,
      "loss": 4.8272,
      "mean_token_accuracy": 0.2370416283607483,
      "num_tokens": 210460167.0,
      "step": 101415
    },
    {
      "entropy": 5.6573200702667235,
      "epoch": 9.199927431059507,
      "grad_norm": 1.53125,
      "learning_rate": 5.718610204684794e-05,
      "loss": 4.77,
      "mean_token_accuracy": 0.24074707478284835,
      "num_tokens": 210470358.0,
      "step": 101420
    },
    {
      "entropy": 5.658379364013672,
      "epoch": 9.20038098693759,
      "grad_norm": 1.4765625,
      "learning_rate": 5.717799216250264e-05,
      "loss": 4.7404,
      "mean_token_accuracy": 0.24278557449579238,
      "num_tokens": 210481720.0,
      "step": 101425
    },
    {
      "entropy": 5.712123966217041,
      "epoch": 9.200834542815675,
      "grad_norm": 1.4453125,
      "learning_rate": 5.7169886782772e-05,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2314532443881035,
      "num_tokens": 210491911.0,
      "step": 101430
    },
    {
      "entropy": 5.671857929229736,
      "epoch": 9.201288098693759,
      "grad_norm": 1.59375,
      "learning_rate": 5.7161785907823625e-05,
      "loss": 4.8139,
      "mean_token_accuracy": 0.23333869725465775,
      "num_tokens": 210502009.0,
      "step": 101435
    },
    {
      "entropy": 5.651693964004517,
      "epoch": 9.201741654571844,
      "grad_norm": 1.546875,
      "learning_rate": 5.715368953782502e-05,
      "loss": 4.8085,
      "mean_token_accuracy": 0.22867684662342072,
      "num_tokens": 210512376.0,
      "step": 101440
    },
    {
      "entropy": 5.638951587677002,
      "epoch": 9.202195210449927,
      "grad_norm": 1.546875,
      "learning_rate": 5.714559767294361e-05,
      "loss": 4.694,
      "mean_token_accuracy": 0.24945977777242662,
      "num_tokens": 210521911.0,
      "step": 101445
    },
    {
      "entropy": 5.737909841537475,
      "epoch": 9.20264876632801,
      "grad_norm": 1.5546875,
      "learning_rate": 5.7137510313346775e-05,
      "loss": 4.9621,
      "mean_token_accuracy": 0.22918384969234468,
      "num_tokens": 210533260.0,
      "step": 101450
    },
    {
      "entropy": 5.746039342880249,
      "epoch": 9.203102322206096,
      "grad_norm": 1.3515625,
      "learning_rate": 5.7129427459201703e-05,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2342885598540306,
      "num_tokens": 210545296.0,
      "step": 101455
    },
    {
      "entropy": 5.7009529113769535,
      "epoch": 9.20355587808418,
      "grad_norm": 1.5703125,
      "learning_rate": 5.712134911067564e-05,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2346903756260872,
      "num_tokens": 210556864.0,
      "step": 101460
    },
    {
      "entropy": 5.645622253417969,
      "epoch": 9.204009433962264,
      "grad_norm": 1.59375,
      "learning_rate": 5.711327526793558e-05,
      "loss": 4.7225,
      "mean_token_accuracy": 0.24835027605295182,
      "num_tokens": 210566738.0,
      "step": 101465
    },
    {
      "entropy": 5.771311187744141,
      "epoch": 9.204462989840348,
      "grad_norm": 1.5546875,
      "learning_rate": 5.710520593114852e-05,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2308598443865776,
      "num_tokens": 210576989.0,
      "step": 101470
    },
    {
      "entropy": 5.700727748870849,
      "epoch": 9.204916545718433,
      "grad_norm": 1.546875,
      "learning_rate": 5.7097141100481335e-05,
      "loss": 4.825,
      "mean_token_accuracy": 0.2392111450433731,
      "num_tokens": 210587233.0,
      "step": 101475
    },
    {
      "entropy": 5.620848655700684,
      "epoch": 9.205370101596516,
      "grad_norm": 1.609375,
      "learning_rate": 5.708908077610079e-05,
      "loss": 4.7038,
      "mean_token_accuracy": 0.25102042853832246,
      "num_tokens": 210597735.0,
      "step": 101480
    },
    {
      "entropy": 5.6657813549041744,
      "epoch": 9.205823657474602,
      "grad_norm": 1.546875,
      "learning_rate": 5.708102495817359e-05,
      "loss": 4.6977,
      "mean_token_accuracy": 0.25027584433555605,
      "num_tokens": 210608220.0,
      "step": 101485
    },
    {
      "entropy": 5.704652976989746,
      "epoch": 9.206277213352685,
      "grad_norm": 1.5703125,
      "learning_rate": 5.70729736468663e-05,
      "loss": 4.801,
      "mean_token_accuracy": 0.23531580567359925,
      "num_tokens": 210618916.0,
      "step": 101490
    },
    {
      "entropy": 5.638085556030274,
      "epoch": 9.20673076923077,
      "grad_norm": 1.703125,
      "learning_rate": 5.706492684234547e-05,
      "loss": 4.7749,
      "mean_token_accuracy": 0.23622104227542878,
      "num_tokens": 210629168.0,
      "step": 101495
    },
    {
      "entropy": 5.719544315338135,
      "epoch": 9.207184325108853,
      "grad_norm": 1.75,
      "learning_rate": 5.705688454477749e-05,
      "loss": 4.7981,
      "mean_token_accuracy": 0.23922067433595656,
      "num_tokens": 210638057.0,
      "step": 101500
    },
    {
      "entropy": 5.685463619232178,
      "epoch": 9.207637880986937,
      "grad_norm": 1.5625,
      "learning_rate": 5.704884675432869e-05,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23657000809907913,
      "num_tokens": 210649183.0,
      "step": 101505
    },
    {
      "entropy": 5.632651281356812,
      "epoch": 9.208091436865022,
      "grad_norm": 1.5390625,
      "learning_rate": 5.704081347116526e-05,
      "loss": 4.7509,
      "mean_token_accuracy": 0.23207817673683168,
      "num_tokens": 210659516.0,
      "step": 101510
    },
    {
      "entropy": 5.701489496231079,
      "epoch": 9.208544992743105,
      "grad_norm": 1.4921875,
      "learning_rate": 5.703278469545335e-05,
      "loss": 4.8201,
      "mean_token_accuracy": 0.2378296047449112,
      "num_tokens": 210670681.0,
      "step": 101515
    },
    {
      "entropy": 5.6149622917175295,
      "epoch": 9.20899854862119,
      "grad_norm": 1.4921875,
      "learning_rate": 5.702476042735901e-05,
      "loss": 4.784,
      "mean_token_accuracy": 0.23646869659423828,
      "num_tokens": 210680260.0,
      "step": 101520
    },
    {
      "entropy": 5.7076903820037845,
      "epoch": 9.209452104499274,
      "grad_norm": 1.4609375,
      "learning_rate": 5.701674066704815e-05,
      "loss": 4.7897,
      "mean_token_accuracy": 0.23708736002445222,
      "num_tokens": 210690792.0,
      "step": 101525
    },
    {
      "entropy": 5.70125904083252,
      "epoch": 9.209905660377359,
      "grad_norm": 1.5859375,
      "learning_rate": 5.700872541468665e-05,
      "loss": 4.849,
      "mean_token_accuracy": 0.23634222894906998,
      "num_tokens": 210701148.0,
      "step": 101530
    },
    {
      "entropy": 5.701562547683716,
      "epoch": 9.210359216255442,
      "grad_norm": 1.53125,
      "learning_rate": 5.7000714670440276e-05,
      "loss": 4.7613,
      "mean_token_accuracy": 0.2393864095211029,
      "num_tokens": 210710669.0,
      "step": 101535
    },
    {
      "entropy": 5.773510837554932,
      "epoch": 9.210812772133528,
      "grad_norm": 1.515625,
      "learning_rate": 5.699270843447466e-05,
      "loss": 4.9423,
      "mean_token_accuracy": 0.22992149591445923,
      "num_tokens": 210721147.0,
      "step": 101540
    },
    {
      "entropy": 5.685460805892944,
      "epoch": 9.211266328011611,
      "grad_norm": 1.46875,
      "learning_rate": 5.69847067069554e-05,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23167216926813125,
      "num_tokens": 210732576.0,
      "step": 101545
    },
    {
      "entropy": 5.6614021301269535,
      "epoch": 9.211719883889696,
      "grad_norm": 1.5859375,
      "learning_rate": 5.6976709488047954e-05,
      "loss": 4.7176,
      "mean_token_accuracy": 0.2390969842672348,
      "num_tokens": 210742271.0,
      "step": 101550
    },
    {
      "entropy": 5.755114507675171,
      "epoch": 9.21217343976778,
      "grad_norm": 1.5,
      "learning_rate": 5.696871677791771e-05,
      "loss": 4.9185,
      "mean_token_accuracy": 0.2288191020488739,
      "num_tokens": 210751914.0,
      "step": 101555
    },
    {
      "entropy": 5.662950944900513,
      "epoch": 9.212626995645863,
      "grad_norm": 1.4609375,
      "learning_rate": 5.6960728576729956e-05,
      "loss": 4.7413,
      "mean_token_accuracy": 0.24059408903121948,
      "num_tokens": 210762689.0,
      "step": 101560
    },
    {
      "entropy": 5.624339389801025,
      "epoch": 9.213080551523948,
      "grad_norm": 1.4921875,
      "learning_rate": 5.6952744884649926e-05,
      "loss": 4.716,
      "mean_token_accuracy": 0.24566102623939515,
      "num_tokens": 210772610.0,
      "step": 101565
    },
    {
      "entropy": 5.664828443527222,
      "epoch": 9.213534107402031,
      "grad_norm": 1.5625,
      "learning_rate": 5.6944765701842665e-05,
      "loss": 4.8096,
      "mean_token_accuracy": 0.22948312312364577,
      "num_tokens": 210781771.0,
      "step": 101570
    },
    {
      "entropy": 5.695682144165039,
      "epoch": 9.213987663280117,
      "grad_norm": 1.53125,
      "learning_rate": 5.693679102847325e-05,
      "loss": 4.768,
      "mean_token_accuracy": 0.23751165121793746,
      "num_tokens": 210792305.0,
      "step": 101575
    },
    {
      "entropy": 5.652436256408691,
      "epoch": 9.2144412191582,
      "grad_norm": 1.5078125,
      "learning_rate": 5.6928820864706565e-05,
      "loss": 4.7311,
      "mean_token_accuracy": 0.24589859396219255,
      "num_tokens": 210802940.0,
      "step": 101580
    },
    {
      "entropy": 5.762735080718994,
      "epoch": 9.214894775036285,
      "grad_norm": 1.5703125,
      "learning_rate": 5.6920855210707404e-05,
      "loss": 4.9063,
      "mean_token_accuracy": 0.23471759259700775,
      "num_tokens": 210813075.0,
      "step": 101585
    },
    {
      "entropy": 5.659088230133056,
      "epoch": 9.215348330914368,
      "grad_norm": 1.4921875,
      "learning_rate": 5.6912894066640546e-05,
      "loss": 4.7833,
      "mean_token_accuracy": 0.24117258191108704,
      "num_tokens": 210824467.0,
      "step": 101590
    },
    {
      "entropy": 5.714759778976441,
      "epoch": 9.215801886792454,
      "grad_norm": 1.421875,
      "learning_rate": 5.690493743267061e-05,
      "loss": 4.8081,
      "mean_token_accuracy": 0.23330031782388688,
      "num_tokens": 210835660.0,
      "step": 101595
    },
    {
      "entropy": 5.650528526306152,
      "epoch": 9.216255442670537,
      "grad_norm": 1.5,
      "learning_rate": 5.689698530896214e-05,
      "loss": 4.7064,
      "mean_token_accuracy": 0.24967444390058519,
      "num_tokens": 210845147.0,
      "step": 101600
    },
    {
      "entropy": 5.69964075088501,
      "epoch": 9.21670899854862,
      "grad_norm": 1.53125,
      "learning_rate": 5.68890376956796e-05,
      "loss": 4.8082,
      "mean_token_accuracy": 0.23989214152097701,
      "num_tokens": 210855298.0,
      "step": 101605
    },
    {
      "entropy": 5.720903635025024,
      "epoch": 9.217162554426706,
      "grad_norm": 1.3984375,
      "learning_rate": 5.688109459298736e-05,
      "loss": 4.8967,
      "mean_token_accuracy": 0.23457927703857423,
      "num_tokens": 210866860.0,
      "step": 101610
    },
    {
      "entropy": 5.616018342971802,
      "epoch": 9.217616110304789,
      "grad_norm": 1.5,
      "learning_rate": 5.687315600104964e-05,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2384447678923607,
      "num_tokens": 210877255.0,
      "step": 101615
    },
    {
      "entropy": 5.751549482345581,
      "epoch": 9.218069666182874,
      "grad_norm": 1.4921875,
      "learning_rate": 5.686522192003066e-05,
      "loss": 4.8668,
      "mean_token_accuracy": 0.22917444109916688,
      "num_tokens": 210887700.0,
      "step": 101620
    },
    {
      "entropy": 5.7004810810089115,
      "epoch": 9.218523222060957,
      "grad_norm": 1.484375,
      "learning_rate": 5.685729235009447e-05,
      "loss": 4.7766,
      "mean_token_accuracy": 0.23731375634670257,
      "num_tokens": 210898500.0,
      "step": 101625
    },
    {
      "entropy": 5.664872121810913,
      "epoch": 9.218976777939043,
      "grad_norm": 1.4140625,
      "learning_rate": 5.684936729140505e-05,
      "loss": 4.7266,
      "mean_token_accuracy": 0.24848500937223433,
      "num_tokens": 210908918.0,
      "step": 101630
    },
    {
      "entropy": 5.7007097721099855,
      "epoch": 9.219430333817126,
      "grad_norm": 1.546875,
      "learning_rate": 5.684144674412632e-05,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22990898489952089,
      "num_tokens": 210919079.0,
      "step": 101635
    },
    {
      "entropy": 5.6895510196685795,
      "epoch": 9.219883889695211,
      "grad_norm": 1.59375,
      "learning_rate": 5.683353070842206e-05,
      "loss": 4.813,
      "mean_token_accuracy": 0.23909665942192077,
      "num_tokens": 210928871.0,
      "step": 101640
    },
    {
      "entropy": 5.638714456558228,
      "epoch": 9.220337445573294,
      "grad_norm": 1.6640625,
      "learning_rate": 5.6825619184455985e-05,
      "loss": 4.7563,
      "mean_token_accuracy": 0.24198397547006606,
      "num_tokens": 210938588.0,
      "step": 101645
    },
    {
      "entropy": 5.712907361984253,
      "epoch": 9.22079100145138,
      "grad_norm": 1.6484375,
      "learning_rate": 5.6817712172391706e-05,
      "loss": 4.782,
      "mean_token_accuracy": 0.24190607964992522,
      "num_tokens": 210948644.0,
      "step": 101650
    },
    {
      "entropy": 5.671142721176148,
      "epoch": 9.221244557329463,
      "grad_norm": 1.609375,
      "learning_rate": 5.680980967239273e-05,
      "loss": 4.7766,
      "mean_token_accuracy": 0.24278903007507324,
      "num_tokens": 210958469.0,
      "step": 101655
    },
    {
      "entropy": 5.677723932266235,
      "epoch": 9.221698113207546,
      "grad_norm": 1.4921875,
      "learning_rate": 5.680191168462249e-05,
      "loss": 4.8131,
      "mean_token_accuracy": 0.2441313773393631,
      "num_tokens": 210969477.0,
      "step": 101660
    },
    {
      "entropy": 5.687431955337525,
      "epoch": 9.222151669085632,
      "grad_norm": 1.5234375,
      "learning_rate": 5.679401820924433e-05,
      "loss": 4.7756,
      "mean_token_accuracy": 0.24630622565746307,
      "num_tokens": 210980617.0,
      "step": 101665
    },
    {
      "entropy": 5.626515293121338,
      "epoch": 9.222605224963715,
      "grad_norm": 1.484375,
      "learning_rate": 5.678612924642149e-05,
      "loss": 4.7435,
      "mean_token_accuracy": 0.23819674849510192,
      "num_tokens": 210992444.0,
      "step": 101670
    },
    {
      "entropy": 5.668351793289185,
      "epoch": 9.2230587808418,
      "grad_norm": 1.4765625,
      "learning_rate": 5.677824479631709e-05,
      "loss": 4.77,
      "mean_token_accuracy": 0.23845761120319367,
      "num_tokens": 211002878.0,
      "step": 101675
    },
    {
      "entropy": 5.738829898834228,
      "epoch": 9.223512336719883,
      "grad_norm": 1.375,
      "learning_rate": 5.6770364859094204e-05,
      "loss": 4.8874,
      "mean_token_accuracy": 0.23185595273971557,
      "num_tokens": 211014238.0,
      "step": 101680
    },
    {
      "entropy": 5.700941133499145,
      "epoch": 9.223965892597969,
      "grad_norm": 1.5625,
      "learning_rate": 5.6762489434915756e-05,
      "loss": 4.8055,
      "mean_token_accuracy": 0.24503110647201537,
      "num_tokens": 211023622.0,
      "step": 101685
    },
    {
      "entropy": 5.670117998123169,
      "epoch": 9.224419448476052,
      "grad_norm": 1.5703125,
      "learning_rate": 5.675461852394465e-05,
      "loss": 4.706,
      "mean_token_accuracy": 0.2455711007118225,
      "num_tokens": 211033615.0,
      "step": 101690
    },
    {
      "entropy": 5.733022260665893,
      "epoch": 9.224873004354137,
      "grad_norm": 1.4453125,
      "learning_rate": 5.6746752126343664e-05,
      "loss": 4.8685,
      "mean_token_accuracy": 0.2355012610554695,
      "num_tokens": 211043920.0,
      "step": 101695
    },
    {
      "entropy": 5.667627000808716,
      "epoch": 9.22532656023222,
      "grad_norm": 1.4296875,
      "learning_rate": 5.673889024227545e-05,
      "loss": 4.8244,
      "mean_token_accuracy": 0.22960906028747557,
      "num_tokens": 211054392.0,
      "step": 101700
    },
    {
      "entropy": 5.67356243133545,
      "epoch": 9.225780116110304,
      "grad_norm": 1.5546875,
      "learning_rate": 5.673103287190264e-05,
      "loss": 4.7622,
      "mean_token_accuracy": 0.23434261828660966,
      "num_tokens": 211063817.0,
      "step": 101705
    },
    {
      "entropy": 5.787474727630615,
      "epoch": 9.226233671988389,
      "grad_norm": 1.484375,
      "learning_rate": 5.672318001538765e-05,
      "loss": 4.9191,
      "mean_token_accuracy": 0.22450658679008484,
      "num_tokens": 211075360.0,
      "step": 101710
    },
    {
      "entropy": 5.7182018756866455,
      "epoch": 9.226687227866472,
      "grad_norm": 1.5234375,
      "learning_rate": 5.671533167289293e-05,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23795755803585053,
      "num_tokens": 211084737.0,
      "step": 101715
    },
    {
      "entropy": 5.710438346862793,
      "epoch": 9.227140783744558,
      "grad_norm": 1.4453125,
      "learning_rate": 5.670748784458078e-05,
      "loss": 4.8012,
      "mean_token_accuracy": 0.2372838333249092,
      "num_tokens": 211095186.0,
      "step": 101720
    },
    {
      "entropy": 5.654373693466186,
      "epoch": 9.227594339622641,
      "grad_norm": 1.4921875,
      "learning_rate": 5.669964853061339e-05,
      "loss": 4.7522,
      "mean_token_accuracy": 0.2394466817378998,
      "num_tokens": 211105734.0,
      "step": 101725
    },
    {
      "entropy": 5.796217012405395,
      "epoch": 9.228047895500726,
      "grad_norm": 1.6171875,
      "learning_rate": 5.669181373115293e-05,
      "loss": 4.9119,
      "mean_token_accuracy": 0.23063729852437972,
      "num_tokens": 211116030.0,
      "step": 101730
    },
    {
      "entropy": 5.648388957977295,
      "epoch": 9.22850145137881,
      "grad_norm": 1.4375,
      "learning_rate": 5.6683983446361394e-05,
      "loss": 4.771,
      "mean_token_accuracy": 0.23514292389154434,
      "num_tokens": 211126453.0,
      "step": 101735
    },
    {
      "entropy": 5.687366199493408,
      "epoch": 9.228955007256895,
      "grad_norm": 1.625,
      "learning_rate": 5.667615767640069e-05,
      "loss": 4.7762,
      "mean_token_accuracy": 0.2423603877425194,
      "num_tokens": 211136747.0,
      "step": 101740
    },
    {
      "entropy": 5.614131307601928,
      "epoch": 9.229408563134978,
      "grad_norm": 1.578125,
      "learning_rate": 5.666833642143271e-05,
      "loss": 4.7216,
      "mean_token_accuracy": 0.24295343160629274,
      "num_tokens": 211147222.0,
      "step": 101745
    },
    {
      "entropy": 5.6871764183044435,
      "epoch": 9.229862119013063,
      "grad_norm": 1.515625,
      "learning_rate": 5.666051968161913e-05,
      "loss": 4.8317,
      "mean_token_accuracy": 0.23944399803876876,
      "num_tokens": 211157583.0,
      "step": 101750
    },
    {
      "entropy": 5.682863569259643,
      "epoch": 9.230315674891147,
      "grad_norm": 1.46875,
      "learning_rate": 5.665270745712165e-05,
      "loss": 4.7501,
      "mean_token_accuracy": 0.24413168281316758,
      "num_tokens": 211167882.0,
      "step": 101755
    },
    {
      "entropy": 5.613162565231323,
      "epoch": 9.23076923076923,
      "grad_norm": 1.5078125,
      "learning_rate": 5.664489974810184e-05,
      "loss": 4.7008,
      "mean_token_accuracy": 0.24983485639095307,
      "num_tokens": 211179423.0,
      "step": 101760
    },
    {
      "entropy": 5.664009237289429,
      "epoch": 9.231222786647315,
      "grad_norm": 1.5546875,
      "learning_rate": 5.6637096554721134e-05,
      "loss": 4.7531,
      "mean_token_accuracy": 0.2433345675468445,
      "num_tokens": 211189099.0,
      "step": 101765
    },
    {
      "entropy": 5.666676998138428,
      "epoch": 9.231676342525398,
      "grad_norm": 1.5078125,
      "learning_rate": 5.6629297877140945e-05,
      "loss": 4.8059,
      "mean_token_accuracy": 0.23031551092863084,
      "num_tokens": 211199079.0,
      "step": 101770
    },
    {
      "entropy": 5.631724119186401,
      "epoch": 9.232129898403484,
      "grad_norm": 1.59375,
      "learning_rate": 5.6621503715522505e-05,
      "loss": 4.6978,
      "mean_token_accuracy": 0.24598273187875747,
      "num_tokens": 211208882.0,
      "step": 101775
    },
    {
      "entropy": 5.664183902740478,
      "epoch": 9.232583454281567,
      "grad_norm": 1.7421875,
      "learning_rate": 5.661371407002702e-05,
      "loss": 4.746,
      "mean_token_accuracy": 0.24608877897262574,
      "num_tokens": 211218046.0,
      "step": 101780
    },
    {
      "entropy": 5.6394508361816404,
      "epoch": 9.233037010159652,
      "grad_norm": 1.4609375,
      "learning_rate": 5.660592894081558e-05,
      "loss": 4.7562,
      "mean_token_accuracy": 0.23901488184928893,
      "num_tokens": 211228316.0,
      "step": 101785
    },
    {
      "entropy": 5.681714391708374,
      "epoch": 9.233490566037736,
      "grad_norm": 1.640625,
      "learning_rate": 5.659814832804916e-05,
      "loss": 4.8186,
      "mean_token_accuracy": 0.23327856361865998,
      "num_tokens": 211238569.0,
      "step": 101790
    },
    {
      "entropy": 5.578644847869873,
      "epoch": 9.23394412191582,
      "grad_norm": 1.453125,
      "learning_rate": 5.6590372231888696e-05,
      "loss": 4.6473,
      "mean_token_accuracy": 0.24959299415349961,
      "num_tokens": 211249676.0,
      "step": 101795
    },
    {
      "entropy": 5.689552688598633,
      "epoch": 9.234397677793904,
      "grad_norm": 1.4453125,
      "learning_rate": 5.658260065249503e-05,
      "loss": 4.8574,
      "mean_token_accuracy": 0.234728866815567,
      "num_tokens": 211261138.0,
      "step": 101800
    },
    {
      "entropy": 5.714961147308349,
      "epoch": 9.23485123367199,
      "grad_norm": 1.6015625,
      "learning_rate": 5.657483359002881e-05,
      "loss": 4.8397,
      "mean_token_accuracy": 0.2342833623290062,
      "num_tokens": 211272049.0,
      "step": 101805
    },
    {
      "entropy": 5.699527215957642,
      "epoch": 9.235304789550073,
      "grad_norm": 1.46875,
      "learning_rate": 5.6567071044650695e-05,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23705617040395738,
      "num_tokens": 211281539.0,
      "step": 101810
    },
    {
      "entropy": 5.711853122711181,
      "epoch": 9.235758345428156,
      "grad_norm": 1.53125,
      "learning_rate": 5.655931301652121e-05,
      "loss": 4.8519,
      "mean_token_accuracy": 0.23578928858041764,
      "num_tokens": 211293284.0,
      "step": 101815
    },
    {
      "entropy": 5.724779081344605,
      "epoch": 9.236211901306241,
      "grad_norm": 1.5,
      "learning_rate": 5.655155950580082e-05,
      "loss": 4.834,
      "mean_token_accuracy": 0.2399736225605011,
      "num_tokens": 211303422.0,
      "step": 101820
    },
    {
      "entropy": 5.632858467102051,
      "epoch": 9.236665457184325,
      "grad_norm": 1.578125,
      "learning_rate": 5.65438105126498e-05,
      "loss": 4.731,
      "mean_token_accuracy": 0.23885987997055053,
      "num_tokens": 211312807.0,
      "step": 101825
    },
    {
      "entropy": 5.692501592636108,
      "epoch": 9.23711901306241,
      "grad_norm": 1.6640625,
      "learning_rate": 5.653606603722847e-05,
      "loss": 4.7904,
      "mean_token_accuracy": 0.23454033881425856,
      "num_tokens": 211323364.0,
      "step": 101830
    },
    {
      "entropy": 5.6641161918640135,
      "epoch": 9.237572568940493,
      "grad_norm": 1.4453125,
      "learning_rate": 5.6528326079696995e-05,
      "loss": 4.7339,
      "mean_token_accuracy": 0.23799585849046706,
      "num_tokens": 211333347.0,
      "step": 101835
    },
    {
      "entropy": 5.627626037597656,
      "epoch": 9.238026124818578,
      "grad_norm": 1.5625,
      "learning_rate": 5.6520590640215394e-05,
      "loss": 4.696,
      "mean_token_accuracy": 0.24850325435400009,
      "num_tokens": 211342198.0,
      "step": 101840
    },
    {
      "entropy": 5.703807592391968,
      "epoch": 9.238479680696662,
      "grad_norm": 1.5625,
      "learning_rate": 5.6512859718943636e-05,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22888579368591308,
      "num_tokens": 211352905.0,
      "step": 101845
    },
    {
      "entropy": 5.665592384338379,
      "epoch": 9.238933236574747,
      "grad_norm": 1.3984375,
      "learning_rate": 5.650513331604164e-05,
      "loss": 4.7585,
      "mean_token_accuracy": 0.24195296168327332,
      "num_tokens": 211363419.0,
      "step": 101850
    },
    {
      "entropy": 5.70406665802002,
      "epoch": 9.23938679245283,
      "grad_norm": 1.515625,
      "learning_rate": 5.6497411431669155e-05,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2371224969625473,
      "num_tokens": 211373348.0,
      "step": 101855
    },
    {
      "entropy": 5.7053056240081785,
      "epoch": 9.239840348330913,
      "grad_norm": 1.46875,
      "learning_rate": 5.648969406598586e-05,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23535334914922715,
      "num_tokens": 211383838.0,
      "step": 101860
    },
    {
      "entropy": 5.727135038375854,
      "epoch": 9.240293904208999,
      "grad_norm": 1.5546875,
      "learning_rate": 5.64819812191514e-05,
      "loss": 4.7562,
      "mean_token_accuracy": 0.24424102753400803,
      "num_tokens": 211394170.0,
      "step": 101865
    },
    {
      "entropy": 5.6666148662567135,
      "epoch": 9.240747460087082,
      "grad_norm": 1.546875,
      "learning_rate": 5.6474272891325234e-05,
      "loss": 4.7235,
      "mean_token_accuracy": 0.24619072824716567,
      "num_tokens": 211404209.0,
      "step": 101870
    },
    {
      "entropy": 5.632785940170288,
      "epoch": 9.241201015965167,
      "grad_norm": 1.671875,
      "learning_rate": 5.646656908266684e-05,
      "loss": 4.665,
      "mean_token_accuracy": 0.24992531836032866,
      "num_tokens": 211413992.0,
      "step": 101875
    },
    {
      "entropy": 5.709248065948486,
      "epoch": 9.24165457184325,
      "grad_norm": 1.3828125,
      "learning_rate": 5.645886979333545e-05,
      "loss": 4.8657,
      "mean_token_accuracy": 0.23471915870904922,
      "num_tokens": 211425971.0,
      "step": 101880
    },
    {
      "entropy": 5.793214559555054,
      "epoch": 9.242108127721336,
      "grad_norm": 1.3671875,
      "learning_rate": 5.645117502349033e-05,
      "loss": 4.9484,
      "mean_token_accuracy": 0.23127335608005523,
      "num_tokens": 211439066.0,
      "step": 101885
    },
    {
      "entropy": 5.716994047164917,
      "epoch": 9.242561683599419,
      "grad_norm": 1.625,
      "learning_rate": 5.644348477329061e-05,
      "loss": 4.8304,
      "mean_token_accuracy": 0.23008424639701844,
      "num_tokens": 211449236.0,
      "step": 101890
    },
    {
      "entropy": 5.767909955978394,
      "epoch": 9.243015239477504,
      "grad_norm": 1.625,
      "learning_rate": 5.643579904289527e-05,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23224915713071823,
      "num_tokens": 211459086.0,
      "step": 101895
    },
    {
      "entropy": 5.63014850616455,
      "epoch": 9.243468795355588,
      "grad_norm": 1.53125,
      "learning_rate": 5.6428117832463366e-05,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2515016540884972,
      "num_tokens": 211470330.0,
      "step": 101900
    },
    {
      "entropy": 5.649033784866333,
      "epoch": 9.243922351233673,
      "grad_norm": 1.5390625,
      "learning_rate": 5.6420441142153655e-05,
      "loss": 4.6977,
      "mean_token_accuracy": 0.2437485247850418,
      "num_tokens": 211481231.0,
      "step": 101905
    },
    {
      "entropy": 5.638731145858765,
      "epoch": 9.244375907111756,
      "grad_norm": 1.5078125,
      "learning_rate": 5.6412768972124916e-05,
      "loss": 4.774,
      "mean_token_accuracy": 0.23734546899795533,
      "num_tokens": 211491381.0,
      "step": 101910
    },
    {
      "entropy": 5.642395639419556,
      "epoch": 9.24482946298984,
      "grad_norm": 1.5625,
      "learning_rate": 5.640510132253583e-05,
      "loss": 4.6946,
      "mean_token_accuracy": 0.2509502097964287,
      "num_tokens": 211501419.0,
      "step": 101915
    },
    {
      "entropy": 5.647286510467529,
      "epoch": 9.245283018867925,
      "grad_norm": 1.4609375,
      "learning_rate": 5.639743819354497e-05,
      "loss": 4.7378,
      "mean_token_accuracy": 0.24556314647197724,
      "num_tokens": 211511409.0,
      "step": 101920
    },
    {
      "entropy": 5.726348447799682,
      "epoch": 9.245736574746008,
      "grad_norm": 1.6875,
      "learning_rate": 5.6389779585310774e-05,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23595012873411178,
      "num_tokens": 211521330.0,
      "step": 101925
    },
    {
      "entropy": 5.741099262237549,
      "epoch": 9.246190130624093,
      "grad_norm": 1.5,
      "learning_rate": 5.638212549799165e-05,
      "loss": 4.9413,
      "mean_token_accuracy": 0.23056703060865402,
      "num_tokens": 211532532.0,
      "step": 101930
    },
    {
      "entropy": 5.681960153579712,
      "epoch": 9.246643686502177,
      "grad_norm": 1.5390625,
      "learning_rate": 5.6374475931745856e-05,
      "loss": 4.8607,
      "mean_token_accuracy": 0.23021851629018783,
      "num_tokens": 211543425.0,
      "step": 101935
    },
    {
      "entropy": 5.655264854431152,
      "epoch": 9.247097242380262,
      "grad_norm": 1.5546875,
      "learning_rate": 5.636683088673166e-05,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23976057022809982,
      "num_tokens": 211554289.0,
      "step": 101940
    },
    {
      "entropy": 5.706612968444825,
      "epoch": 9.247550798258345,
      "grad_norm": 1.6171875,
      "learning_rate": 5.635919036310708e-05,
      "loss": 4.8265,
      "mean_token_accuracy": 0.2366013303399086,
      "num_tokens": 211564328.0,
      "step": 101945
    },
    {
      "entropy": 5.607343626022339,
      "epoch": 9.24800435413643,
      "grad_norm": 1.7265625,
      "learning_rate": 5.6351554361030174e-05,
      "loss": 4.6327,
      "mean_token_accuracy": 0.24804240018129348,
      "num_tokens": 211574418.0,
      "step": 101950
    },
    {
      "entropy": 5.700932216644287,
      "epoch": 9.248457910014514,
      "grad_norm": 1.5546875,
      "learning_rate": 5.6343922880658835e-05,
      "loss": 4.7463,
      "mean_token_accuracy": 0.25119208693504336,
      "num_tokens": 211584921.0,
      "step": 101955
    },
    {
      "entropy": 5.6490741729736325,
      "epoch": 9.248911465892599,
      "grad_norm": 1.515625,
      "learning_rate": 5.6336295922150876e-05,
      "loss": 4.7332,
      "mean_token_accuracy": 0.24486998319625855,
      "num_tokens": 211594888.0,
      "step": 101960
    },
    {
      "entropy": 5.673582935333252,
      "epoch": 9.249365021770682,
      "grad_norm": 1.4375,
      "learning_rate": 5.6328673485664055e-05,
      "loss": 4.7335,
      "mean_token_accuracy": 0.23844024688005447,
      "num_tokens": 211605655.0,
      "step": 101965
    },
    {
      "entropy": 5.702453708648681,
      "epoch": 9.249818577648766,
      "grad_norm": 1.578125,
      "learning_rate": 5.6321055571356014e-05,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22729630321264266,
      "num_tokens": 211616900.0,
      "step": 101970
    },
    {
      "entropy": 5.777866411209106,
      "epoch": 9.25027213352685,
      "grad_norm": 1.5703125,
      "learning_rate": 5.631344217938425e-05,
      "loss": 4.871,
      "mean_token_accuracy": 0.24098589271306992,
      "num_tokens": 211627209.0,
      "step": 101975
    },
    {
      "entropy": 5.617551517486572,
      "epoch": 9.250725689404934,
      "grad_norm": 1.53125,
      "learning_rate": 5.630583330990622e-05,
      "loss": 4.694,
      "mean_token_accuracy": 0.24974294602870942,
      "num_tokens": 211636718.0,
      "step": 101980
    },
    {
      "entropy": 5.766848611831665,
      "epoch": 9.25117924528302,
      "grad_norm": 1.46875,
      "learning_rate": 5.629822896307932e-05,
      "loss": 4.9296,
      "mean_token_accuracy": 0.22997950464487077,
      "num_tokens": 211648295.0,
      "step": 101985
    },
    {
      "entropy": 5.6786445617675785,
      "epoch": 9.251632801161103,
      "grad_norm": 1.5234375,
      "learning_rate": 5.6290629139060745e-05,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2376089170575142,
      "num_tokens": 211659329.0,
      "step": 101990
    },
    {
      "entropy": 5.733236312866211,
      "epoch": 9.252086357039188,
      "grad_norm": 1.421875,
      "learning_rate": 5.628303383800767e-05,
      "loss": 4.7719,
      "mean_token_accuracy": 0.23619751632213593,
      "num_tokens": 211669158.0,
      "step": 101995
    },
    {
      "entropy": 5.654819488525391,
      "epoch": 9.252539912917271,
      "grad_norm": 1.5703125,
      "learning_rate": 5.6275443060077246e-05,
      "loss": 4.7191,
      "mean_token_accuracy": 0.24379921406507493,
      "num_tokens": 211679554.0,
      "step": 102000
    },
    {
      "epoch": 9.252539912917271,
      "eval_entropy": 5.411006993190195,
      "eval_loss": 4.907995223999023,
      "eval_mean_token_accuracy": 0.2389736551184595,
      "eval_num_tokens": 211679554.0,
      "eval_runtime": 20.7469,
      "eval_samples_per_second": 1704.35,
      "eval_steps_per_second": 213.044,
      "step": 102000
    }
  ],
  "logging_steps": 5,
  "max_steps": 110230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.592674240441344e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}